Accident Prone: A History of Technology, Psychology, and Misfits of the Machine Age 9780226081199

Citation preview

ACCIDENT PRONE

ACCIDENT PRONE A History of Technology, Psychology, and Misfits of the Machine Age

JOHN C. BURNHAM

t he uni v ersi t y of chic ag o pr ess chic ago a nd london

j o h n c . b u r n h a m is research professor of history at Ohio State University and the author of many books including, most recently, What is Medical History? The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London © 2009 by The University of Chicago All rights reserved. Published 2009 Printed in the United States of America 17 16 15 14 13 12 11 10 09

1 2 3 4 5

isbn-13: 978-0-226-08117-5

(cloth)

isbn-10: 0-226-08117-6

(cloth)

Library of Congress Cataloging-in-Publication Data Burnham, John C. (John Chynoweth), 1929– Accident prone : a history of technology, psychology, and misfits of the machine age / John C. Burnham. p.

cm.

Includes bibliographical references and index. isbn-13: 978-0-226-08117-5 (cloth : alk. paper) isbn-10: 0-226-08117-6 (cloth : alk. paper) 1. Accidents—History. 2. Accidents— Psychological aspects. 5. Industrial safety.

3. Accidents—Prevention. 4. Technology—Social aspect.

6. System safety.

7. Safety education.

I. Title.

hv675.b76 2009 363.1—dc22 2008040174 The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ansi z39.48-1992.

CONTENTS

Acknowledgments vii Introduction 1 1. Before Accident Proneness 13 2. German Origins 36 3. British Origins 51 4. Preparing the Way: Transport Operators 67 5. The Streams Come Together in the Late 1920s and Early 1930s 86 6. Consolidation and Development, 1930s–World War ii 99 7. How Psychiatrists Did Not Adopt and Medicalize Accident Proneness 122 8. The Mid-Twentieth-Century High Point 145 9. Eclipse of the Idea Among Experts 166 10. Bypassing Accident Proneness with Engineering 193 Conclusion 219

Notes 235 Brief Bibliographical Note 313

Index 319 v

ACKNOWLEDGMENTS

In a largely pioneer project like this book, an author becomes more indebted than usual to kind colleagues who have helped conceptualize, understand, and investigate often unfamiliar materials. My debts are indeed too great to enumerate in the detail that they deserve. I tried as the research proceeded to give my heartfelt personal thanks for each act of assistance at the time it was rendered. In a number of notes to the text, I have also been able to identify acts of generosity. Yet some of the help that I have received should receive additional public acknowledgment. The project was initiated with institutional support from the Department of History of Ohio State University and then fully launched when I had the honor to receive appointment as a Bye-Fellow in Robinson College, University of Cambridge, and Visiting Scholar in the Department of History and Philosophy of Science at Cambridge. Much of the actual writing was then supported when I served as Scholar in Residence in the Medical Heritage Center in the Prior Health Sciences Library at Ohio State University. There George Paulson and the late Charles Wooley were wonderful colleagues, and Judith Wiener and Kristin Rodgers provided many kinds of assistance, both technical and intellectual. German Berrios of Cambridge University provided me with enthusiastic, sustained encouragement as well as substantial insight. I also greatly benefited from enormous encouragement and substantial suggestions from Joel Tarr of Carnegie-Mellon University. John Forrester of Cambridge University, too, provided me with ideas and supportive discussion. Other colleagues who should receive special recognition and thanks include not only referees of the University of Chicago Press but James Capshew, Roger Cooter, Jacalyn Duffin, Ernst Falzeder, Alexandre Métraux, John Mollon, Annette Mülberger Rogele, Judith Peters, Carroll Pursell, Graham Richards, Mark Rose, John Sauer, William Rothstein, and Deborah Thom. vii

viii

Acknowledgments

I also owe many thanks to archivists, librarians, and libraries—a debt reflected in the annotations to this work. Special mention should be made of the libraries of Ohio State University and the OhioLink system, the libraries of the University of Cambridge, and the Wellcome Library. Material from this book was presented in several venues, and in each one colleagues offered helpful comments and suggestions: the history of medicine program at Queen’s University, Kingston, 2005; the joint meetings of Cheiron, the International Society for the History of the Behavioral and Social Sciences, and the European Society for the History of the Human Sciences, Dublin, 2007; the History of Science Society, Arlington VA, 2007; and the American Historical Association meetings, Washington DC, 2008. I am very grateful for permission to include in this book substantial amounts of material from two articles: John C. Burnham, “Accident Proneness (Unfallneigung): A Classic Case of Simultaneous Discovery/Construction in Psychology,” Science in Context, 21 (2008), 99–118 (through the generous courtesy of Science in Context and Cambridge University Press); and John C. Burnham, “The Syndrome of Accident Proneness (Unfallneigung): Why Psychiatrists Did Not Adopt and Medicalize It,” History of Psychiatry, 19 (2008), 251–274 (through the generous courtesy of the editor of History of Psychiatry, German E. Berrios).

INTRODUCTION

Technology unforgivingly demands uniformity from human beings who encounter it. People encountering technology, however, differ from one another. The history that now follows centers around that disjuncture. For decades, as Europeanized people of the twentieth century tried to contain the awful consequences of injuries that occurred at the interface between humans and machines, experts and thinkers used an idea that soon became part of folk wisdom: accident proneness, or the tendency of a particular person to have more accidents than most people. The origins of the idea in the World War i period immediately demand attention because they provide a striking example of the phenomenon of “simultaneous discovery.” The fading away of accident proneness in the late twentieth century, however, shows that it had an additional dimension of fundamental significance: where once leaders striving for safety had worked to manipulate human beings who might have accidents, by the end of the century, experts and policymakers were engineering safety for everyone by using technological fi xes. My history of an idea and the programs in which it was embedded therefore provides evidence of great shifts in implicit social strategy that took place in the twentieth century. Leaders in the safety movements of the early twentieth century through either voluntary efforts or governmental regulation attempted to institute physical protections, such as guards on machines or traffic controls. They also, in a second strategy, attempted to educate and discipline people to behave safely. Accident proneness was part of a third important strategy that then appeared: separating particularly vulnerable and dangerous people from risky situations. At the end of the century, however, this third alternative diminished dramatically when physical or engineering means overwhelmingly came to dominate safety, obscuring even the continuing programs to educate people to behave so as to avoid accidents. 1

2

Introduction

My story therefore starts with a close examination of the origins and development of the idea of accident proneness. The end of the story, however, shows that the rise and fall of accident proneness was a clear sign of not just one, but also a second broad shift in social strategy among Europeanized countries.

acciden t proneness a n ex a mpl e of simu lta neous discov ery The very first appearance of a named, systematic concept of accident proneness therefore marked a change of great historical significance. Moreover, the fact that the idea emerged independently and at the same time in different countries in itself signals that the idea had a special quality and importance in Westernized cultures. Accident proneness indeed turns out to offer an unusually pure example of independent, concurrent discovery (or construction) of an idea. Moreover, in more simultaneity, during the year 1926 both the English term, “accident proneness,” and the German, Unfallneigung (inclination to accident), were introduced into the literature (figure i:1).1 Any time that a new idea appears spontaneously in two different places, intellectual historians and historians of the various sciences take special notice.2 Simultaneous discovery suggests to scholars the existence of some impelling shift in the intellectual community and the surround-

figure i:1. In 1926, these two publications appeared completely independently in Germany and Britain, introducing defi nitive descriptions of accident proneness or Unfallneigung.

Introduction

ing culture, either a general cultural configuration or a point at which knowledge in some way had reached a state that stimulated an inevitable next step, as both Newton and Leibniz in the seventeenth century had to devise calculus (or something like it), independently and at the same time. No matter that historians have raised many points of doubt and in general want to reject a mystical idea, Hegelian or otherwise, that the course of events is inevitable. One could of course naturalize the process. That is, simultaneous discovery shows that historical events, or at least main trends, are causally predetermined. 3 The idea of simultaneous discovery therefore fi rst originated in the internal history of ideas, in which each advance appeared as a logical next step. But externalists, sociologists and social historians, soon adopted it as well.4 On a practical level, causes operate in a society and in the world of science. Of course a single set of causes can bring forth the same results from two disparate sources simultaneously. People in a culture share habits of thought that will give rise to similar ideas in different parts of the culture. Believers in a spiritual Zeitgeist or cultural or economic determinists all end up in the same place: simultaneous discovery (or independent social construction) of an idea. Some scholars, who wish to emphasize human agency, have tried to deconstruct the idea of simultaneous discovery. Indeed, some of the most able deny that the conventional phenomenon of scientific discovery even existed. 5 Yet in the case of accident proneness and other multiple discoveries, people immediately look back and recognize the phenomenon, which has been described many times over the centuries. Three substantive questions come up in discussions of simultaneous discovery. First, was the idea new? Second, did events of discovery actually occur at the same time? And third, did the independent discoverers actually construct the same concept? Philosophers can raise questions about whether or not any event can have an equivalent. Historians, however, can simply look for the best evidence, and the evidence in the case of accident proneness is striking. The parallels in the two discoveries or constructions of accident proneness are particularly compelling. The formulator in Germany, psychologist Karl Marbe, became an advocate. In Britain, another psychologist, Eric Farmer, emerged from a group as the chief advocate. It was Marbe and Farmer who gave the phenomenon its fi nal and definitive names in major independent publications in 1926. Each, however, had already in an obscure and local publication announced the name in, again, exactly

3

4

Introduction

the same year, 1925. Such persistent and recurring coincidence suggests that there was indeed “something in the air,” intellectual, material, or cultural. Moreover, as I shall note, the naming signified a shift in the intellectual site and a subtle but substantial innovation in the idea. The extent to which the German and English discoveries of accident proneness concerned exactly the same phenomenon will be suggested in many passages below in which evidence of two kinds is noted. First, concerned figures at that time held the two to be equivalent. Second, the readiness for the idea is striking in that so many people took up the formulations immediately, declaring that the concept fitted the functional ideas that they held but had not articulated. That readiness offers eloquent confirmation that accident proneness was a socially important idea. In succeeding decades, the idea of accident proneness therefore spread throughout the Europeanized world. The high point of attention and interest came in the 1950s. And so accident proneness has a birth date, even a date when it was named, and a time of full development. Such a profile is very promising for the biography of an idea. Yet there is no death date. Instead, accident proneness gradually receded in importance in programs to prevent accidents. It was at that point that experts, reacting to new social conditions, turned their attention instead to general safety efforts and particularly to engineered safety measures. 6 And yet accident proneness, as an idea, persisted on the popular level and in some of the technical literature.7 Indeed, the pattern of persistence suggests that the idea is a model for the ways in which any technical idea can persist long after it has been largely abandoned by experts. And since accident proneness was exquisitely the expression of thinkers’ trying to deal with the interface between humans and technology, it perhaps serves as a better model for the history of ideas in the twentieth century than some other models, such as phrenology or even Lamarckianism, that came from the nineteenth century.

acciden t proneness a s a s y mbol of impl ici t soci a l st r at eg y Accident proneness was, therefore, not just a miasmatic sign of changing times. It has another type of historical significance. The idea of accident proneness illuminates brilliantly the ways in which humans in the West interacted with technology. Indeed, the idea raises fundamental questions about human nature, human equality, and the limits of

Introduction

social discrimination. Most basically, how should social units deal with someone who shows a pattern of inadvertent but sometimes dangerous destructiveness? Intriguing and significant as was the double birth of accident proneness, the slow fading away of the concept reveals a tectonic shift in the whole Western strategy of coping with injuries and damage that came out of the human-technology encounter. What happened was that in the last decades of the twentieth century, experts dealing with mechanization in many settings all moved away from a focus on the careless or cursed individual who caused accidents. Instead, they now concentrated, to an extent that is remarkable, on devising technologies that would prevent damage no matter how wrongheaded the actions of an individual person, such as a worker or a driver. In the withering away of the idea of accident proneness, it is possible to see the growing dominance of a new social strategy, solving safety problems by controlling the technological environment. It is ironic that destruction that came from using technology came to be addressed so exclusively with “technological fi xes.” 8 Beyond this shift in social strategy, accident proneness also stood as an alternative path for social belief and action in a world given to medicalizing problems. Intriguingly, accident proneness was originally conceptualized in medical terms, particularly “susceptibility” to accidents and “predisposition” to have accidents. But psychologists, who invented the idea, after some implicit struggle managed to keep it out of the hands of the doctors and in the hands of psychologists, albeit mostly industrial and other applied psychologists.9 The idea of accident proneness was therefore a concept that existed for a long time at the edges but never was pulled into the mainstream of medical thinking. Despite the fact that much of the original language was medical language, in the end, the fate of accident proneness remained a curious exception to what appeared as the inevitable process of medicalization in the modern world. Eventually, as my narrative will show, another professional group, the engineers, came more completely into possession of the field of accident control.10 Even late twentieth-century psychologists working on “human factor” problems were devoted to making technology that was compatible with human reactions and thinking. But by that time, the notion of accident prone people had entered into general popular thinking, and not even the engineers could make the idea go away. From the history of this one idea it is possible to trace not only efforts to popularize a technical concept but to see how levels of

5

6

Introduction

popularization developed as the idea fostered more research and took on more complexity. The process of popularization had apparently unstoppable ramifications. In the late twentieth century, indexers included under the heading “accident prone” fiction that did not mention the term and movies like Mr. Hulot’s Holiday.11 As late as March 2003, a letter in the London Times was labeled “Accident-Prone Lorries.” Two months later, a reporter from Boston in the United States referred to a barge company as “one of the most accident-prone.” 12 Other references in the Anglophone press early in the twenty-first century described individual people such as the actor Tom Cruise as accident prone.13 The internet contained a large number of references to accident proneness and hundreds of items in which someone used the phrase. There was even a “punk rock” group by that name. Internet searches could also produce some evidence that writers continued to use the German-language Unfallneigung. The ubiquity of the phrase is suggested by a book that appeared in 1987. A British husband-and-wife team, writing under the pen name of John Penn, produced a murder mystery entitled Accident Prone, in which a victim was made to appear accident prone before fi nally being knocked off. The characters in the book used the term and even noted that it was popularly understood: “accident prone, as they say.” 14

t he ide a of acciden t Accident itself is a standard category in Western thinking.15 Like all standard categories, philosophical discussions and definitions exist, even though the idea has long been generally understood (figure i:2). Indeed, the idea of accident persisted from the idea of misfortune to, as Judith Green points out, “the late twentieth-century discourse of risk factors and responsibility.” 16 Yet accident was always a marginal phenomenon: marginal to a rational worldview and marginal to medicine (or any other standard modern category). The accident as a marginal category appeared in the seventeenth century, Green shows, but only in the twentieth century was it possible to talk about accidents in the modern sense.17 Over many decades, writers in the modern safety movement did all agree on two elements, however. First, accidents come by surprise and occur unexpectedly even in situations of great danger. And, second, the actual, specific event is unanticipated. Humans involved somehow do not plan for an accident to happen or do not believe that it will happen to them.18

Introduction

figure i:2. The philosophical and legal problem of defi ning what constitutes an accident has persisted for a long time. The actuality of injuries or damage is, however, usually obvious. Source: Travelers Standard, 25 (1937), 154, reproduced with the kind permission of Travelers Insurance.

Historians in particular, however, have had very little to say about accidents, except, sometimes, in terms of factory laborers and liability law. Medical historians have written about repairing and rehabilitating patients—but only after the patient had suffered an accidental injury. Not until 1997 was there a general book available on the history of accidents.19 Nevertheless, enough literature now exists to see that accidents changed as humans had more and more interaction with technology, particularly machines and powered devices. By the twentieth century, “the accident” had become a social institution, part of the world of modernity.20 Of course not every accident involved technology. People could suffer from encountering the natural environment around them. Lightning could strike. On 28 June 1591, a bargeman, age twenty-six, was, according to the coroner’s report, “washinge himself in the River of the Theames where the streame taking him he was drowned.” On 19 May 1615, a boy of seven or eight “was drowned in Goodmans ffeilds in a Pond, playing with other Boyes there and swymming.” A “Visitation of God” or an “Act of God” represented general as well as legal thinking.21 A secular or naturalistic view of accidents, however, also existed and spread as technology and society changed. The accident was an event that was not foreseen. Yet, afterward, people could explain in a rational way how it came to happen.22 As early as 1813, Newton Bosworth, in a book for young people, spelled out this thinking. Accidents, he wrote, are “sudden and unexpected events, generally of a calamitous kind, to which we are all more or less exposed . . . They are only called accidents, because previous circumstances did not appear to indicate them, or, in simpler terms,

7

8

Introduction

because they came upon us unawares.” Of course, he went on, because they have causes, they are, nevertheless, a part of the operations of nature.23 This naturalistic worldview, plus the growth of transportation and manufacturing technology, led to attempts to get at the causes and so to prevent accidents. Eventually, at the turn of the twentieth century, the preventive activities consolidated into safety movements.24 Those safety efforts finally, in the last half of the twentieth century, became framed in the idea of risk.25 The safety movements of the twentieth century therefore provided a context and changing background for the idea of accident proneness.26 Safety advocates were attempting to carry out two types of reform simultaneously. The fi rst was to change the physical environment so that human beings would have fewer chances to have an accident, indeed, in the case of safety switches and machine guards, to make it impossible for a factory worker to have an accident (figure i:3). This strategy is what I generally refer to as engineering safety. The second type of safety reform was aimed at changing the behavior of human beings. In the workplace, managers could educate and discipline and even discharge workers.27 Outside of the workplace, education could reduce actions that were believed to be hazardous. In the case of children’s encounters with traffic in the United States in the 1920s, for example, the results of a public educational campaign were spectacular in reducing injuries and deaths.28 In the second half of the twentieth century, many experts recommended a completely noncommittal view of accidents and began referring to “injuries,” not “accidents,” so that the question of cause was left open. If one speaks of an “accident,” one is implying that it could not have been prevented. Therefore, workers in the safety movement, who believed

figure i:3. Engineering safety could be very simple and sensible, like this adjustable, portable fencing to keep people from falling into an unexpected hole in factory flooring in Germany in 1931. Source: C. Haide, “Four Years’ Work as a Safety Engineer: A Review of Safety Activities in a Smelting Works,” Industrial Safety Survey, 7 (1931), 66.

Introduction

figure i:4. Different kinds of hand injuries incurred in the lumber industry in Germany, 1919. Source: Osw. Heller, “Eignungsprüfung und Unfallvorbeugung in der Holzindustrie,” Industrielle Psychotechnik, 1 (1924), 99, reproduced with the kind permission of Springer Science and Business Media.

that humans could in fact, with enough effort, prevent accidents, wanted to substitute the term “injury.” Thinking in terms of mere accidents, they feared, would close off any possibility of social action to keep people from getting hurt.29 Historian Roger Cooter has identified a defining “moment” around the end of the nineteenth century when injuries changed from private concerns to public concerns. An organizational society with a military model, with insurance concerns, and with philanthropy shifted the perspective of leaders in at least British society. Anson Rabinbach notes that the “shocking statistics” that came out of Germany in the last quarter of the nineteenth century made accidents into “a problem of crisis proportions.” Indeed, between 1885 and 1908, there were over two million industrial accidents in that country alone (figure i:4). 30 The resulting safety movements, often led by enlightened business managers for business reasons, were a natural outcome of bourgeois (if not modernist) attempts to make society work better by controlling human actions—including a person’s actions that the person could not consciously control. “When men live in society,” wrote the American jurist Oliver Wendell Holmes in 1881, “a certain average of conduct, a sacrifice of individual peculiarities going beyond a certain point, is necessary to the general welfare. If, for instance, a man is born hasty and awkward, is always having accidents and hurting himself or his neighbors, no doubt his congenital defects will be allowed for in the courts of Heaven, but his slips are no less troublesome to his neighbors than if they spring from guilty neglect.” 31

9

10

Introduction

impl ic at ions of t he history of acciden t proneness Accident proneness represented a transition from carelessness and recklessness, which people believed could be controlled or countered, to identifying human actions that were in some sense outside of the idea that all activity could be deliberate. The concept permitted a new way to view both groups of people and individuals. Generalizations about an idea that was cultural and scientific should not obscure the fact that accidental injuries have always constituted a profound and tragic human problem. Perhaps because accidents by definition are unexpected and therefore appear to be beyond control, they continue to attract disproportionately little attention in the public arena. Why bother? Yet from all countries from all years, the figures are astonishing. 32 As early as 1923, when motorcars were very limited in number, British authorities counted almost 3,000 people who died that year in road accidents alone, plus more than 86,000 with nonfatal injuries. As late as 2000, a study in the United States found that in one year, for every 100,000 population, 18,135 injuries of all kinds were medically treated. Another authority estimated that in the whole world at the beginning of the twenty-first century, 1,200,000 people died each year in traffic accidents alone, and an additional 50 million were injured. 33 The economic burden of injury and disability therefore continued to be enormous, with the cost of production errors on top of that. The personal and human costs one can infer. This additional context of the costs of accidents adds to the history of accident proneness an unspoken but haunting dimension that helps explain the urgency with which experts of another day tried to understand the human tendency to make damaging mistakes. As will be shown below, the idea of accident proneness had substantial presence in evolving professions and communities of experts in the twentieth century, sometimes illuminating particular aspects of the history of psychology and medicine. Dealing with accident proneness also became a major part of the efforts of many experts in the safety movement. Yet with the exception of two books, 34 historians have largely neglected the powerful and important shaping force of safety efforts in modern life. 35 Scholars in general have also not described how central the idea of accident proneness was in the ways in which people in economically advanced countries in the twentieth century were trying to deal with the costs of technology. The history of accident prevention in mass transport

Introduction

industries, especially, suggests that there was a clearly identifiable Zeitgeist within which people in many areas were prepared and ready to use the concept of accident proneness. Finally, around the beginning of the 1930s, all of the independent lines of thinking came together so that accident proneness had a definition and a body of research to carry the idea of an unconscious and deviant pattern of behavior into the safety movement and popular thinking. In psychology and safety, where the idea was most fully developed and used in the middle of the twentieth century, experts gradually, as the decades passed, moved on to other ways of viewing accidental injuries and damages. It was at that point that accident proneness became less and less useful as experts conceptualized injuries in terms of epidemiological categories, errors, and, finally, risk groups. The history that follows proceeds simultaneously through three main dimensions. First, it is the history of an idea important among Europeanized people in the twentieth century. Second, accident proneness was not just an idea that arose and had powerful cultural implications. In the history of ideas, accident proneness provides a challenging instance of an idea that had a striking, spontaneous origin and then an even more challenging instance of decline as it lost force among socially important groups. Finally, in the third place, this history inadvertently reveals how Western societies adapted to the seldom-discussed underbelly of increasing mechanization: the cost in injuries and deaths and damage done to productivity as well as to human bodies.

t echnolog y These events were taking place, however, through a period in which technology moved beyond mechanization in the factory and transport. The sites of encounters expanded in homes and public spaces, and more technologies appeared, such as household synthetic chemicals, pressure devices, and radiation. Lasers and everyday pneumatic nail guns were not in sight early in the twentieth century. 36 The conventional definition of technology had been artifacts that humans devised to manipulate the natural environment—not only obvious tools and machines but structural elements like stairs and walkways or piping and wiring. In 1951, a chimpanzee to be used in a movie was insured “against accidents which may result from trying to act like a human.” His trainer reported, “While he was swinging around in trees I didn’t worry. But in two days of movie work he bumped his head something awful in a

11

12

Introduction

fall from his high chair, almost strangled himself on a string of blocks in his crib, and shoved his hand into a vacuum cleaner.” 37 More and more often, historians have seen that in the twentieth century especially, technological artifacts operated as elements in social systems. Technologies turned into modalities of transport or production or service, or networks of communication and reaction. 38 I have not tried to describe these new perceptions and orientations in great detail. I have, however, tried to indicate briefly that it was in such a world that the fading away of the idea of accident proneness signaled new human adjustments to technologies that still brought traumatic injury and destruction. Most writers on technology have focused on innovation. New devices and arrangements have indeed, over the generations, proven dangerous. Ordinary accidents, however, the ones that aggregated into a major social problem, came from human contact with everyday technology. As David Edgerton vividly reminds us, it was the extent of use, rather than intellectual excitement, that determined the significance of technologies for the fates of human beings. 39

1.

BEFORE ACCIDENT PRONENESS

As human beings increasingly often came into significant contact with technology in the nineteenth century, accidents assumed a more and more conspicuous social profile. The cold statistics such as those cited in the introduction were always shocking. By the middle of the twentieth century, young people in some countries were dying more frequently from accidents than from infectious disease. Later in the century, one team of investigators figured that in the United States, “the injured occupy more hospital beds than do patients with heart disease and four times more than do cancer patients.” 1

de v elopmen t of awa r eness of sa f e t y In the late nineteenth century, the rise of major insurance arrangements to cover factory workers and workers in mines and on railroads caused governments and private firms to accumulate extensive accident statistics. The statistics defined in a material and understandable form the ways in which human encounters with technology and natural law moved among educated people, beyond personal experience, to assume the guise of a social problem.2 And even before World War i, that is, before the formal idea of accident proneness appeared, there was some evidence that safety measures were tending to contain or reverse the steady increase in very serious accidents in industry in major Western countries. 3 For occupational injuries, Anglo-American and other writers could 13

14

ch a p t er 1

cite the experience of the Germans from the 1880s. Then worker compensation and other regulatory arrangements elsewhere made it possible subsequently to classify and count accidents on an immense scale.4 Lawyers and courts in each country were able to establish boundaries between what was an accident and what was not. 5 By the 1920s, the general configuration of the accident problem in the West (and by extension, other places where humans encountered technology) was reasonably well established, and accidents were moving from the category of personal risk to the responsibility of business or the state. In Germany, legislators began to mandate proactive safety efforts. 6 At first, the major category of social concern was “industrial accidents,” which would include any injury incurred in the pursuit of a person’s occupation, chiefly mines and factories (figure 1:1).7 Railway injuries were for some time conceptualized as occupational, but as the public became more involved, a special category of accident crystallized. Courts were treating rail and transit employees as “servants” of the operating company, and so any carelessness or negligence on the part of employees that caused members of the public to be injured could be costly to the company.8 Especially with the rise of street railways and automobiles, which could injure non-employee passengers and pedestrians, another major

figure 1:1. A reenactment of an accident that occurred when a baker in Denmark was cleaning a large industrial kneading drum while it was in operation. He leaned into the drum too far and was “caught by the rotating kneading arm which broke his jaw, his left hand, two of his ribs and a vertebra in his neck.” Source: “Official Reports, Etc.,” Industrial Safety Survey, 23 (1947), 23.

Before Accident Proneness

category and concern arose. Pre–World War i publications of all kinds contributed to consciousness of this new menace, and the problem got worse after the war. Already, in 1906 alone, 374 people died in automobile accidents just in the United States. But with the increase in numbers of motor vehicles, by 1919 the number for the year was 10,000. “Practically every man, woman and child who walks out of his front door is exposed to this new and giant hazard which stalks down our streets and highways,” as one writer put it.9 It was a major change in the everyday environment that no one had asked for. Streets, which had once been understood to be public spaces in which anyone could walk, became contested areas. Not only did vehicles contend with each other, but pedestrians and children playing in the streets did not easily give way to the drivers of cars and trucks who tried to make public spaces into motorways and to blame pedestrians for unsafe behavior just as employers blamed employees for getting caught in machinery. Already in the year 1911, according to one authority, traffic accidents accounted for 129 deaths in Berlin, 236 deaths in Paris, 410 in London, and 532 in New York.10 Clearly it was no longer possible to confine concern over accidents to the workplace. All kinds of personal misfortunes came under the formal social category of “accident”: children at school, people at play, people in their homes. The social problem, as opposed to the legal problem, was usually not defined by the site of the accident except insofar as it contributed to an established subcategory. Within a very few years after the fi rst, tenuous appearance of heavier-than-air flight, for example, concerned people were using the category of automobile accident as a model upon which to conceptualize still another category, the “airplane accident” (figure 1:2).11 It is tempting to speculate that the automobile accident, which focused on the action of the driver, prepared people to think of individuals who were accident prone. In the factory, where accident proneness was actually discovered, safety workers could still be preoccupied with the plant supervisor who was responsible for dangerous conditions and personnel. The automobile accident, by contrast, was almost invariably the work of an individual car or truck driver. And in the 1920s, people found auto accidents increasingly alarming.12 Those seeking to prevent accidents used a number of approaches to try to get at the naturalistic causal chain that led to injury. As late as 1926, for example, the leading German industrial psychologist, Walther Moede, described the combination of statistics and accident analysis as the naturalistic approach to studying accidents.13 If an accident was a nat-

15

16

ch a p t er 1

ural event, with causes, then any human agency that set off the accident could be identified. So what were the personal factors that explained accidents before accident proneness came along? One approach of experts in the early twentieth century was the epidemiological: find out what classifications of jobs or situations created hazards.14 Thus, for example, a statistician could identify the occupation of railroad locomotive fi remen as “extra hazardous”—more dangerous even than that of engineer.15 Very early, too, analysts also knew about the “Monday effect,” which caused disproportionate numbers of both accidents and production errors on the first working day of the week. People also found that the first hour of a shift brought more injuries than later hours.16 Another approach was to concentrate on the material elements involved: the machines, the tools, the steps and ladders, the containers of hazardous fluids, the materials left in walkways and work areas. The editors of a safety magazine in 1913 scornfully printed a photo of a work area

figure 1:2. Wreckage of British air pioneer S. F. Cody’s Mark VIA near Aldershot in 1913. Spectators gathered at the site of this airplane accident in which Cody and his passenger both died. Source: Imperial War Museum, reproduced by permission.

Before Accident Proneness

figure 1:3. The sarcastic comment of a safety journal editor: “The management of this plant was insistent that ninety-five percent of all accidents are due to the carelessness of the men.” Source: Safety Engineering, 26 (1913), 385.

that was dangerous because it was very heavily littered with cast-off boards and other obstructions (figure 1:3).17 And another pre–World War i writer pointed out that “an unguarded bursting emery wheel” could injure the plant supervisor as well as the mechanic who was tending the wheel.18 Large industrial establishments were accumulating statistics about the number and nature of accidents. Even more private and public bodies, too, were developing tallies of different kinds and totals of injuries. Was the cause an explosion? A falling object? A vehicle? In particular, individual plant managers analyzed which work areas and divisions had the most incidents and therefore where educational safety efforts would pay off. Clearly by the World War i period, every element was in place also to examine the records for individual workers’ accident records. But no one reported doing so.19

per sona l fac tor s Increasingly, however, those concerned with safety in Europe and America spoke of the importance of “the human factor” in accidents.20 Americans, working in a culture that emphasized education and environmentalism, came to the human factor earlier and more easily than did many Europeans. Continentals, especially, focused more frequently on protective design and regulations.21 Yet they too were aware of personal factors.22 Wrote the American expert H. B. Rockwell in 1905, pure accidents, events that cannot be anticipated, seldom happen. Rather, “some one has blundered, some one has disobeyed an order or undertaken to reverse a law of nature.” The human errors he could think of just concerning street railways could include improperly laid tracks, badly designed or faulty equipment, failure to inspect equipment, employees’ negligent or careless actions, and passengers’ or pedestrians’ thoughtless or risky behavior. Or errors could act in combination, as when one night a streetcar

17

18

ch a p t er 1

motorman let a woman passenger distract him, and he did not brake soon enough to prevent the car from running over a drunk lying on the tracks. Rockwell thus considered, beyond workers’ actions, bad engineering and faulty management to be human errors—a view often shared by lawyers who could characterize even a corporation, through its employees, as “negligent” when they failed to take action to warn, discipline, or remove hazards and provide safeguards.23 A British writer in 1919 describing “the human factor in accident occurrence” showed that managers were well aware of possible hazards in both groupings of people and individuals (in this case workers). Too little attention is given . . . to the personal factor—the worker—as a contributing element in accident occurrence. The worker, just because he is a human being, is subject to certain influences which may seriously increase the likelihood of his being injured. This human factor manifests itself . . . principally, perhaps, in the higher accident rates of the inexperienced worker . . . lacking in the training which teaches him control and caution. It shows itself in the higher accident rates of men working under unfavorable conditions, such as at night or when physically unfit, or in extremes of heat and cold. Thus the worker through some lack or defect or through the influence of unfavourable surroundings, may contribute to the accident which causes him injury, and as such may be regarded himself as a cause.

The writer went on to note that stress and fatigue in workers may increase accident rates, and particularly that workers who are not used to working in an industrial setting incurred many accidents.24 From other viewpoints, however, had come a more focused idea, the idea of individual differences in behavior in general. In medicine and biology, as J. Andrew Mendelsohn and others point out, people of the late nineteenth century were thinking of the inequality of bodies, and some psychologists and psychiatrists began to extend the idea to differences in people’s psychology and character. By measuring those differences, whether of reaction time or intelligence, investigators reinforced tendencies to think that each person differed from other people—and yet could be grouped or arranged on a scale with the others.25 And one of the many groupings of humans to emerge in the decades around 1900 was that of people who suffered injuries and caused damage. Accident proneness had a particular background in the young science of psychology, on the one hand, and, on the other hand, in the realm of

Before Accident Proneness

general knowledge and belief. Managers in the industrial system drew on both expertise and common ideas to understand why a person might have one or more accidents. But while the idea of accident proneness had major roots in the concerns of managers, others, too, were deeply concerned, and the concept quickly played into general thinking about people who made errors and initiated or suffered injuries.

c a r el essness a nd r eck l essness Traditionally, people had clear ideas of what a person would do actively to set off an undesirable sequence of happenings. The core beliefs were stated in terms of familiar character flaws, which could be temporary or persistent: carelessness and clumsiness. People who acted carelessly were morally, if not legally, blameworthy. Indeed, careless or reckless people could even be willful, as was particularly clear in children’s literature. In 1830, in The Book of Accidents, the anonymous author provided pictures of different kinds of dangerous activities and the resulting accidents that could befall children, such as “tumbling down stairs” (figure 1:4), “playing with fire,” “drowning,” “playing with knives,” “falling out of a window,” or “crossing streets.” In his or her injunctions to children, the author used terms such as “careless,” “dangerous,” “wicked and malicious” [in the case of throwing stones], “without caution,” “foolish,” “heedlessly,” and “thoughtlessness.” 26 In the middle of the nineteenth century, an author of children’s stories wrote about a boy, “Troublesome Tom,” who came into misery and danger because he was naughty and disobedient. The moral to the story was “duty is safety.” There was no doubt that if Tom had acted responsibly and as he knew he should, his life would have been much happier—and safer, for “when I gave up my heart to be undutiful, I was always in some dreadful danger.” In this case, there was no doubt that accidents—although the author did not use the term—resulted from personal moral lapses.27 Generations of writers in industrial and safety publications before the end of World War i hammered away at the need for managers and employers to try to overcome the problem of employees’ carelessness or heedlessness or, in an extreme form, risk taking.28 In all of these publications, the assumption was that an individual person was responsible, as Rockwell, quoted above, insisted, because he or she was disobedient or careless or reckless. The editors of Safety Engineering, for example, for years ran a series, “The Man on the Job.” Readers furnished many examples of the carelessness of the man on the job. “One of the men engaged in unloading

19

20

ch a p t er 1

cars stepped on a nail which protruded from a board on the ground . . . A blacksmith was driving a wedge into a hammer when a chip flew into his eye, although safety goggles were at hand.” Discipline and “education” were invariably prescribed to reduce accidents. Catchy signs were a staple: “Machinery Does Not Think; Men Do; Therefore Be Careful.” “Machinery Is Active; Some Brains Are Not. Beware!” 29 Of course commercial and monetary interest caused some groups to blame workmen or users. Lawyers and courts taught business people to think about negligent behavior and legal responsibility. Employers became involved in such lawsuits in all countries. Yet it was not difficult to show that carelessness caused fires, for example. Or to show that a workman “forgot to shut a trap door leading to a meter; a child fell through the door.” 30 As the president of the U.S. National Safety Council concluded

figure 1:4. Source: Anonymous, The Book of Accidents, Designed for Young Children (New Haven: S. Babcock, 1830), 22.

Before Accident Proneness

figure 1:5. Educational poster, Germany, 1927: “Protect yourself! Wear tight-fitting clothing!” Source: Kurt Seesemann, “Psychotechnical Studies on Industrial Safety Propaganda,” Industrial Safety Survey, 3 (1927), 72.

in 1920, “the greatest risk of injury that a careful man or woman runs on the streets, in the schools, in our homes and industries, is the risk of being hurt by some heedless, careless, or reckless person.” 31 By 1920, then, safety crusaders were trying to separate people into two groups: the careful and the careless. Social pressure to coerce people to enforce carefulness and to be careful themselves was effective in many parts of society (figure 1:5). Statistics showed some accident-free workplaces. Safety crusaders could point to successes.

clu msiness But what if there were people who were different, particularly those who were different through nature, that is, through no fault of their own? And part of their nature was, unlike that of most people (who could choose normality or non-deviance), not to be able to respond to safety campaigns and, rather, to persist in having accidents?

21

22

ch a p t er 1

Traditionally such people were not blameworthy when they did not pay attention or when they could not carry out a task properly. These were people who were naturally clumsy. Clumsy people were often considered to be so by nature, as distinct from those who chose to deviate from careful, dutiful behavior or those who failed morally otherwise, for instance by making themselves clumsy by drinking. 32 Accident proneness, it turned out, was, like clumsiness, a natural characteristic of a person. And while carelessness and clumsiness both continued to appear as common explanations for unfortunate events, to some extent (and I shall spell out the limitations) accident proneness was a reconceptualization of persistent natural clumsiness in a person. Although the idea of clumsiness is a basic and homely one, indeed, an obvious one, the subject of clumsiness in human affairs is something that has just not attracted any academic interest. 33 The idea is not even listed in German- or English-language dictionaries of the history of ideas. 34 In 2006, however, Judith Peters provided a chapter of background history of the recently defined clinical syndrome, developmental coordination disorder, focusing on sensorimotor or movement dysfunction. She points out that lay language has consistently overlapped with efforts of professionals to find terms that correspond with “the familiar temporary experience of motor difficulty” and the difficulties confronted by clinicians. So there was not only a person who was clumsy but those who were “awkward,” “uncoordinated,” “gawky,” “bumbling,” “butterfingered,” and so on. 35 Not surprisingly, clumsiness has equivalents in languages other than English, and assembling versions of a precursor concept reveals that clumsiness represents a deficiency or decrement in the clumsy person. The idea of accident proneness, by contrast, had an additive connotation not commonly found in other pathological syndromes of that time. Bilingual colleagues and dictionaries, particularly historical dictionaries, disclose some basics inherent in this predecessor idea, clumsiness. Peters has assembled a list of twenty-eight equivalents for “clumsy” from thirteen Western languages. 36 In German, the common word is unbeholfen, and it is a specifically negative word. One of the connotations is lack of skill. And the word of which it is, derivatively, the negative, behelfen, means to make do, to improvise, to be ingenious. Those are attributes that an unbeholfen person lacks. But the adjective beholfen, without the “un,” is not used in German; only the negative or decremental meaning is commonplace. The examples can be extended, for example to the French maladroit, which is transparently negative and decremental in either French or Eng-

Before Accident Proneness

lish. But the most convincing evidence that clumsiness refers to a decremental attribute comes from the English word itself. Clumsy, which according to the OED appeared in the language only in the seventeenth century, comes from a Scandinavian word that means benumbed by cold. In other words, a person acts as if he or she, when one’s hand is frozen, had lost a dexterity that one once had, indeed, that anyone should have. 37 It thus becomes obvious what accident proneness had that clumsiness, in almost any language, did not have: accident proneness is an added attribute, an active tendency, not a lost or missing attribute, talent, or ability. 38 The evidence therefore suggests that around World War i, a functional gap appeared in Western culture that could be filled with the additive attribute of accident prone. And there was something else—beyond a positive function—that differentiated accident proneness from clumsiness. Clumsiness was an attribute that anyone could observe, including the clumsy person, who would often excuse an action by saying, “How clumsy of me.” Accident proneness, by contrast, was a hidden attribute. In a group of people, no one could immediately pick out the ones who were accident prone—as one could soon detect someone whose movements were clumsy. Even over a period of time, at most one could identify a person who repeatedly had accidents as someone who suffered bad luck, an affliction that came from outside the person’s identity. But in the twentieth-century age of psychologization (as I shall remark again below), hidden motives to injure oneself or make mistakes cast accident proneness as a surreptitious but distinctive part of someone’s personality or operational self. 39 Indeed, so hidden was accident proneness that by the 1930s, psychologists found that even motor skills tests could not identify people who would suffer an abnormally high rate of accidents. Observable clumsiness later, at the end of the twentieth century, took on another meaning, in clinical neurology and child development. Neurologists had long dealt with movement disorders of various kinds, particularly in children and in the aged (often “ataxia”). Clumsiness, however, was a term avoided in medicine and allied disciplines for a very long time. Motor debility (débilité motrice), for example, was suggested in France in 1909 to cover a range of motor inadequacies.40 Then throughout the second half of the twentieth century, workers in many different fields, but especially neurology, child development, education, and physiotherapy, wrote more and more extensively about persistent clumsy actions, or movement disorders, as signs or symptoms or syndromes. The terms in use were astonishingly numerous, chiefly, in English, “clumsiness,” “dys-

23

24

ch a p t er 1

praxia,” and “developmental co-ordination disorder.” 41 Clumsiness, for example, became a differential sign in Asperger’s syndrome. A Harvard neurologist, C. M. Fisher, in 1967 wrote of “the dysarthria-clumsy hand syndrome.” 42 Beginning especially in the 1960s, then, the term “clumsy” began to appear in English-language technical articles and books to describe children who had movement disorders.43 Eventually scales of clumsiness appeared for different diseases—inspired at least in part by bureaucratic demands that patients meet tests of impairment, for example in Friedreich’s ataxia.44 In the 1970s, especially, works began to appear for parents, teachers, and clinicians that were labeled specifically to focus on concern with “the clumsy child.” The authors spoke of “managing” the child’s difficulties and reducing social consequences for the youngster. They set up programs of training to minimize clumsiness. All of these efforts reflected psychological and medical conceptions of what in plain language could now be called clumsiness.45 When clinicians of various kinds in the late twentieth century wrote and spoke of clumsiness, however, their focus was entirely on helping the individual child or patient. Their concern was one or more discrete cases that might be treated or “managed.” So even in technical usage, clumsiness, or “the clumsy child syndrome,” continued to be an individual, not a social problem.46 People framed accident proneness, by contrast, in a social context and as a social problem. Earlier clinicians and educators of course had noticed clumsy people. At the turn of the twentieth century, insofar as people were clumsy, those people were considered to lack graceful movement and motor skills. And that was the form in which thinkers portrayed clumsiness in a systematic way.

ph ysic a l educ ator s Two groups were particularly concerned with motor skills. One group consisted of physical educators, who worked with children, especially, but also with adults. The other group was psychologists, who were trying to account for error and individual variations in performance in work. Turn-of-the-twentieth-century writers on physical education assumed that they were dealing with normal people or at least potentially normal people. Their goal therefore was to train children to develop in a normal manner or to offer exercises that would correct any imperfections in a person of any age.

Before Accident Proneness

Development was relatively straightforward. Infants and children went through stages of normal development in which they started out making clumsy movements, that is, they lacked “neuro-muscular” coordination. At the stage of learning to read, for example, wrote one expert in 1913, “there is the evident lack of muscular control in eye and hand at this stage in both reading and writing.” Wrote another expert advising physical educators, training works to shape and form “the factors of the development of the power of voluntary movement in the infant . . . it is clear that they (and others, perhaps) underlie the same process in the man and woman.” 47 Indeed, one could recommend “psycho-physical exercises for strength and grace, and for special ailments and deformities” as well.48 Physical educators therefore viewed developmental processes and clearly pathological processes as all subject to achieving mastery and coordination. The leading American physical educator of the time, Luther Halsey Gulick, noted that “conditions of atrophy which are due to long continued abstinence from actions which are dependent upon traumatism are . . . peculiarly susceptible to gymnasium treatment.” Wrote one exercise expert concerning patients with nervous system disorders, Many are unable to walk in a normal manner, not that they have not the power but because they have ataxia. I use the term ataxia to mean the inability to perform the ordinary movements made by a normal individual in the normal manner because he does not know how to do so. That is, ataxia is incoordination. The first time we try a new game, such as billiards, we are painfully aware of our ataxia. It is not that we do not possess the power but that our nerves and muscles are not properly trained.

It was in this context of clumsy actions—actions that could be corrected— that physical educators wrote about positive goals, not clumsy people per se. “The test of the gymnastic value of a [dance] step,” wrote an Indianapolis physical educator, is, “Does it promote grace and skill?” 49

ps ychologists Much later, and in a different context, as will appear below, accident proneness as such, beyond coordination and gracefulness, came to concern physical educators. The specific idea of accident proneness, however, originated and grew first in the discipline of psychology as it developed in the late nineteenth and early twentieth centuries.

25

26

ch a p t er 1

Psychologists before World War i studied two particular phenomena that came to be applied to accident prevention: motor learning and fatigue. Before 1900, psychologists were overwhelmingly studying sense perception and cognitive processes, trying to establish laws by which human mental processes operated. It is true that Hugo Münsterberg, then in Germany, had observed how emotional states could affect the production of errors in mental work. And just at the end of the nineteenth century, a U.S.A. psychologist, R. S. Woodworth, studied accuracy in voluntary movement. Woodworth measured errors made by adults in handwriting, in playing the violin, in singing on key, in hitting a small surface with a hammer, and other motor activities. He found, not surprisingly, that errors could be reduced by practice. And he even, at one point, referred to awkwardness—noting, however, that to some extent practice could reduce that awkwardness and improve the person’s control over his or her motor activity. 50 In subsequent years, other studies followed. Psychologists continued to search for general natural laws that governed how psychological processes worked. Could, for example, distraction from a task be measured in the ways in which noises slowed a person’s work down or caused a person to make errors? Indeed, was motor ability even a quality that could be measured and scaled as intelligence was?51 It was not a big step to trying to match workers to their tasks. The studies of error in motor tasks therefore showed a bias toward fi nding appropriate workers. Frederic Lyman Wells, in using typewriting as an experimental task to study psychomotor errors in 1916, remarked that this was an extremely “industrial” task. 52 One particular line of study very rapidly ran into industrial application: the study of fatigue, which was widely researched by psychologists.53 Originally, particularly in the nineteenth century, physiologists had investigated how work energy of the human body could suffer from fatigue. By the 1890s, some investigators were crossing over into mental fatigue and showing that physical and mental fatigue operated in the same continuum. Pathological fatigue affected white collar brain workers particularly, but ordinary workers could also suffer. Fatigue therefore led into industrial psychology in various ways as investigators attempted to identify fatigue that could impair the productivity of a worker. The early twentieth-century debate over how many hours in a workday would lead to the most productivity was thus well launched on an empirical and sometimes scientific basis. In an increasingly technical economy, mental fatigue as well as muscular fatigue became very important. 54 One

Before Accident Proneness

list of the costs of fatigue in industry included “health, longevity, safety, labor supply, employment stability, industrial contentment, productive efficiency,—i.e. alertness, speed, accurate work, minimum waste—as well as output and profits.” 55 Fatigue experts at the beginning of the twentieth century were already pointing out that fatigued workers had more accidents. They also connected fatigue to the finding, mentioned above, that more accidents occurred on different days and at different points in the workday (figure 1:6). 56 In France and Germany especially, statistics were showing that accidents increased at the end of the morning (before the meal break) and at the end of the day, although as early as 1844, a British reformer had noted a general impression that “a large proportion of [mill] accidents—

figure 1:6. In the early 1920s, a German safety poster warns workers to be especially careful at certain times: “When do most accidents happen? From 12 to 6 p.m. Mondays and Saturdays. In summer. In the ages from 26 to 40. In these dangerous times, be very careful! Think: prevent accidents!” Source: K. A. Tramm, “Die Verhütung der Unfälle durch Propaganda,” Industrielle Psychotechnik, 1 (1924), 153, reproduced with the kind permission of Springer Science and Business Media.

27

28

ch a p t er 1

particularly those which may be denominated the minor class, such as loss of fingers and the like—occur in the last hours of the evening, when people become so tired that they absolutely get reckless of the danger.” 57 Fatigue, therefore, from the beginning represented a factor outside of the immediate control of a person, who might in consequence become “reckless.” Continental authorities could read the German numbers and French studies to show that “the length of working time and progressive uncertainty of control over muscles, as well as the relaxation of muscle tone, is made clear in the statistics.” One could even formulate a law, as did an Italian expert: “All work has a limit beyond which, if effort continues, attention decreases and tends to disappear completely.” 58 As Josephine Goldmark concluded in her exhaustive compendium in 1912: “Fatigue of attention and muscular control . . . play a subtle part in the occurrence of industrial accidents.” Of course fatigue was affected by speed, by rhythm, by the pressures of piecework. Workers particularly feared and resisted the notorious “speedup.” But the end product of different kinds of coercion was the same: injuries. 59 And both muscular coordination and attention, often understood as the ability to react rapidly and appropriately, were subjects that psychologists were investigating. The stage was therefore set for psychologists to study how mental fatigue and human behavior more generally led to accidents and errors that affected industrial efficiency. And this interest in industrial accidents could, and did, pass easily into the arena of non-industrial accidents, particularly of transport passengers and auto accident victims who were becoming so conspicuous, as I have noted. In combining mental and physical fatigue, psychologists could point out that all measures of intellectual fatigue and error involved both mental effort and some muscle movement (as in reaction times or repeated key tapping), and so the study of cognitive error or muscle control fell into the realm of thinking and behavior that psychologists claimed. And they were well aware that their work could apply to industrial efficiency and output. 60 Further study of mental fatigue in fact suggested that it did not operate in the same way as did physiological fatigue. In memory retention, mental fatigue was not the same strong predictor of error that physical fatigue was.61 To search for the source of error—and, hence, accident— psychologists were therefore poised indeed to look for new dimensions in “the human factor.” On the Continent, the quest for efficiency led to the field of “psychotechnics.” In the Anglo-American world, essentially the same subjects were classified as applied psychology. 62

Before Accident Proneness

indi v idua l dif f er ences Meanwhile other psychologists had been developing a science of individual differences. It is not surprising, therefore, that the search for the causes of errors and accidents moved into a study of people who made errors and had accidents. And it is not surprising, either, that this factor moved into attempts of psychologists to devise tests to screen employees, especially those in the transportation field. Trying to fit the worker to the job had led to aptitude testing. Psychologists therefore embodied experts’ attempts to intervene practically into the autonomy and individuality of workers whom one could visualize as human machines. The specific approaches included all of the work on fatigue, on employee selection to match the worker to the task, on ways to keep workers happy and productive, on many versions of taylorization (labor efficiency and time-andmotion studies), and on various ways in which managers and workers could affect the human factor to increase industrial efficiency. 63 Münsterberg, the German psychologist who moved to the United States, wrote extensively and influentially on psychotechnology/industrial psychology and dismissed group generalizations as too facile and unhelpful, particularly when psychologists could identify individual characteristics—part of the personality—that would make a worker fit for a job. He expressed doubt about the usefulness of group characteristics: “We may know that the inhabitants of a special country are rather alert, and yet the particular individual with whom we have to deal may be clumsy and phlegmatic.” He, like others, wrote specifically of individual differences, of individual dispositions and tendencies and traits, and he cited laboratory tests and scales for abilities in such characteristics as attention. 64 Psychotechnologists were all looking at various aspects of human nature, especially as they took up the task of devising tests that might help match a worker with an occupational niche. Psychologists could offer vocational guidance to an individual, or they could offer employers testing to assist employee selection. In each country, the emphasis on what would be tested seemed to differ somewhat. According to one 1921 writer, the French emphasized the physiologies of workers, the Germans the intellects, and the Spanish the temperaments.65 Psychologists found that other professionals, too, were taking an interest in finding an efficient and happy worker in any job. A variety of physicians offered diagnosis and advice. Psychiatrists early in the twentieth century contended that maladjusted employees were costly to industry—

29

30

ch a p t er 1

simply by reason of causing expensive labor turnover, if not inefficient production. Psychiatrists were concerned that the workers be not only adjusted but happy, as well, with the undertone that this might diminish industrial unrest. But such writers were dealing with the whole personality of the worker, just as they would any patient. Or with the whole neuropathic constitution of the person. 66 Many physicians insisted that each employee and prospective employee should undergo a medical examination. In that examination, the physician, these experts insisted, should be able to detect physical conditions that would impair the functioning of the employee. It seemed a good idea, for example, that people with epilepsy or cardiac problems not be serving as train or tram drivers. In this case, belonging to a diagnostic category should affect the kind of work one did—both because asthmatics should not work in dusty places for their own health and because eye problems in a machine shop would lead to errors and accidents.67 The psychologists who became interested in accidents worked within the same framework as the doctors but with a different focus. Psychologists used the synthesizing concept of individual personality—but instead of working with the whole personality, they talked about traits that made up the personality. Some of those traits had a physiological basis, such as reaction time and, as I have suggested, fatigue, and so physicians could also claim them. But to a surprising extent, the individual personality trait or characteristic came within the expertise of psychologists. And obviously traits that affected worker productivity were the focus of industrial psychologists, who particularly attempted to measure and scale individual traits and groups of traits. Psychologists, like psychiatrists, could utilize global areas of functioning, such as “intelligence.” “A manager tells of a woman whom the tests showed to be mentally under age, but who, when she fi rst came into the shops, did very well at a simple operation. At the end of the year, however, she was promoted to another kind of work and did not succeed. At last she was put back to the original simple work, where she did very well.” 68 In using the idea of personality traits or characteristics, like clumsiness, to explain accident patterns, psychologists were following the model of specific kinds of physiological functioning. The classic example of individual differences was reaction time. In the early nineteenth century, astronomers found that a “personal equation” could account for slight differences in recording exactly how long it took a star to pass past observational lines. Later in the century, psychologists worked with reaction times as involuntary, “personal” differences between experimental

Before Accident Proneness

subjects, assuming that the differences between individuals were determined by individual physiology. By 1916, a careful American scientist was finding marked variations in reaction time patterns in people who were diagnosed with alcoholism, hysteria, epilepsy, and paresis. His results underlined the individual nature of a factor, reaction time, that many experts believed could cause one person to have more accidents than other people.69 In the realm of accident specifically, there was yet another early and influential model: color blindness.70 In the 1850s, George Wilson of the University of Edinburgh published warnings about the dangers of hiring color-blind workers in the transportation industry especially. He suspected that many ships had been lost because some color-blind pilot or sailor had not seen colored signals correctly. Wilson found that, fortunately, there had already in Her Majesty’s service been some informal screening for ability to discern colors. But Wilson was especially critical of the British railways, which had begun using colored signals, especially red and green. “If the present system of signals is continued,” he went on, “no colour-blind person should be admitted as a railway servant [employee]. I am happy to say, that the publication of my papers has induced the Great Northern Railway Company to require that, in future, all their porters shall be tested as to their freedom from colour-blindness before they are admitted.” 71 In 1875, officials traced a serious railroad accident in Sweden to color blindness. A Swedish scientist then devised practical tests for railroad employees, and others used and improved on those tests. Only after many more such accidents all over the world, however, did testing for color blindness become general.72 Where physicians like Wilson had long served to do any screening of employees, psychologists at the end of the nineteenth century began to use their laboratory testing to claim the ability to fi nd employees with physiological traits—and specifically visual defects and slow reaction times—that would affect occupational safety. It was, therefore, not a great step, as more holistic psychologies developed in the twentieth century, to try to identify behavioral traits that might lead to elevated accident rates. Given the growing emphasis on individual differences, not only in physiological functioning, but in aggregate or behavioral personality traits, it can much more easily be understood why the idea of accident prone individuals crystallized when it did. People, and especially psychologists, were thinking often in terms of individual differences that could

31

32

ch a p t er 1

be defined in terms of traits. They were also thinking in terms of personality, which was a persisting structure of habitual behaviors in which one could discern an individual’s traits. And in the 1920s, just as accident proneness was appearing, psychologists suddenly produced a very large number of tests for both intellectual and non-intellectual personality traits.73 Writing as early as 1919, the British psychologist Charles S. Myers observed, “None but those who have had experience in psychological tests can realise what a wealth of information in regard to the ‘character’ of a subject is incidentally gained from tests systematically and individually applied.” 74 In 1918 in the United States, the National Association of Employment Managers formed to carry over into civilian business and industry the personnel system of the wartime army, which included using psychological testing for placement.75 While accidents were becoming a social problem, error also gained attention in psychological writings. Error in industrial production was of course a problem. But psychologists took much interest in intellectual and stimulus errors, such as errors of speech and muscular movements. Many of these errors could pass for clumsiness. Others clearly were examples of actions that a person could not control and of which the person often was not conscious. Educators wanted to eliminate such errors from pupil performance. Industrial managers deplored the effect of errors that spoiled goods in production. Even before errors in personal typewriting (mentioned above as a subject of psychological investigation), typographical errors in the press were the subject of humor in publications devoted to the humorous—in which the clumsiness or lack of control of a person produced a comic result. Moreover, there was a suggestion that such errors involved a hidden pattern of motivation—probably unknown to the erring person but expressive of his or her personality. Arthur Schopenhauer, for example, noticed that people who made errors in making change more often than not did so in their own favor.76 In 1901, Sigmund Freud in Vienna produced a whole book devoted to unconsciously motivated errors and treated them, indeed potentially all accidents, explicitly as “psychopathology,” for, as he noted, “there is a sense and purpose behind the minor functional disturbances in the daily life of healthy people.” Freud found an “inner unity” in acts of forgetting, in misremembering, in slips of the tongue and pen, in bungled actions, in errors of all kinds. He wrote: “If slips in speaking—which is clearly a motor function—can be thought of in this way, it is a short step to extend the same expectation to mistakes in our other motor activities.” 77 Although Freud’s work was gaining both professional and popular circulation be-

Before Accident Proneness

fore 1920, and although his approach emphasized the individuality of errors in action and expression, it was, as I shall explain later, some time before motivated accidents came into the discourse about accident prone people.78

common observat ion Finally, before the explicit idea or term “accident proneness” came into professional or public attention, it had been (and still is) a folk observation that some people have a lot more bad luck (a concept that includes accidents) than most other people. A character in Anton Chekhov’s Uncle Vanya was actually named “Accident.” Major Greenwood and Hilda M. Woods in 1919, in the fi rst British publication describing accident prone people, noted that there was a common explanation for individual susceptibility to having accidents: “carefulness or carelessness; as one says, there are people whose fingers are all thumbs[,] and there are others who are neat fingered; or again some people are scatter-witted and others circumspect . . . Do our results amount to more than an arithmetical verification of this?” 79 Another, and, again, contemporary, example of common observation was provided by a Swiss military surgeon, Chs. Widmer, also in 1919. He wrote specifically about “Unfalldisposition” [accident disposition]. Any physician, he observed, has treated trauma, and in each case the standard anamnesis, finding out “how did this happen?” brought out the failure of the patient to react automatically so as to avoid the accident. Moreover, with extensive experience, such as he had with the same group of military personnel, one would notice a small number of those who did not react appropriately and who showed this failure by repeatedly becoming injured. Anyone with extensive clinical experience such as he had, Widmer wrote, would identify people who were likely to suffer injuries. One could, he concluded, see an “accident neurotic” long before the accident came to pass. 80 In 1920, a member of the State Safety Board in of Washington State, who had experience in investigating industrial accidents, struggled to describe and explain what he had observed and had concluded concerning individual differences in the realm of accidents. He wrote of “many unfortunate human beings engaged in very hazardous work who, through intemperance, hereditary mental disturbances, domestic troubles and generally speaking ‘all around hard luck victims,’ have grown disgruntled, careless and indifferent, thus jeopardizing their own lives and the lives of those surrounding them.” 81

33

34

ch a p t er 1

Other managers and safety experts had noticed that some individuals were a constant source of errors and injuries. In 1913, William H. Tolman and Leonard B. Kendall, early leaders in the American safety movement, all but gave accident proneness a name, and they anticipated later discussions around the idea: “It is always a good plan to watch out for men who are hurt frequently. It is hard to discharge them, but this is better than running the constant risk that their clumsiness will endanger their fellows.” 82 All of this common experience, lay and professional, in all Western countries, set the stage for the idea of accident proneness to appear explicitly in Germany and Britain. Later, in 1933, the medical director of the Selby Shoe Company testified how he personally came to embrace the idea once he heard of it. Anyone connected with an industrial clinic or hospital soon becomes acquainted with workers who make frequent visits because of accidents or illness. During the past seventeen years I have repeatedly remarked, “Here comes the fellow who is always getting hurt.” At short intervals he reports with a cut, bruise, sprain or burn. Interspersed with his numerous minor accidents will be an occasional serious accident. Often you hear him remark, “I don’t believe anybody has more bad luck than I have.” He is the type of fellow who attracts the attention of safety engineers and safety committees. He and other such workers have been classed as “accident prone.” . . . The accident prone workers are interesting people because they present so many problems and because the best authorities today admit they know so little about them. 83

By the early 1930s, then, the formal concept of accident proneness was widely familiar—and, as this witness shows, he, like many others, could immediately recognize the phenomenon from his own long experience with men who encountered machines, experience that antedated even the naming of the idea of accident proneness. Perhaps the most curious example of anticipating the idea of accident proneness was a book, Mental Causes of Accidents, by a U.S. management consultant, Boyd Fisher, which appeared in 1922, just as the idea of accident proneness was taking shape but had not been differentiated and named. Fisher reviewed various kinds of individual actions and habits that could lead to accidents and errors, and he summarized them as: “ignorance, predispositions, inattention, preoccupation, and depression.” Some were temporary, some were persistent character traits. At one point, he showed

Before Accident Proneness

that he was aware of current beliefs, influenced by the new dynamic psychiatry, that “men are not always responsible for their own acts,” a belief reflected in no-fault worker compensation laws. Fisher wrote of “an age tendency to accident” in the young and “the special accident liability of old men.” Then at the end, in suggesting questions to ask in analyzing what happened in an accident, Fisher started right out with, “Is he a ‘repeater?’ i.e., prone to accident?” 84 Although there is no internal evidence, Fisher could well have been aware of the pioneer British work of 1919–1921, and he devoted his concluding chapter to advising managers about what up-to-date psychology books to read, including not only standard texts but Freudian works on motivation. In addition, Fisher was a major figure in the National Association of Employment Managers. Although it is entirely possible that Fisher had seen some of the early British publications, he certainly did not see there the terms “accident prone” or “repeater.” His apparently natural use of the terms “repeater” and “prone to accident” presaged how easily they would come into common use in English. It remained for Farmer actually to introduce the English term “accident prone” so that it acquired a technical meaning. 85 All of these observations, and others to be noted below, suggest that the phenomenon and some idea of a category of a person who had repeated accidents was present in developed societies. The German origins of the term Unfallneigung, inclination to accident, are notable in confirming this impression and show how European thinkers were prepared for the idea of accident proneness when it appeared.

35

2.

GERMAN ORIGINS

Karl Marbe (figure 2:1), the German originator of the idea of accident proneness, had already in his psychological research shown an interest in errors in speech and writing in connection with his interest in linguistics. Marbe investigated the associations between words to try to establish the nature of normal thought processes. When his experimental subjects did not make the usual associations in a timely way, he conceptualized their variation from normal as a faulty or contaminated thought process—by implication a deviation by the subject.1 In Marbe’s psychological work, therefore, he had already been writing about variations from a normal pattern of reaction. He was thinking, however, in terms of experimental subjects rather than in terms of individual differences in a general population. Differences between individuals, as I indicated in the previous chapter, was an orientation that was developing elsewhere in psychology in that same time period.

m a r be’s rou t e to “discov ery ” Marbe was acquainted with psychological research that suggested variations on the statistical normal curve. He was also aware of the work of others on individual differences. But all such works he subordinated to his philosophical belief that phenomena grouped themselves together— including the phenomena in human thinking processes (demonstrated, he pointed out, not least in the more general phenomenon of simultaneous discovery in science!).2 Marbe himself, however, traced his thinking on the subject of accident 36

German Origins

figure 2:1. Karl Marbe in 1909. Source: Selbstbiographie des Psychologen: Geheimrat Prof. Dr. Karl Marbe in Würzburg, ed. Emil Abderhalden (Halle [Saale]: Deutsche Akademie der Naturforscher Leopoldino, 1945), frontispiece, reproduced with the kind permission of the Deutsche Akademie der Naturforscher Leopoldino.

proneness as such back to his school days. The only child of a German merchant initially operating out of Paris, Marbe was born in 1869 and attended school in Freiburg i. B. in Germany after his father became ill and moved back to Germany. 3 Marbe later recalled that as a boy attending the Volksschule [elementary school], it occurred to me that several of my fellow students again and again would suffer small accidents, while other students remained wholly free of them. That several pupils also suffered different illnesses, while others in the same time period remained quite healthy, was also observed by many teachers. And in life one meets a number of people who have broken every possible limb, while, again, others never have such a misfortune.

Marbe went on to note that there are capitalists and bankers who never fail to incur losses, while others produce good outcomes from their ventures. And in general, some people have bad luck all of their lives, and others, “Glückskinder” [lucky people, children of fortune], enjoy success.4 Marbe, who was one of the key figures in the history of the Würzburg School in psychology, was not a likely figure to originate an idea of accident proneness as part of applied psychology. He was a defender of scientific and rigorously experimental psychology. But he was drawn into applied psychology through some pioneer work that he did in forensic psychology. In 1911, he gave expert opinion in a case of sexual abuse, and

37

38

ch a p t er 2

figure 2:2. Photo of wreckage after a train ran off the track and smashed into the station at Müllheim in Germany in 1911. Psychologist Karl Marbe determined the human error that caused the tragedy and subsequently gave much attention to the psychology of accidents. Source: E. Nohl, “Die Eisenbahnunglück am Hauptbahnhof Müllheim i. Baden am 17. Juli 1911,” Zeitschrift für Samariter und Rettungswesen, 17 (1911), 132, through the generous courtesy of the Staatsbibliothek zu Berlin.

in 1912, he got involved in two more court cases. 5 The next year, he published a report of his opinions concerning the psychology of employees involved in a terrible train accident near the Swiss border (figure 2:2). The engineer had a history of unreliable behavior, and the guard who might have stopped the train did not have the mental capacity to make an appropriate judgment. These conclusions served as models for Marbe’s interest in psychological factors in accidents for many years. 6 With this court expert experience, he and his students were fairly launched into applied psychology, including now the psychology of accidents.7 Marbe first announced his new surmise incidentally in a long work on philosophy in 1916. He had decided that the accident statistics showed that the probability was that someone who had suffered one accident must expect to suffer further accidents. Or, as he fi nally put it, the probability of an accident was greater for the person, A, who had had an accident, than for the person, B, who had not. And Marbe meant this as conventional statistical probability, specifically beyond chance distribution. For empirical evidence, Marbe alluded generally to the German insurance statistics, which he assumed were available to anyone. 8

German Origins

Marbe viewed statistical grouping in terms of his philosophic belief that, in nature, similarities occur and tend to group together. With regard to human beings, Marbe held that they would behave consistently in similar circumstances—as they did in word association tests. He even referred to the law of repetition by which an individual person would show consistent behavior. The basis of the consistent behavior Marbe held to be a combination of heredity and acquired habit, a combination that produced what he characterized as personality—a term coming into common use in psychology at that time.9 One of the themes running through the work of Marbe and his students was that of errors in otherwise uniform or similar performances.10 One student, for example, investigating errors in writing in 1913, focused on the psychological processes and regularities in writing—and in making errors. He was not looking particularly at the people who were making errors but rather the total psychomotor process going into writing.11 Yet Marbe was acquainted with and cited the work on association of ideas carried out by Emil Kraepelin and Gustav Aschaffenburg, who found mental patients with distinctively disturbed associations that were different from those of most people. Marbe also knew the early twentiethcentury work of C. G. Jung, who believed that deviant responses on word association tests revealed constellations of disturbed thinking in patients.12 But what seemed to catch Marbe’s attention particularly was not the idea of deviation from a norm as much as the regularities in errors— and then accidents, which appeared more consistently in certain people, in individuals. In 1924, to describe the “personal factor” in accidents, Marbe was using a chemical metaphor, Unfallaffinität or accident affinity.13 Affinity was of course consistent with Marbe’s idea of uniformities grouped in nature.

in t roducing a t er m In 1925, in a short paper in the proceedings of a local society, Marbe began referring to this probability of a person’s having accidents as Unfallneigung, which may be translated as inclination (or strong inclination) to having an accident.14 This paper of Marbe’s came out a year before his more publicized (and cited) use of Unfallneigung, in 1926. Marbe in this way first introduced a new name for the phenomenon in an obscure publication, and, as I have noted, in exactly the same year as did Farmer, the English namer. Marbe in his earlier formulations had used other familiar terms. As late as 1923, he did not use the term Unfallneigung when he reported on in-

39

40

ch a p t er 2

dustrial accident cases analyzed by his student, Nico Greiveldinger. There was, wrote Marbe at that time, a personal factor that made some people disposed (disponiert) to accidents. Accident-burdened individuals had different personalities (“die verschiedenen Persönlichkeiten der Unfälle erleidenden Individuen”). He relied most on the medical term, “disposition.” By the last part of the article, Marbe was referring casually to the term in various forms: “Unfalldisposition” and “Unfällen disponierenden Faktor,” and “unfalldisponiert,” and he ended up with “Disposition zu Unfällen,” a wording that he used repeatedly.15 Marbe’s definitive publication on accident proneness (Unfallneigung) was his book of 1926 summarizing his thinking. He was already able to cite four articles published by others in 1924 alone on the subject of people who had accidents, but for the most part his book was a synthesis of his own previous publications. He explained in detail how consistencies in behavior were based on both inherited constitution and upbringing and mental habits and training. Those consistencies inclined an individual to have accidents (“individuelle, persönliche Unfalldisposition oder Unfallneigung”). Moreover, he pointed out again, the strength of the inclination was proportional to the number of accidents previously suffered. Evidence from those who consistently had no accidents was even more convincing, Marbe noted, than that of those who had a record of accidents.16 He thus had grouped people into two categories: those who did not suffer accidents (Nichtunfäller), and those who had accidents (Unfäller) in proportion to the number of their accidents in a given time period. There was one element that was new in 1926: Marbe explicitly extended the idea of accidents from personal injury to production errors, picking up on his previous work on psychomotor errors. Errors made by children sorting peas showed the same incidence pattern found among workers who suffered injuries. Finally, Marbe emphasized accident proneness as an “endogenous” factor, and, despite earlier words about training, he wrote pessimistically. These were not ordinary people. The statistics for criminal convictions, Marbe pointed out, showed the same pattern as for accidents: the more convictions, the more chance of recidivism. Both showed the law of repetition of behavior.17 By shifting his terminology away from “disposition,” Marbe was not just re-naming accident proneness. He was signaling his perception that inclination to accident did not belong in the realm of medicine. Unfallneigung therefore was a phenomenon that belonged in psychology, not medicine—and, as will be noted below, Marbe’s British counterparts made a similar transition at the same time, 1925–1926.

German Origins

Moreover, from the beginning, Marbe made clear that his intended audience was not just theoretical psychologists. He saw that his ideas should have concrete application to insurance, industry, and transportation. An accident prone chauffeur, Marbe contended, needed to find another kind of occupation. In industry specifically, as the right worker was found for the right job and production damage was minimized, costs would go down. He ultimately invoked social values, beyond profit, using locomotive drivers as a dramatic example. He had two immediate practical suggestions. One was psychological testing of workers. The second was the idea that each worker should have a personal card on which accidents and errors would be noted—an idealistic scheme that was not followed up, by Marbe or by others.18

pu bl icizing t he ide a Given Marbe’ new practical interests, it is not surprising that he publicized his ideas in a number of different venues in the mid-1920s.19 The most important was the world of European, and especially German, psychotechnology. In that rapidly expanding world, taylorization techniques were already very important, but devising personnel tests had become a major concern. Between their preoccupation with management techniques and aptitude tests, even applied psychologists, who knew Marbe’s reputation in the general discipline, had not been raising the question of accidents when he began writing about his theory.20 As a British observer of German psychotechnology explained, Experimental psychological methods have been used especially with female workers in the electrical industry, with apprentices entering a trade, and, in some cases, with fully trained workmen . . . The strong elements in the character and physique of an individual are registered as exactly as possible in reference to the weak, so that his real position in the workshop can be found. The exigencies of modern mass production have caused greater emphasis to be laid on precision of the senses, attention and skill in action, and closer co-ordination of intelligence and physical movement.21

Fitting the worker to the job, however, did offer an opening for Marbe’s ideas. The problem was that accident proneness was not a desirable trait for a worker—the kind of trait that most psychological or practical tests might identify in an employee or job candidate—but rather a trait that

41

42

ch a p t er 2

employers would want to avoid. Likewise, studies of output did not usually include accidents and breakages but, rather, general working conditions, such as lighting.22 By 1926, however, Marbe could point out that some industrial policies already were implicitly recognizing the problem of accident proneness. Most conspicuous (as I shall describe in chapter 4 below) were the personnel practices of some German railways, which included psychological testing as early as 1917.23 Indeed, Marbe used evidence from some of these industrial practices to vindicate his theory. The managers of transport systems were well aware of the potential for injury and property damage from employees who might in some way tend to have accidents. But most of their tests were traditional—for intelligence, reaction time, vision, manipulative ability. All of these traits, physical and mental, were implicitly, if not explicitly, designed to ensure efficient and safe operations of transportation systems. This same kind of testing could be applied to other occupations, and already was, sometimes—such as woodworking, in which workers often sawed off a finger. Could the worker see? Could the worker react? 24 All of the explications in such discussions tended to be stated in terms of what was desirable to make a good and safe worker. Moreover, the dangers were specific: a signal not detected, a sudden danger not reacted to, a switch not thrown. What Marbe, by contrast, was describing was a person who was not disqualified by some specific disability such as color blindness or indecision, but who, on the contrary, could be identified simply because he or she had suffered one or more accidents.25 Already in 1925, when he could announce the forthcoming publication of his book, Marbe had devised his own history of his ideas. He confirmed how even before World War i he and other psychologists had turned to applied psychology and how they had applied psychology to prosthetic devices for the war wounded and how they had adopted psychological testing, particularly psychological testing that might assist in occupational choice and employee selection. He specifically called attention to the pioneer testing of locomotive engineers and other railway employees by Walther Moede and by K. A. Tramm (see below) and to his own recent advocacy of testing of railway personnel, which he published in 1924. And then Marbe pointed out how all of this work fitted in with his ideas about the ways in which natural phenomena grouped themselves.26 What Marbe was advocating in the field of accident-inclined individuals was a theory, and as early as 1923, he was referring to “my theory.” Workers in the field of psychotechnology did eventually become ac-

German Origins

quainted with the theory, but they kept it at some distance from their practical, task-oriented vocational testing. As a reviewer of his book put it, “Marbe erläutert seine Unfällertheorie” [Marbe explains his theory of those who have accidents]. The reviewer noted that the theory was convincing and was of general interest.27 But it was still only a theory.

ex t ending k now l edge of m a r be’s ide a s What kind of impact did Marbe’s work have? The evidence indicates that most European applied psychologists came to know his work, but citations of Marbe diminished after the early 1930s. In Germany, the major publications explaining and expanding his theory came from his students. As early as 1924, for example, before the terminology was in place, Maria Schorn in Marbe’s research institute wrote a paper titled “Unfallaffinität und Psychotechnik,” where she explored the idea of testing for characteristics that would prevent accidents. After reviewing the specific tests, as for strength and reaction, she went on to point out that keeping accident statistics to identify individuals with an affinity for accidents would serve to prevent accidents. Even among schoolchildren, her investigation showed, one could separate those who were having accidents from those who did not (figure 2:3). Schorn of course mentioned Marbe’s ideas.28

figure 2:3. This table, published by Marbe’s student, Maria Schorn, in 1924, shows clearly how the author was already publicizing Marbe’s idea of dividing those who had accidents (Unfäller) from those who did not (Nichtunfäller). Source: Maria Schorn, “Unfallaffi nität und Psychotechnik,” Industrielle Psychotechnik, 1 (1924), 160, reproduced with the kind permission of Springer Science and Business Media.

43

44

ch a p t er 2

Perhaps the most striking work confirming Marbe’s theory was carried out by a German railway administrator, E. Schmitt, who employed extensive statistics on railway accidents. A 1924 presentation by Marbe set Schmitt off on his investigation, he reported, and he was still using the term “accident affi nity” from this earlier work of Marbe’s. But Schmitt was able to answer two possible objections to Marbe’s theory. First, he found that the phenomenon of the tendency to have accidents was unrestricted: it was strikingly obvious among a large number of various kinds of railway workers over several years. Second, he found that even if one took into account the common finding of differing rates of accidents at different times of day, on different days of the week, and under different atmospheric conditions—there still were the usual groups of those with consistently safe records and those whose accidents foretold further accidents in the future.29 In 1929, in a paper out of Marbe’s Würzburg institute, Anton Grüb summarized the previous work and further confi rmed Marbe’s theory. But Grüb now attempted to find what elements in the workers’ personalities could be associated with an elevated incidence of accidents. He examined one hundred railway workers for various characteristics, and the most important personality traits that he could associate with accident rate were such things as alertness, attentiveness, concentration, and prudence. Grüb pointed out that it would be possible to test employees for these characteristics, and he also pointed out that these were characteristics that involved habits beyond the inborn personality. Safety instruction and propaganda did work, he pointed out, as other writers had already shown in the psychotechnology literature. 30 These publications out of Marbe’s Würzburg institute and by the railway official who heard Marbe speak suggest that, by 1930, a small but substantial literature existed on Marbe’s theory. The authors cited not only Marbe but each other. Sometimes the influence went further. As early as 1927, Fritz Giese of Stuttgart in a textbook on business and industrial psychology mentioned Marbe’s theory in a short section on industrial accidents, noting that accidents and personality were connected in a regular way susceptible of investigation. 31

fa il ing to at t end to m a r be’s t heory Yet there were many German-language writers who did not refer to Marbe or to the idea of Unfallneigung. It was not just that the authors of psychotechnics and safety publications did not take up the idea. The jour-

German Origins

nal in which much of the Würzburg material appeared, Industrielle Psychotechnik, for years at a time in the 1920s and 1930s carried no relevant material. In particular, many experts writing explicitly on the psychology of accidents did not find Marbe’s theory of interest. They could write on related subjects such as making errors or differences in reaction time without connecting their work in any way with the idea that some people would have more misfortunes than others. The latter was an additional consideration that they just did not utilize. 32 There also were changes in the position of psychotechnology in Germany, beginning in the mid-1920s, just as Marbe most energetically publicized his idea. As David Meskill points out, the mass-production organizations for which accident statistics were most appropriate were losing ground to screening for quality work and production. The character of a potential worker became of more interest than elementary skills such as reaction time that psychological testing could test. 33 Therefore even if accident proneness was part of personality, German employers (outside of mass transit) tended to focus on attributes in a worker other than an inclination to have accidents. Aside from adapting to the German industrialists’ turn to quality, when psychotechnologists discussed the psychology of accidents, they (like many others, particularly in the Anglo-American world) mostly had in mind long-standing campaigns to use psychology in connection with educating or manipulating workers (and sometimes the general public) so as to elicit safer behavior. 34 Or they had in mind standard tests for reaction time and perception such as were already in place for selecting transportation workers. Or they advocated commonsense measures, such as training workers and urging them to use foresight and intelligence. 35 In 1927, in a dissertation from Bonn, a neurologist reviewed the literature on children’s accidents, noting for example that a company attempting to write accident insurance for children found that 1 percent suffered injuries, a rate so high that premiums had to be increased. While she did note that epileptics and hysterics and those with other kinds of illnesses could constitute safety problems, her exposition, like that of many, many other writers, was largely about the kinds of injuries and sequelae that children suffered. Nowhere did she bring up the idea of accident repeaters. 36 The insurance company was following standard procedures for the industry everywhere in the early twentieth century: to pick out a group to be insured and to make an estimation of the charge for the insurance that would cover the total injuries in that group. Where experience showed

45

46

ch a p t er 2

that the estimate for the injury rate of the group was too low, the insurer raised the premium rate for children as a group. At that time in Europe, as I shall note further below, accident insurance was generally not written for individuals, and Marbe’s idea was simply not directly relevant to industrial insurance practices. It is true that Marbe had started his empirical work with insurance statistics. It is therefore ironic that identifying accident prone Unfäller did not affect practices in industrial insurance, either in Central Europe or elsewhere. In 1930, a Swiss writer on accident insurance cited Marbe’s 1926 monograph—but for the purpose of emphasizing that nonenvironmental factors caused accidents. This writer would go no further than naming inappropriate vocational choice, and there is no sign that the point of Marbe’s monograph, accident proneness, was relevant at all. Accident insurance, the author pointed out, assumed that the insured was a normal and healthy person (not, for example, an epileptic). In the world of insurance, the accident would be a “natural” happening. In fact, the idea of insurance was to diminish concern about individual cause and blame. 37 During the 1920s, then, Marbe attempted to call attention to his theory, and he and his students gathered empirical evidence to confi rm and apply the theory. Yet, in the end, not many psychologists or workers in industry or insurance reacted to the idea that identifying individual agents of error and accident was an idea worth developing, extending, and using. Nevertheless Marbe had succeeded in bringing into applied psychology and safety literature some familiarity with the concept of accident prone people. Writers on the Continent, particularly, cited his work for decades after the 1920s, albeit at a diminishing rate. Marbe was active in the international psychotechnology organization in the 1920s, and in 1927 he headed a committee on the psychology of accidents. Then in 1928, none of the members attending the meeting of the psychotechnologists in Utrecht could have missed knowing about the idea of accident proneness. At a session specially on accidents, not only did Marbe speak, but so did his student, Maria Schorn. Also on the program were J.-M. Lahy of France, who later wrote a book that dominated what there was of Continentals’ writings on the subject of accident proneness, and Morris Viteles, an American whose importance in publicizing the idea among psychologists is also taken up below. 38

German Origins

a va r ie t y of r e ac t ions to m a r be’s wor k As early as 1928 in Germany, however, another type of reaction appeared: criticism of Marbe’s theory. The basis of the opposition was that the identity of an accident repeater was not valid when applied to individuals, and this theme continued in many discussions of accident proneness for all of the decades afterward. The psychotechnologist who began the attack was H. Hildebrandt of Berlin. He published an article devoted to questioning the validity of Marbe’s work, and he laid down what turned out to be the main lines of later arguments. 39 On the basis of the records of 200 female workers, Hildebrandt showed that there was a very large number of exceptions to the idea that accidents occurred to people in proportion to their earlier accident records. And the number of exceptions, Hildebrandt held, was so great as to deprive Marbe’s theory of an acceptable level of validity. Moreover, Hildebrandt found six youngsters, three of whom could be classified as accident-free, and three of whom had accident records. First he had them examined by a physician. The results showed no “constitutional” basis for differences between the two groups. Sensory acuity, for example, was greater in the accident repeaters. Then he subjected the youngsters to psychological tests for such attributes as attention and reaction time, tests that were used then for such employment as streetcar driver. Again, no basis for classification existed. The accident-free children had reaction times slower than the accident prone. Hildebrandt was aware that his numbers were small, but the results were so contrary to what Marbe contended as to raise profound doubts: the theory of inclination to have accidents was just not useful in judging any particular individual.40 Marbe’s theory had, therefore, become known among Germanspeaking experts—to the point that they now could reject it. In 1929, for example, the Göttingen psychologist, Narziss Ach, well known for his experimental work on reaction times, wrote about preventing automobile accidents. He was concerned with speeding, which in crashes brought the laws of physics into play to the disadvantage of the human body. But at one point, he took pains to say, without mentioning Marbe’s name but using his terms, that even Nichtunfäller, who had superior psychophysical characteristics so that they were “not inclined to have accidents,” could not escape the laws of physics. Clearly, Ach expected his readers to understand his sarcastic allusion.41 Nevertheless, many people concerned with accidents simply ignored Marbe’s ideas. German publications on safety continued for the most

47

48

ch a p t er 2

part unaffected by Marbe’s concerns, although human factors and even psychology became more important in the German-language literature. An observer in 1930, for example, who characterized Germany as “the motherland of technical accident prevention,” noted that personal and psychological factors now were becoming much more prominent, that is, in practical efforts to motivate and educate workers. Writers increasingly examined the role of alcohol and even tobacco in accidents, to say nothing of carelessness, fatigue, and all of the other human factors that an American-style educational safety campaign (as Europeans saw it) might affect.42 One key figure suggests how Marbe’s work appeared to many of his German colleagues—and how they could translate his findings into confirming their preexisting conceptualizations. The figure was Moede, who was also editor of the main psychotechnology journal in Germany and author of a defining textbook in the field. As I have indicated, he published articles by Marbe and his students, and he attended meetings at which Marbe and others spoke about accident proneness. In his 1930 standard textbook on psychotechnology, Moede did list Marbe’s 1926 monograph as a reference, but he did not discuss Marbe’s theory in either the relevant passage or elsewhere in the book.43 Finally in 1934 in a major article, Moede explained his stand on Marbe’s theory. Yes, he wrote, there were Unfäller and Nichtunfäller in industry. But Moede enthusiastically claimed that standard psychotechnology could solve the problem of excesses of accidents by addressing the components of accident proneness. Psychotechnologists, he explained, started by doing a detailed analysis of accidents and errors to fi nd out exactly what was happening. Workers with reaction and attention deficiencies could be detected by psychotechnological testing, either before or during employment. Testing, especially aptitude testing, could also identify people with other personality traits that made them predisposed to have accidents. And at that point, wrote Moede optimistically, psychotechnologists could provide either ways to screen out unsafe job candidates, or they could suggest safety training that would counter undesirable traits. Thus the problem and the cure were already being handled by psychotechnologists.44 Outside of the German-speaking world, writers on accidents and safety, much less psychology, did occasionally refer to Marbe’s ideas, but often they did not.45 In the International Labour Office 1930 encyclopedia of hygiene, pathology, and social welfare, the authors of the article on industrial accidents and “the human factor” did not refer to Marbe and did not take up accident proneness per se except (conceivably) by listing psy-

German Origins

choneurosis as a possible factor in individual accidents. Instead, the authors kept to the standard factors of fatigue, lighting, temperature, inattention and carelessness, age and experience, reaction time, and the like. In one set of statistics, showing that some workers accounted for many accidents, the disparity came from the fact that “certain individuals fi nd adaptation to a given type of work impossible.” Marbe’s idea of Unfäller as such was just not present.46 Yet the potential for introducing the concept existed, and in succeeding years, the idea appeared in significant publications on the Continent, if not so often in Germany (under circumstances explored further in chapter 6). In the years after World War ii, Marbe’s name continued to come up in the reports of various investigators. What they remembered and remarked on was Marbe’s original categories, separating those who had accidents from those who did not. What later writers left out was the idea of a personal attribute of inclination to have accidents. Marbe’s term, Unfallneigung, ultimately did not survive well, indicating that the idea did not, either. The last extensive use of Unfallneigung that I have found is in an up-to-date 1961 book on occupational psychology by a Braunschweig specialist who cited Marbe only incidentally.47 Both German and other Continental writers after World War ii, however, tended to use cognates of accident disposed or predisposed rather than accident inclined. As late as 1955, a writer who cited Marbe’s work still used Unfallaffinität rather than Unfallneigung. Two German-language writers of that same year simply used the English word “proneness,” albeit in quotation marks. A 1959 Swiss expert named Marbe as the originator of the idea of Unfäller, those who have accidents, but she confirmed that in “today’s scientific terminology,” the wording would be the one that Marbe had abandoned in 1926, affinity for accidents—now constructed for modern use as Unfallaffinität.48 In the 1920s, however, both Marbe and his theory received a significant amount of attention from colleagues interested in the psychology of accidents. And immediately after the appearance of Marbe’s defi ning book in 1926, there was one particularly important international recognition. A British journal representing industrial psychology carried a review of the book, starting with: “That individuals vary in liability to accident has been frequently asserted; but Dr. Marbe’s book, like Dr. Greenwood’s work in this country, raises what was once merely a ‘view’ to the status of an established fact.” The rest of the review summarized Marbe’s evidence.49 The review was unsigned, but there is good reason to believe that the author was the editor of the journal and the leader of industrial psychol-

49

50

ch a p t er 2

ogy in Britain at that time, C. S. Myers. Myers deliberately played a central role in the field and had effectively been the founder of the National Institute of Industrial Psychology. 50 The other person who might have written the review was Eric Farmer, who later in 1927 published his own article on “personal differences in accident liability.” There he reviewed the British work to date and at the end also referred the reader for further information to Marbe’s book.51 At this point, then, in 1927, for the first time the German and British explorations of accident proneness, or Unfallneigung, began to come together.52 Simultaneously, a variety of writers in Europeanized countries for their own reasons took up some version of the idea of accident proneness. British experts developed the idea of accident proneness in ways that found better institutional support, or at least receptivity, than did Marbe’s initiative. Except among administrators in the transport industries, Marbe’s work remained largely theoretical, while the British attempted more generally practical applications of the idea. And the implications in Marbe’s work, that a personality that included an inclination to accidents might have an inherited basis, often did not play well, as I shall note later, in other national cultures.

3.

BRITISH ORIGINS

Where Marbe distinctively made inclination to accident “his” theory, in Britain accident proneness initially came out of an organization. Despite the fact that they were describing the same phenomenon, the German and British principals followed different paths in the ways in which they advocated the concept. Indeed, until Eric Farmer became the chief advocate, the British workers seemed to find the idea of an accident prone employee an awkward, isolated empirical finding.

t he process of discov ery in br i ta in British investigators first reported their findings in 1919. The circumstances were these. During World War i, great pressure developed to increase industrial output.1 In an effort to maximize munitions production, the British government set up a Health of Munitions Workers Committee in 1915. As early as 1916, it was apparent that authorities were coming to appreciate the contributions of physiological and even possibly psychological experts. “Physiological management,” wrote one official in 1916, was a great improvement over philanthropists who were clashing with laissez-faire employers. Instead of such philosophical issues, industrial fatigue “for many months has been prominently before the Health of Munitions Workers Committee.” 2 At the end of 1917, the government appointed a committee “to investigate the subject of Industrial Fatigue on comprehensive and systematic lines,” with regard to “industrial efficiency and to the preservation of health among the workers.” Officials developed particular concern about 51

52

ch a p t er 3

figure 3:1. Women in a munitions plant during World War i, Woolwich, UK, wheeling boxes of cartridge cases. The records of these and other women munitions workers led to the discovery of unequal individual incidences of industrial accidents in Britain. Source: Imperial War Museum, reproduced by permission.

labor unrest and the health of women workers in the munitions industry (figure 3:1). The working group on fatigue continued in 1918 and then, even during and after the postwar depression, as the Industrial Fatigue Research Board, a unit of the Medical Research Council. 3 At some point that cannot now be ascertained, the question of fatigue was extended from production and health to accidents. Accidents, according to the investigators, “may be taken as indicative of fatigue.” 4 Eventually the accident studies, and a substantial proportion of the personnel involved, moved into the Industrial Fatigue Research Board. In the first accident publication out of the wartime investigations, H. M. Vernon, an Oxford physiologist and psychologist, revealed that he had accident statistics from four wartime munitions factories, and women workers were indeed a special concern. They were working from 55 to 75 hours a week, some, at one point, an average of 77 _41 hours. The connection between the hours of work, fatigue, and output was obviously of importance to factory managers after the war as well as during

British Origins

the conflict, and the hours of work were a constant problem in labormanagement relations. Vernon therefore set about to identify the connection between output and accident in the more general context of fatigue and hours of work. 5 Vernon amassed a record of 50,000 personal injury accidents. He concluded, on the basis of his statistics, that “Accidents depend, in the main, on carelessness and lack of attention of the workers.” Fatigue, lighting, and temperature all contributed to those human failings. In particular, Vernon found that “speed of production” could produce higher or lower rates of accidents. 6 In Vernon’s work, not a hint of persistent individual differences in accident rates appeared. Yet that same year, 1919, in another publication, his colleague investigators with the Industrial Fatigue Research Board, Major Greenwood and Hilda M. Woods, reported individual differences as one of the findings of the British investigators. Greenwood was an epidemiologist (his university title was medical statistician), a pioneer on his way to becoming a star in the field (figure 3:2).7 Woods was characterized as a staff assistant. They summarized their findings by writing, “the bulk of the accidents occur to a limited number of individuals who have a special susceptibility to accidents, and suggests that the explanation of this susceptibility is to be found in the personality of the individual.” 8 Greenwood and Woods noted that their work was based on extensive records held by the Ministry of Munitions, but they did not reveal who first noticed that a small group of workers produced “the bulk of the accidents.” Moreover, Greenwood and Woods added to the munitions workers’ records those of another large factory. By 1919, their published

figure 3:2. Major Greenwood. Source: Biometrika, 38 (1951), 2, reproduced with the kind permission of Oxford University Press.

53

54

ch a p t er 3

discussion of alternative explanations and additional hypotheses suggests strongly that the profi le of workers who had accidents had stimulated much thought among investigators associated with the fatigue work.9 Moreover, Greenwood and Woods saw other implications of their new way of viewing the data. Twice they mentioned practical consequences— beyond the dangers of speeding up production: “both initial selection of recruits and also a rapid elimination of those sustaining multiple accidents should have a great effect in reducing the casualty rate of the hospital,” and “by weeding out susceptibles the accident rate would necessarily decline.” Also they referred to “unsafe people” and “a more dangerous person.” Accidents, they concluded, did not occur just by chance. Chance they portrayed as the explanation that was alternative to their finding. Rather, they believed that “varying personal liability” was an important factor in accidents.10 Altogether, the conclusion, that one worker was someone who had accidents while another was free of accidents, was remarkably parallel to Marbe’s categories of Unfäller and Nichtunfäller. Greenwood returned to this problem in a paper he published in 1920 jointly with a statistician, G. Udny Yule. The two together examined the methods of calculation by which the accidents in munitions factories were examined for probability. They concluded that the accident rate for workers should not show anything like the distribution that they found. That is, the distribution of accidents was not a chance distribution. The means of calculating the chances, however, were very complex, and so they justified a simpler method to reach the same end. But in this article they focused on method, not on the conclusion that the more accidents a worker had, the more likely was that worker to incur more such incidents.11 In 1927, Greenwood recalled the origins of the British work in the only record that exists as to who, among their group, actually noticed the workers who had a markedly disproportionate number of accidents. When he and Woods first published, a critic noticed that it was trivial work, for everyone knew that “liability to accident varies.” But Greenwood, working with Yule during the war, saw a theoretical interest in accident statistics. I was on a Committee appointed to advise on certain medical aspects of the Flying Service. Inter alia we had to consider the nature of tests which would be suitable to determine the aptitude of entrants for the Flying Service, and to determine whether persons who had broken

British Origins

down, or who had been sent back from the front for wounds or other reasons, were fit to return to duty. In the course of the investigation, the point arose whether a man who had had a small accident not sufficient to damage himself or his machine seriously might yet have sustained such a shock that it would be undesirable for him to fly again. The problem emerged whether it would be possible to distinguish by analysis of the frequency-distributions of accidents between the three possibilities: (1) That accidents are pure accidents in the sense of being “simple” chance events; (2) that the distribution of first accidents is that of “simple” chance events; (3) that the distribution whether of fi rst or subsequent accidents differs in a specific way from the “simple” chance scheme.

Since Air Force personnel were subject to “unequal exposure,” data about individuals would not be comparable. Greenwood and Yule therefore turned to the records available to them, trivial accidents in munitions factories.12 So it came about that “two statisticians” noticed the unequal distribution of accidents, as Greenwood recalled at the end of his life. The phrasing suggests that the statisticians were Greenwood and Yule. Clearly Greenwood was one of the two who saw the statistical distribution, even if, given the gender relations of the time, it might have been Woods and not Yule who was the other “statistician.” 13 The fact remains that the idea was taken up by the whole research group. Thus a theoretical question in statistics, plus common knowledge (like Marbe’s) that some people have more accidents than others, led a British group to systematize the idea of accident repeaters and to see that the theoretical findings had practical implications. The fi ndings then were extended by another group of workers, who had personal contact with the originators within the same institutional setting, the Industrial Fatigue Research Board and its successors.14

ex t ending t he f indings In 1921 in a general book on The Health of the Industrial Worker, Greenwood repeated and expanded his conclusion that “in industrial life we may remove from risk those who show a tendency to accidents.” He reported on the ways in which varying conditions could change “the human factor” in accidents, but he then took up the findings he reported jointly with Woods, that multiple accidents happened to the same individuals. To the earlier findings, Greenwood added, from further research by the group,

55

56

ch a p t er 3

confirmation that some individual workers have a “personal predisposition to accidents.” 15 While Greenwood did not himself pursue further this particular line of investigation, he had clearly launched the idea of personal predisposition to accidents in an important general book on industrial medicine.16 Meanwhile, in 1920, the first annual report of the Industrial Fatigue Research Board included a substantial recognition of this line of research: “The inferences drawn [by Greenwood and Woods], though new and of great interest, are admittedly based on slender data, and cannot yet be regarded as finally established. The Board, however, consider them of sufficient importance to justify a continuance of the investigation on wider lines, so as to include men and boys, and hope to extend it by psychological inquiry into the causation of accidents.” Then later on the same page, the tone of the report became more assertive, suggesting that “accidents are very largely due to a special susceptibility inherent in the personality of the individual, so that the bulk of accidents occur amongst a limited group of individuals”—although, the anonymous author added, a single accident has no significance.17 The members of the Board and of the Medical Research Council included many, if not most, of the leading figures in British medicine and especially members of the professional psychology community in the U.K. The first chairman of the Board was Charles Sherrington, the preeminent neurophysiologist. Surviving official correspondence shows that members of the Board and even the Council reviewed publications carefully—and of course the official reports.18 Such evidence makes it clear that, from the beginning, the idea of accident proneness was becoming familiar to an important group of experts. The Board continued to direct research into accidents. Some investigation emphasized physiological causes and trying to find physical conditions that led to accidents. Vernon continued some of this work, and the Australian pioneer industrial psychologist Bernard Muscio, then working in Britain, extended it with psychological experiments. The investigation also, however, proceeded in the period 1920–1924 with statistical and psychological work with the object of determining (i) how far a characteristic ratio between notifiable accidents and accidents of all kinds exists for different selected occupations; (ii) the correlation of accident incidence with experience, age, and sex;

British Origins

(iii) the extent to which special susceptibility to accident exists; (iv) the primary causes of such susceptibility.19

en t er er ic fa r mer A year later, the Board reported that the first of the four inquiries was impractical to pursue, and so the investigators concentrated on the last three—which moved accident proneness to a major focus of the Board. There were now three types of inquiries into industrial accidents, according to the Board. The first was statistical, carried out by Ethel Newbold and based on accident statistics of twenty-two large industrial fi rms. The second was that of Eric Farmer (“assisted by Mr. E. G. Chambers”), trying to select and devise tests to identify people who were having accidents.20 A third line was the continuation of the work of Vernon and Muscio. Farmer made such progress that by 1925, his first report, the one that incorporated the term, “accident proneness,” appeared as an appendix to the Board report (the 1924 report was published in 1925).21 Farmer, who had joined the Board in 1919, had been working mostly on time and motion studies (figure 3:3). His work on accidents, however, had begun at least by 1923. He was born in Northampton in 1888 and was a 1914 graduate of Trinity College, Cambridge, where he read moral sciences. Charles Samuel Myers, the central figure in British industrial psychology, recalled Farmer as one of the research students who passed through his experimental psychology lab at Cambridge. From 1935 to 1953, Farmer was reader in industrial psychology in the University of

figure 3:3. Eric Farmer. Source: Photo graciously furnished by Professor John Mollon from the Department of Experimental Psychology, University of Cambridge.

57

58

ch a p t er 3

Cambridge, a special post fi nanced by the Medical Research Council for him personally. Students remembered him as a clear lecturer but not as an innovative thinker. He was formal in manner, reserved, and had memorably military bearing and attitude. He clearly was not an easy mixer.22 His influence in spreading the idea of accident proneness, then, was destined to be formal, rather than personal, much less charismatic. Farmer’s short article, “The Method of Grouping by Differential Tests in Relation to Accident Proneness,” was a progress report on the research sponsored by the Board to find psychological tests to identify accident prone people. “We already know from work the Board has published,” wrote Farmer, “that people differ in their degree of proneness to accidents. This does not mean that there is a group who have accidents and a group who do not, but that people exposed to equal risk will tend to sustain accidents according to their degree of proneness.” 23 It was this idea of proneness about which Farmer, along with Chambers, wrote in the 1926 report that entered the scientific literature at the same time that Marbe’s monograph did. At that time, they, like Marbe, were using the medical concept of individual susceptibility. Thus a person could suffer repeated attacks of an illness under the same or similar circumstances, while the disease might pass other people by. And since the investigators had originally been working with labor in the munitions industry, their viewpoint was clear: “Trivial accidents are indicators of unsafe people whom the record of the ambulance room can be employed to discover.” 24 In later reports, the same group continued to use the formulation of individual susceptibility, or, at most, “tendency” to having accidents, until Farmer and Chambers introduced “accident proneness” and “accident prone” people.25 “ ‘Accident proneness,’ ” they wrote, “is a narrower term than ‘accident liability’ and means a personal idiosyncrasy predisposing the individual who possesses it in a marked degree to a relatively high accident rate. ‘Accident liability’ includes all the factors determining accident rate: ‘accident proneness’ refers only to those that are personal.” 26 Several years therefore passed between the time that the British investigators observed the phenomenon (during World War i) and the time that they accepted a term to identify it. In 1920, in the fi rst annual report of the Industrial Fatigue Research Board, the term used for personal factors was “susceptibility”: “accidents are very largely due to a special susceptibility inherent in the personality of the individual, so that the bulk of accidents occur amongst a limited group of individuals.” 27 Greenwood and Woods explicitly compared the susceptibility to accidents to

British Origins

the susceptibility to disease: “A person may acquire some disease by the merest accident, but passing through the attack will profoundly modify his chance of acquiring it again when the original conditions are, in all other respects, reproduced; he may be practically immune or conversely he may be much more sensitive to infection.” 28 Farmer, then, emerged as the proponent of the term “proneness” in 1925–1926. And he may have been the originator. Other publications from Board workers did not include it. Indeed, as late as 1926, the statistician who was working on the project, Ethel M. Newbold, published a report in which she referred to “the personal factor in accident causation.” In other passages, she wrote of “special individual susceptibility” and, less frequently, “individual tendency to accident.” 29 It was Farmer (sometimes with the assistance of Chambers) who introduced the nonmedical concept of “proneness” rather than the medical terms “susceptibility” or “predisposition.” British publications on accidents did attract attention from industrialists and those interested in safety. The attention was of course in the context of familiar ideas about physical conditions that led to accidents and about times and other circumstances that drove the accident rate up. 30 All such knowledge, people hoped, would help control and reduce the number of injuries. In a continuing series of various publications, Vernon publicized his point of view that workers’ carelessness and inattention as well as speed of production contributed to the number of accidents. In 1921, he published a whole book, Industrial Fatigue and Efficiency, rehearsing his ideas. There, for the first time, he alluded, albeit in a short paragraph, to Greenwood’s and Woods’s findings that some individuals had more accidents than others. “Apparently they possessed an innate clumsiness or carelessness. The output of these susceptible workers was not below the average,” he continued, but he expressed concern about “such of them as happened to be employed in a dangerous occupation . . . where an accident might cause frightful disaster.” 31 Concerned British experts also watched reports from the American and other contributors to the safety movement. The one important theme about which almost all of them commented was that there was one particular group of employees who amassed a notably high rate of accidents: new employees. All commentators continued their concern about fatigue, lighting, and alcohol, among the other factors, but it was clear that the experts were disagreeing about what factors were most important and which were easiest to control. 32

59

60

ch a p t er 3

a dd in: t he nat iona l inst i t u t e of indust r i a l ps ycholog y It was in this context that the continuing contributions from the Industrial Fatigue Research Board were understood in the decade after World War i. In 1921–1922, a new factor entered British research into accidents: the founding of the National Institute of Industrial Psychology (NIIP). The chief mover was C. S. Myers, who, as noted above, left Cambridge to head the Institute. Virtually all leading British psychologists were involved with the Institute at one time or another, and for two years Farmer was part-time with the Institute, until the Fatigue Board reclaimed him for full time. The Institute did much contract work for industry, and accidents were at times one of the subjects of research reported by Institute investigators. 33 From the beginning, the Institute published a journal and included abstracts of the literature. Accidents appeared as a regular category in that section. In the early 1920s, abstracts dealt with factors such as speed and lighting and tests used for transport workers. 34 One particular investigation was of particular interest: psychological factors that might affect breakage in a factory, for which, in addition to injuries, the term “accident” was also used. Investigators found that circumstances, rather than individual carelessness, most often accounted for breakages. By altering physical arrangements and reducing psychologically irritating or frustrating or fatiguing factors, breakage rates could be reduced dramatically. 35 The success of this work obscured any attention to groups of workers who might be accounting for a disproportionate number of breakages—other than new workers. Among experts, accidents of whatever kind clearly had become an important issue. In Britain, however, few people had shown interest in psychological tests for rail and road drivers. The NIIP—probably Myers—did, however, publicize the pioneering work of German, French, and American psychologists and transport officials (figure 3:4). Again, motor vehicles generated an interest that had unusual significance for promoting the idea of accident prone people. 36 Moreover, British industrial physicians were always aware of industrial accidents. In 1927, for example, a textbook on industrial hygiene included a special category of accident: “That unknown factor which seems to cause otherwise conscientious workers to sustain repeated minor or even major injuries.” 37 Altogether, then, the English-speaking world had developed an idea that not only general psychological factors, but personal, individual fac-

British Origins

figure 3:4. A 1926 British publication of an American simulation device for testing competence of tram drivers, “The Viteles Motorman Selection Test.” Viteles was becoming acquainted with both European and American innovative vocational testing using simulation apparatus. Not responding to distraction was a major component of this test. Source: S. V. Keeling, “Recent Tests for Competence in Tram Driving,” Journal of the National Institute of Industrial Psychology, 3 (1926), 90.

tors might contribute to accidents. Vernon in Britain, for example, published “The Human Factor and Industrial Accidents.” At the same time, an American was speaking to the Foremen’s Safety School on “the personal factor in accident prevention.” Both writers took note of accidents caused by dangerous machines but both also distinguished and emphasized incidents depending upon the human factor—“physical, mental, or moral defects” in individuals. Each expert of course had his or her own list of possible worker shortcomings. 38 The situation was summarized in an article in the Journal of the National Institute of Industrial Psychology, in which the author rehearsed the various factors, including age, experience, lighting, dexterity, emotional distraction, and the like, concluding, “That liability to accident may be increased considerably by loss of self-control will probably be conceded.” 39 Clearly, readers of such discussions were ready for the idea of personal “liability to accident.”

61

62

ch a p t er 3

t wo pu bl ic at ions in 1926 It was in that context that in 1926 Newbold published her statistical analysis of the British data and that Farmer and Chambers published their monograph naming accident proneness. Newbold had at her disposal not only the statistics used by Greenwood and Woods but accident reports from a number of producers of different kinds—thirteen large firms in all. She was interested primarily in the statistical questions by which one could ascertain and specify the “human factor” in industrial accidents, and she wanted, she wrote, a large data set of both “accidents and freedom from accidents.” First she determined that accidents originated with “a comparatively small number of workers, and that the distributions among the workers are far from chance ones.” She tested this idea by a number of statistical considerations, concluding, for example, that “the observed coefficients of variation are very large, much larger than the equal risk theory would lead us to expect and such as to indicate a much greater amount of lack of homogeneity than observation of the conditions of work would suggest.” 40 Newbold particularly teased out factors that might account for elevated accident rates among individual workers: age, experience, night or day shifts, sex differences, and reporting practices. She found that age was a small factor. Yet the fact remained that some employees had a “personal tendency” to have accidents, and that was the term that she used to describe her basic finding.41 Where Newbold noted that “tendency to accident” was “a rather vague quality,” Farmer and Chambers were set on a different mission: to see if they could find ways of analyzing and testing for the qualities that made some people have more accidents than others, a “special susceptibility inherent in the personality of the victim.” They therefore moved from “susceptibility,” introduced by Greenwood and Woods, to “accident proneness,” bypassing the “personal tendency” term that Newbold used. She herself had noted that the term “tendency” “has no meaning at all until the measure chosen to represent it is clearly defined and scaled.” That variable, she noted “is not meant to represent any mysteriously vague quality called accident or sickness tendency,” which could well have been an allusion to at least some of Farmer’s expositions.42 Farmer and Chambers started right out by distinguishing accident proneness from accident incidence. Incidence, they wrote, turned out to mean carelessness: “those who say that accidents are due to carelessness . . . when asked to define carelessness do so in such a way as to leave

British Origins

little doubt that by carelessness they mean having an undue number of accidents.” Farmer and Chambers administered a series of psychological tests to more than six hundred workers. They reached a number of conclusions, including confirming Newbold’s finding that people who reported minor illnesses also reported having accidents. But, fundamentally, they found that “personal measurable qualities” could distinguish between those employees who were accident prone and those who were not. Farmer and Chambers tested reaction times under varying conditions, such as distraction. They tested subjects for dotting on a pursuit meter (a moving target), for ocular muscle balance, and for other physiological processes, such as tremor. Making correlations of test results with accident proneness showed that intelligence, as measured by standard testing, was not a factor in accident proneness. But other investigations yielded positive results for tests of “aestheto-kinetic co-ordination” [a term introduced to avoid the established implications of standard “sensori-motor” or “neuromuscular”] and nervous instability. Farmer and Chambers noted the preliminary nature of their findings, but they expressed the hope that psychological testing could suggest groups of people who likely would have unduly high accident rates. In spite of the implications of their findings, they denied that they could identify a particular person who would be accident prone.43

t he ide a in t he a nglophone wor l d Within two years after these 1926 publications, English-speaking safety experts had begun to incorporate the British literature and use the term “accident prone.” 44 Most significant was Morris S. Viteles of the University of Pennsylvania in the USA, for he wrote the periodic summaries of world industrial psychology for the Psychological Bulletin. By 1928, Viteles, in his summary praising the scientific rigor of the studies, had put together the work of Marbe and other Germans with that of Newbold and Farmer and Chambers and other British authors, along with pioneer work begun in the United States on accidents of taxi drivers. By 1929, Viteles was reporting his own research into “proneness to motor vehicle accidents.” 45 Like the Germans, other Americans quickly began to take an interest in transport operators in the late 1920s. As early as 1927, two psychologists reported a study of motormen and bus drivers of the Boston Elevated Railway. The study, they declared, was initiated by the firm’s general manager, and they found that certain individual train operators were “suscep-

63

64

ch a p t er 3

tible” to having accidents. Of course they went on to try to discover why, and they came up with certain signs beyond obvious problems like faulty vision: older men with abnormal blood pressure, younger men with little experience, men with generally bad records (reflecting attitude usually), and men who did too much braking and accelerating. By 1929, the personnel superintendent involved in a similar study in Cleveland was using the term “accident-prone,” and a study of New York State drivers by the National Safety Council was explicitly paralleled to the work reported by the British group at the Industrial Fatigue Research Group. The latter specifically recommended the “weeding out of those [drivers] who are unable to avoid accidents.” 46 By 1930, the work of the British investigators had become written into a standard history. The implications of the idea of accident prone people had been developed. And advocates believed that by testing, the idea had practical outcomes. Farmer, writing alone, set out the history. Greenwood and Woods in 1919 identified an abnormal distribution of accidents among workers. Some workers “are inherently more liable to sustain them than others.” Then Newbold with more data confi rmed Greenwood’s and Woods’s conclusions, “which may now be regarded as established in view of the extensive data analysed.” Moreover, Newbold found factors such as high accident rates among younger and less experienced workers. And she eliminated conditions affecting all workers so that there was left “a small group of workers having an unduly large number of accidents.” And then Farmer and Chambers devised tests, tests that included “industrial proficiency,” or productivity, which correlated negatively with accident proneness so that using these tests to screen employees could “serve a dual purpose and increase industrial efficiency in two different ways.” 47 Chambers, writing on his own, confirmed this narrative of how British investigators found that accidents were not distributed randomly among a group of workers. Newbold established this skewed distribution definitively, and so “the majority of industrial accidents . . . are due to personal proneness.” Chambers then went on to observe that proneness is a “personal quality” but not a single quality. Rather it is “a group of co-operating functions.” This complexity, Chambers observed, made for difficulties in practical research on the problem. The only choice left, and the one followed by Farmer and Chambers, was to apply a whole battery of psychological and physiological tests to try to devise a score for each individual’s accident proneness. Indeed, “although the individual has a certain proneness to accidents, it may be that he has no special ability

British Origins

for their avoidance.” In any event, concluded Chambers, research into accident proneness could have great practical value. 48 That same year, the assistant director of the National Institute of Industrial Psychology, G. H. Miles, gave a broader and even more optimistic reading of the work that had been coming out of the Industrial Health Research Board. Miles made a strong claim for the practical advantages of the work on accident proneness by recounting the experience of a NIIP investigator in a large textile mill: Our investigator had obtained accurate records of all accidents for a considerable period, and he found one instance of a boy who had already sustained five minor accidents during the period. As the boy was working in a particularly dangerous job, he was selected, together with several others who also showed high susceptibility, for transfer to other work. Before the transfer could be made, however, he was caught and lost his hand in the machinery he was cleaning. Up to this point, there had been considerable doubt on the part of the managing director as to the need for such a detailed accident survey. Needless to say, those workers who sustained multiple accidents in future were quickly removed from the danger zones.49

Already in 1930 Miles, like American experts, was taking special notice of one area from which accident prone employees could be removed with great profit: transport (as I detail further in the next chapter). He was able to cite a number of examples from Germany, France, and the United States in which transport managers had improved their safety records dramatically by identifying accident prone employees or giving tests that would eliminate before hiring those with faulty judgment and reaction times. After presenting statistical examples, such as selected tram operators in Paris who had 16.5% fewer accidents than did the unselected group, Miles concluded: “Such results require no comment.” 50 Miles and other writers thus at the end of the 1920s showed that as the British work on accident proneness expanded, a pattern was developing. First, practical German results, such as those with transport operators, were being incorporated into the British discourse. Second, accident proneness clearly was an international phenomenon, now observed and reported in many areas of the Western world. Finally, awareness of accident proneness generated two types of reactions. The first reaction was to call attention to a class of individuals who caused more accidents than did others. Moreover, these individuals now

65

66

ch a p t er 3

had a name: accident prone, or susceptible, or predisposed. Careful industrial records could identify the accident repeaters, who could then be transferred or retrained—as opposed to the usual custom at that time of discharging such employees. The second reaction was to subject people to psychological screening (including physiological screening) in order to keep them out of sensitive positions, particularly in transport and dangerous occupations. Screening procedures, as both Farmer and Chambers noted, were as yet imperfect. 51 Nevertheless, investigators were already trying to identify exactly what accident proneness was, beyond just measuring it. For some, then, accident proneness was to be discovered after the fact, but before more facts (accidents) occurred. The defi nition of accident proneness, as these early writers pointed out, could appear circular, for it consisted of past acts of having accidents. For other writers of that time, the hope was to identify accident prone people ahead of time and prevent their coming into hazardous situations. These two distinct approaches were of course not incompatible, but they did complicate the question of what people at the opening of the 1930s meant when they wrote or spoke about accident proneness. By the end of the 1920s, the campaign of Farmer and Chambers and their colleagues to spread knowledge about accident proneness, including use of the phrase, had made substantial progress in Britain and had carried over at least to the United States. Certainly other writers from a variety of backgrounds got the point that “a small group of workers are always found monopolizing an undue proportion of the accidents.” At first, managers and those concerned with management were most sensitive. “The hope of the future,” wrote one expert concerned with industry, “appears to lie in the possibility of diagnosing proneness to accident and excluding susceptible individuals from those industries in which they would be a menace to themselves and their fellow-workers.” 52 At least among Anglophone thinkers, and especially psychologists, accident proneness was a concept that was becoming familiar.

4.

PREPARING THE WAY: TRANSPORT OPERATORS

At the end of the 1920s and the beginning of the 1930s, publications on and around the subject of accident proneness were producing a standard idea that would last until after World War ii. It was not just the British and German psychological streams that came together, however. Another, parallel stream became integrated into the idea and strongly reinforced it: urgent programs already established to ensure the safe operation of mass transit and other transport, the beginnings of which I have already referred to. The existence of this interest in safe operators of mass transit constitutes evidence of a notable readiness for the idea of accident proneness. If a Zeitgeist produced the idea of accident proneness, the transportation stream provides remarkably tangible evidence that that a Zeitgeist or cultural equivalent existed. The concern of managers of mass transit systems antedated the formal conceptualizations of psychologists. Moreover, the interest came from managers who had to deal every day with practical issues of injury and damage from the use of expensive machinery. These managers in fact were in the process of, implicitly, identifying some human factor or factors that were not immediately obvious before an accident occurred.1

g ood a nd ba d employ ees As mechanization and industrialization came into Europeanized countries, managers of transportation systems did notice that some employees 67

68

ch a p t er 4

figure 4:1. As early as 1920, motion pictures were used in a simulation apparatus in Dresden to test reaction time and other practical attributes in employees driving trains. Source: [A.] Schreiber, “Das Prüflaboratorium für Berufseignung bei der Eisenbahn-Generaldirektion Dresden,” Praktische Psychologie, 2 (1921), 232.

were superior to others. What did it mean to be a good vehicle operator? One measure of superiority was those who had fewer accidents than others (figure 4:1). But there were other criteria, such as smooth and efficient starting and stopping, or courteous and thoughtful treatment of passengers, and of course following rules, dressing neatly, and taking instructions.2 In the early twentieth century, in two countries, especially, France and the United States, managers who were trying to control the “efficiency” of their employees—including accident damage—sometimes turned to psychologists to see if the occupational aptitude tests that the scientists were devising could be adapted to identifying job applicants who might incur accidents. This concern arose in the framework of managers’ attempts to apply psychology in order to select suitable, productive employees: “the right man for the right job,” or “employee selection.” 3 Although some of these selection efforts began early, the responsible managers and their consultants continued to think in terms of grading employees or employees’ fitness for the job. Unlike Marbe and his English counterparts, pioneers in transit accidents did not conceptualize a trait of personal accident proneness as such. In 1908, the Compagnie des Tramways de l’Est Parisien in France asked a psychologist, J.-M. Lahy, head of the experimental psychology laboratory of l’École pratique des Hautes Études, to see if he could identify or devise tests for their drivers. Lahy had been working on physiological

Preparing the Way: Transport Operators

measurement, such as tests involving blood pressure. In the new work, Lahy sought tests to identify applicants who could economize in the use of electricity as they used the operating controls in tram cars. He first published his results in 1913.4 Economizing was Lahy’s idea, and that of his employers, of efficiency. But already they showed that, like many others at the time, they were making discriminations between good tram operators and those who were not “superior.” This traditional distinction between satisfactory and unsatisfactory employees was the first step in conceptualizing how employees might be assorted into groups, groups that might be detected by psychological means even before they appeared in statistics taken from performance records. 5

m ü nst er berg In 1913, the very same year that he first published, as he later recalled, Lahy found a report of other tests in a pioneering book on applied psychology just published by the German-American psychologist then at Harvard University, Hugo Münsterberg. Münsterberg gave as an example of applied psychology his devising tests not only for operators of electric railways but for officers on ships. He had been approached by a shipping line for help in selecting suitable ship captains who specifically would not cause accidents or damage. In describing work on streetcars, Münsterberg’s own account summarizes what was happening everywhere: Last winter the American Association for Labor Legislation called a meeting of vocational specialists to discuss the problem of . . . accidents under various aspects. The street railways of various cities were represented, and economic, physiological, and psychological specialists took part in the general discussion. Much attention was given, of course, to the questions of fatigue and to the statistical results as to the number of accidents and their relation to the various hours of the day and to the time of labor. But there was a strong tendency to recognize, as still more important than the mere fatigue, the whole mental constitution of the motormen . . . The companies claimed that there are motormen who practically never have an accident . . . while others relatively often experience accidents of all kinds . . . They can hardly be blamed, as they were not careless, and yet the accidents did result from their personal qualities; they simply lacked the gift of instinctive foresight [to avoid an accident] . . . the Association suggested that I undertake an inquiry into this interesting problem. 6

69

70

ch a p t er 4

Indeed, at the meeting, Martin J. Insull, general manager of the Louisville & Northern Railway, spoke about how his road was increasingly concerned about accidents and how he had reached a limit in reducing accidents by education and persuasion and by current means of employee selection. “I found,” he reported, “that high marks in an examination on rules did not necessarily indicate a high degree of safety of operation in the motorman or conductor examined . . . We had entirely overlooked that very important factor, the mentality of the men on both ends of the car.” And he cited the experience of other rail executives who had come to the same conclusion.7 It was under these circumstances that Münsterberg thereupon tried to locate or devise tests that would differentiate response characteristics of different individuals who might be employed by mass transit organizations. He started with two groups, the “reliable” and the “unreliable” motormen upon whom to test his apparatus. The good and the bad were defined by how many accidents they incurred. Much of Münsterberg’s effort went into the analysis of factors involved in actual operation of the streetcars (and, in his other project, of ships) and how those factors could be measured in a laboratory situation. 8 Münsterberg did not do much work further on the problem of identifying safe transit operators, but it was taken up here and there by others in the United States and in Europe. An American observer believed that testing for streetcar drivers arose in Germany in several centers “almost simultaneously,” and avoiding accidents soon became a central concern of transit operators.9 A Dallas street railway official reported in 1916 that his company in 1914 had instituted strict screening procedures for motormen. After eliminating about 80% of the candidates for more or less obvious reasons (slowness, inability to read instructions, physical disqualifications), the applicants took psychological tests for attention, observation, and judgment, all focused on accident prevention. The last two tests the officials had modified from tests suggested by Münsterberg. The results of the new hiring procedures led to a dramatic 50% decline in turnover between 1912 and 1915. Few motormen who survived the testing had to be discharged for unsatisfactory performance (which officials conceived of as “incompetency and recklessness”).10

ps ychologic a l scr eening in f r a nce a nd ger m a n y Meanwhile, Lahy had by his work commanded the respect of transit executives in the Paris area, and in 1921, after World War i, he became founding

Preparing the Way: Transport Operators

director of a psychological laboratory set up by the Société des Transports en Commun de la Région Parisienne, a system that included a staggering number of employees, trains, streetcars, and buses. Lahy and his staff instituted a series of tests for candidates for tram and bus drivers that was widely admired and copied. The staff now tested applicants for speed of reaction and for attentiveness as well as the ability to use the controls in an economical way. While the idea of accident proneness was not explicit, already both managers and investigators were, like Münsterberg, using accident records as one sign of a good employee or a bad or mediocre employee. Thus classification of employees by groups, and knowledge that accident records could be characteristic of individuals, set the stage for these French investigators to think in terms of accident prone employees. Moreover, Lahy was well aware of the potential to use selection tests and classifications in other types of mechanized industry.11 By the early 1920s, mass transit administrators elsewhere in the Europeanized world, particularly in Germany, were introducing tests with which they screened candidates for positions or marked for attention individuals already employed. Very often the officials collaborated with or utilized psychologists, and together they produced many publications reporting tests and testing apparatus.12 The general pattern was that initiatives to use tests began with street railways and then spread to railways and motor buses generally. In 1917, the Dresden divisional president of the German State Railways set up psychophysical testing for employees, with extensive and elaborate apparatus and grading procedures.13 Czechoslovakia was using tests for railway engineers by 1923.14 Beyond the work in Paris, probably the most important and most widely publicized psychological testing for rail employees occurred in Germany, such as that to which I have already referred. Tests were at first for reaction time and alertness, and then for ability to estimate distance and speed and for quick decision-making. At some point, the obvious tests for color blindness and night blindness, in addition to optical and auditory acuity, were added. Finally, special talents were tested, such as physical strength for a switchman to throw a switch and motor and manipulative abilities for various tasks (figure 4:2). Indeed, by 1920–1921, a substantial German-language literature on the subject was appearing, albeit the testing was usually framed in terms of general vocational testing, rather than specifically safe operation.15 But accidents became more and more conspicuous in transit executives’ discussions. In Berlin, K. A. Tramm, the engineer-manager of the Berlin Street Railway, devised a number of psychophysical tests for motor-

71

72

ch a p t er 4

figure 4:2. This portable device was used to test the body strength of candidates for a position with German railways as switchman. Source: C. Heydt, “Eignungsprüfungen für den Rangierdienst,” Industrielle Psychotechnik, 1 (1924), 144, reproduced with the kind permission of Springer Science and Business Media.

men, designed especially to test applicants’ reactions to dangerous situations. Although Tramm did not have a professional identity as a psychologist, in designing the tests, he was duplicating the role that psychologists were playing in the United States and France. Tramm’s tests were very practical, including speed of reaction and judgment. These tests were, moreover, particularly designed to screen out individuals who might have accidents. Tramm was obviously inspired by vocational testing for factory workers, but that testing did not yet include the factor of accidents in the way that Tramm included it because of his special interest. Tramm reported his testing apparatus as early as 1915, and it was widely admired and copied. Moreover, Tramm continued to extend and publicize his work, which he came himself to view as a contribution to psychotechnology.16 During World War i, for both the armed services and war industry, psychotechnologists offered a variety of selection tests. Authorities (as noted already) were particularly anxious to find airplane pilots who were not going to crack up the planes. Psychotechnological vocational testing therefore developed a substantial momentum that set the stage for further development after the war.17 As an observer wrote in 1922, “There is a growing tendency in Germany at present to increase the efficiency of workers in all walks of life by submitting every candidate for a profession or calling to elaborate psychological tests. This movement is especially

Preparing the Way: Transport Operators

marked in the case of lines of work such as airplane pilot, locomotive engineer, automobile driver and street-car motorman where property and human lives are jeopardized by unskillful operators.” 18 No one at that time seemed to notice anything special about the pervasive assumption that there were “unskillful operators.” Instead, experts focused on the issues of safety and the resulting economy if managers kept “unskilled” people out of positions in which they could cause accidents. Writing in 1925, a British psychologist noted that “it is now recognized that physical, and more particularly mental, fitness is absolutely essential in all who have control of rapidly moving vehicles”—by which he meant particularly employees of large transit and transport organizations. He recalled the World War i experience, when “scientific selection” eliminated candidates “unsuitable” to pilot airplanes and reduced the “human error” factor in crashes from 80% to 20%. But for him the model program was that of Lahy in dealing with mass transit in the Paris area. “To-day the scientific selection of personnel has become as much a necessity . . . as the selection of suitable material for rolling stock (figure 4:3).” 19

figure 4:3. A streetcar was climbing a steep hill when a loose manhole cover disabled the brakes. In an instance of human failure, the motorman neglected to institute an emergency procedure using the power of the motor, and the car rolled wildly backward and off the tracks. Source: “A Street Car Accident,” Insurance Engineering, 24 (1912), 293.

73

74

ch a p t er 4

ex pa nsion of scr eening for acciden ts The negative task of eliminating unfit or dangerous transit and transport operators therefore became a major project of managers in advanced economies. It was only a matter of time before concerned experts would notice the work of Farmer and Chambers and Marbe and apply their formulations to people who were associated with accidents. But there was some hesitation before the psychologists and the practical transportation specialists brought their efforts together with the British and German accident proneness streams. Just as Moede, the German psychotechnologist, in 1926 had referred casually to “der zu Unfällen disponierte Mann” [the person disposed to accidents], so in a very similar way, Lahy in 1927 referred to “les non accidenteurs” without any explicit sign that he had heard of Marbe’s just-published categorization of Nichtunfäller. The “non accidenteurs” were just another class like those whom he called “economizers.” 20 We know, however, that German specialists in transportation personnel had been exposed to Marbe’s thinking, even if not all of his terminology, as early as June 1924, when the psychotechnologists concerned with German rail transport met in Würzburg, Marbe’s home base. Marbe in fact spoke to the group, and he introduced the idea of “ ‘Unfäller’, d. h. der besonders zu Unfällen neigenden Personen” (“persons specially inclined toward [having] accidents”). Moede was present, along with many railway officials.21 The process by which they switched to thinking in terms of Marbe’s formulation was gradual, over a number of years. With some, however, the adoption of the idea of accident proneness may have been more rapid, since they had already started to conceptualize a class of people whom they associated with accidents.22 Such appears to have been the case with Schmitt, for example, whose sudden conversion to Marbe’s theory at the 1924 meeting was noted above. During the 1920s, testing employees and employee candidates continued to spread widely among transit and rail systems on the Continent, and accidents became ever more the focus as traffic and speed on the lines increased.23 Although the British did not follow the Continental example, in the United States, programs that were focused on accidents grew out of the same problems as those in Europe.24 As early as 1920, Morris S. Viteles, whose later articles provided key publicity for accident proneness, became consulting psychologist to the Milwaukee Electric Railway and Light Company. He was working on vocational selection of employees for the auditing department, the subject of

Preparing the Way: Transport Operators

his doctoral dissertation at the University of Pennsylvania.25 He was also, he wrote later, trying to devise tests to select motormen, beginning in the summer of 1920. At that time the author had at his disposal only the description of Münsterberg’s experiment and an account of the work done by Gerhardt with the Münsterberg and other tests. Descriptions of the work of foreign investigators were not then available. It appears now that the apparatus devised by the author . . . is similar, in some respects, to that used by German investigators. This similarity is altogether a fortuitous one. It is the result of an accord on the part of independent investigators on the analysis of the abilities underlying the safe operation of a street car.26

Company officials in Milwaukee had earlier tried to use Münsterberg’s tests, but they did not work. Viteles made a different analysis of the requirements for safe operation of a streetcar, and he based his tests on his own analysis—again similar to that reported later from the Continent.27 Viteles in 1922 left to spend a year abroad, but others continued the testing in Milwaukee and developed it further.28 Just at that time, too, in 1922, the main light rail trade journal in the United States, the Electric Railway Journal, carried a number of articles describing tests and testing programs for streetcar operators. One reported work parallel to that in Milwaukee but carried out in San Francisco by a “practical railroading” man.29 In every case reported then and later, those in charge rated the operators and grouped them so as to separate the satisfactory from those who were not satisfactory. The best publicized case—except for the earlier work of Tramm in Berlin—came from Boston, where “a railway company controlling transportation by subway, street car, and bus of a million passengers per day discovered on investigation among its operators a certain proportion who were particularly liable to have accidents.” Indeed, it turned out that 27% of the operators caused 55% of the accidents. The general manager, Edward Dana, as early as 1927 was aware of various earlier attempts to use psychological tests, including obviously the work of Münsterberg with his own organization, and Dana decided to try to identify the factors that made for accidents and then to screen any new employees and to remediate older employees. The psychologists working on the project knew at least the English work on differential accident rates. Their fi rst printed reports in 1927 included the terms “proneness to accidents” and

75

76

ch a p t er 4

“individual accident proneness.” These investigators utilized standard tests of visual functioning, reaction times under different conditions, fatigue, attention, and movement sensitivity. Of eleven tests, seven proved significant, and a very low score suggested that an accident prone person had been identified. Reports of the subsequent reduction of accidents by 26.6 per cent gained attention, as did the news that, as in Europe, motormen who were more efficient and used less power also were remarkably accident free. 30

using acciden t proneness ex pl ici t ly Two of the investigators, W. V. Bingham and C. S. Slocombe, continued to write and speak about accident proneness for years afterward, often using data from the Boston study. Bingham, a very prominent psychologist, in particular publicized the Boston Elevated Railway work. Moreover, he understood the implications. He noted repeatedly that the great improvements obtained in Boston, “so important for the company, the men and the public alike, have been secured without recourse to discharge or transfer or scientific selection.” 31 In 1930, the Cleveland Railway Company issued a report explicitly titled “The Accident-Prone Employee.” The Accident Prevention staff by 1928 found that their educational efforts had not entirely succeeded in reducing accidents, and they studied one particular station area with many accidents. There they discovered that a relatively small group of motormen were involved in a large percentage of all accidents reported, although operating under practically identical conditions as the other larger group of trainmen whose accident records were, in most cases, excellent. This finding apparently confi rmed statements of psychologists and psychiatrists that accidents do not distribute themselves by chance but that they happen frequently to some men and infrequently to others . . . those possessing the necessary physical and mental requirements show little susceptibility to accident. There is an increasing tendency, therefore, to consider the occurrence of accidents as an individual problem rather than as a situation to be controlled entirely by means of educational work among a group at large. 32

The investigators clearly knew about and used the concept of accident proneness. Moreover, they came to believe that each case was exquisitely

Preparing the Way: Transport Operators

individual: “no two cases were found to be similar.” Some accident prone motormen had “characteristics” such as bad judgment and impulsiveness. Others had “habits,” such as starting and stopping abruptly. The company officials did intensive retraining and individual counseling so that they did not have to discharge most of the accident prone employees. For company officials, the idea of accident proneness was useful and represented an observable reality. 33

e x t ension to t r a f f ic acciden ts gener a l ly All of this work on mass transit carried over easily to another area that was growing very rapidly in the early twentieth century: traffic accidents. 34 Trains and streetcars in fact often collided with automobiles and trucks. The actions of drivers of cars and trucks became an acute social problem throughout the world. By the 1920s, just as the idea of accident proneness was developing, automobile traffic in itself was becoming a major social issue. In Berlin in just two years in the 1920s, the number of motorcar accidents more than doubled. For the year 1923, the British could tally 2,979 fatal road and street accidents and 86,122 nonfatal injury accidents. In the United States, 12,557 traffic fatalities were counted already in 1920, but 34,400 in 1931 (with a million people injured). 35 At the opening of the twentieth century, public officials were concerned with chauffeurs and truck drivers, who were employees, comparable to streetcar operators. But each year, more and more private vehicle operators also posed substantial hazards. Private automobiles were at first basically playthings of the rich. More ordinary folk whose lives and property were placed in danger resented misbehavior of the wealthy auto owners and sought to curb recklessness and rudeness by means of laws and civil damages suits (figure 4:4). 36 Some operators of road vehicles were clearly deviant in the ways in which they drove. Employers could discharge obviously bad or incompetent drivers—as was the case with bus drivers in Paris and operators of fleets of motor trucks. But individual owners posed another kind of problem. Laws against speeding and recklessness sought to control automobilists’ behavior, that is, to use education and discipline where the behavior was willful. But what about clearly incompetent drivers? In various constituencies, licensing of drivers developed. A person had to demonstrate some level of competence to take a vehicle on the public roads. By 1900, French licensing was the model. Licensing of drivers came very slowly, however, to many parts of the world. As late as 1938, five of the forty-eight

77

78

ch a p t er 4

figure 4:4. In this choice photo from 1910, a sportsman’s private car blew a rear tire while racing. The blowout kicked up dust and dirt while throwing the car into a dangerous skid. At this time, tire blowouts were a major cause of accidents, injuries, and damage. Source: Fred J. Wagner, “Preventing Accidents,” House Beautiful, July 1912, suppl., xxvi.

U.S. states still did not license drivers. And the problem of certifying competence became ever more acute as many different kinds of people joined the rich sportsmen whose shortcomings endangered the public. Chemnitz, Germany, for example, at the end of the 1890s found two groups who should not be allowed to drive: persons under fifteen years of age and those ignorant of how to operate a motor vehicle. In 1904 in Milwaukee, the only requirements for a license to drive a car were to be eighteen years of age and to have the use of two arms. 37 In Detroit, a court set up a “Psychopathic Clinic” that as early as 1921 began to examine and deal with traffic offenders, on the presumption that frequent offenders were not “competent or safe drivers.” The judge was quoted as saying, “I believe there is a mental weakness in the driver who speeds recklessly through crowded traffic, not caring how many he may kill or injure.” Both mental and physical defects among the offenders suggested that the presumption that certain people represented a deviance such as mental illness that was not wholly voluntary was not out of line. This independent pioneering effort came of course too early for any reference to a syndrome of accident proneness as such, but as a practical measure prepared the way for the idea. 38 Most requirements to drive represented common sense, and reckless operation was clearly deviant (figure 4:5). Yet the observation that some people consistently drove as if crazy seemed to suggest that some mental

Preparing the Way: Transport Operators

or psychophysiological factor might be operating—not least, for example, defective vision. Many of the same people who were concerned about accidents in general, and many who had identified accident proneness in transit systems, easily carried the idea of accident proneness over to drivers on the road. The bad employee had a counterpart, the bad driver. Psychologists began studying why some drivers were better than others, just as tram drivers had been studied. 39 By 1930, a number of writers, using the idea of accident prone drivers, were suggesting that there were drivers who should not be permitted on the road. Thus was the transfer from streetcar to bus to truck to automobile an easy one, and it was under way just as the formulation of the idea of accident proneness was coming in. In the United States, one psychologist, working with the process of selecting taxicab drivers, actually arrived at some formulations similar to those of Marbe and Farmer—but in 1925, a year before Marbe and Farmer published definitively on accident proneness/Unfallneigung, and he did not cite even the early British work. The psychologist, Adolph J. Snow, with a

figure 4:5. Near the beginning of the 1920s, the Baltimore Safety Council placed signs at important street intersections. “This method has brought excellent results.” This fi rst sign emphasized the responsibility of individual motor vehicle drivers. Other signs emphasized the responsibility of pedestrians to behave safely. Source: “Local Councils Advertise Public Safety,” National Safety News, September 1923, 21.

79

80

ch a p t er 4

doctorate from Columbia University, was working in applied business psychology. In trying to find psychological tests appropriate for taxi drivers, he studied their work and immediately focused on the high desirability of drivers who would not have accidents. He tied his research directly to a general category, “transportation pilots”—whether “a passenger or freight motor vehicle, a street car, a subway . . . a steam or electric railway.” Eventually Snow and his staff tested 12,000 taxi drivers or driver job applicants.40 Like Greenwood and Marbe, Snow found that some drivers had many accidents, and others few or none. He analyzed the distribution of accidents just as others did later who used the term and concept of accident proneness. He found that in one year at the Yellow Cab Company of Chicago, “46% of the accidents were caused by 18% of the drivers.” Subsequent selection processes led to a dramatic decrease in accidents.41 Snow’s conceptualization of drivers who had accidents was different from that of those who said that accidents predicted accidents. Instead, he thought in terms of what tests could show, and they showed that people who were “mentally subnormal” or totally “mentally unfit” were those who had accidents. He cited a Massachusetts law that could deny driving privileges to “mentally defective” persons. Snow, however, was going beyond intelligence to tests for characteristics such as carelessness and emotional stability. Performance on such tests could indicate serious shortcomings in a driver, Snow concluded. He wrote in terms of a person’s total mental inferiority, and he generalized his findings: “If we permit mentally incompetent individuals to drive automobiles, nothing under the sun can prevent them from causing accidents.” 42 Not until well into his original work, in 1926, did Snow indicate that he might have found out about other workers in the field. Unlike Snow, some of them eventually came to use the idea of accident proneness.43 The most notable was Viteles, who after pioneering testing in Milwaukee, became a publicist of the concept of accident proneness. At that point, Viteles was one of those who specifically made the transition from streetcar motormen to automobile operators. By 1929, he was reporting on accident records of women taxicab drivers.44 Bingham, who also started out with mass transit, in Boston, soon was publishing persistently on accident prone drivers, and in 1929–1930, he was citing the work of the British investigators who “early proved that accidents do not just happen; they are not distributed according to the laws of chance,” as he put it in a Science Service radio talk and article. By 1932, he was writing specifically about “the accident-prone driver,”

Preparing the Way: Transport Operators

continuing to publicize the findings of those who had worked with mass transit and extending that work to ordinary private motorcar drivers. “A few drivers—relatively a very few, I believe—are born into the accidentprone class,” he wrote. “They are temperamentally too excitable, tending to get flustered and to do exactly the wrong thing in an emergency.” And he proposed various measures to get them off the highways. 45

t r a f f ic acciden t concer n spr e a ds t he ide a The work of another American psychologist, Donald Laird, illustrates clearly the way in which transportation employee selection facilitated the spread of the idea of accident proneness in traffic. He published a journal, Industrial Psychology Monthly, beginning in 1926, and the very first issue carried a report from Snow of his work—but that, as I have noted, did not embody the concept of accident proneness. Laird also carried in his journal brief notices of other transport worker selection programs in Europe and the United States, and he advocated testing of private automobile drivers. By 1927, Laird had discovered Marbe’s monograph and described “individuals susceptible to accidents.” Marbe’s investigations, wrote Laird, “indicate that there are individual differences in liability to accidents, and that the degree of liability to accidents tends to be constant.” 46 In 1928, Laird included a report of the work of the British Industrial Fatigue Research Board, and accident proneness—using that term—appeared conspicuously in that article and elsewhere in the journal.47 It was clear, however, that Laird and others who were working with transit operators and automobile drivers already had known that a small group of people consistently caused a disproportionate number of accidents. These psychologists and managers could therefore adopt the term and idea of accident proneness easily and naturally. Of course some psychologists without previous experience with mass transit operators or motor fleets also turned their attention to the slaughter on the roadways. The most notable American was Alvhh R. Lauer, who in 1928 began work on traffic accidents and in subsequent years made a career of testing the competence of drivers.48 As early as 1925, David Wechsler of the Psychological Corporation, in work strikingly parallel to that of Snow in Chicago, carried out tests for a Pittsburgh taxicab company to eliminate “unsafe and incompetent” drivers. Wechsler in his 1926 report (composed before the appearance of the critical publications of Marbe and Farmer and Chambers), was already able to list four other American projects testing drivers.49

81

82

ch a p t er 4

The work on drivers and mass transit operators in the United States therefore paralleled work in Paris and in the German-speaking countries in the 1920s. And in the details of that American work appeared traces of a psychological-medical model. In Milwaukee, after several years of investigation and screening, the company organized a regular “accident clinic” to deal with “individual motormen who have had rather heavy accident records.” Among the group, clinic operators could show an 81% reduction in accidents. 50 For some years in the 1920s, then, research on transport safety proceeded without naming accident proneness. Once the term was in the literature, however, managers and investigators more and more frequently made the connection. Indeed, it is hard to know how the label might come in. In 1929–1931, the Dayton Power and Light Company, using company drivers’ “accident curves” that showed “repetition of accidents,” was able to reduce dramatically the accident rate for company vehicles. The managers worked “on the theory that a rather limited group of drivers might be responsible for a large percentage of the traffic accidents” and employed psychological equipment from Ohio State University. The 1931 writeup of this experience in the National Safety News contained in the headline “Accident-Prone” but in the text only one time did the phrase “prone-to-accident drivers” or the equivalent occur. To what extent it was the editors of the journal who introduced the term, one cannot say, but the writeup suggests that those who carried out the work did not habitually utilize the label, even if they were familiar with the idea, so well publicized among experts by 1929. 51 At least part of the Dayton study was initiated by Lauer, who was working with the preeminent psychologist, A. P. Weiss. These authors, who during 1927–1929 were supposed to be investigating the role of vision in automobile accidents, in fact reported on much broader considerations about how to test for “good” drivers. Weiss and Lauer wrote of accident prone drivers and of non-accident drivers and attempted to find tests to detect which drivers were members of which group. 52 Tests for drivers therefore appeared in both Europe and the United States with remarkable rapidity during the 1920s. These tests, as the various reports suggest, grew in part out of the mental testing movement as psychologists were attempting to make psychology socially (and commercially) useful (figure 4:6). The presumption always was that obviously bad drivers, both public and private, should be identified before the accident records showed that they were bad drivers—that same record that would constitute the basis for a label of accident prone. 53

Preparing the Way: Transport Operators

figure 4:6. Rather than tediously testing candidates in actual operation or simulation devices, psychologists in Amsterdam in the early 1920s attempted to abstract qualities needed to drive a car and devised laboratory tests for chauffeurs. In this one, the candidate held a metal pointer in each hand and tried rapidly and carefully to negotiate a way between the unpredictably charged wire needles without incurring a very unpleasant shock or reacting to the shock when it did occur. Source: J. M. Van Went, “Die städtische psychotechnische Prüfstelle Amsterdam,” Industrielle Psychotechnik, 3 (1926), 305, reproduced with the kind permission of Springer Science and Business Media.

Within the next few years, it became obvious that people who were concerned about traffic accidents assumed the idea of accident proneness whether or not they called it that. One expert a few years later, for example, referred to a “repeater control plan” to diminish auto accidents. 54 Such practical quests for safe operation of transportation technologies represented a material embodiment of the constellation of outlooks and thinking that produced the simultaneous and independent formulation of an idea that there are accident prone people. Particularly with motor vehicle accidents, another element in the Europeanized societies showed how circumstances prepared people to take up the idea that accident prone people existed: these were the casualty insurers who came to offer automobile and accident insurance. Just before the turn of the twentieth century, the first automobile insurance policies appeared on both sides of the Atlantic. At first, the policies varied according to what kind of car was being insured. By the World War i period, this basis was clearly breaking down. One of the suggested bases for rating insurance, for example, was the neighborhood in which the car was driven, because, in the minds of many people at that time, some neighborhoods had populations who had the trait of persisting in getting run over by vehicles!55 Also in the 1920s, insurance companies launched a movement for general vehicle inspection in order to cut down on accident losses

83

84

ch a p t er 4

from mechanical failure. But in the United States, it was 1936 before the companies developed the first standard policy for both automobiles and drivers. 56 Meanwhile, the insurance companies began to offer group accident policies, without regard to the site or means of accident. This insurance covered either employees in a given plant, worker compensation schemes (in the Anglo-American world), or, as typically on the Continent, state worker insurance. In the very early twentieth century, some individual accident policies were sold. Insurance experts argued continually for decades over which was better: prospective estimates of group ratings (as was typical in some property casualty insurance, where a building could be judged beforehand, for example for fire danger) or retrospective corrections after the company had had experience with the groups or the individuals. 57 The truth was, however, that insurance companies had a very longstanding practice of canceling policies that incurred losses above what the company officials thought was the average. Therefore, a dangerous factory with many claims or a group or individuals who suffered too many injuries would fi nd their policies canceled and the balance of their premium payments returned. As one officer commented about individual policies in 1920, “As for the use of the cancellation clause, we all understand about it . . . You take a man who has paid twelve years under a given contract, paid a $60 premium—$720 in premiums altogether—and he puts in a claim for four weeks [missed work], gets a $100 check back and a cancellation notice without explanation; he doesn’t like it.” 58 For the relatively few individual accident insurance policies in the opening decades of the twentieth century, then, there was an automatic tagging of accident repeaters—without giving them a categorical designation. As the authors of a 1929 insurance textbook commented, “If a risk [insured person] becomes impaired by disease or develops an unusual accident frequency, the company must take steps to have such a risk removed from their books or, at least, modify the coverage . . . risks of an impaired character or unprofitable history . . . must be guarded against if the [basic insurance] law of great numbers is to produce expected results.” 59 Likewise another expert, writing in 1930, noted that “The unsafe automobile driver; the employer who takes no interest in safety; the careless person who . . . is notoriously lax in the protection of his jewelery; these and many others are ‘bad risks’ for the insurance company because they prevent the proper functioning of averages, and introduce greater probability of loss into the insurance transaction.” 60

Preparing the Way: Transport Operators

Insurers’ retrospective identification from accident records of “bad risks” and “unsafe” drivers was only a first, pragmatic step. It did not constitute a concept of an accident prone person such as Marbe and the British extracted from similar retrospective numerical records.61 Instead, in the area of vehicular safety, withdrawing insurance came to be understood as enforcement of driving safely. In this way, in the 1930s American comprehensive safety slogan of “education, enforcement, and engineering,” disciplining individual drivers came in as a parallel with the traditional idea of disciplining workers in the workplace. 62

85

5.

THE STREAMS COME TOGETHER IN THE LATE 1920S AND EARLY 1930S

As events described in previous chapters show, by 1930, material and organizational structures had given rise to thinking that was well prepared for some version of the idea of accident proneness.1 In Germany and Britain, psychologists had conceptualized and named the idea. They had situated it as a personality trait and were moving to apply it to industry by means of psychological tests. Meanwhile, employers in mass transit, who were already aware that some employees incurred more accidents than others, implicitly and explicitly embraced the idea and helped extend it to the growing problem of traffic accidents. It was as all of these streams came together that the idea of accident proneness developed a standard profi le. The safety staff of the Cleveland Railway Company in North America made a list of factors that they now classified as “primary causes of accident-proneness” in their rapid transit motormen (figure 5:1). The list shows how it had become possible at the beginning of the 1930s to reconceptualize many accident causes as elements in a general category, accident proneness. Some causes were indeed familiar personality traits such as “faulty attitude,” “impulsiveness,” and “irresponsibility”—not to mention “intemperance.” Others were strictly psychophysical, such as “defective vision,” “fatigability,” and high blood pressure. Other factors were more difficult to classify such as judgment, nervousness, and attention. The different causes were familiar, but they all added up to statistics that 86

The Streams Come Together, Late 1920s–Early 1930s

figure 5:1. “Primary Causes of Accident-Proneness, Percentage Distribution Among Fifty Motormen, Woodhill Division, the Cleveland Railway Company.” Source: [Metropolitan Life Insurance Company], The Accident-Prone Employee (Cleveland: Cleveland Railway Company, 1930), 4, reproduced with the kind permission of MetLife.

showed these rapid transit managers that some operators consistently incurred more accidents than did others. The idea of accident proneness thus shifted many safety experts’ attention to specific individuals rather than abstract factors such as carelessness in the workforce that they were already identifying and addressing through education and discipline.

t he sa f e t y mov emen t During the 1920s, the safety movements described in the introduction became better organized and financed. Most important, safety leaders were increasingly sophisticated in their analyses. They were searching out causes of accidents. As a safety engineer wrote in 1928, “the reduction of accidents can only come about through knowledge of basic accident causes.” And by basic, he meant working out how employers could use not only physical guards and education but could see how prevention could work out in practice. To this end, he already was putting substantial emphasis on urging supervisors to identify employees who were

87

88

ch a p t er 5

mentally or physically unfit or imperfectly trained—before they suffered an accident.2 Safety experts were also insisting on having statistics about accidents on which to base their analyses. Since both German and British investigators had discovered accident proneness in statistical data, and the work of safety experts in the field of transportation was decidedly statistical, the demand for better statistics in the advancing safety movement thus in another way provided a nourishing environment in which ideas of accident proneness could grow. 3 Readiness for the idea of accident proneness therefore took a very specific form in the safety movement as the search for causes led investigators to individuals who in one way or another initiated accidents. In 1929, in a typical article, a spokesperson for the coal mining industry in the American South reported that their group had eliminated the word “accident” from their reports and instead were using the word “failure.” The fact was, he reported, that accidents, that is, wholly chance incidents, rarely occurred “at the mines or for that matter anywhere else, that accidents are nearly always caused by failure on somebody’s part.” 4 Such thinking in terms of causes of accidents demonstrated how the safety movement had matured. Safety advocates continued to sound the alarm about how many deaths and injuries and industrial costs resulted from “accidents.” But increasingly they were able to assume that everyone knew about the problem and now could focus on the causes of the accidents, which turned out to be mistakes. It was this “human factor” in causing accidents that particularly continued to constitute a growing concern—so much so that writers on safety could often ignore the small percentage of mechanical factors and concentrate on the human beings. The usual figure cited by writers was 25% or less, mechanical, and 75% or more, human. 5 Another sign of maturity was the growing emphasis that safety advocates placed on generating records of accidents—chiefly in industrial settings, of course, but in general to assemble facts about each accident. On the one hand was the demand from public health officials who were correctly alarmed by the number of people, including children, who were injured or killed each year. There were, for example, officials of an insurance company surprised by the number of accident claims growing out of injuries to people participating in sports. On the other hand was a variety of people in industry and management who believed that full records would help them uncover why accidents were happening. One of the major elements in the safety literature became articles suggesting how to

The Streams Come Together, Late 1920s–Early 1930s

gather detailed and accurate information about each accident—and in a format that could be put into standard statistical form that would permit data to be compared. Of particular importance, a number of writers were insisting that minor accidents as well as major accidents could be of great significance to managers and safety experts.6

per sona l fac tor s Accident proneness was of course originally a statistical phenomenon. But in further analysis and in searching for “human” causes of accidents, safety experts were utilizing a series of categories. These categories prepared the way for thinking first in terms of individuals and then in terms of individuals who were prone or inclined to have accidents. As a 1928 author, emphasizing the stereotypical American concern with individual differences, observed, “Manual dexterity, number checking, and the solution of a problem requiring mechanical analysis, show individual differences as striking as do physical characteristics.” 7 Psychophysical causes continued to rank high quantitatively in the technical literature as authors identified human factors in accidents. Some were the still obvious constitutional factors such as faulty eyesight or the more elusive factor of reaction time. Such factors could be very simple and straightforward. In 1927, British consultants reported that “a large amount of defective paper was being sent out from the fi nishing department of a certain mill. Eyesight tests were therefore given to all the workers, of whom 25 per cent. were found to need glasses. The glasses were provided, and no further reports were received about the quality of the paper.” 8 Motor coordination and ability to concentrate and pay attention were other factors that physicians or psychologists or both could measure, as was general intelligence. Naturally safety writers paid the most attention to factors, like sensory acuity, for which tests already existed. One investigator, for example, already in 1928 found among boys in a shop class in New Jersey that accidents incurred over a twenty-month period confirmed that youngsters with low scores on IQ tests had markedly elevated rates of accidents. He was able to demonstrate in statistical tables, therefore, the “relationship between intelligence and accident proneness of repeaters.” 9 In addition, there were all of the factors associated with fatigue and working conditions, such as work rhythms and lighting and temperature. And each person of course had a different set of physiological capacities, such as the ability to withstand fatigue of different kinds.10

89

90

ch a p t er 5

It is easy to see how this interest in individual psychophysical and physiological factors that caused accidents prepared the way for the concept of the accident prone individual. Two Italian psychologists, for example, in 1933 interpreted Marbe’s findings to mean that, based on their psychophysiological attributes, individuals varied in their abilities to adapt to their environments. So street accidents happened because people could not see, did not judge distances accurately, or did not react rapidly and appropriately.11 An American physician summed up one way of thinking in 1929: “Because there is no place for them in industry, the feeble-minded, the epileptic, the constitutionally frail or inferiors, or the sick (arteriosclerotics, alcoholics, cases of Bright’s disease, anemia, tuberculosis, cancer, skull fracture or brain concussion, or the senile) . . . had best be pensioned off or worked in segregated groups.” 12 In 1930, there occurred an incident showing in a transparent way how considering psychophysical factors prepared the way for thinking in terms of accident proneness. A London physiologist, G. P. Crowden, read a paper to the Royal Society of Medicine, reviewing research in Germany, Russia, the United States, and especially Britain that showed the effects of fatigue on industrial workers. Crowden included as fatigue factors nutrition, monotony, temperature and ventilation, general health and fitness, length of the working day, and training—or, in general, areas for “the application of medical science to the human problems of industry.” The paper stimulated Major Greenwood, who was present, to comment that his now considerable experience “leads me to put much more weight upon the psychological than the purely physiological elements.” Obviously Greenwood, with his background as discoverer and his continuing connection with the Industrial Health Board, was informed by the idea of accident proneness as he advocated psychological thinking, even if he did not find it necessary to mention accident proneness by name on that day.13

per sona l c auses of acciden ts Psychophysiological factors that were not inherent but were often temporary also set the stage for experts to think in terms of individual variations that might set off a series of accidents. One was the notorious influence of alcohol that might affect a person so as, for a time, to generate dangerous behavior. As I noted above, that category was like carelessness—something one could blame on the person’s willful actions—but it still brought a focus on the individual who had accidents.14

The Streams Come Together, Late 1920s–Early 1930s

Other temporary conditions went into the lists of involuntary factors that might cause one or more accidents. Writers mentioned disabilities from various infections, such as sight affected by tertiary syphilis in the brain. Or they could mention serious problems that followed from diabetes and other endocrine disturbances. Many departures from good health included disease symptoms that affected a person’s senses, consciousness, and judgment, or rendered one particularly subject to fatigue. “A tooth ache may make a person highly nervous, in others it produces bad temper and surliness, in others carelessness and indifference or sometimes a combination of qualities and actions which are the reverse of safe when working.” 15 And of course looking closely at causes of accidents also turned up temporary mental states, similar to psychophysiological factors that rendered someone unsafe. Wrote one safety expert, “There are all around us victims of great mental agitation . . . These people are not, strictly speaking, in their normal mental condition, their mind is absorbed to such an extent that the every day occurrences do not register with them . . . [and they are] not capable of exercising the alertness and interest and observing carefulness as they would if their mind was not so possessed with one subject.” 16 As the Cleveland Railway Company chart reproduced earlier in this chapter shows, investigation of accidents turned up, beyond ignorance and breaking rules, people who acted without proper attention and foresight, perhaps with haste or some other mental pressure that was not usual. It was but a short step to identifying as accident prone people whose mental condition was not just a passing state.17

cl a ssif y ing peopl e in t er ms of acciden ts Still one more type of thinking that prepared for the idea of accident proneness was the tendency of various kinds of experts to group not only causes and types of accidents but people who had accidents. Of course one way of identifying a cause was to categorize the human beings who were the agents who set off accidents. Already such classification could be done by character: identifying those who were careless. And there were other commonplace ways in which safety experts classified people into groups that appeared “natural”—just as those who were accident prone, as disclosed by statistics, came to appear to be a natural group. Earlier investigators had already identified some classifications of people at hazard to have accidents. People with epilepsy, as has been noted already, posed a great danger in the minds of many experts—part of the

91

92

ch a p t er 5

usual discrimination against that particular group.18 There were people of various ages, particularly the elderly, who often needed protection from accidental harm. And of course everyone knew about children and about reckless young men. By the opening of the 1930s, each was a well identified category, members of which, in the opinions of investigators and safety experts, could require special treatment.19 One particular grouping that appeared in classifications of almost all kinds was women. In this case, categorization often passed over into stereotyping, In the case of whether or not women suffered more accidents than men, the evidence mustered on either side implicitly addressed the continuing question of the extent to which women should have some special treatment in societies in which men set the standard. Nevertheless, around 1930, the statistics on women’s accidents were still incomplete and inconclusive, and so femaleness (and implicitly maleness) was not a very useful category.20

a bsor bing a nd spr e a ding t he ide a By the early 1930s, then, personal accident proneness was an idea that could be easily assimilated into the thinking of Europeanized people. And in fact it was assimilated. Much additional evidence suggests how rapidly many people, beyond those concerned with transportation accidents, understood and used the idea. Certainly the assumption was commonplace that some automobile drivers were consistently dangerous drivers—whether or not the term “accident prone” was used (figure 5:2). Asked one 1930 writer, “Why do some drivers have ‘bad luck? ’ If some persons are constitutionally unfitted to drive,” he continued, “how can they be singled out and barred from the road? ” Yet he did not use the term accident prone or refer to that literature.21 One example of the transition from preparation to recognition and acceptance was the work of a safety engineer with a US accident insurance company, H. W. Heinrich. In 1929, he urged employers to pay attention to “non-injury” accidents because they indicated settings in which the potential for serious accidents existed. After all, he pointed out, for every losttime accident, there were twenty-nine incidents, from which one could learn. “When a person stumbles, falls, and sustains an injury, is it safe to say that he never fell before? ” And Heinrich went on to say that “in making a survey of one hundred typical manufacturing plants, we found that in the majority of them the causes of the serious injuries . . . did not fairly picture the unsafe conditions needing first attention.” None of this expo-

The Streams Come Together, Late 1920s–Early 1930s

figure 5:2. Citizens of Europeanized countries were becoming more and more used to seeing terrible automobile accidents, and by the beginning of the 1930s, well informed people all over the world had become greatly concerned. Source: Aetna Life Insurance Company ad, Safety Engineering, 57 (1929), 4.

sition suggested that Heinrich was using the idea of accident proneness. Two years later, in 1931, however, Heinrich in his book Industrial Accident Prevention made accident proneness a major focus of his attention, and he recommended detailed examination of any worker who suffered an accident to see what factors like defective eyesight might have been involved. In a worker who was dropping tools and work objects, psychological examination could be used, Heinrich noted, to find and eliminate the cause of this behavior. Heinrich was already starting an investigation to see if workers who made production errors also had accidents. His successive publications show how easy it was for him to accept the conceptualization and practical implications of accident proneness when he heard about it.22 Elsewhere, publications reporting investigations of accident proneness were abstracted and reviewed in survey articles. The work of Farmer and Chambers in particular, but also other writers on the subject, was covered regularly in safety, psychological, and medical journals, both specialized and more general.23 Particularly important were the summaries written by the Philadelphia psychologist, Viteles, whose earlier work was

93

94

ch a p t er 5

noted above. He continued into the 1930s in his periodic surveys of industrial psychology to call attention to work on accident proneness, surveys that experts in the field all over the world read. In those review articles, Viteles mentioned work from Marbe’s students as well as the British investigators, and Viteles continued to put the two together along with new American work. In this context, research on the subject of accident proneness constituted therefore a small but standard part of the rapidly expanding field of industrial psychology.24 By the beginning of the 1930s, textbooks, too, frequently included descriptions of workers, especially, who had repeated accidents. One author, for example, did not use the term or cite works by such authors as Farmer and Marbe, but he did give statistics and clearly described the phenomenon—only later to ascribe accidents to carelessness: “Those employees who already have accident records are almost invariably worse than those who have clean records, in spite of the fact that individual examination . . . seems to show that the particular accident was unpredictable and therefore pure chance.” 25 An American author of a 1931 industrial psychology textbook devoted substantial parts of his chapter on accidents to individual tendencies to have accidents and to tests for “accident susceptibility,” “accident liability,” and “tendency to repeat” having accidents. He included extensive descriptions of psychological tests that had been devised, and he cited the work of both Marbe and Farmer and Chambers, as well as that of other writers. He just did not happen to use the term “accident proneness.” Another writer from that era, who did use the term, expressed his belief that accident proneness was a developing research area.26 Sometimes, of course, the idea did not appear in places it might have, but at other times it appeared in substantial detail, usually when authors summarized the growing number of researches that had been reported. Despite his central role in publicizing the idea among industrial psychologists, Myers in Britain, for example, was remarkably restrained in bringing the subject up himself. In a book in 1926, he devoted much space to factory breakages that could be reduced by reorganizing production, but he did comment, We now know from statistical inquiry that accidents are far from being distributed uniformly among the workers, but that some workers are much more liable to them than others. We still await the detailed report of a psychological examination now proceeding of those persons who are specially prone to accidents. The characteristic is obviously a psychological one . . . Those who are especially prone to accidents

The Streams Come Together, Late 1920s–Early 1930s

clearly need early detection and removal to places of safety where they are no longer a danger not only to themselves but often also to their fellow-workers.

A second edition in 1933 cited Farmer’s and Chambers’ continuing work and other literature, but the text was virtually unchanged. Meanwhile, Myers had edited a volume that in one essay included accident proneness in a major, fundamental way, and elsewhere at the beginning of the 1930s, he was including the idea and term in expositions of industrial psychology.27

mor e m a nager s join in Managers were certainly becoming aware of the problem. Not all welcomed an idea, or any idea, that came from psychologists. A writer in Engineering, a British publication, in 1930 commented that “by no stress [sic] of imagination can the great part of this [applied psychological] work be regarded as psychological.” The psychologists, he commented, “have raided boldly the entire field of management.” 28 Yet elsewhere in that same journal in that same year, a writer grudgingly admitted that accident proneness was becoming more credible with the wide-ranging experiments of Farmer and Chambers to find indicators of accident proneness, although the tests had not as yet reached practicality.29 In the United States, between 1931 and 1933, the main general safety journal, the National Safety News, carried at least five major stories explicitly on accident proneness. Moreover, the director of the National Safety Council Industrial Safety Division wrote an article elsewhere in which he construed accident proneness as the major industrial accident factor. He was not without hope, however, for he pointed out that employees could be warned, have their equipment changed, be transferred, or be discharged. “Collectively these reports prove that accident-proneness does exist and that most cases may be corrected by intelligent treatment.” 30 Although the Germans, especially, had pioneered testing train and tram operators to eliminate those who might have accidents, as the idea of accident prone automobile drivers came in, as noted in the last chapter, it was American managers and investigators who conspicuously came to participate in studies of transport and traffic accidents generally. Moreover, for those safety experts working in other areas, such as factories, the statistics out of transport studies were startling and were used as powerful, classic demonstrations that there do exist accident prone people. At the beginning of the 1930s, then, those concerned with industrial

95

96

ch a p t er 5

safety and efficiency, those concerned with transportation accidents, and those in the safety movement had to some extent been exposed to the idea of accident proneness. That idea consisted of recognizing the existence of a phenomenon: some individuals were responsible for more accidents than were other individuals under the same circumstances. Some groups seemed particularly at risk, such as those with defective eyesight. But there was one particular group of people who, with the best of intentions, for whatever reason simply incurred more accidents than did comparable people. The literature was beginning to suggest that some testing might identify those accident prone people before their records would identify them. In that case, they could be removed from risky activities or they could be retrained; the state of knowledge was as yet inconclusive. Indeed, it was unclear how applicable managerial experience in factories or transportation was to more general social situations. And meanwhile, psychologists, who were trying to extend psychological testing in industry, and had not made great headway in the late 1920s, found the idea of testing for accident proneness held substantial promise. 31

a r a nge of l e v els of ex pert ise During the 1930s and 1940s, the idea of accident proneness continued to appear in various technical and popular or semi-popular publications. Some authors’ explanations were very brief and elementary. Others discussed differential accident rates in great detail or described diagnostic testing programs at length. Some writers publicized programs that brought improved records of safety by concentrating corrective measures on people who caused the most accidents. All such types and levels of information continued into the midcentury period. Often elementary information would appear not only simultaneously with, but in the same publications as, advanced discussions—a pattern characteristic of popularization and communication of technical ideas. Catherine Covert named this distinctive pattern the upward spiral because over a period of time, the average level of discussion in popularization typically rises. 32 In the case of accident proneness, the most elementary level tended for decades to persist, right alongside the most sophisticated or refined reports of advanced investigators. In various social groups, the idea of thinking statistically advanced only slowly. And safety always involved a constant education and re-education process as new people entered into changing technological systems and managerial

The Streams Come Together, Late 1920s–Early 1930s

circumstances. Just getting people at all levels of responsibility to be continuously safety conscious was a persistent problem. 33 Within the large context of concern with accidents, older ideas of course persisted as well as anything new. The idea of accident proneness had to compete with the traditional wisdom that carelessness or incompetence could explain why humans suffered injuries and made errors. As one factory manager wrote, We have always figured we had too many accidents, and we have tried hard to cut them down. We bought guards and made guards and replaced guards and invented guards—and still had accidents. We had safety committees and safety inspectors and safety meetings and safety movies—and still had accidents. Then we got ‘hard-boiled’ and, in place of sympathy over a minor injury, began to ask the employee why he didn’t sleep nights instead of sleeping on the job—and though we still have accidents, they are steadily growing less, even with a considerably larger crew and with a decidedly faster working tempo. 34

figure 5:3. Education to heighten concern about safety consciousness continued into the 1930s and ever afterward. This French safety poster from 1937 shows two frisky workers roughhousing, with serious consequences for another worker, and warns, “Do not play in the workplace.” Source: G. Ichok, “La Prevention des accidents du travail,” Le travail humain, 5 (1937), 469, reproduced with the kind permission of Presses Universitaires de France.

97

98

ch a p t er 5

Many safety experts believed that education offered the one practical way to reduce accident rates. For them, the idea of accident proneness could be irrelevant or marginal. A Göttingen psychologist alluded to Nichtunfäller, Marbe’s term, assuming that his audience knew about “people disposed to accidents,” a phrase that appeared in his abstract. But in the article he concentrated entirely on vocational testing (particularly of reaction time) and safety education. 35 Indeed, many writers interpreted “the psychology of accidents” to be devising educational programs and safety propaganda. Safety education was in fact very conspicuous in the 1920s and 1930s. 36 And yet advocates of the idea that there are accident prone people—particularly accident prone drivers—could claim that education aimed at a critically hazardous group of people was particularly effective. By the beginning of the 1930s, it was possible for writers to include the term “accident prone” and to speak of specially susceptible people casually even while continuing to advocate traditional education and discipline. Or experts could just include the finding and treating of accident prone people as one item in a list of safety measures. Or the term could be generalized and applied to types of behavior, as in “the person prone to horseplay” in the workplace (this usage is from 1933) (figure 5:3). 37 All levels of use represented the infiltration of notions of accident proneness. While the general pattern of acceptance and occasional advocacy persisted, as times changed in economic depression and war, new advocates and investigators helped consolidate the place of the idea of accident proneness among Western experts and publics.

6.

CONSOLIDATION AND DEVELOPMENT, 1930S–WORLD WAR II

From one point of view, the idea of accident proneness did not change in the Great Depression–World War ii era. Heinrich’s textbook, for example, in a new edition in 1941 carried about the same exposition of the idea that it did in 1931. Yet over a period of years, much did change. Evidence from that period shows that many more people heard about accident proneness and reacted to it. In addition, the range of levels of understanding expanded.1 Experts refined the idea and the ways in which they could apply it especially to transportation and factory work.

t he spr e a d of t he ide a As I indicated earlier, in those troubled times, the idea became less and less conspicuous in Continental European publications.2 Where the idea of inclination to accident persisted in Europe, managers, rather than psychotechnologists, were often the most interested. In Britain, psychologists continued to show interest, and they succeeded in getting official government reports to recognize the problem. 3 In the United States, psychologists and especially those active in the safety movement carried the momentum at first.4 Eventually the ranks of those who for any reason were furthering the use of the concept of accident proneness came to be populated largely by Americans, including some managers. 5 99

100

ch a p t er 6

By the beginning of the 1930s, then, not only psychologists and safety experts but managers in general were becoming familiar with at least the idea that some people, usually workers, had more accidents than others. An industrial psychologist, C. A. Oakley, writing in wartime Britain in 1942, could already recall that “in the early ‘thirties’ . . . few industrial administrators and higher executives could have been in ignorance of the existence of tests of accident proneness.” 6 “We are all familiar with the expression ‘Accident Proneness,’ ” wrote an Indianapolis factory executive in 1936.7 And the same, as I have indicated, could be said of those in the transportation fields. After the 1920s, the safety movement provided the major social context in which the idea of accident proneness took root and grew. 8 Even Vernon, the British investigator, in his mid-1930s textbook, Accidents and Their Prevention, took up the question at great length and occasionally used the term of his sometime colleague, Farmer, “accident proneness.” Vernon preferred older phrasing such as “susceptibility to accidents,” and he otherwise indicated many reservations about applying the idea in the workplace.9 But, whatever their reservations, throughout the Western world, safety experts like Vernon, along with advocates, were the most effective agents in spreading the idea. The centrality of the safety movement explains in part why it was in North America that much of the publicity for the idea of accident proneness appeared. The safety movement on the Continent, as noted above, continued to lead in advocating regulations and mechanical means for protecting workers.10 The safety education and training programs of the United States, however, were the models for the rest of the world. And there, as well as in Britain, the human factor in accidents—including accident proneness—came to play a notably important part in the safety movement.11 Heinrich, as a major figure and eyewitness in American industrial safety, in 1937 recalled the changes in the safety movement, including new attention paid to symptomatic minor accidents: The 1937 brand of safety is far different from that of 20 years ago. Industrial accident prevention in 1917 as a science or an activity subject to rules and defi nitions did not exist. It was comprised almost entirely of eliminating, guarding or correcting hazards of a mechanical or physical nature. Today such hazards are corrected as a matter of routine good practice, but greater emphasis is placed on correcting the unsafe practices of persons. This is because it has been demonstrated by research that more than 75 per cent of all accidents are caused by the personal factor.12

figure 6:1. In 1939, highways as well as people became “accident prone,” as described by safety experts in Ohio. Source: State of Ohio, Department of Highways, Summary of Traffic Accidents on State Highways Outside Municipalities (Columbus: Ohio Department of Highways, 1940), 25–27.

102

ch a p t er 6

This continuing general shift to the human factor was symptomatic of another change. More and more, managers and experts were treating labor less as a commodity and more as a personnel problem. One notable symptom was the fact that beginning in the 1930s, and markedly after World War ii, fatigue studies diminished dramatically in the safety literature—just as psychological investigations increased. Not least among the factors in this shift were the famous Hawthorne studies.13 It is striking that in the 1930s, experts who published on the subject of accident proneness were consistently divided evenly, year after year, between those concerned with industrial accidents and those focused on transport and street accidents. During World War ii, industrial accidents, which increased substantially, dominated the literature, but by 1945, the balance was back in place.14 At any time, however, publications on accident proneness as an abstract, general problem were few. Instead, psychologists, managers, and safety experts all tended to apply to practical problems what they had heard or read on the subject. Application and practicality therefore marked discussion of the idea. In 1940, a landmark event occurred indicating how extensively the idea had penetrated. For the first time in print, an author used the term “accident-prone” to refer not to one or more persons, but to inanimate objects. The writer was an anonymous civil servant in the Division of Traffic and Safety in Ohio who gave the label to areas with high “accident frequency” on the state’s roadways. That is, more accidents were occurring there than in other areas in the highway system: “almost one-third of all reported accidents are concentrated on 4.5 per cent of the total state highway mileage.” These were the accident prone areas. Employees in the division, it turned out, had even charted the hazardous areas on what they referred to as “the accident prone map,” a copy of which they published in their report, and the “accident prone locations” were indeed indicated strikingly (figure 6:1). In addition, an abstract of the report appeared in the widely read professional/public service journal, Public Works, under the headline, “Accident-Prone Areas.” 15

popu l a r izing t he ide a This generalization of the term to a new generic status, used to designate not only people, but any node of high accident frequency, reflected the extent to which the idea had spread and was diffused. How far popularization had proceeded by 1940 is suggested by another change. In the early 1930s, some writers and editors had treated accident proneness as a tech-

Consolidation and Development, 1930s–World War ii

nical term unsuitable for a popular audience. Reporting the work of investigators who unmistakably used the term and idea of accident proneness, popularizers resorted to such descriptive euphemisms as “driver fitness” or “physical attributes” or “defect.” 16 But by the World War ii period, many people, at least in the now primary arena, the United States, had used and were using the term.17 As technical discussions continued to appear, so did popularizations, including many elementary explanations, coexisting as in the Covert spiral model of popularization.18 Safety publications reached many general readers, and in 1942, the U.S. news magazine, Time, carried a story explaining how an expert on traffic accident repeaters in Detroit had persuaded executives of the industries of that city to launch a campaign based on the idea that “some men are more prone to industrial accidents than others, and the way to prevent plant accidents is to cull these men out.” One Packard [Motor Car Company] employee had nine traffic and 14 plant accidents in five years. The foreman found him a hothead who liked to ram drivers he saw violating traffic regulations, a fast worker with excess time for practical jokes (which often backfired). Switched to a responsible maintenance position, his traffic and plant accident scores dropped abruptly.19

The war indeed provided a major opportunity for advocates of the idea of accident proneness to gain hearers among officials and the members of the public who were concerned about wartime productivity. In Britain, the Industrial Health Research Board summarized a generation of research in a report on “the personal factor in accidents.” “The war,” the authors wrote, “has greatly aggravated the conditions conducive to a high accident rate, both on the roads and in the factories.” On roads, the blackouts and numbers of new drivers constituted special hazards. “In industry a large number of people have undertaken work to which they are not accustomed, and women and young people have been put on to work normally done by men. Long hours have to be worked, and the speed of production has risen.” And so the Board set out the causes of accidents, such as fatigue, temperature and light, worker and management attitudes—and accident proneness, “an unfortunate characteristic of a section of mankind.” 20 The Board’s report gained an unusual amount of exposure in the English-speaking world. The editors of the leading British engineering journal reprinted large parts of the report and ran another story on the

103

104

ch a p t er 6

accident proneness section specifically. The editors of Nature also ran a story summarizing the report, and half was given over to accident proneness.21 The U.S. government publication, Monthly Labor Review, carried a summary, more than half of which, again, was devoted to accident proneness, and that article was noticed in Industrial Medicine. Meanwhile, an additional summary of the British report appeared in the Medical Journal of Australia.22 The war also presented the opportunity for another figure to publicize the idea of accident proneness: Helen Flanders Dunbar, an American physician who was the leading figure in the new field of psychosomatic medicine. Her particular agendas and impact I take up in the next chapter, but in publishing on accident proneness in both technical and popular printed media, Dunbar repeatedly called attention to the phenomenon. Her popularizations of the idea that some people were accident prone appeared widely in American media so that the term, at least, could have become familiar to many additional Anglophone readers.23 What is striking is the extent to which in all countries, writers in even the technical media from the 1920s right to the end of World War ii were repeating basic findings by Greenwood and Woods and Marbe. These publications conveyed the excitement of discovering the phenomenon of accident proneness. Any possible refi nement of the idea was seldom a part of these publicizing publications except insofar as the refi nement might be applied as part of a formal campaign to prevent accidents. If one judges by content of the publications, it was the practical application noted above that made the idea exciting.

m ajor f igu r es a s r epr esen tat i v e f igu r es Some idea of the way in which the notion of accident “repeaters” and “proneness” developed before the end of World War ii can be suggested by the efforts of three key figures, the two original psychologist advocates, Farmer and Marbe, and Lahy, the French psychologist who ultimately wrote the definitive French statement on the idea. Eric Farmer continued working with E. G. Chambers after their 1926 publication that launched the term “accident proneness.” Under the aegis of the Industrial Fatigue Research Board, later the Industrial Health Research Board, the two of them continued to attempt to find psychological tests that would identify potentially accident prone workers. By 1929, they had devised a battery of weakly predictive tests that correlated with each other, “suggesting that accident is dependent on many dominant

Consolidation and Development, 1930s–World War ii

factors and not on a single predominant one.” Their tests at that point did not relate to sickness records, but only accident records, they found, but they could report that worker productivity “is associated with low accident rate.” 24 Just as the intelligence testers of that day were trying to identify a general factor, g, for intelligence, Farmer and Chambers were attempting to find a common general factor for accident proneness that a battery of tests would produce. For decades, various writers made different assumptions about whether accident proneness was a single trait or factor, or whether it was a combination of traits and factors. No consensus ever emerged.25 In 1930, Farmer indirectly took note of the idea that character analysis that produced types might have predictive qualities for accident proneness. He analyzed some of his test subjects and found that he could divide them into controlled workers, good workers, thinkers, inept workers, and miscellaneous. But he concluded that his analysis was not particularly useful, although he allowed that someone else might come up with a character typology that would be better than psychophysiological testing.26 Farmer summarized his work in three lectures published in 1932 as a book on the causes of accidents. And of course he recounted the tests he had worked out to predict both favorable and unfavorable profi les. Those with consistently bad accident records needed special treatment, he concluded. They are people who “react unfavorably to modern industrial conditions. They find the strain of their environment too much, and react by inefficiency and by having an undue number of accidents and sicknesses.” 27 Farmer continued work on testing for accident proneness, and he continued to urge that tests be used to select employees and further be used in conjunction with accident records in employees’ probationary periods to ensure a minimum amount of risk in the workplace.28 By 1934, Farmer was acting in full as a publicist of the idea of accident proneness. His address to the National Safety Congress in London he published in an American safety journal. Statistical records permitted one to identify those who were the source of “a majority of accidents . . . whenever the data were adequate,” and, Farmer continued, “it [the factor of accident prone people] has always been confi rmed.” 29 In 1937, Farmer made a sudden jump from industrial accidents to the other main stream in the accident literature. With Chambers he published an investigation to determine if the same profile of persistent accident proneness showed up among drivers of various kinds of motor vehicles.

105

106

ch a p t er 6

Again, they tried to use psychophysical testing to predict an individual’s driving record. What was striking was how similar the results were to those found in years of investigating industrial workers’ accidents. Farmer and Chambers found the phenomenon of persistent accident records as stable among drivers as among industrial workers. Psychophysiological tests again correlated only weakly with accident proneness but could be useful when combined with accident records. The tone in Farmer’s and Chambers’ new publication, however, in places had become strident: Accident proneness is no longer a theory but an established fact, and must be recognized as an important element in determining accident incidence. This does not mean that knowledge of the subject is complete, or that the liability of any particular individual to accident can with certainty be predicted. What has been shown, so far, is that it is to some extent possible to detect those most liable to sustain accidents.

Farmer’s last publication on the subject of accident proneness, at the end of the war, was also on traffic, urging testing to reduce the number of accident prone drivers on the road. Farmer maintained a strident tone. He summarized previous investigations that “have shown conclusively that some people are more prone to accidents than others exposed to equal risks. 30 Because of his persistent publicizing, writers on accidents cited Farmer at that time and for many years afterward. Even in non-English-language publications, authorities often repeated the British version of how the idea of accident proneness developed. Marbe’s record after the 1920s was not comparable with that of Farmer. As I have noted, Marbe did publicize “his theory,” and he had students who helped with that project. Yet by the mid-1930s, Marbe had to retire, and he sought inconspicuousness to protect his wife, who was classified as nonAryan. Nevertheless, his differentiation of those who had accidents from those who did not, the Unfäller and Nichtunfäller, as I have already suggested, was remarkably persistent and served better to carry the idea of accident prone people than either “accident proneness” or Unfallneigung. Indeed, Marbe himself deviated from the latter term in places after 1926 and reverted to “disposition.” At least two factors explain why Marbe’s ideas did not spread significantly beyond the basic idea that some people had a history of accidents and that other people did not. First, not only did he emphasize theory, but the theory did not fit easily into the practical programs in psychotechnology. Moreover the theory invited theoretical argument, as I suggested ear-

Consolidation and Development, 1930s–World War ii

lier (chapter 2). And, worse, the theory involved some ideas that were not fully acceptable by the 1930s, particularly the idea of accident proneness as a character or personality trait. 31 During the 1920s, German psychotechnologists tended to shift from applying tests to match the worker to the job to making sure that the character of the worker made the worker a good worker. In such an atmosphere, the characteristic of inclination to having accidents should have fitted very well. The tests for this one personality characteristic, however, came from the battery of tests suited to fit the worker to the production line, such as reaction time or reaction pattern. In such a context, general vocational aptitude and tests for good worker characteristics such as intelligence and persistence rendered accident proneness marginal to psychotechnology. 32 If one argued that there was an accident prone personality, or a distinctive trait, then the argument became problematic. Still, Marbe did succeed in calling attention to the phenomenon of an individual’s inclination to accident. Continental writers cited his work. A few extended it. In 1935, one of his students published on accident proneness among children. Also in that year, another expert applied Marbe’s ideas to workers in the iron and steel industry. Another member of Marbe’s institute published—still in 1935—a whole book, Warum Unfälle? [Why accidents?], in which he placed Marbe’s theory in the context of the broader preoccupations of psychotechnologists such as fatigue and psychophysical testing—noting incidentally that the British investigators working with the idea of accident proneness still did not use the German literature. 33 Finally, in that same year, 1935, Marbe himself published a special article in English in the organ of Myers’ National Institute of Industrial Psychology, briefly summarizing his ideas, an article in which Unfallneigung was consistently translated as “accident proneness.” But Marbe was no longer offering new data or new theory. 34 Moreover, since German psychotechnology, which had provided Marbe a forum, diminished rapidly during the 1930s, so did Marbe’s work. 35 In 1943, a Viennese expert, in calling attention to the problem of accidents, mentioned Marbe’s ideas. In that same year, however, Moede, in his textbook on vocational aptitude testing, now omitted Marbe’s work except for a small trace. 36 In 1948, Marbe again published a brief summary of his much earlier work, but he offered no new data, and he more strongly than ever tied the phenomenon to his philosophical idea about phenomena that grouped in nature. 37 The reaction of Lahy, the dominant writer on the psychology of safety in France, and a powerful founding figure in psychotechnology, illustrates another reason why Marbe received credit for identifying Unfäller

107

108

ch a p t er 6

and Nichtunfäller but did not win a following for his idea that accident proneness was a personality trait: Marbe wrote about personality in terms of types and constitutional inclinations. 38 In 1936, Lahy, with Suzanne Korngold, published a major work on the psychological causes of accidents. 39 The authors cited work showing that most industrial accidents had “human” causes, and then they reviewed the work of Marbe, Greenwood and Woods, and Newbold, and even some recent work of Farmer and Chambers and the Americans, Bingham and Slocombe. In the end, Lahy and Korngold accepted the idea that some people had more accidents than other people, using Marbe’s labels or translations for those who did and did not have histories of accidents—Unfäller and Nichtunfäller. Where Lahy and Korngold took issue with Marbe, however, was with the idea of predisposition, which they held was implicit when it was not explicit in Marbe’s writing.40 Lahy and Korngold subjected two groups of workers, those who had accidents (“sujets fréquement blessés”) and those they labeled “normals,” to a series of psychological tests (many taken from Farmer and Chambers) to see if they could disentangle the factors that made one group different from the other. They concluded that combinations of tests for motor abilities and for emotional or affective reactivity could serve the practical purpose of screening out unsuitable candidates for employment. Other characteristics did not differentiate those likely to have accidents from those not. Lahy and Korngold showed for example that those having accidents and “normals” all fatigued at the same rates, so that where fatigue might affect a whole workforce, the fatigue factor did not complicate differentiation of employees who had accidents and those who did not.41 Lahy and Korngold particularly concluded that accident prone people did not adapt to new rhythms of work. This absence of the characteristic of “functional plasticity,” however, was not part of a fi xed “biological constitution,” or predisposition, according to Lahy and Korngold. That constitution could be modified not only by transient factors at any time but by education that might change a person’s habits. Lahy and Korngold in this way dissented from Marbe’s suggestion that human beings sorted themselves into fi xed types.42 Thus did a stream of environmentalism develop explicitly in the history of accident proneness, as opposed to the implied nativism of Marbe. Lahy and Korngold denied that individuals having accidents did so as a lifetime characteristic. They emphasized the different work environments that caused accident rates to go up for all members of a work group. And particularly they emphasized the bad working and living conditions

Consolidation and Development, 1930s–World War ii

that caused the fatigue and physical weaknesses that would make anyone accident prone on a physiological, and often temporary, basis. In short, they objected to blaming the victims. Nor was such a pro-labor stance surprising. Lahy was a committed man of the Left and the Popular Front, and he apparently died in 1943 as a martyr because of his alignment with the French Underground.43 Although Lahy did not follow up on his publications on accidents, as I have noted, they were widely cited for years afterward, especially in European countries.44 Farmer in England immediately published in English a notice of the work, for it in part paralleled his own and, he believed, offered confirming experiments. Lahy and Korngold’s testing, he wrote, yielded “at least a presupposition in favour of suggesting that the complex sensori-motor tests are dependent more on nervous control than on sensori-motor co-ordination, as it is ordinarily understood.” 45 And the hope that accident prone individuals could be identified and retrained played into the rationale of other safety campaigners. As a commentator from the USA observed, even without citing Lahy’s work, “Most repeaters can be cured of their accident habits. They must be if the accident situation in America is to be improved.” 46 Lahy had, however, identified underlying attitudes that conditioned the ways in which experts, at least, took up the idea of accident proneness. It is true that the whole idea of individual, personal factors was not always welcome. A Soviet writer, for example, in 1931 explicitly denied Marbe’s contention that some people had a special trait of proneness to accident. Rather, he maintained, accidents grew out of the organization of work, or of a failure to screen out workers who were just not conscientious.47

a n inbor n t r a i t The idea that a special group of humans existed who had the trait of accident proneness raised further problems in the 1930s and 1940s.48 A number of experts with different points of view agreed generally that such a group existed, but they disagreed about the cause. In 1937, some French investigators reported that individuals predisposed to have accidents did constitute a group, and probably the predisposition was based on a group of biological factors.49 This was roughly the same approach of the American auto driver safety tester, Alvhh Lauer, who wrote that “men who have accidents are ‘safety sick’. They may be chronically affl icted, or they may be only acutely ill.” 50 The belief of Farmer and other psychologists that mental tests would show that accident proneness would be related to

109

110

ch a p t er 6

psychological traits could segue into general personality profi les or types. Such viewpoints were usually not the same as the hereditary typology that Lahy saw in Marbe’s work. Nevertheless, the idea of accident proneness as a persisting inborn trait underlay much thinking on the subject. As an expert writing in 1939 put it, there were in the idea of accident proneness many problems of qualification and complexity, but “that does not mean that there is no such thing as congenital proneness to occupational accidents. The existence of such proneness is not disputed, but this relation of cause and effect is not the only one, and certain counteracting relations are important too.” An American occupational psychologist believed that “we should at least remain open-minded on the question of whether or not the apparent predisposition to accidents is innate. The evidence is too impressive to dismiss the concept of accident-proneness as a mere figment of the imagination.” 51 A number of thinkers before 1945 expressed more certain ideas. One industrial expert made a telling comparison: “Just as a typhoid carrier is a menace in the kitchen, through no fault of his own, so certain persons, because of congenital peculiarities, are menaces to themselves and to their fellow workers in industrial plants employing high-speed machinery. It is no reflection on their diligence or their moral character; they are the misfits of a machine age and just happen to have been born that way.” 52 Merely defining accident proneness as inherent, however, was not sufficient. If accident proneness was a trait (or Anlage, tendency, as Marbe suggested), it could be defined operationally. As an American psychologist put it, accident proneness in an individual “implies the presence of a combination of qualities which have been found from scientific investigations to lead to an undue number of accidents.” 53 Sometimes, too, the formulation was in terms of a “type,” and there were many advocates at the time who believed human beings could be classified as “types.” It was easy, therefore, to characterize the accident prone person as a type, with the presumption that the type was an inherited characteristic. 54 So a French scholar in 1937 who published extensively on human typologies wrote of the biological nature of the predisposition of a person to have accidents, noting, however, that tests devised until then had still not established the general factor (facteur général) of predisposition to accidents. 55 In any case, investigators of the 1930s period attempted to use many psychological tests to try to figure out if accident prone behavior was the result of a combination of personal factors or if there was, indeed, a gen-

Consolidation and Development, 1930s–World War ii

eral factor, or trait, comparable to the g in intelligence. A Stuttgart investigator, for example, in 1935 used Marbe’s classifications of those who did and did not have accidents and tried specifically to find consistencies in the situations or types of accidents suffered by the accident prone. He found none, and his conclusion was that accident proneness was some kind of biological disposition, an insufficiency in the person when he or she had to cope with reality.56 One of the questions that challenged investigators before and during World War ii was whether accident proneness was a temporary or a persistent tendency. As one expert noted, “Many accidents are due to temporary emotional reactions that even a psychiatrist might have trouble in locating. There probably is a much closer relation between such common matters as unpaid grocery bills and industrial injuries than ever has been suggested in the accident records of any plant.” Another expert offered as an example from his own cases the worker who “rushes to work without breakfast and on a sultry August morning faints while working in a manhole.” Other commentators remarked on temporary personal upsets that preceded accidents, with the assumption, often justified with happyending case histories in which managers intervened, that the tendency to have an accident was temporary. The injured workman who had argued with his wife made up with her and did not have any further industrial accidents. 57 Alongside expositions about accident prone people appeared other publications, usually by industrial managers who, in taking a commonsense approach to accident prevention, included, sometimes casually, the idea of accident prone people. Typically, accident prone “repeaters” constituted just one of a list of factors that writers asked supervisors to take into account. A mining safety advocate, for example, in 1940 urged operators to consider psychological approaches to safety, including “a study of occupational selection and guidance, rest pauses, enlightened supervision and the removal of accident-prone workmen to more compatible occupations.” 58

f u rt her in v est ig at ion of t he probl em Two major types of expositions therefore existed: those in which safety advocates placed accident proneness in a broader context of accident prevention, and those in which investigators, usually psychologists, attempted to examine the phenomenon of accident proneness to find out what elements went into it. The ultimate purpose of the investigators

111

112

ch a p t er 6

was to devise some way of identifying people who were accident prone and to find the causal elements in this condition. With this knowledge, the experts hoped to render harmless the proclivity to cause injury and damage. As early as 1936, Lauer was able to declare bluntly that “safety engineers are rapidly coming to realize the need for other measures of accident-proneness than the crude observational methods used in the past.” He went on to write that “there is no one cause of accidents. There are not two, three, or four causes. There are numerous causes. A test of reaction time used alone is about as indicative of accident-proneness as pulse rate is diagnostic of one’s need of medical attention.” 59 Nevertheless, experts continued to explore the phenomenon of accident proneness with a variety of individual tests—medical examinations, psychophysical tests, and psychological tests, including measures of both cognitive and emotional functioning. 60 Some tests came from Marbe’s laboratory. Others were based on the work of Farmer and Chambers. An investigator from Marbe’s department found that the typical Unfäller did not do well on accident tests included in the standard vocational battery.61 Other experts kept adding to the battery of tests or subtracting from it. A French industrial psychologist, for example, used standard tests from Lahy, from Moede, and from the United States, and then he added his own turning cylinder tests. He measured the results of this testing by the number of days workers missed, since this figure, he found, corresponded well with accidents. And of course using his battery of tests to screen out employees brought the absence rate (and presumably the accident rate) down by half. 62 Increasingly in the literature, psychologists who abstracted tasks from the workplace to the laboratory predominated. Instead of having a candidate for a job stand working dummy controls, psychologists had more easily measured follow-the-dot or react-to-astimulus exercises (figure 6:2). Beyond psychophysical tests were tests of personality. Many employers depended heavily upon judging character in applicants for a position, but psychologists had devised a variety of tests other than those for psychomotor abilities. Typically, the source for any practical testing was simply that of diagnostic testing for mental illness or character type. 63 Systematic empirical observation also contributed, such as extensive evidence that accident prone children tend overwhelmingly to be adventuresome. Or employers used social characteristics—not only the traditional age, but marital status, social class, or “nationality.” 64

Consolidation and Development, 1930s–World War ii

figure 6:2. A psychological laboratory test instead of a simulation apparatus: rating hand-eye coordination with a zigzagging disc. Source: Robert E. Martin, “Are You Fit to Drive Your Car?” Popular Science Monthly, April 1930, 55.

From the welter of tests and observations, experts sometimes attempted to make generalizations. After reviewing the effectiveness of a number of tests, especially those measuring manipulating and seeing quickly, a West Virginia psychologist, for example, developed a hypothesis: “Individuals whose level of muscular reaction is above their level of perception are prone to more frequent and more severe accidents than are those individuals whose muscular reactions are below their perceptual level.” 65 Or another veteran industrial psychologist, Slocombe, identified accident prone people as those who did not conform to social standards (people who a few years later would be labeled those who show social deviation). Most safety campaigns, he pointed out, had been based on the idea that workers willed, that is, actively chose, to obey safety rules, and so the aim was to get workers to want to act safely. The accident prone, however, Slocombe noted, did not have the ability to conform, and so willingness was beside the point. He labeled “accident repeaters” as “non-conformists.” They required special, individual treatment, he concluded. 66 Virtually everyone working with the problem, as I have noted, was deeply concerned with useful changes, especially in the context of the safety movement. A safety official from Colorado, for example, compared the “shotgun method” that had to cover all employees as opposed to the “rifle method” that singled out “repeaters” for attention. For his company, he believed that it might be more efficient to target just the 20% who were causing 76% of all of their accidents.67

113

114

ch a p t er 6

One of the substantial changes that just the idea of accident proneness helped bring to the safety movement of the 1930s and after was calling experts’ attention to the importance of registering (and counting) minor as well as major accidents. Moreover, accident proneness, according to a 1934 report, “applies not only to factory accidents but to those at home, such as falls, burns and accidental cuts with knives. Still more surprising is the same rule for people who get things in their eyes.” Safety experts pointed out that in the factory, the number of dispensary visits was very significant, and the problem was to keep employees from failing to report small accidents if they knew there might be consequences such as those recommended by one expert: the “transferring, or other disposition, of all employees who have accumulated a record which shows them to be ‘subject’ to injury.” 68 Almost all of the applications of the concept of accident proneness turned out to be negative in operation. The whole idea of identifying an accident prone person had the purpose of removing that person from a situation in which he or she would be a danger to him- or herself or to others. This screening or “weeding out” process applied to both of the two main practical objects of the idea of accident proneness, drivers and workers. As noted in chapter 4, first mass transit officials noticed that there were bad operators. Then as the automobile accident rate became alarming, almost everyone became conscious of the existence of bad drivers. In the 1930s, as they had in the late 1920s, psychologists continued to move in on these practical categories of transportation operatives to suggest that those who incurred an excess of accidents could be conceptualized as accident prone.69 With or without psychologists, or the terminology, however, civil authorities who had to deal with an alarming social problem increasingly focused on people who could drive competently and those who could not. Even without the term, when “sixty per cent of traffic accidents are caused by ten per cent of the drivers,” concentrating on the 10% was just common sense.70 One of the other great pieces of common sense that was finalized in the 1930s was the idea that drivers of cars and trucks, not vehicular failures, caused the sensational increase in injuries associated with motor vehicles. In most cases, the problem was not any defective component in the machinery but “the nut behind the wheel” (figure 6:3). Accident proneness provided an effective way to focus on faulty drivers. In 1938, a major governmental study in the United States, based on almost 30,000 drivers over a six-year period in Connecticut, showed that only 19% of the drivers had any accidents at all. Moreover, the 3.9% of the drivers who had two or more accidents accounted for 36.4% of all accidents.71

Consolidation and Development, 1930s–World War ii

figure 6:3. One of many references to bad drivers, “the nut behind the wheel,” as the cause of accidents. This one is a poster of the U.S. National Safety Council, 1938. Source: National Safety News, November 1938, 44, reproduced with the kind permission of the National Safety Council.

Accidents per Driver During First Period (1931–1933)

Accidents per Driver During Second Period (1934–1936)

0 1 2 3 4

0.1 0.2 0.3 0.5 0.7

A frequently reprinted chart highlighted the possibility of repeater accidents. These repeater figures, as an investigator emphasized in 1942, put to rest the idea that accident proneness was a minor factor in damages, injuries, or deaths. As a social problem, accident proneness was persuasively established.72 Clearly it was time to get these menaces off the roads. As the authors of the U.S. government report concluded, “Since there exists a clearly defi ned class of accident repeaters, whose group behavior is fairly consistent from one period of time to another, it may properly be assumed that abatement of the accident rate may well begin with attention to this class.” 73 The chart showed how drivers with a record of more accidents in one period on average incurred a correspondingly greater number of further accidents in a later period.

115

116

ch a p t er 6

a n il lust r at i v e mid -1930s inciden t The idea of accident proneness came up in Britain in the mid-1930s, when a Medical Research Council report was used to support legislation to institute the examination of drivers and the disqualification of drivers who had a record of bad driving. The report contained strong statements that some people showed a persistent pattern of having accidents and that such a person should “be removed from a position in which he is a danger to himself and to others.” The report also touted psychological tests devised by Farmer and his associates—even before Farmer published about his interest in traffic accidents.74 News of the report caused a large insurance broker to approach David Munro, the distinguished secretary of the Medical Research Council’s Industrial Health Research Board, with the idea of testing the drivers working in large companies for which the broker had written insurance. The attempt was to gain insurance “experience” to see if eliminating drivers who did badly on the tests actually affected insurance claims.75 Munro immediately wrote to a friend, Ernest Bevin, the general secretary of the Transport & General Workers Union, who subsequently read the Medical Research Council memo and met with the broker representative. Bevin then wrote to Munro, “the difficulty I am in with regard to psychological tests is that there is no protection for the men who will be prejudiced and, of course, as I have already indicated to you, if there were psychological tests for employers I am sure that more than half of them would be out of business.” Bevin was concerned about the approach, which, he wrote, was “penal” rather than constructive. “Reports both on physical and psychological grounds might be presented, but the men would not have any opportunity of knowing what was in the reports; they would just find themselves out of work.” He was, naturally, deeply concerned in that period of acute unemployment about any driver’s being discharged, and he essentially rejected any program of testing.76 Given British institutions of power at that time, while licensing and a penalty system eventually came into place, psychological testing for accident proneness did not. This is the only serious confrontation of explicit labor interests with the practical application of accident proneness that I have discovered, but the incident shows the potential where there was a strong union presence. Also, as I shall emphasize again below, insurance executives’ interest in the concept was marginal even when it would have been logical, and this attempted project of one broker was, at best, extremely atypical.

Consolidation and Development, 1930s–World War ii

But the matter did not rest there. In 1935, the Minister of Transport suggested organizing statistics from insurance companies for the purpose of studying accident proneness, using the Industrial Health Research Board to do so. Company representatives replied that they still insured owners, not drivers, of vehicles, and so their statistics would not be appropriate and anyway did not included unreported accidents. Munro replied that they were interested in getting insurance ratings that would in a practical way eliminate accident prone drivers from the roads. And he hotly contested with evidence the dismissive statement that “the phenomenon of accident proneness depends upon . . . ‘a frequently repeated assertion.’ ” 77 When representatives of the Council (Munro, Greenwood, and Farmer) subsequently met with insurance company representatives, it became clear that the records of individual private drivers were just not available. But implicitly the companies were strongly pressured to switch to insuring drivers, rather than owners, and they certainly were educated further about the evidence supporting the idea of accident proneness—what Greenwood referred to privately as “Farmer’s motorious researches.” 78 Moreover, stories appeared in the press of that day, airing the issues involved. One attacked “accident-prone test—a new third degree,” as the headline put it. The author included both industrial workers and individual drivers who could suffer, and he or she explained the idea and the research and concluded: “There is no proof of the perfect reliability of accident-proneness tests. There is no satisfactory proof of the existence of accident-proneness.” What about social conditions? he/she asked. But in the course of his/her denunciations, he/she managed an extensive explanation of the idea.79 Other writers were interested and often concerned about exceptions and fairness. Even Greenwood was quoted as having reservations about eliminating too large a percentage of drivers from the roads. 80 For the time being, British drivers had to work with performance, not psychological tests.

discr iminat ing aga inst acciden t prone peopl e As illustrated by the British case, the speed and means by which licensing and testing of automobile drivers proceeded varied greatly all over the world. In Spain, the methods of selecting bus drivers carried over to automobile drivers, and in Barcelona psychological testing began for all drivers except private auto operators in 1933. In the UK, in 1943, a committee of the House of Lords made the recommendation for screening in terms of accident proneness:

117

118

ch a p t er 6

There are certain drivers who are constitutionally unfitted to drive a car. It is difficult to analyse the reason for this. It is probably due to a combination of vagueness, slowness in decision, nervousness, deficiency in road sense and judgment. It is not necessarily due to recklessness, or to bad road manners, or to active fault. If a driver is found to be accident prone, through defects either of capacity or of temperament, or by being involved in a series of accidents, the Committee consider that, in the interests of all concerned, he should be disqualified from driving. 81

In the United States, a number of experts came to specialize in tests that might identify “problem drivers.” How effective such testing might be, or was, remained controversial. What was not controversial was that there was an irreducible percentage of drivers who were, to quote the title of a popular article, “born to crash.” But as a British expert wrote dryly in 1945, “The ideal motorist may not yet have been bred, for not until a century’s statistics are available will the laws of inheritance of driving-ability be known; by which time the emphasis will have shifted to the aeroplane.” Of course testing to screen out accident prone airplane pilots also proceeded in the 1930s and during World War ii as it had since World War i, but now the idea of accident proneness was finally explicitly applied to them. 82 It is clear that experts writing about accident proneness in the 1930s had begun to face openly the social-ethical conflict that the negative consequences of accident proneness generated. That is, to protect society, accident prone drivers and workers needed to be removed from activities in which they generated hazards. One could even argue that industrial breakage and work absences represented unnecessary inefficiency and economic loss. 83 But what about the people on whom the negative screening or removal fell? Anyone who argued the practical consequences of accident proneness was faced with a classic confl ict between social gain and individual cost. Many writers of course did not comment in published works. Many others used the formulation that the worker must be suitable for his or her occupation and that it would be a protection for such people to be screened out of positions in which they would be at risk. This was most often the argument of European experts. “Since the tests showed also a certain correlation with the proficiency of the worker, an element of safety may be incorporated in the very definition of fitness to work. A good worker is not only one who possesses the required experience, abili-

Consolidation and Development, 1930s–World War ii

ties, and skillfulness, but also one who is not prone to have accidents,” wrote a psychologist in 1943. 84 The question of automobile drivers was more sensitive. With rigorous testing of commercial drivers, for example, “rejecting a candidate meant that he could not go on performing his usual job and made it impossible for him to obtain a similar occupation. Thus a heavy responsibility was laid upon the examiner, particularly when he was dealing with candidates already dependent on driving for their livelihood.” 85 In an industrial setting, observed a number of writers, detecting and dealing with accident repeaters was the responsibility of supervisors. The supervisors could re-train, transfer, or discharge the employee. The content and word choice in injunctions to employers suggest a range of attitudes. One was “weeding out those who are specially prone to accidents and by transferring them to employment where there is as little danger as possible.” While “weeding out” was a strong metaphor, the more usual term was an impersonal suggestion to “eliminate” the employee. “Elimination of those whose employment in the factory is likely to lead to accidents, to themselves, to their fellow-workers, or to those whose safety depends upon the perfection of the material produced, is a point the value of which cannot be disputed,” asserted a wartime British writer, who recommended “selection of the individual and careful discrimination and supervision by the shop manager or foreman.” 86 “Selection,” “elimination,” “weeding out,” “cull,” and similar phrasing reflected the harsh realities of business. But it also reflected the attempts of well-meaning people to prevent injury as well as economic loss. Of course when people attempted to do the right thing, complications always arose. One industrial psychologist told of the great success a firm operating a fleet of vehicles had when it was able to screen out accident prone drivers. First they discovered that when transferred, the employees had accidents in the jobs to which they transferred. Thus the managers and consultants were aware that even if they discharged the accident prone worker, they had not solved the social problem of the dangerous drivers. “Probably all these repeaters went on driving—for themselves or for some other employer. Those with whom [the company] kept in touch after it released them continued to have accidents about as before . . . While [such] companies have cut their losses heavily, they have merely transferred them to others.” 87 Psychologists, particularly, who were working with accident prone drivers and workers often expressed sensitivity to the individual as well as social considerations that came with identifying someone as accident

119

120

ch a p t er 6

figure 6:4. This 1939 safety poster from Australia urged the proper training of new employees. Source: “New Posters,” Industrial Safety Survey, 15 (1939), 59.

prone. In 1941, the managing director of the U.S. National Safety Council questioned labeling a worker as accident prone: “Frequently an unfair case is made against the employee—that he is a misfit, accident-prone or possessed of poor attitude, when actually such diagnosis is wrong and the real trouble is something no more mysterious than a foreman who hasn’t ‘done right’ by the new employee.” 88 Such conscientious people could, at the least, point out that it would save money in a large organization to avoid discharging a worker if at all possible. Business wisdom was that the hiring process was extremely expensive. 89 The alternatives to discharge were transferring the employee to a nonhazardous position or careful education, training, and supervision (figure 6:4). Of course the education and training alternative was bothersome and expensive, but many psychologists held substantially environmental beliefs and thought that the training alternative might work. One psychologist advocated identifying the accident prone while they were still in school so that the tendency “can be conditioned in the school just as any habit can be corrected or formed.” 90 Viteles was a notably pro-labor writer, for example, who strongly advocated transfer and retraining of accident prone workers. He cited examples from his consulting practice in which transfer had succeeded. Or as an investigator at a large American firm who identified accident prone employees asserted strongly, “Re-

Consolidation and Development, 1930s–World War ii

search of this type should not be followed by any criticisms or discipline of those employees concerned.” 91 Another form of humanitarian attitudes in dealing with accident prone people appeared especially conspicuously during the World War ii period. This was emphasizing treating each worker or driver as an individual and his or her accident proneness as an individual matter. Slocombe, for example, began writing about how irrelevant were all of the tests and mass appeals to act safely: “Each case of non-conformity of workers to the safety principle presents its own problem, which must be diagnosed, and a suitable treatment or remedy worked out.” 92 The immediate and obvious response was to treat each case of accident proneness as an individual neurosis. A psychologist who had become personnel director of a public utility had come to believe that “future prediction of accidentprone employees will depend not so much on single psychological tests as it will depend on a detailed study of the whole personality by means of the clinical method.” 93 This response, as will be detailed in the next chapter, was appropriate for American culture, at least, which, as noted above, was rapidly becoming “psychologized.”

121

7.

HOW PSYCHIATRISTS DID NOT ADOPT AND MEDICALIZE ACCIDENT PRONENESS

This is a chapter about something that might have happened, and even started to happen, but, in the end, did not. How this potential was not realized highlights what actually did happen. This chapter, then, is about how accident proneness did not become a standard disease syndrome and was not incorporated into medicine by way of psychiatry. That is, psychiatrists in particular did not adopt and use the syndrome of accident proneness as a distinct disease or a way of defi ning human differences.1 Instead, as I shall show in the next chapter and elsewhere, accident proneness was sometimes used in psychiatry and other parts of medicine and psychology as a sign that some other disease process was operating, of which inclination to accident was only a subordinate part or a symptom. In part this subordinate status derived from the nature of accident proneness. A single accident could be a symptom. A series of accidents, i.e., symptoms, collected and connected together into a history of accidents, would constitute a syndrome, but not necessarily a disease.2 Mentions of the syndrome of accident proneness, it is true, appeared in medical publications, which should have constituted a step in medicalizing it. That is, a few psychiatrists and some other physicians wrote descriptive articles or applied the concept to their particular areas of ac122

How Psychiatrists Did Not Medicalize Accident Proneness

tivity. 3 But accident proneness as such did not become an integral part of psychiatry and, thereby, medicine.

acciden t proneness a nd ps ychi at ry Accident proneness appeared precisely at the time when psychiatrists were moving out into society to deal with non-institutionalized people who were deviant. Neurologists had first dominated the realm of the neuroses, but psychiatrists joined them in the early twentieth century. As part of the mental hygiene movement, a number of physicians specializing in nervous and mental diseases were working to understand, prevent, treat, and cure behavioral deviations. Typically, then, psychiatrists would have acted as the agents for medicalizing accident proneness.4 Physicians, and psychiatrists in particular, had medicalized other patterns of behavior, such as alcoholism and much juvenile misconduct, not to mention mental retardation and hysteria—and even, amusingly, some phenomena that were probably not deviant, such as menopause and sexuality. One such psychiatric medicalization of a behavioral pattern from a later period, for example, with some striking parallels with accident proneness, was chronic fatigue syndrome. 5 Some commentators accused psychiatrists, again in particular, of expanding their domain to an extreme and thus medicalizing many areas of human existence. 6 The case of accident proneness, then, stands as an exceptional negative instance in which psychiatrists did not extend their professional and cultural authority over an area of human difficulty. Moreover, a number of other concerned professionals, as has already been noted, attempted, first, to identify personality configurations that carried the aberrant syndrome and, second, to find the cause of the deviance, which had such ominous social importance. Accident proneness therefore raises the question of where psychiatrists would draw the line at what they considered a medical condition. It also raises the question of who else might step in and utilize the syndrome. Indeed, it was the case that psychologists, who had formulated the concept in the first place, retained their possession of expertise concerning accident proneness.7 Psychiatrists could find accident proneness of interest on several counts. 8 First, the syndrome might serve as a sign or symptom of some specific organic disease, particularly one affecting central nervous system functioning. Second, work with mental illnesses might provide concepts that would make the trait of accident proneness understandable. Finally,

123

124

ch a p t er 7

psychiatric treatment, especially psychotherapy, might offer a means to cure a person of his or her tendency to suffer injuries and make errors. At the time of World War i, when the concept appeared, there were of course already specialists available with an explanation for why people made errors. These experts were the psychoanalysts, who were largely physicians who worked clinically and theoretically with disabling and damaging personality traits.9 Indeed, Marbe himself, in his 1926 monograph, cited Sigmund Freud’s Psychopathology of Everyday Life (first published in 1901) as authority to suggest that dreams could put a person in a frame of mind to have an accident, just as, Marbe thought, suggestion and fatigue could.10 And one stream feeding into psychoanalysis was individual differences in psychophysiological testing of mental patients, as for example differences in reaction times in different kinds of mental patients, a stream that, as I have indicated, Marbe knew well.11

men ta l he a lt h in t he m achine age By the beginning of the twentieth century, physicians in general had come at the problem of accident victims from a number of different points of view—beyond the obvious need to patch up and treat injured bodies. Particularly with the growth of railways and factories in the nineteenth century, physicians became involved in the insurance and legal as well as medical aspects of accident cases. As time passed, large industrial units hired physicians to look after their employees, and industrial medicine became a recognized medical specialty. In particular, beginning in 1884 in Germany, social insurance organizations brought physicians in to determine whether or not claims for injuries were legitimate. Late in the nineteenth century, neurologists and psychiatrists, especially, were therefore called upon to pass judgment not on the injury itself, but on the effects of the trauma of being injured. Some of the damage appeared to be general damage to the nervous system, causing “functional” diseases such as the notorious “railway spine.” And some of the aftereffects of an accident appeared to be functional but in the nature of a traditional diagnosis of hysteria or shock or, more scornfully, “pension neurosis.” 12 Such cases concerned the sequelae of injuries, however, rather than what caused or explained accidents.13 Nevertheless, recognizing the psychological factors in disproportionate aftereffects of injuries helped prepare the way to extend the idea that psychological factors might be operating in causing an injury.14 Heinrich Meng, a psychoanalyst in Frankfurt in 1927, for example, believed that

How Psychiatrists Did Not Medicalize Accident Proneness

children who developed neurotic disorders after an accident had neurotic processes already in place before the injury, but he did not mention the new idea of accident proneness or connect the neurosis directly as a cause of the accident. By 1932, a New York neurologist, James H. Huddleson, was also convinced that neurotic tendencies predisposed some injured people to suffer post-traumatic difficulties. He could reconstruct the psychological dynamics in such cases and posit types of personalities subject to disabling neuroses after they were injured. Although Huddleson was acquainted with the phenomenon of accident repeaters, and although he cited Freudian ideas of repetition compulsion, he still made no connection except to say that people prone to accidents should be weeded out of sensitive positions like that of bus driver—and to note further that a tendency to have accidents was evidence that a neurotic tendency was present that would result in a neurotic reaction after an injury.15 As noted earlier, industrial physicians, even if not specializing in mental and nervous afflictions, in their everyday practices certainly did repeatedly run across people who suffered a series of accidents. One North American railway surgeon in 1904, for example, reported the case of a railway worker who suffered a series of minor injuries at work and developed a paralysis of the lower limbs. Eventually the worker received a settlement from the company, slowly recovered, and resumed work elsewhere. The physician concluded, “Now was this purely a case of malingering or was it hysterical? ” 16 It is no wonder that when competent clinicians were faced with such cases, particularly when more than one accident was involved, they could, without resolution, think of both fraud and some mental or psychosomatic disorder (hysteria).17 Beginning in the 1920s, a few medical specialists began writing about what they called industrial psychiatry. They distinguished industrial psychiatry from psychology precisely because of the clinical focus of psychiatrists—as opposed to the psychologists’ use of mental testing to classify employees into general categories or their use of experiments on the general effects of working conditions on productivity. That is, unlike psychologists, psychiatrists proposed to solve problems by working with individual workers who had mental problems or mental diseases. In 1931, psychiatrist Kenneth E. Appel of Philadelphia offered the following instance of the way in which a psychiatrist focusing on personal problems of individual patients could solve a workplace problem: A young girl at work became unable to concentrate, and from being a superior office worker became careless and inefficient. A nervous

125

126

ch a p t er 7

break-down followed. It was not the work that she could not stand, but the incessant and selfish demands of a mother who could not see her child growing independent of her, and becoming free and happy. Giving help and understanding to this girl enabled her to return to work on her old level of efficiency.18

In such ways, advocates and practitioners of industrial psychiatry, who might particularly have had an interest in accident prone people, in fact had another approach to upgrade the safety of workers: they tended to embrace the idea of solving all workplace problems by improving the mental health of individual workers.19 Any workforce contained people who were dissatisfied, unsuccessful, unproductive, and even disruptive. Industrial psychiatrists claimed that by targeting such cases for treatment, they could enhance overall morale and increase output.20 As the director of the Boston Psychopathic Hospital explained in 1929, psychiatrists could focus not upon “impersonal processes” such as disease, but upon the personal problems of workers, just as “in dealing with delinquents or criminals the tendency is not to be content with treating the delinquent act merely by the application of traditional methods, but rather to study the individual delinquent and the situation to which he has responded and to consider the best procedure in the individual case.” 21 Psychiatrists in industry saw cases that were no different from those that would be seen in an ordinary dispensary. In that sense, industrial psychiatrists had no special role other than, like any physician, general maintenance of the health of the workforce. In the early years of the new concept of accident proneness, that new concept simply did not penetrate into industrial psychiatry to any significant extent. In 1927, for example, Millais Culpin, a British psychiatrist who wrote about adjustment and other dynamic approaches, actually collaborated with Eric Farmer. One of the symptoms that Culpin noticed in patients was that some repeatedly had accidents—but in his view this happened only in the context of many other signs of psychoneurosis. Or the director of the Mental Health Clinic in Pittsburgh, psychiatrist Edward E. Mayer, as early as 1929 expressed special concern about accident proneness in a context of safety programs: “Recently I asked a mill surgeon what happened to a workman whom he recognized as accident-prone. His answer was that he had the man discharged. His organization makes no attempt to fit a man for another job or to adjust him to his work.” Other fi rms, Mayer continued, used industrial psychiatry and psychology to improve working conditions. He cited the Cleveland rapid transit system, which reported “how accidents can be

How Psychiatrists Did Not Medicalize Accident Proneness

reduced and employees made efficient, even if they have been repeaters in accidents and seem to be accident-prone.” But, again, Mayer was emphasizing safety programs and the general mental health of workers and did not construe accident proneness as a special psychiatric problem.22 Despite their awareness, then, psychiatrists were operating under a disadvantage. Where psychologists could speak very specifically about matching the employee to the job, nervous and mental disease specialists in medicine talked about the employee’s “mental status.” Where an employee’s negligence was blamed for an accident, noted one physician, “might we not find the responsibility to be partly on the part of the employer in not having predetermined the right type of work for the man’s particular mental capacity? ” 23 But since predetermination of suitability of the worker already lay primarily in the hands of psychologists, who had tests for vocational abilities, vague individual psychiatric assessment of idiosyncratic workers did not offer an effective leverage to keep accident proneness in the realm of psychiatry and medicine. Indeed, in 1929, one psychologist used accident proneness as part of the evidence for his argument for privileging industrial psychology, which was objective and general, over industrial psychiatry, which dealt with the clinically abnormal individual.24 It was possible, of course, for a psychiatrist to use the psychologists’ tests within the physician’s clinical examination.25 Such was particularly the approach of an American psychiatrist, V. V. Anderson, the author of a widely cited pioneer work, Psychiatry in Industry (1929) and a physician for the Macy department store in New York. He used psychologists’ tests for vocational placement, and he wrote of accident prone employees. But Anderson did not use these employees as paradigmatic examples of the general problem, an individual tendency to incur accidents.26 Psychiatrists like Anderson could, therefore, portray psychologists’ work, including the idea of accident proneness, as subordinate to psychiatry. Anderson wrote about “how valuable psychological measures [tests] are as an aid in enabling the diagnostician to gain certain sidelights on his case. But never in any instance do we believe they can be trusted to select and place people for us—a program involving the diagnosis and understanding of the total personality, and the adjustment possibilities of the whole individual.” 27 Moreover, psychiatrists were at a disadvantage relative to psychologists in another way. Psychiatrists could talk about the misfits in industry who would make errors and have accidents, but they often characterized lack of fitness, or lack of adjustment, as part of a person’s personality. The

127

128

ch a p t er 7

problem was that psychologists at that time by offering tests were effectively making the idea of personality part of psychology. Personality was, in fact, as I have noted, exactly the concept that Marbe, particularly, used to frame accident proneness. The models that many psychiatrists were using in the 1920s and 1930s did lead them to sort people into general personality groups or even types, as has already been mentioned was the case with other professionals who mentioned accident proneness. The German psychiatrist, Eugen Kahn, for example, in the 1920s wrote that both constitutional factors and developmental and environmental influences went into the formation of a person’s personality. The personality therefore consisted of a number of Anlage or dispositions toward types of actions. He was not thinking of having accidents, however, but rather of more general patterns of behavior so that the psychiatrist would have to deal with the abnormal versions of the anxious, the excitable, the affectively cold, the impulsive, and many other kinds of people.28 In a similar way, Herman M. Adler of Boston found that people with employment difficulties fell into three categories: the paranoid personality, the inadequate personality, and the emotionally unstable.29 Such psychiatric writers as Kahn and Adler in this way used the idea of personality—personality type—to delineate people who were in some way deviant in their social functioning. 30 Accident proneness could thus be an element in some personality types utilized by psychiatrists. It was of course possible, as I have suggested, to write about patients whose mental deviance manifested itself as maladjustment to work. A Johns Hopkins psychiatrist, Esther Loring Richards, in 1934 wrote in general terms about people who were personally unsuited to their vocational positions and who consequently developed mental problems. 31 But it was also possible for a psychiatrist to focus on individual maladjustment so as to function in the interest of economic efficiency or, more directly, in the interest of increasing production by managing workers in the workplace. In the 1920s, output, worker turnover, strikes, and dissatisfaction were therefore managerial concerns to which psychiatrists might apply their special qualifications. Indeed, such major labor concerns tended to distract physicians’ attention away from the question of accidents and the people who had accidents. Instead, psychiatrists wrote about general mental hygiene among workers as a means of creating a happy workforce. 32 If they did mention accident proneness, there was an undercurrent in any discussion of the subject: how could managers, advised by psychiatrists,

How Psychiatrists Did Not Medicalize Accident Proneness

deal with workers whose deviance took the form of having accidents and making errors? Often in this time period, as I have suggested, discharge of such workers was taken for granted—but not necessarily by psychiatrists. One early writer, for example, did not advocate weeding deviants out of the workforce but, rather, treating or in some way adjusting them. In his one brief mention of accidents, he noted that “epileptics are always a menace to themselves and others. Should be carefully watched and kept at work in a place where he will not increase the accident hazard of his fellow workmen.” 33 Clearly the potential for discharge was present when psychiatrists, working with individuals, identified problem workers to management. Despite this potential, possibly interested psychiatrists did not in fact set about identifying accident prone workers for either treatment or transfer or discharge. Nor did psychiatric writers take up the syndrome in other contexts as they might have. That is, they could take some interest in accident prevention, but, over time, psychiatrists just did not write about people who repeatedly had accidents. 34

sel f -dest ruc t i v e beh av ior a s a n opening in to ps ychi at ry There was a category of patient that physicians had long claimed, one that might have served as a model with which psychiatrists would investigate accident prone people: individuals who injured their own bodies apparently deliberately rather than unintentionally. Suicide tended to be a special category, but for centuries physicians also had reported with remarkable (if not morbid) interest cases in which patients swallowed bizarre objects and substances or mutilated themselves (figure 7:1). A Bristol surgeon in 1921 reported the case of a woman of eighteen whose behavior was unfathomable: that she for two years could “inflict such injuries upon herself as to push a thin bone needle from the right iliac fossa into the bladder, simply to escape work or to excite sympathy (of which she got very little . . .), seems hard to believe, yet the only evidence to be found of unsoundness of mind is furnished by these actions. Her behaviour and conversation were otherwise normal.” Moreover, such cases were marked by repeated, rather than single, instances of self-mutilating or damaging actions—hence over the years in medical publications the fascinating depictions of collections of objects removed, for example, from stomachs. 35 Sometimes physicians speculated about motivation in such cases, as, in the case just quoted, perhaps attempting to “excite sympathy.” “A Form of

129

130

ch a p t er 7

figure 7:1. In 1900, a medical journal carried the case report of a young seamstress who was admitted to a hospital in Aachen with great, irregular blisters on her hand, as shown in the photo. She also had had such blisters elsewhere on her body. The physicians fi nally found a thread covered with caustic ointment that she had covertly used to cause the blisters. When it was taken from her, the blistering ceased. The diagnosis was an established medical category, self-destructive acts caused by hysteria or, more loosely, mental unbalance. Source: J. Eversmann, “Ein Fall von Selbstbeschädigung auf hysterischer Grundlage,” Münchener medizinische Wochenschrift, 47 (1900), 290.

Self-Mutilation of the Penis in Young Boys,” from1915, carried an implicit suggestion as to why a habit persisted. But for psychiatrists, anything beyond manipulative malingering could be included under a general selfdefining term such as “pathological impulse” or psychopathic behavior, as was suggested by another writer on the subject, a physician from the state hospital at Görlitz in Germany. 36 Such thinking, as I shall explain, did in fact eventually lead a few psychiatrists to consider accident proneness. Nor is it any wonder: the idea that people’s “subconscious minds” caused unintended, damaging behavior was commonplace in Europeanized countries, as Lauer, the psychologist who was testing drivers, as early as 1934 pointed out in connection with accident proneness. 37 By the 1930s, then, physicians, and psychiatrists in particular, who became aware of the problem of accident proneness had two distinctive ways of approaching the syndrome: that of the psychologists, with their tests, and that of the psychoanalysts. In each case, however, physicians ultimately focused on individual patients so that any wisdom or treatment was clinical and individual in nature, rather than greatly advancing ideas about the general phenomenon of people who repeatedly made errors and suffered injuries. Otto Löwenstein of Bonn, for example, in 1934 directly addressed the problem of “individual accident proneness” from the point of view of neurologists and psychiatrists. He noted that there was “no considerable

How Psychiatrists Did Not Medicalize Accident Proneness

literature on the subject,” and so he reported on his own work on the topic. His work turned out to be oriented toward industrial accidents, and he used psychological and physical tests to try to determine what would cause an individual to have repeated accidents when others had different “psycho-physiological qualities” and did not suffer repeated accidents. Löwenstein traced the problem to “individual differences” in “muscular tone,” and he devised instruments to measure this psychophysiological quality, using means that were identical to those that psychologists had been using for some time (figure 7:2). Among his subjects was a painter who kept falling off his ladder and a young mechanic who had a series of “serious accidents.” Löwenstein concluded that accident proneness could be detected, whether or not the cause was constitutional or a disease or incipient disease. In any case, he wrote, the “diagnosis and treatment are matters for the doctor.” And whether or not therapeutic measures would help— depended upon the individual patient. Regardless, he concluded, accident proneness was a matter not only for medicine but specifically for neurological/psychiatric study and treatment. 38 But Löwenstein was almost unique as a physician who would take up the subject. His work did not attract the attention of his colleagues. Two years later, a psychiatrist who was becoming part of the Freudian community in the United States, Karl Menninger, wrote about the theme of mysterious self-destructive actions, and he came close to making the

figure 7:2. How a psychiatrist ended up using somatic signs to detect a syndrome that could have been psychiatric: apparatus to measure physiological responses to see if an individual had a “muscle tone” that would predispose to having accidents. Source: Otto Löwenstein, “The Nature of Accident Proneness,” Industrial Safety Survey, 10 (1934), 3.

131

132

ch a p t er 7

case for accident proneness as a standard element in psychoanalysis. He had first published a paper on suicide, showing how suicidal people had unconscious (dynamic) motives that mobilized impulses to kill and to be killed as they turned guilt and hatred back upon themselves. 39 Then in a second paper, he took up reports of self-mutilation—the phenomenon that, as I have noted, continually fascinated physicians. As an emerging psychoanalyst, Menninger contended that acts of self-mutilation were psychologically meaningful. The motives for self-mutilation, he wrote, were unconscious and symbolic, and, in that form, the actions could be understood as the person’s punishing himself or herself.40 As he discussed self-mutilation, Menninger was able to draw upon some psychoanalytic case material, particularly concerning psychotic patients. He also cited psychoanalysts’ analyses of religious and folk customs reported in anthropology, such as rites of circumcision and castration. Western patients frequently injured their genitalia as a punishment for actually or in fantasy having committed some forbidden sexual act. Another common version of the same theme was patients who damaged their hands because they had masturbated with those hands—but were unaware of the connection between the impulse to mutilate the hand and ideas of guilt connected to the fingers that had carried out the dreaded act. Or in one case, a woman in a psychotic state had murdered her baby with a hammer and then later had escaped and arranged to be run over by a train in such a way that the forearm and hand that committed the murder were cut off on the train track—after which the woman recovered.41

ps ychoa na lysis of t he singl e acciden t It was therefore but a small step to thinking that people who suffered injuries that appeared to them and others to be without meaning, injuries produced simply by chance or “bad luck,” in fact were carrying out unconscious wishes in symbolic form. Menninger in his third paper directly confronted such so-called accidents, “which upon analysis prove to have been unconsciously purposive. The paradox of a purposive accident,” he continued, “is more difficult for the scientifically-minded person to accept than for the layman who in everyday speech frequently refers sardonically to an act as done ‘accidentally on purpose.’ ” 42 Menninger’s paper on accidents has two important aspects. First, he was claiming a special place for psychoanalysis in understanding why accidents happen, that is, they are unconsciously symbolic actions. Second,

How Psychiatrists Did Not Medicalize Accident Proneness

as an aggressive and enthusiastic analyst, he was claiming the phenomenon of accidents for not only medicine and psychiatry but psychoanalysis. In one place, he wrote of “the way in which the individual may be obliged to carry out the dictates of the super-ego through the utilization of ‘accident.’ ” Only at the end of the paper did Menninger quite incidentally observe that the motive for an accident could continue and cause a person to have additional accidents. Even then, however, Menninger did not refer to the by-then common idea of accident proneness. His focus was on the motive that caused a person to have any accident.43 Menninger traced the idea of purposeful accident to Sigmund Freud’s Psychopathology of Everyday Life (1901), the book that Marbe had cited in 1926. There Freud made the case that all human actions have determining causes that psychoanalytic investigations can uncover, whether or not the human is conscious of his or her motives: “Anyone who believes in the occurrence of half-intentional self-injury . . . will be prepared also to assume that in addition to consciously intentional suicide there is such a thing as half-intentional self-destruction (self-destruction with an unconscious intention).” Freud differentiated between bungled actions (clumsiness), including “falling, stumbling and slipping,” and “ ‘chance’ actions.” Chance actions have no avowed reason—they just happen, “accidentally.” And errors, Freud added, could fall under the same category as injuries. Having already established mistakes in speaking—slips of the tongue—as motivated actions, Freud simply extended the concept to human activities of all kinds.44 Freud drew on the 1895 work of Rudolf Meringer and Karl Mayer on lapses in language and slips of the tongue—the same work that Marbe cited in his work on linguistics. The lapses and slips, Freud believed, could be explained by motivated interruptions of mental association processes. He wrote, “If slips in speaking—which is clearly a motor function—can be thought of in this way, it is a short step to extend the same expectation to mistakes in our other motor activities.” And he therefore included motivated breakages along with slips and accidents. But Freud’s successors like Menninger did not usually pursue all of Freud’s psychological fine points.45 What is striking is that, even though he could quote Freud, Menninger had great difficulty in finding cases in the psychoanalytic literature to illustrate his contentions about motivated accidents.46 Menninger did find a case reported by the early Berlin psychoanalyst, Karl Abraham, of a young woman who suffered so many self-injuries that she raised suspicion that they were not mere accidents. Abraham gave other instances of un-

133

134

ch a p t er 7

consciously motivated injuries, and he tied them to instances of indirect sexual gratification by means of a variety of psychological and physical traumas. Such incidents, in Abraham’s view, were secondary symptoms of more primary psychological processes. What he found most striking was the fact that such experiences often happened to patients not just once, but repeatedly: “frequent traumatic experiences (not only of a sexual kind),” thus distinguishing those individuals, even beyond those who claimed a pension, for “it is not uncommon for persons who have just had an accident to meet with another.” But Abraham did not pursue this line of thinking further, and it lay fallow in the psychoanalytic literature.47 Menninger, in his article, explicitly tied three cases of unconsciously motivated accidents, which he was reporting, to self-mutilation. But most of his illustrative material concerning accidents, he was abashed to confess, had to come from anecdotes extracted from press reports, and Menninger then had to add to each one his inferred interpretations of guilt or hatred or other unconscious motives that led to an apparently chance or accidental event. Yet Menninger had clearly opened the subject of motivated, “purposive” accidents, labeled as such, in the psychoanalytic literature. He prefaced his discussion of motivated self-injuries with a memorable personal account: “I recall that I was once seated at a formal dinner by a woman for whom I had some dislike, which, however, I resolved to blanket completely so as not to spoil the conviviality of the party. I believe I succeeded quite well until an unfortunate piece of clever clumsiness on my part resulted in upsetting a glass of water over her gown into her lap.” Menninger’s dismay was intensified because he knew that the lady also believed that accidents were unconsciously motivated.48 In the foregoing incident and in the instances that Menninger and others cited, they were able to trace single accidents to a particular set of motives. Yet they did not explicitly show that a series or pattern of accidents or injuries came from the continuing influence of that set of motives. In short, they were all writing about accidents, not about accident proneness. Abraham, for example, wrote that “the very frequent anxiety-dreams of persons injured accidentally fall into line with Freud’s theory of wish-fulfillment. The unconscious is untiring in its efforts to give expression to a complex.” It made no difference to Abraham that the accident was repeated in a dream, not real life. He was focusing on the motive.49 It still remained, therefore, for a psychoanalyst or a dynamic psychiatrist (typically one influenced by psychoanalytic thinking) to put together the idea of a motivated injury and “repeaters”—and thus claim the idea for

How Psychiatrists Did Not Medicalize Accident Proneness

medicine. One problem, of course, was that psychologists, as well as physicians, used dynamic constructions of the causation of injuries and errors, repeated or not. Another problem was that, as in the cases of Abraham and Menninger, the search for motive obscured any pattern of a repeated behavior—essential of course for accident proneness. From a dynamic point of view, an accident could constitute just another psychopathological phenomenon, like accidents of speech or clumsy actions. 50 Multiple or repeated accidents did not constitute a category. 51 At best, dynamic psychiatry thinkers simply wrote about accidents as isolated phenomena. 52 If, typically, dynamic psychiatrists did not write about multiple accidents as a category, yet the potential was always there. Wrote an industrial psychiatrist with the Metropolitan Life Insurance Company in 1940, “I have come to believe that an accident is a telltale symptom of emotional illness.” She discussed dynamic and medical factors in industrial accidents, but only by the way and near the end of her article did she show that she assumed the idea of accident proneness. At that point, she suggested targeting accident repeaters as such, as opposed to looking at the psychological circumstances of each individual accident. 53 Again, this was an isolated passage without noticeable effect in the literature.

du nba r, ps ychosom at ic medicine, a nd t y pes The potential for a dynamic approach to accident proneness ultimately worked out, however, in an unexpected area. In the mid-1930s, when Helen Flanders Dunbar was doing the work that established her as the most obvious originator of psychosomatic medicine, she and her collaborators studied a very large group of cardiac and diabetic patients who were hospitalized. For a control group, they chose fracture patients, who, colleagues assured the team, were the most “normal” patients available. The investigators found—obviously without any intention—that their control group manifested strong psychological determinants for their injuries and, in addition, a psychoneurotic tendency to have more than one accident. As alert scientists, they connected their findings with some of the literature on accident proneness. Accident proneness thus became a significant stream in the literature on psychosomatic medicine and, by this means, in a small way in psychiatry. 54 After that, Dunbar, in the midst of her other activities, published repeatedly on the subject of accident proneness, usually as another example of the way in which dynamic psychiatry could assist in the treatment of organic illnesses such as ulcer and diabetes—including now the standard

135

136

ch a p t er 7

organic affliction of injuries from “accidents.” By 1939, Dunbar was citing not only major statements from proponents of the idea of accident proneness but Menninger’s article on purposeful accidents. During World War ii, she published specifically on the subject of accident prone people, to whom she now extended a tendency to make errors. Such people manifested, she wrote, “the ‘syndrome of accident proneness.’ In older terminology,” she continued, “the syndrome was called ‘accident habit,’ which is really more accurate, because a person may be occasionally accident prone as a result of a specific situational and neurotic crisis without having an accident habit.” 55 The use of the term “habit” permitted Dunbar to include accident proneness unambiguously in the personality, for habit was traditionally an essential part of the idea of personality. In psychosomatic medicine, then, Dunbar and others pointed out how personality factors could be very important in the development of apparently organic diseases. As Robert C. Powell points out, Dunbar represented a holistic approach to psychosomatic medicine. Although she passed at the time for a psychoanalyst, she traced diseases to personality types. More than the psychologists, she included dynamic interpretations, but she drew on ideas of psychobiological holism and, like some German physicians (examples of whom were noted above), she developed the typological point of view. For her, psychosomatic medicine indicated how a physician might intervene with a patient who had a personality type that would lead to disease. By contrast, other analysts, like Menninger, focused on the etiology of particular symptoms in particular patients. And, as I noted earlier, still other psychologists and psychiatrists emphasized, at least in accident proneness, somatic flaws such as eyesight and reaction times.56 In 1943, Dunbar set out a full case for the accident habit and personality as factors in fractures—and, by inference, other accidents. In a long chapter in a book on psychosomatic diagnosis, she described at length how dynamic factors set up patterns of having accidents. For her purposes, a single accident would make the point—except that as part of a personality pattern, psychological determinants were more convincing when they fell into a pattern such as having multiple accidents. She cited the studies that showed that some individuals—typically in an industrial setting—had many accidents and others had none. And then she gave numerous illustrative case histories to impress her point upon the reader.57 Again and again in her case histories, patients who were injured turned out to have a history of accidents. A woman who was conflicted, for example, fell off some steps and broke her hip so that she could avoid

How Psychiatrists Did Not Medicalize Accident Proneness

confronting her actions. Then Dunbar added, “Although this was the first accident requiring hospitalization, she gave a history of many minor injuries.” Another case “came in with a compound fracture of the left elbow . . . At 14 he had accidentally stuck his left index fi nger into a machine . . . At 8 he had fallen off his bicycle and broken his left wrist. In general he was known as the kind of person who was always getting hurt.” Dunbar clearly was suggesting that having accidents was parallel to the process by which other patients transformed psychological conflicts into persistent somatic symptoms, as in hysteria.58 Dunbar divided the patients with accident habit into four groups. The first group consisted of those who kept suffering injuries to the “same member, whether by burning, cutting, or fracturing.” The second group “showed a definite predilection for accidents of a specific type, such, for example, as automobile accidents whether as driver or pedestrian.” A third group were those described as careless, unlucky, or “all thumbs,” incurring a wide variety of injuries. And the fourth group consisted of those who had a history of falling and of taking risks: they might claim that they had fallen down stairs “twenty-five times before and never got hurt till now.” 59 Dunbar then noted that although generally well adjusted, for example socially and sexually, nevertheless these patients had personality patterns and neuroses that translated into an accident habit. They tended to have some conflicts with authority, albeit well masked. “When thwarted, deprived, or subjected to some strain such as unemployment or a motherin-law living in the family, patients who developed ‘accident proneness’ had the tendency to do something either to modify the situation or to get away from it instead of just keeping their anger bottled up and boiling inside.” Dunbar found the patients sociable but restless. They could express their aggressions in an accident habit.60 For years, Dunbar continued popularizing the idea of accident prone people. Her writings also appeared in translation in many languages. In 1948, the editors of Science Digest, a magazine of science popularization, ran an excerpt from her new general book on psychosomatics, including an account of Nick, an accident prone dog. 61 Colleagues in psychiatry and other medical specialties cited Dunbar’s work and her ideas about accident prone people, sometimes using her term, “accident habit.” 62 In the realm of psychoanalysis proper, for example, the authoritative handbook for the orthodox, Otto Fenichel’s The Psychoanalytic Theory of Neurosis (1945), carried a noncommittal one-sentence paragraph in the section on character disorders, “Accident-prone patients have been studied by Dunbar as a specific type of personality.” 63

137

138

ch a p t er 7

Despite the efforts of dynamic psychiatrists, typological or otherwise, the syndrome of accident proneness therefore still did not fi nd a home in psychiatry except as an occasional element in eclectic psychiatric practice.64 As Powell points out, by 1945, Dunbar’s approach to psychosomatic medicine was being superseded by a non-holistic psychosomatics based on specific etiology of individual cases—in the style of most psychoanalysts. Writers cited Dunbar for the general idea of accident proneness and the existence of the phenomenon, but not usually for any detailed content or method of interpretation. 65 Popularizations of psychoanalytic interpretations of accident proneness continued to appear, however, as an incidental part of the popularization of dynamic thinking.66 During the rest of the twentieth century, within mainstream psychoanalysis itself, motivated accidents continued as an occasional theme. 67 What is striking is that psychoanalytic writers did not usually refer to the idea of accident proneness even when they described patients who repeatedly had accidents. A 1950 French practitioner described “women who fall,” even tracing the pattern back to their childhoods—but with no reference to the accident proneness idea or literature.68 In the mid-1960s, a group of Southern California psychoanalysts studied psychodynamic patterns involved in fatal accidents. They were aware of the idea of accident proneness, but they distinguished “suicide prone,” that is, self-destructive, from “accident prone” individuals. More commonly, those people who suffered fatal accidents were impulsive, and their fatal impulsivity was transient and did not, according to the investigators, constitute a consistent life pattern. Their psychoanalytic colleagues generally reverted to seeing each case as simply the outcome of unique psychodynamics, with general psychodynamic patterns not tied specifically to accident behavior.69

a bsence of t he s y ndrome f rom ps ychi at ry mor e gener a l ly After midcentury, the term “accident proneness,” and the idea, continued to appear occasionally in general psychiatric writings.70 More often, however, it was missing.71 At the very least, as a syndrome, accident proneness might have appeared in the second half of the twentieth century in formal medical descriptions of symptoms and illnesses. But it did not. That was a good indicator of the failure of the idea to penetrate into the specialty.

How Psychiatrists Did Not Medicalize Accident Proneness

The World Health Organization Manual of the International Statistical Classification of Diseases, Injuries, and Causes of Death in 1948 did not list the syndrome of accident proneness. Accidents per se were classified by either the agent of the injury (like an automobile or carbon monoxide) or the nature of the injury (to the arm, the bronchi). Psychiatric categories in the manual included various psychoneuroses, including character and behavioral disorders, but not a behavioral pattern of inclination to have accidents. The 1965 revision had more general headings under personality disorders, but accident proneness was not specifically mentioned. The 1975 revision, even where there were much more detailed descriptions of personality disorders, did not include accident proneness there or anywhere else. In the 1952 Diagnostic and Statistical Manual of the American Psychiatric Association (editions of which came to be widely used), there was a category, “personality disorders,” that could easily have included accident proneness: “cases in which the personality utilizes primarily a pattern of action or behavior in its adjustment struggle, rather than symptoms in the mental, somatic, or emotional sphere.” 72 This dynamic formulation, however, did not generate a mention of accident repeaters as such. By DSM-II in 1968, there was a category, “inadequate personality,” that could have included accident proneness—but accident proneness again was not specified. The potential was clearly there, however. One stipulated disorder in the pediatric section, for example, was a behavioral pattern, “runaway reaction of childhood.” DSM-III (1980), however, moved away from dynamic formulations, and it would have been harder to find a home for accident proneness in this new psychiatric nosology, which tended to be descriptive rather than pathological. One historian wrote of a “narrowing of the psychiatric gaze” in DSM-III that seemed permanent. Indeed, the concept of neurosis barely survived in DSM-III. Kleptomania, for example, was described as a failure to control an impulse, not a behavioral deviation based on personality such as accident proneness had become.73 In none of the formal psychiatric classifications, then, did accident proneness appear as a pathological syndrome. Moreover, it did not even appear as a sign or symptom of some other disease classification. Where some textbook writers had once included the syndrome, very few did in the closing decades of the twentieth century. In one major American psychiatric textbook, for example, a special chapter on accident proneness lasted through the 1975 and 1980 editions and then disappeared from later editions.74

139

140

ch a p t er 7

Even if accident proneness did not have a place in formal psychiatric nosology, a few dynamic psychiatrists continued to work to refine the concept. Two Detroit physicians in 1963, for example, reported that in industry, workers with illness and accident records needed special attention both for their own good and for reducing disability claims. Before the accident occurs there is a state of confl ict and anxiety within the patient. As a result of this condition the worker fi nds a selfdestructive, injury-producing act which causes his “death” as a worker. From this moment the patient reacts exactly as do other psychiatrically ill people, except for the character of his symptom. Instead of having a presenting complaint of anxiety, depression, or other classical psychiatric symptom, he has the physical disorder which is the result of his accident . . . The glib labeling of this [self-destructiveness] as accidentproneness fails to take into account the real events in an individual’s life. The original concept made the assumption that certain people are always prone to accidents. Our findings do not indicate this at all. They indicate that under special circumstances when certain conflicts exist many individuals tend to cause their own accidents and probably will hold onto the injuries they sustain, because the accidents solve their life problems.75

Moreover, over many years, dynamic psychotherapists sometimes treated patients whose chief presenting symptom was a record of accidents. For those practitioners, official disease classifications were irrelevant. They wanted to find and treat the psychological dynamics that led to the patient’s being injured.76

w h y acciden t proneness fa ded ou t of ps ychi at r ic w r i t ings For decades after about 1950, the patterns that had been established before midcentury persisted. The main area of psychiatric and medical literature in which accident proneness appeared was one fading branch of psychosomatic medicine.77 A number of holistic thinkers in medicine did mention the idea, at least as a phenomenon and as a condition for which practitioners should watch. Some pediatricians, especially, wrote with concern about accident repeaters among the young, as will be noted in the next chapter.78 Occasionally, as I have noted, dynamic or eclectic

How Psychiatrists Did Not Medicalize Accident Proneness

psychiatrists, especially those working in industrial psychiatry, would mention the subject. An International Business Machines psychiatrist in 1963, for example, noted that workers did sometimes suffer a series of injuries, but he was skeptical of any generalizations, for “there are as many accident-proneness patterns as there are individual accident repeaters.” Karl Menninger himself in 1959 emphasized happy environmental conditions as much as individual treatment for accident proneness: “The more mental health, the fewer accidents.” 79 English-language publications overwhelmingly dominated the subject in general, and dynamic formulations in particular. Continental medical writers who alluded to accident proneness did so largely along lines of thinking laid down in the Anglophone discussions of the subject. 80 In both technical and popular psychiatric writings, accident proneness functioned as part of the psychologization of parts of Western culture. 81 In the middle of the century, two New York bus officials (one a physician) wrote of driver selection, “It has been our practice to reject a candidate who presents a psychoneurotic picture, because we have found that under stress these men become accident repeaters.” A 1951 story from a safety engineer was summarized, “Supervisors do not have to be psycho-analysts but they should have an understanding of the reasons why some people are prone to accidents.” 82 All of the familiar types of discussion of accident prone people became quantitatively less and less significant in psychiatry and medicine in the last third of the twentieth century. Partly this shrinkage reflected the remarkable increase in psychiatric publications, so that a number of articles could represent only a very slight visibility. The subject diminished absolutely, however, even for those interested. By 1981, Joseph Connolly of Westminster Medical School in London could comment as an eyewitness, “The literature in English about the subject [of accident proneness] has been quiet for approximately a decade.” 83 One obvious reason for the declining amount of attention was the fact that the midcentury popularity of dynamic psychology-psychiatry itself began to fade. The notorious romance of Americans with Freud came to an end. Opponents of dynamic thinking came to flourish everywhere. An Italian scholar in 1974, for example, denied the validity of the very concept of the Freudian slip and explained errors in terms of linguistics and various kinds of forgetting—not personality patterns. 84 In psychiatric practice, as in the theoretical and clinical literature, accident proneness had little place in the new styles of treatment that dominated the concerns of specialists.

141

142

ch a p t er 7

t ests for a s y ndrome By the time of the American DSM-III in 1980, as Allan Young points out, psychiatrists were using a hierarchy of tests of disease classification. The lowest in the hierarchy of tests was face validity, when the disease entity would fit “the clinical impressions and experiences of experts.” The next step up in the hierarchy was the predictive validity of the description, if and when the disorder developed in a particular pattern that could be foretold. And the third was independent validity, which came when a cause (such as a specific bacterium) for the disease could be demonstrated. 85 In this hierarchy, accident proneness did not by any means have a cause established, for types were not convincing, and dynamic explanations were idiosyncratic for each patient. The personality traits that psychologists established and tested were of course not unique to psychiatry. The question of predictability turned out to be particularly problematic, and not just in psychiatry. By definition, accident repeaters could be predicted to have more accidents. But statisticians by the 1940s had begun to show that one could not predict that a particular individual would have an accident—a subject to which I shall return in the next chapters. The most that could be predicted was a statistical aggregation in a group. These statisticians’ publications eventually caused a substantial decline in attention to accident proneness among psychologists, a discouraging factor that carried over to well-informed psychiatrists. An industrial psychiatrist in 1959, for example, concluded that “accident-proneness as a statistical concept remains very elusive,” and he reverted to advocating personal, case-by-case treatment. 86 Still another problem for psychiatrists was that accident proneness was not easily amenable to treatment other than discharging or transferring workers or denying driving licenses to motorists. In an age in which medicine in general became distinctively more technological, no applicable technology appeared that would fit accident proneness—not even a chemical such as those that were applied to psychotic and then some neurotic patients beginning in the 1950s. Franz Alexander, a psychoanalyst and pioneer psychosomaticist with a different approach from Dunbar’s, as early as 1949 pointed out the difficulty: The psychiatric interview, conducted by an expert, which reveals the whole previous life history of a person is the most, if not the only reliable method. The accident habit develops early in life and manifests itself in the youngster in a noticeable inclination to contract physical

How Psychiatrists Did Not Medicalize Accident Proneness

injuries, even if only minor. Also, the combination of excessive resentment and guilt manifests itself in early childhood in various ways familiar to the trained psychiatrist. To alter such an ingrained emotional pattern as is characteristic for the accident-prone individual by psychotherapy is a major therapeutic task. It requires prolonged treatment and is, therefore, in the present state of psychiatric facilities, of no practical significance. 87

This harsh, if pragmatic, judgment was shared by other physicians, especially after dynamically oriented practitioners lost out to other kinds of psychiatrists. In 1965, two California public health physicians asked bluntly, “Is ‘accident proneness’ a useful concept for medical practice? ” The concept was altogether too popular and general to be practical, they concluded. They pointed out tellingly that because of the difficulties, “another term, ‘accident susceptible’, has been coined to recognize the concept of excessive accident experience, but to avoid the psychiatric overtones” of motivated misfortunes. Indeed, these public health experts ironically medicalized the phenomenon precisely by removing it from the realm of psychiatrists. There were, the authors decided, other factors that explained the phenomenon: changed exposure, statistical chance, “man-machine incompatibility” (the small driver in a big car), physical handicap, inadequate training and experience (the common observation that new workers always have more accidents), indifference to safety, temporary emotional upset, and abuse posing as accident. They did admit that there were some individuals who still had an excess number of accidents—but one should assume that there was some incipient cardiovascular or other disease or physical cause. Anything else was merely “a source of speculation.” 88 In this way, in such thinking, the appearance of accident proneness was medicalized right out of psychiatry and the more general idea simultaneously dismissed as “speculation.” 89 Since physicians were not claiming the idea of accident proneness, psychologists were relatively free to continue to work on accident proneness as an aspect of personality. They could also devise tests and advise parents and the rest of the public about the phenomenon of accident repeaters. Even the question, “Are psychotics accident prone? ” was answered in 1963 not by psychiatrists, but by psychologists. And it did not take a psychiatrist to investigate the socially important question of how soon mental patients could have their driving privileges restored—and whether or not they in fact constituted a significant risk group.90

143

144

ch a p t er 7

That mainstream psychiatrists as professionals and specialists did not in the twentieth century attend to the phenomenon of the person who suffered repeated injuries and made repeated errors suggests one conclusion, however: contrary to the claims of anti-medical alarmists, there were definite limits to what psychiatrists, at least, could medicalize in the twentieth century.91 The fate of accident proneness therefore lay largely outside the realm of psychiatry. For psychiatrists, the problem with accident proneness went beyond their professional competition with psychologists. The syndrome consisted of a series of happenings over a period of time. The syndrome therefore did not fit into the descriptive conventions that, as German Berrios has pointed out, in the late twentieth century tended to exclude the time factor from psychiatric nosology. (Hyperactivity, for example, which was at least partially medicalized, can be observed in the clinic in a way that a history of accidents cannot.) Even dynamic psychiatrists tended to use conventional nosologies, but when they went beyond them, dynamic thinkers combined current mental content with the form of the patient behavior.92 Nor did accident proneness fit into the older standard categories based on faculty and association psychology. A developmental model was possible in the hands of dynamic psychiatrists and psychoanalysts, but with the exception of psychosomatic medical advocates, they did not accommodate their individual motivational analyses to the statistical phenomenon of repeated accidents.93 In the end, then, one major explanation for why psychiatrists did not pay attention to accident proneness was simply that the idea did not fit into psychiatric discourse as it developed in the 1920s and after. Even without professional competition from psychologists, psychiatrists were not set to adopt, explain, and—as Franz Alexander noted—treat accident proneness. Later in the twentieth century, accident proneness did not show up in reports from nervous system visualizing technology. And there continued to be no pharmaceutical specific for accident proneness that could confirm that such an entity existed. For generations, many or most physicians were on some level aware of the orphan idea of accident proneness. But it was a phenomenon that psychiatrists, at least, were not diagnosing, explaining, or classifying. If psychiatrists had defined their disease entities or symptoms differently, accident proneness might have been medicalized. But they did not. Thinking in terms of individual, personalized neuroses avoided the generalizing that other professionals embraced.

8.

THE MIDTWENTIETH-CENTURY HIGH POINT

Quite apart from psychiatrists, then, among people concerned about injuries, the idea of accident proneness reached a high point of acceptance and use in the years after World War ii. For decades in the second half of the twentieth century, the subject continued to interest psychologists, safety experts, and, as I have noted, some physicians.1 By one count, up to 1960, 3,000 articles had been published on the human factor in just motor vehicle accidents. A very large proportion of such publications contributed to the peak of interest in accident proneness.2

t he prof il e of post wa r pu bl ic at ions These publications created a profi le that ran through all levels of discussion. Experts refined the idea and then raised doubts about it. Eventually, among the well-informed, with a delay in the reaction of experts on children, accident proneness changed from a statistical phenomenon to, at most, a clinical problem. At the same time, in textbooks in advanced fields and in other indicators, it is possible to trace clearly how the concept was fading away after the postwar high point. Even as explicit popularization appeared less and less frequently, however, the idea of someone prone to have accidents persisted in general use in the language. After the peak, as the slow decline in total publications on accident 145

146

ch a p t er 8

proneness was just beginning to get under way in the 1960s, the general literature on the subject had become less fully dominated by American writings. Nevertheless, the prominence of American publications continued to be striking. Indeed, postwar Americans themselves seldom paid any attention to non-American writings on the subject. So many experts in the United States took up the idea that South African investigators Lynette Shaw and Herbert S. Sichel were moved to write in 1971 “that in no other country does one find widespread evidence of the sort of belief which apparently gained currency in America, that a handful of accident-prone people must be responsible for the majority of the accidents.” 3 What was most striking about all of the writings at midcentury, however, was the extent to which publications on the subject soon came to be concentrated in the area of traffic accidents and road safety.4 A comprehensive survey from 1971 showed that already about half of the publications on accident proneness were on traffic accidents. That survey also revealed that industrial accidents were attracting a markedly diminishing proportion of the research publications. At any time, a scandalously tiny percentage of effort went into home accidents, which were now accounting for almost half of the actual injuries.5 In traffic, the practical problems of dangerous drivers—possibly accident prone people—persisted and of course grew with the greater use of automobiles. Decline of concern about industrial accidents also came in the context of social change. Factory and mill work diminished in the post-industrial world of many advanced economies. In those countries, more discussions focused, rather, on children’s tendencies to have accidents. And as the social and behavioral sciences blossomed after World War ii, the levels of discussion of accident proneness became more sophisticated. In the end, as I shall describe in the following chapters, systematic accident analysis facilitated a resurgence of emphasizing engineering protection in safety campaigns, an emphasis that additionally drew experts’ attention away from the accident prone person. But that outcome was hard to foresee in the years immediately after 1945.

t he high poin t in t he post wa r y e a r s Evidence of a high point of the idea of accident proneness in the earlyto-mid 1950s was, of course, the number and ubiquity of published treatments of the problem. These publications reflected both professional and popular interest. As a Glasgow industrial psychologist reported in 1952, “Discussions at meetings on Accident Prevention in Industry show that

The Mid-Twentieth-Century High Point

figure 8:1. The “accident prone” came up specifically for a managers’ discussion at the Sperry Gyroscope Company in 1951, pictured here. Their decision: trained supervisors can help accident prone workers. Source: Fred P. Kiefer, “Help for the Accident Prone,” Safety Maintenance & Production, October 1951, 21.

much interest is being taken in this country over the possibility of some people having characteristics that make them more prone than others to have accidents” (figure 8:1). 6 Publications and presentations came largely from safety experts and three different areas that were part of, or overlapped, safety literature in general. The three areas were, much as before the war, medicine (but now particularly pediatrics), psychology, and popularization. So technical had the subject become that popularizations came to include attempts to alert and inform experts, particularly within formal safety efforts, as well as the general public, about the idea of accident proneness.7 One particularly tangible sign of a high point of general interest was a book published in 1954 that was entirely about accident proneness. 8 The author, Norman Roberts Lykes, reported that he had spoken with many American experts in the field, and he quoted most of the standard publications.9 He was able to assemble the knowledge of the day, and on the basis of that knowledge, he advocated a program of laws and individual and group actions to counter the influence of individuals’ proneness to have accidents. Lykes summarized “what we know about accident proneness”: 1. A comparatively small number of people have most of the accidents. 2. The more accidents a person has had, the more likely is he to have others. 3. Accidents are not distributed on a chance basis. Most of them don’t just happen. They are caused by numerous factors.

147

148

ch a p t er 8

4. One of the most important contributing factors in most accidents is an unconscious need or desire to have accidents . . . In a small minority of people a neurotic accident habit has developed. 5. The accident habit or pattern is set in motion by emotional stress, conflicts, or problems which the accident-prone person attempts to solve on an unrealistic, irrational basis—by having an accident.

Lykes went on to enumerate persisting motives that might be involved in bringing on accident proneness in a person, motives such as hostility, guilt, and dependence on others. He urged his readers to work to diminish injuries by focusing on the accident prone person. Both psychotherapy and social actions, including propaganda and punitive laws (especially traffic regulations), he believed, would be effective.10 Lykes offered his readers commonsense advice in simple language: keep a safe home and workplace and do not undertake hazardous activities such as driving, swimming, or operating machinery when one is emotionally upset—or when one has been having a series of accidents. He urged the transfer of employees with accident records or, if transfer was impractical, discharging the employee so that the person would work in a less dangerous industry. “A motor vehicle in the hands of an accidentprone driver is just as lethal as a gun in the hands of a homicidal maniac. The taxicab driver who killed Margaret Mitchell, author of Gone with the Wind, had a whole string of accidents and violations (22) on his record. He didn’t deliberately kill Miss Mitchell. He couldn’t help killing her. He was accident-prone.” 11 Another sign of a high point of attention to the idea was the publication in 1955 of a technical, rather than popular, review of more than a thousand publications on motor vehicle accidents. Wrote the authors, “A rapid glance over the titles in bibliographies on safety and on accidents reveals the prominence which has been given to the concept of accidentproneness. This emphasis is apparent not only in efforts directed towards the analysis of the causes of accidents, but in suggestions for the control and prevention of accidents as well.” The authors then devoted approximately half of their discussion to ways in which accidents tended to appear in significant clusters in certain people.12

r ef inemen ts by t he ex perts Over the years, experts trying to master the technical literature concerning accident proneness began to draw up systematic categories into which

The Mid-Twentieth-Century High Point

their colleagues’ work fell—a sure sign that the concept had become more complex. Especially after the 1950s and 1960s, specialists tended to focus on small areas within the general subject—exactly as one would expect in societies dominated by complexity and specialization. Post–World War ii investigators therefore inherited from earlier workers some basic concepts and hypotheses, mostly those summarized by Lykes. Accident proneness was a part of personality that produced patterns of behavior. The characteristic was persistent, and the victim would not outgrow it.13 Indeed, just after Lykes’s book appeared, another major book, by a Cincinnati physician, Morris Schulzinger, appeared, The Accident Syndrome. There Schulzinger described at length the qualifications and complexities that were beginning to surround the subject. Schulzinger included accident proneness in his survey of why people have accidents, but he also included much else from the standard literature, such as time of day, physical impairment, and circumstance. “In most of the accidents in which psychological factors play either minor or major roles,” wrote Schulzinger, “it is not the accident or the injury per se that the individual is seeking. The accidents are merely an indirect outgrowth of circumstances or situations that the individual gets into because of psychological demands.” 14 One of the first things that showed up in postwar investigations was distinguishing even more strongly than before between accident repeaters and accident proneness.15 Accident proneness, the personality characteristic, was abstract and controversial. Repeaters, people who repeatedly had accidents, by contrast constituted a much more easily verifiable phenomenon (although, as I shall note, many statisticians and safety experts eventually did not see it that way). This shift in language reflected perhaps the most striking modification of accident proneness in the 1950s and 1960s: a growing agreement on the idea, already broached in the 1930s, that although a history of accidents was indeed a characteristic of a person, the condition could be temporary. Statistical studies in particular inspired this new understanding of accident proneness as a transient condition. Yes, one could show in a factory that some people were having more than their share of accidents. But at a later time, using the same group of employees, there was an almost total change of personnel in the accident prone subgroup: the original repeaters stopped having accidents and were replaced by a new group who were incurring a disproportionate number of accidents. “The relatively small percentage of the population that contributes a disproportionate number

149

150

ch a p t er 8

of accidents is essentially a shifting group, with new persons constantly falling in and out of the group,” as Schulzinger explained in an article in 1954.16 It was this idea that shifted investigators’ attention from permanent personality characteristics to temporary personal upset or—later— stress that might cause them to have accidents.17 Seeking tests that could identify people who temporarily or permanently were accident prone continued for decades. The results were still often unsatisfying and, worse, the investigators produced results that were often inconsistent with the results of other competent investigators. As Finnish investigator Sauli Häkkinen noted in 1958, after surveying the findings of his predecessors, “On the whole the results are not too convincing.” One 1955 group of investigators, for example, found that, contrary to other research, “pilots who have a greater number of aircraft accidents than expected . . . tend to be younger, better adjusted to their jobs and status, and more skillful than pilots who have fewer accidents.” 18 Investigators tried to correlate accident frequency with any element for which psychological tests had been devised, such as “empathy,” and some for which special tests were invented, just as had been done before World War ii. They utilized special tests for intellectual and cognitive abilities, for mechanical and psychomotor abilities, for judgment and distractability.19 They tried using whole batteries of personality and psychiatric tests, which were being well developed in that period, such as the famous MMPI (Minnesota Multiphasic Personality Inventory). To a substantial extent, then, investigators fitted the concept of accident proneness to the tests that were available, particularly personality tests.20 Were accident prone people neurotic? Or, as noted earlier, extraverted? Did the individual have spare capacity to avoid accidents? 21 As late as the 1970s, individual ability to tolerate monotony and to process information constituted characteristics that investigators found distinguished low-accident people from high-accident people.22 Three other types of factors particularly went into the post–World War ii concept. One was the medical factors noted earlier, such as sensory perception and neurological and muscular fitness. The second was material from personal examinations of individuals, including not only medical histories and accounts of life course but expert evaluations of mental status. Of particular importance, because they involved tangible evidence and correctable behavior, were specific diseases such as heart disease and the use of medications.23 Epilepsy had long been a problem, and diabetes and stroke could still also cause sudden unconsciousness, notably in drivers. Brain tumors could cause motor and sensory problems that resulted

The Mid-Twentieth-Century High Point

in accidents. Reports of accidents involving these diseases were especially noticeable in areas in which motorcar traffic was increasing.24 The third new emphasis, which also grew with the traffic accident literature, was the conspicuous role of alcohol. Experts had already mentioned the subject repeatedly before World War ii, but after the war, for decades, alcohol was a growing feature in the literature on injuries. All over the world, drunks got injured and injured other people, and their actions became more conspicuous in society, along with accidents in general, as infectious and other disease rates declined.25 Specialists of the postwar period also paid attention to a few relatively new factors to which experts attributed accident frequency. One was the trendy new formal concept, stress. Stress fitted well into the idea that a tendency to have accidents was a temporary condition.26 “It has been well established that the majority of accidents are due to personal patterns of reaction to stress,” wrote a clinician in 1955 who was identifying “transient situational maladjustments predisposing to accident.” 27 Another new emphasis was bringing in social factors that correlated with elevated accident rates. General social class had already attracted investigators’ attention. Poor people, of course, got injured more often than those better off.28 A new type of social factor, however, was the relationship of the accident prone person to other people. Investigators at the time, for example, were struck by the fact that involvement with social agencies such as the welfare and prison systems was substantially correlated with accident rate (figure 8:2).29

figure 8:2. “Social background of accident free and accident repeaters.” Social agency data in 1952 suggested that accident prone people tended to be social deviants. Source: Ross A. McFarland, “Human Factors in Highway Transport Safety,” in Road-User Characteristics (Bulletin 60, Washington: Highway Research Board, 1952), 41.

151

152

ch a p t er 8

But beyond that was broader social analysis, such as peer group dynamics. Two American experts using a sociogram found that workers in a steel mill who were most unpopular suffered the most accidents. Or the morale of a whole work group or even a whole plant would be predictive of accident incidence—an idea that came in part from the continued impact of the famous Hawthorne studies. 30

medic a l r econcep t ua l iz at ion: stat ist ic a l v er sus cl inic a l In fact, however, in the middle of the twentieth century, accident proneness was splitting into two streams, as was suggested in the last chapter. As early as 1951, A. G. Arbous and J. E. Kerrich, South African investigators, recognized that something had happened. Originally, statistical analysis had revealed a group who were responsible for a disproportionate number of accidents (figure 8:3). But, Arbous and Kerrich wrote, “the term ‘accident-proneness’ is now no longer used in the original sense. It is now defined on the basis of a clinical and not a statistical diagnosis, and, as such, is largely a term of convenience, rather than of precise mathematical definition.” 31 One stream of investigators, largely out of psychology, did still use the term even in the original statistical sense, and it was among that community of experts, many of whom were engaged in psychological and occupational testing, that the concept in its original, statistical form eventually ran into a great deal of skepticism and rejection (see next chapter). The clinical modality persisted better, in some ways, however. In both medical and practical settings as well as laboratories, physicians and some psychologists were trying to identify and treat individuals who were accident prone, just as described in the last chapter. Out of the clinical stream, however, came a shift in the status of accident proneness. More and more frequently, first in medicine of course, as noted in the last chapter, but then in other areas, accident proneness came to mean not a condition or disease but simply a diagnostic sign or symptom—one of the standard uses of the idea of a syndrome—“the recurrence of signs and/or symptoms that appear with fair regularity” in a clinical disease state. 32 A clinician at Michigan State University in 1980 wrote of “repeated trauma as the presenting symptom of a pathological grief reaction.” 33 And a later team could add a further concern for physicians. “Repeat accidents” not only increased the chance of further accidents, but they ominously raised “the possibility that the child’s injury may have been intentional.” 34

The Mid-Twentieth-Century High Point

figure 8:3. Distribution of accidents in a General Motors factory in 1948, demonstrating the existence of an accident prone cluster of workers. Source: Ernest Wolff, “Accident Proneness: A Serious Industrial Problem—What Can the Industrial Physician Do About It?” Industrial Medicine and Surgery, 19 (1950), 419, reproduced with the kind permission of Occupational Health and Safety.

It was in the form of a descriptive medical symptom or syndrome, found in an individual, then, that accident proneness had the most appeal to clinicians. Two specialists in 1965 spelled out for a general medical audience how physicians could use the accident prone syndrome: “The concept of ‘accident proneness’ can be meaningful if it is used, not as an etiologic diagnosis, but to designate a sign or symptom indicating a greater than average accident experience, and requiring additional medical analysis.” 35 In 1976, for example, a St. Louis physician in the course of introducing an ordinary case report wrote: “The patient is a nine-yearold boy who was originally referred for persistent accident-proneness as a consequence of which he had broken many bones and several times almost killed himself.” 36 The clinical, or individual, approach meant that an accident history was only part of a unique case, in medicine generally as in psychiatry. A physician facing a patient would be oriented toward treatment and

153

154

ch a p t er 8

would deal with all aspects of the life of the patient—“the individual, his family, his environment and his behavior,” as Schulzinger put it in 1956. Schulzinger contended that “the new opportunity in accident prevention strikingly becomes medical and the need is for an integrated clinical approach.” He explicitly contrasted a multifactorial approach to treating a patient with the idea of focusing on accident proneness. Viewing accident proneness as a symptomatic syndrome in a general disease process, he wrote, has the “advantage of replacing the essentially fatalistic concept of ‘accident proneness’ by a more hopeful theory of accident causation.” 37 Nor was the individual, clinical approach to an accident problem necessarily impractical in an organizational or social setting. In 1968, for example, a New York psychiatrist reported conducting a study of clerical employees by means of psychological testing and a psychiatric interview, still attempting to identify what made the difference between those with an accident history and those who had been accident free. At the end, the most important finding by far was that, regardless of group, the number of lost time days declined from 11% to 3%. Clearly just telling employees that the investigators were studying accidents had a powerful effect, as in the Hawthorne experiments. Moreover, the effect on those employees was a lasting effect. 38 As I have noted, others had reported that one-on-one clinical work with people who had unusual numbers of accidents had paid off, although of course it is necessary to make discount for a tendency of clinicians to report favorable outcomes. Altogether, by the late twentieth century, the literature offered substantial support for treating accident proneness on an individual, clinical basis. 39 As a public health official observed as late as 1999, “injury ‘repeaters’ represent a targetable and high risk sample . . . who could benefit from interventions that exist within currently operating programs.” 40 There was, however, no special therapy to target accident proneness as if it were a distinct disease. It therefore remained a symptomatic syndrome of some other disease entity. Accident proneness in this way moved into mainstream medicine in a small and limited way. Obviously psychiatrists could also use the syndrome or sign as it showed up in a patient, but any physician should be alert for it. “If a person has repeated falls, bumps, cuts, or similar injuries,” wrote a U.S. physician in 1948, “he suffers from a predisposition to accidents and should seek medical help, just as he should if he suffers from joint pains, chronic indigestion, or a neurosis.” This early formulation still had the suggestion that accident proneness was a medical

The Mid-Twentieth-Century High Point

condition, but then this author pointed out that many factors might be involved, and he cited the case of a boy who was falling repeatedly because he had “knock knees and weak ankles.” While exercises strengthened his feet and legs and stopped the falling, it turned out that the basic cause was malnutrition, caused by his parents’ not feeding him properly.41 Physicians and other professionals who worked under the public health umbrella were often part of the larger safety community, and they tended to approach accident proneness from the point of view of epidemiological expertise.42 In 1972, for example, a Council of Europe health committee wrote about the public health problem of children’s accidents, focusing on the epidemiology of fatalities. The committee members urged education and legislation to try to curb the alarming numbers of injuries. In their approach, the committee had no need to posit the idea of accident proneness, and the only time that they used the term, they used it in a neutral descriptive sense and without the connotations that the phrase had acquired: “Children between 4 and 10 years of age are the most accident-prone, and accidents are more frequent among boys.” 43 Clinicians in medicine, and to some extent in psychology, however, could easily apply the syndrome of accident proneness to their patients. And they did. Moreover, insofar as accidents were a social problem and accident prone people caused the accidents, physicians of all kinds, and, again, not just psychiatrists, could contribute to safety by curing individual people who were the source of accidents. In particular, a physician need not believe that all accidents were the working out of unconscious motives. Clinicians could deal with the many complex factors, both personal and environmental, that might cause a person to be accident prone—including, as a California public health physician noted, bad vision, inadequate housing, and fatigue.44 Moreover, other irrational factors continued to come from psychiatry and medicine into the safety movement. Wrote one group of physician investigators in 1965, “The three ‘E’s,’ engineering, education, and enforcement, have been the bases upon which successful accident prevention programs have been erected in the past. Our attention is now directed to a fourth ‘E,’ emotions, as we attempt to understand the emotional aspects of accident behaviour.” 45 Just after World War ii, however, at a time when other changes were taking place, medical experts were pointing out that accidents, or “injuries,” were different from both older public health problems and from traditional medical problems.

155

156

ch a p t er 8

t he del ay ed impac t in pedi at r ics Beyond the psychiatrists’ work that I described in the last chapter, there was a another special line of medical interest that appeared after World War ii and that contributed substantially to the high point of interest in accident proneness. Beginning at the end of the 1940s and into the 1950s, pediatricians in particular began to call attention to the problem of accident proneness in children.46 Although some psychiatric specialists seconded this small movement in which physicians were taking a special interest in accident repeaters, applying the idea to children was a special-case area in which, for a time, medicalization of accident proneness made some progress.47 By the 1960s, however, psychologists rather than physicians were producing most of the relevant research reports about children who had accidents, and by the 1970s, the idea of accident proneness in children was in decline just as it had been in decline among other kinds of experts for more than a decade. In the psychologization of Western societies (most notably the United States) referred to in the last chapter, children played a special role because of the general assumption that events in childhood had a decisively formative effect upon adult life. Yet the dynamic psychological interpretations of Menninger, Dunbar, and others who wrote about motivated accidents did not for some time get applied specially to children. Indeed, it was well into the 1950s before a significant number of writers on accident proneness took up children as a particular population group. When in 1951 an investigator reviewed the literature on accident proneness in children, she found very few specific references on which she could draw at that time.48 The first few post–World War ii experts to write about very young accident repeaters came from the child psychiatry tradition and were interested in personality traits.49 Then in 1948, drawing on the work of Dunbar especially, New York pediatricians Ruth Morris Bakwin and Harry Bakwin wrote explicitly to call the attention of their pediatric colleagues to the problem: “Accident proneness should be suspected in children who have frequent accidents or who have the [distinctive] personality make-up and home environment . . . The accident habit,” they continued, “should be looked upon as a ‘behavior problem’ ” that physicians could undertake to treat.50 This growing interest among pediatricians was important because accident proneness could also be part of the pediatric practice that general medical practitioners undertook as well. 51 What happened particularly to stimulate interest in children who

The Mid-Twentieth-Century High Point

had accidents was that with the coming of antibiotics and other medical advances in the World War ii era, specialists dealing with children particularly had far fewer diseased patients, and suddenly they and others noticed that accidents were remarkably more serious as medical problems than were the infections that had previously sickened and carried off countless youngsters. For years, therefore, articles about childhood accidents almost invariably began with recitals of how much more important, for children, tragic accidental injuries were than serious illnesses. In the 1960s, each year 5,000 children were dying of accidents in the U.K. and 12,000 in the U.S.—and that did not count those injured and disabled. As a result of this new perspective, pediatricians led dramatic and sometimes effective efforts to reduce children’s accidents. 52 When investigators began to work on children’s injuries, they turned of course to the existing literature on accidents in general. And there they found all of the work on accident proneness. Wrote a 1948 researcher, “There appears to be considerably more than age, intellect, sex, or isolated personality traits operating to produce the injury-prone or noninjury-prone child. These hints as to the complexity of the injury-proneness problem are in accord with the findings of workers seeking to identify and describe the accident-prone adult.” 53 Another team commented that “in spite of the lack of reported studies of children there has been a tendency to talk about the accident-prone child and to attribute to him the personality characteristics and motivations found in the adult accident repeater.” Moreover, even if they found only a surprisingly few accident prone children, wrote two New York public health experts reporting their study in 1954, findings concerning those children could be of importance as showing in exaggerated form the patterns of accident causation in general. 54 By 1960, children’s accidents and accident proneness could appear to offer a prototype for one new direction that pediatric medicine would be taking: “The complex interrelationship of social, behavioral, and biologic factors which underlie many childhood accidents may also be operative in producing emotional disturbances and other behavioral problems, including juvenile delinquency and drug addiction.” 55 And of course physicians dealing with both adults and children were drawn into this broadening scope of medical thinking that included psychosocial factors, paralleling the introduction of social factors into discussions of accident proneness in the psychological and safety literatures. Social analysis, already begun in general discussions of accident proneness (as noted above), was the most distinctively different empha-

157

158

ch a p t er 8

sis to appear in investigations of children’s accidents after World War ii. Since children could be found in groups in classrooms and similar settings, research into social patterns was relatively easy. As early as 1951, one study found that children in a classroom who tended to be socially marginal also tended to suffer more injuries. And in the 1960s, some writers found it possible to report social class differences, with the not-surprising finding noted above: children from unfortunate circumstances also got hurt more frequently than those from more fortunate families. Moreover, deprived children did not learn safe habits even after they had accidents—but of course they had less parental care and fewer models of carefulness. Indeed, on both sides of the Iron Curtain, it was possible to discuss children’s injuries in terms of bad family or bad company or other environmental circumstances—without mentioning accident proneness at all. 56 Most of the pediatric experts, however, wrote about the psychology of accident proneness. Again this was not surprising in the era of psychologization (and the preponderance of American publications on the subject). Personality types and characteristics taken from the work of Dunbar and other investigators appeared frequently in discussions of children’s accident proneness, for example. A British pediatrician quoted an American expert who held that “accidents are more likely to happen to children who are over-active, restless, impulsive, unpopular, resentful, hostile, immature and without parental supervision.” The pediatrician then asked, “Does this get us anywhere? ‘Over-active, restless, impulsive, immature’—well, that’s what children are like, so we seem to be back to square one. I don’t know about ‘unpopular, resentful and hostile’— perhaps that holds mainly for American kids—or does it? ” This pediatrician ended pessimistically by maintaining that if people permitted dangers to persist, there was not much to do to prevent children from getting hurt. But then he added, “It’s up to the psychiatrists to discover why some people are accident prone, and why most people don’t care.” 57 In the literature on children’s accidents, the way in which dynamic psychological interpretations could reinforce the idea of accident proneness appeared particularly clearly. In 1970, a New Hampshire physician noted that many commentators on accident proneness had found conceptual and methodological problems in the idea: “We are persuaded that to postulate a trait such as proneness to accidents is not to explain why some individuals incur repeated accidents and others do not. This is akin to the common error of trying to explain traits such as aggressiveness by postulating a ‘need to aggress.’ ” Yet this expert, too, turned to family dynamics

The Mid-Twentieth-Century High Point

to explain why children had accidents, so that each case involved unique circumstances that would fit just that one child. 58 Very typically, investigators started out with two groups already identified, the accident prone and the accident free, as had investigators of adult accident proneness. Finally, just as Marbe as a boy had drawn conclusions from observing a classroom of children, so physicians and other professionals who worked with groups of children reported over and over the common experience of seeing individual children who had all of the accidents. Commented Helen Ross of Chicago in 1952, “Years ago, before I knew the term ‘accident proneness,’ I recognized in my summer camp the tendency among certain children to have more than their share of accidents. ‘There goes Jane again,’ or ‘Who do you suppose fell out of the boat? Mary, of course.’ ” 59 As in medical and safety writings about adults, the idea that there are child accident repeaters did survive for some time, but in only a weakened and altered condition. As the author of several 1970s articles explained, “The term ‘accident proneness’ is now part of everyday language and is confused with increased liability to accidents and usually offered, with no justification[,] as an explanation for accident repetitiveness . . . Recurrent injuries in children should be regarded as a symptom complex rather than a diagnosis of ‘accident proneness.’ ” 60 In children, then, the idea of accident proneness ended up, with some delay, much as it did with adult populations and medical practices. 61 As in all science, communication and popularization did not work instantaneously.

popu l a r iz at ion In any communication of technical ideas, as noted above, elementary expositions always existed simultaneously and side by side with advanced research reports. Such was the case with accident proneness after World War ii as well as before. Moreover, popularizations also changed to reflect the work of the specialists who were addressing different audiences, some of whom, as I have suggested, were themselves expert, notably in the fields of safety and applied psychology. Just as in the discussions of accident proneness in children, so, as the years passed, the proportion of elementary expositions in the general literature declined. Partly the writers assumed that their audiences knew basic ideas. And partly the authors of advanced treatises were focused on the more complex and advanced findings and were mostly addressing specialist audiences. 62 Or, for example, in medical journals, articles of various levels of sophistication

159

160

ch a p t er 8

appeared, designed to keep physicians updated on recent changes in the technical literature. 63 I have tried to find and identify the last of the basic popularizations of the idea of accident proneness that were so commonplace before and also after World War ii. The most likely candidate was an excerpt of an article by British pediatricians that appeared in a safety journal, Care in the Home, in 1973. “Children with repeated accidents are not uncommon,” they wrote. But they counseled that professional intervention could help break the pattern of behavior. The liability to repeated accidents is clearly multifactorial. The children we have observed, however, often show two features which are of importance, an extrovert personality, and involvement with and confusion by other family problems. These children react to a difficult family situation by hurting themselves . . . There is also evidence from general practice that the accident-prone family is also the sick-prone family. 64

In the years after World War ii, popularization followed the streams in the technical literature, particularly accident proneness as a personal, clinical problem and, in occupational and traffic injuries, a social problem. In the late 1940s, as opinion leaders’ preoccupation with war was receding, the same kind of popularization that appeared in the 1930s became commonplace again. Both experts and journalists addressed managers in factories and businesses with the news that some of their employees would have more accidents than others. Some of the expositions included recent research findings, such as the social factors in accident proneness or the transitory nature of accident proneness. But for a few years, writers still fundamentally addressed someone who needed to learn for the first time about the uneven distribution of accidents: “Some Folks Have All the Accidents,” or “Discovering the Accident-Prone Employee,” or “Who Are Your Accident-Prones, and Why.” 65 Other strands from an earlier time continued, particularly a concern about dangerous, accident-prone drivers. 66 Immediately after the war, however, one of the new streams from the technical expositions emerged in a striking way: accident proneness as a personal problem. The content of such articles revealed particularly that by the late 1940s, Dunbar’s attempts to popularize her ideas about the personalities of accident prone people had made a discernible impact. Dunbar’s own work, as noted earlier, was framed in 1948, for example, to emphasize the personal: “Do You

The Mid-Twentieth-Century High Point

Have the Accident Habit? ” Even without referring to Dunbar, however, the theme of accident proneness was commonplace. A 1951 general article in the American Medical Association popularization journal, Today’s Health, was called directly “Are You Accident Prone? ” 67 Along with the personality types, such personal approaches often included two other basic elements from the propaganda of psychologization: dynamic psychological descriptions of unconsciously motivated accidents, and advice to get professional help, with the suggestion that accident proneness was curable. One author, writing for parents, summed up the mindset after she had described Dick, an eight-year-old who over a short period of time among other things bumped his head, broke his arm, and broke his leg: It is highly important that the mothers of children like Dick, who show unusual proneness to accidents, should consult a psychiatrist as to how they should handle the problem because it is not a trait that is likely to be outgrown. The accident-prone child becomes the accident-prone adult whose existence has long been recognized by insurance companies and safety engineers in large industrial plants. 68

These various kinds of elementary popularizations, like Lykes’s book, continued to appear in the 1950s, but in diminishing numbers after the middle of the decade. But, as I have noted, for years longer, the idea of accident proneness found ready acceptance among those interested in children, and particularly among those sensitized to a psychologized view of the world. In 1958, a writer in Safety Education reported that accident prone children had twice as many unmet emotional needs as did accident free children. An article in the National Parent-Teacher as late as 1960 was titled “Is Your Child Accident Prone? ” (figure 8:4).69 Popular and semi-popular articles indicated therefore that experts had a high level of awareness of the idea of accident proneness in children. The subject ultimately showed up in advanced textbooks, and it was obligatory, for example at a symposium on childhood accidental injury as late as 1966, to have a paper on accident proneness.70 As a leading U.S. pediatrician wrote, “The question of accident proneness is raised in most discussions of accidents, whether of children or of adults. What studies are available . . . indicate that some children are accident-prone.” 71 In the midcentury decades, then, well informed people, at least in the Englishspeaking world, would very likely have repeatedly run across the idea of accident repeaters.72

161

162

ch a p t er 8

figure 8:4. This article appeared in the U.S. general magazine, National Parent-Teacher, in 1960, part of the continuing popularization of the idea of accident proneness and especially accident proneness in children. Source: Lawrence and Eda LeShan, “Is Your Child Accident Prone?” National Parent-Teacher, March 1960, 16. Image by permission of Robertstock Images.

At the same time, accident proneness did not appear in some popular venues where it might well have come to the attention of general readers. The most important example is perhaps the famous Dr. Spock book on infant and child care, where even sections on accident prevention did not alert readers to the idea that some children have more accidents than others.73 Such instances suggest the extent to which people who were particularly oriented to safety as such were the primary sources of popular knowledge.

t ex t book s conf ir m t he prof il e In still one more dimension of popularization, education, by the 1950s, accident proneness had begun to appear as a standard subject in advanced tertiary-level textbooks. And, of course, once in a textbook, the segment persisted. One public health textbook continued to carry a discussion of accident proneness in different editions from 1956 to 1973.74

The Mid-Twentieth-Century High Point

Both in the history of science and in the history of popularization, textbooks play a special role in indicating broad community interest and acceptance. The collective place of accident proneness in U.S. textbooks on occupational safety, safety education, and similar advanced subjects in fact confirms the overall profile of what happened to the idea after World War ii.75 One particular textbook, first issued in 1942 and reissued in 1947 (with little alteration) was of special significance because other writers on both sides of the Atlantic so often cited it. This was Joseph Tiffi n’s Industrial Psychology. In 1947, Tiffin devoted more than thirty pages to accident proneness, citing all of the familiar studies, beginning with Greenwood and Woods. Moreover, he included under accident proneness virtually all material directly pertaining to safety, including vision problems, dangerousness of the industrial plant, age and experience of the employee, mental abilities, and screening and training employees.76 It was no wonder that specialists complained about unnamed writers—very possibly Tiffi n—who used accident proneness in a loose way. In 1946, Norman R. F. Maier, provided a second founding text, including twelve pages—much better focused than Tiffi n’s—on accident proneness. “The inescapable conclusion” that Maier came to, in the face of the evidence he cited, was that “people differ in the degree to which they are accident-prone in the same way that they differ with respect to other traits.” 77 And in 1949, a third founding English-language textbook, by Thomas Willard Harrell, appeared, with explicit discussions of accident proneness scattered throughout the book (there were fourteen separate entries in the index).78 These texts set the model for others that followed. Tiffin’s book had so many editions that it passed through the hands of two more authors, but the treatment of accident proneness, and the citations, changed remarkably little until 1975. Other authors’ works, however, sometimes were sensitive to changes in the literature.79 In the last part of the 1950s and early in the 1960s, the tone in a number of the textbooks changed. Authors’ treatments of accident proneness tended to be ever briefer and more negative.80 It is true that into the late 1960s, some writers still maintained substantial sections on accident repeaters, if not accident proneness. 81 By the 1970s, however, accident proneness either dropped out of textbooks or remained as only a faint ghost from earlier times, even in the sixth edition of the book that Tiffin had started in 1942. 82 Altogether, the change in the textbooks over thirty years was remarkable and signaled how experts backed away from the idea of accident proneness. 83

163

164

ch a p t er 8

This whole record of popularization shows that by the 1960s, just as predicted by Covert’s ascending spiral model, the basic, introductory popularizations had largely disappeared. Instead, writers were popularizing refinements, such as the idea, noted above, that the accident prone group was not a permanent part of the population: “That 15% of the population suffering 85% of the accidents is a shifting group with new persons constantly falling in and out of it.” “The day will come,” wrote one hopeful writer publicizing the idea that changing life circumstances could make a person accident prone, “when people will go beyond saying something like ‘You drive, I have a headache,’ to saying ‘You drive, I’m mad at the boss.’ ” 84 In a similar way, a number of writers tried to bring to one kind of general public or another news of doubts that experts raised about accident proneness, such as the warning that “back of this current emphasis on accident-proneness as a major factor in accident causation is an almost unbelievably complicated structure of theory and half-true assumptions.” 85 A 1972 author noted specifically for health and safety workers that there were differences in quality in research in the field, but in this case he restricted himself to the cautious statement, “If we consider only the results of the better studies, the evidence is strongly in favour of proneness as a real factor in accident causation in a number of jobs, particularly complex jobs such as bus driving.” 86 As a technical idea, then, accident proneness faded out of public discourse in Europeanized countries except for just a certain amount of general usage. In 1982 in a collection of forty-six celebrities’ accounts of mishaps they had suffered, only two mentioned that they were accident prone—and yet those two clearly expected readers to understand the term. 87

e v idence f rom l a nguage There was abundant evidence that the term “accident proneness” became “part of everyday language” in the decades after World War ii. A Swiss investigator in 1959, for example, noted that Marbe’s term Unfäller (a person who has accidents, literally an “accidenter”) had entered colloquial German because of “the horrible increase in traffic accidents.” 88 Yet another Swiss investigator in 1973 was reporting a prospective study in which the accident rate for women who were skiing was significantly higher than for men. In translating his abstract, he used the English term “accident proneness” for Unfallgefährdung, or danger (risk) of accident. 89

The Mid-Twentieth-Century High Point

By the beginning of the 1960s, it had become obvious that the term “accident prone” had introduced a conceptual paradigm into the English language, at least. In 1975, when the editors of an Australian medical journal listed the medical problems that came from smoking, the category of “accidents” was the only one in which they used the word “prone.” 90 Not only did the term apply now to inanimate objects, but the very construction of the phrase spread to many areas—communicating meaning taken from the original formulation. Writers referred to accident prone and injury prone people and even “burn-prone” people, and also to “pain-prone” patients, “poison prone” children, and [law] “suit prone” doctors and hospitals.91 Some 1978 investigators discovered “the zoonosis-prone veterinarian.” 92 An abstracter of an article on pilots in aircraft accidents noted that some were “anxietyprone”—but avoided the term “accident prone.” 93 At the end of the century, it was possible for health experts to speak of “disease-prone personalities” in people who coped with stress by inducing illness. A 1970 investigator into patterns of absence from work found in the literature “sickness repeater,” “sickness-proneness,” “absence-proneness,” and “tendency to sickness absence.” 94 Even the whole phrase, “accident prone,” specifically could be used in generalized, often inaccurate ways. A pediatrician in the UK used standard accident analysis categories in 1984 to suggest that “the environment rather than the child may be ‘accident prone.’ ” 95 A student of early English texts in 1988 wrote ironically of accounts of a type of unfortunate injury under the heading of “An Accident-Prone Anglo Saxon.” The preeminent British historian, A. J. P. Taylor, at one point introduced the idea of contingency in history under the heading, “Accident Prone, Or, What Happened Next,” and recounted this story: A Galician priest was explaining to a peasant what miracles were. “If I fell from that church tower and landed unhurt, what would you call it? “An accident.” “And if I fell again and was unhurt? ” “Another accident.” “And if I did it a third time? ” “A habit.” 96

165

9.

ECLIPSE OF THE IDEA AMONG EXPERTS

The profile of experts’ writings on the subject of accident proneness parallels the general picture outlined in the last chapter. That is, after the high point, expert interest declined gradually after the 1950s. While popular accounts did in fact reflect many technical developments, experts who worked on the idea of accident proneness carried on among themselves a vigorous discussion after World War ii. In the post-peak phase, some experts tried actively to kill off the idea, while others, on a variety of grounds, defended using the idea of accident proneness. Shaw and Sichel, writing in 1971, already could trace to the 1950s the appearance of two groups of experts: those who wished to abandon the idea, and those who wished to save it.1 But as it turned out, the overall waning of the idea had more significant implications than did specific expert disagreements. As those concerned with accidents and safety turned away from the idea of accident proneness, they turned to another strategy, one that was not just negative: they substituted for accident proneness other ways of thinking about how humans differed in how often they incurred errors and injuries. The decline of accident proneness therefore occurred simultaneously with the rise of a new epidemiology and the idea of risk groups.

166

Eclipse of the Idea Among Experts

stages in t he process of decl ine The stages in the experts’ retreat from accident proneness represented a common, if not usual, sequence in changes in thinking in any expert community when thinkers are shifting from or abandoning an idea.2 As I have sketched earlier, after World War ii, the idea of accident proneness started out as a widely known set of beliefs so well established that they functioned as common sense among those concerned about accidents. The first step in disestablishment came as, here and there, an occasional doubter spoke up, sometimes timidly or occasionally with bravado, as did the first statistical critics. Simultaneously, as I suggested in the last chapter, investigators were adding complexity to the idea, to the point that at least some of the force of a simple idea became lost in detail and reservations. For example, in one major collection of essays on accident research, several authors were at least skeptical, and one asserted, “One model that we do not feel has proved productive in the past, despite its widespread use, is that of ‘accident proneness.’ ” Yet another author went so far as to suggest in connection with children’s injuries that there were whole residential neighborhoods in which youngsters were accident prone. 3 Eventually enough partial skeptics had voiced their concerns explicitly that a second stage developed. Critics were now brave enough to assert, for example, that accident proneness depended upon much flawed research. Yet these experts were willing to admit that there probably remained a core of accident prone people, or people who were accident prone part of the time. Despite the cautions of such authors, others could well see how opinion within the interested community was changing. When the editor of Industrial Medicine and Surgery in 1953 reprinted in shortened form one of the first major criticisms (published in 1951), he noted that the article “might have been entitled ‘Exit Accident Proneness.’ ” Clearly he perceived that a shift in expert opinion might already be under way even at that early date.4 The third stage therefore came with all-out attacks on the whole idea of accident proneness, including attacks on the logic and research that had gone into it. These attacks, which appeared to reinforce each other, were often openly aggressive, as if the experts were trying too hard to kill off an idea of the previous generation. Moreover, many experts simply did not refer to the subject any more. 5 Finally, in a stage evident by the late 1970s, rejection of accident proneness and pronouncing it dead became the new commonly accepted belief and approach. “We now know,” wrote an Australian safety expert in 1978, “that ‘proneness’ is neither

167

168

ch a p t er 9

permanent nor immutable, but is a transient phenomenon.” A New Zealander the same year pointed out that continuing to try to identify accident prone individuals distracted from the more important concern with the safety of the environment. And a 1981 safety writer concluded, “The old concept of ‘accident proneness’ (i.e., a lifelong personality trait) has been largely abandoned today as an explanation of accident behavior.” 6

ex perts encou n t er stat ist ic a l probl ems Statisticians in particular were the source of doubt about the whole concept of accident proneness. Statistics as a field began a fresh phase of development and extension before World War ii.7 Not only did new thinking fuel statisticians’ doubts about accident proneness, but those doubts were doubly effective because psychologists, the group who were still the main carriers of the notion of accident proneness, became more statistically oriented after World War ii. The centrality of statistics to accident proneness of course came from the fact that Marbe and Greenwood and Woods extracted an idea of accident proneness from statistical material in the first place. Moreover, statistics as part of epidemiology was also involved in a major shift of thinking that took place in that field in the middle of the twentieth century. One of the pioneer figures was none other than Major Greenwood, and, to complicate things more, the work on accident proneness was of general theoretical interest to specialists in statistical theory and technique who had no interest in accidents per se. In 1970, for example, C. D. Kemp of Belfast concluded that statistical work on accident proneness, while not very helpful in the safety field, had been very productive of statistical theory and had, he believed, a development remarkably parallel with that of actuarial risk theory. 8 The person who effectively started off the series of publications that raised doubts was Percy W. Cobb, an investigator with the Highway Research Board of the U.S. National Research Council. His article appeared in 1940, but other psychologists did not follow his work up until after World War ii. Cobb raised the problem that collections of cases that others had chosen to analyze could not legitimately be compared to the curve of chance distribution that the investigators had been using. After the war, H. M. Johnson of Tulane University took Cobb’s argument further. “In very few instances, indeed,” wrote Johnson, “has any evidence been submitted of the statistical reliability of these procedures; in still fewer has their validity been meaningfully discussed.” Johnson cited Lahy’s work as particularly flawed, including Lahy’s employing “an illicit statistical proce-

Eclipse of the Idea Among Experts

dure.” The best such work, Johnson continued, “could be usefully applied in the construction of an actuarial table; better selectors have long been available; it is practically worthless as a means of predicting the safetyclassification of any individual, as such.” 9 In 1950, Major Greenwood published what turned out to be his last paper, “Accident Proneness,” in which he defended statistical assumptions of investigators who attempted to devise tests for accident proneness. In the real world, Greenwood noted, empirical data about accidents would always be limited.10 Meanwhile, from 1949 until 1956, a flurry of technical papers appeared in which the authors (all from Englishspeaking countries) argued furiously and usually abstractly about whether or not the statistics that workers in the field of accident proneness had used were methodologically valid—indeed, if such an entity as accident proneness actually existed statistically. Two New York psychologists started by denying the validity of the method of “percentages” (such as 5% of the workers cause 36% of the accidents). The percentages method, they pointed out, requires that, in order to find a deviation, everyone expects the same number of accidents. The factor of chance, however, would depend not only on the distribution of accidents in the ideal group (the Poisson curve that better investigators had been using) but on the changing overall conditions involving the people who were being examined for accident proneness. Without denying the existence of accident proneness, the authors concluded cautiously that “in many instances personal accident proneness, which is but one of the components of accident liability, has been an overemphasized factor.” 11 Some South African statisticians then chimed in. One claimed that even the chance (Poisson) distributions did not furnish adequate comparisons to determine accident prone cases and that precise time measurements were also necessary. Two more experts, employing elaborate statistical analysis, concluded argumentatively that “the evidence so far available does not enable one to make categorical statements in regard to accident-proneness, either one way or the other; and as long as we choose to deceive ourselves that they do, just so long will we stagnate in our abysmal ignorance of the real factors involved in the personal liability to accidents.” 12 Others, from South Africa, Britain, and the United States, contributed more to the controversy. First, correlation techniques and methods of statistical comparison came up, and then once more the question of practicality in predicting accidents.13 In ensuing discussions, the same questions kept recurring: formulas to

169

170

ch a p t er 9

tease out relationships, the importance of time intervals, and the quality and relevance of the data being manipulated. In the mid-1950s, the statistical discussions of accident proneness split into two general categories. In one, statistical theory was at stake. In the other, questions of statistical treatment of real data concerning possible accident proneness were at issue, such as the conclusion of one statistician that accident proneness did exist, but in order to reduce accidents by 25% among a group of bus drivers, half of the drivers would have to be discharged. Experts with either orientation did not necessarily communicate directly with the other.14 Within a decade, in the 1960s and after, review essays made it clear that in discussions of accident proneness, the standards of evidence, sampling, and analysis had risen dramatically. “It is a pity,” wrote a 1964 book reviewer, “that such a good discussion of past work and of general principles is supported by statistical material of perhaps inadequate volume and analysed in an unsatisfactory manner.” 15 The statistical critiques, however, also eventually produced two modifications in the conceptual content of the idea of accident proneness. The first was to reduce investigators’ ideas of the number of the accident prone people in any given group. The statisticians had simply shown that too many factors other than accident proneness had to be considered. The second was to encourage still further the idea that accident proneness was a temporary condition, a complication that I noted already in discussing the growing complexity of the idea. Again and again, investigators insisted that accident proneness did not persist through a long period of time in a person’s life. In the end, accidents simply did not necessarily foretell more accidents.16 In 1970, Peter Froggatt of Belfast published a study of employee absences from work that utilized the statistical questions that accident proneness studies had raised. He found that statistics might not show causes, but they could predict whether a worker was likely to take short absences from work—“one-day absence-prone” and “two-day absenceprone.” “Proneness,” Froggatt concluded, “is a markedly successful hypothesis” in the case of absences. The worker’s absence characteristic, however, was much more stable and identifiable than had been demonstrated in the case of accident proneness.17 There, as the statisticians and critics had shown, difficulties in measuring, in standardizing data, even in obtaining precise information concerning accidents and accident incidence, made accidents very different from something direct and simple like absences—to the point that the scientific outlook for accident proneness was becoming unfavorable.

Eclipse of the Idea Among Experts

cr ack s in t he st ruc t u r e w iden All of these changes set the stage and provided the background for the attacks on the idea and the defenses of the idea that eventually worked out in the way in which accident proneness faded out of scientific discussions. Moreover, doubts and reservations spread beyond statisticians and psychologists to all of those involved in safety. Even those concerned with children became affected. While some investigators of injury patterns in children, for example, reported negative findings when they tested for accident proneness, for some time, others found exactly the opposite (“It would appear that the repeater concept in accidental poisoning is established”).18 Increasingly during the 1960s and after, authors in technical literature dealing with children suggested that there might be a few accident prone youngsters in various populations, but writers concerned with safety tried to look at more general epidemiologies, with or without dynamic interpretations.19 Or as a British child specialist finally concluded, “It has been remarked that preoccupation with accident-proneness may well divert attention from other more rewarding aspects of accident causation.” 20 By the end of the 1960s, it was obvious that skepticism concerning both the validity and relevance of accident proneness was causing greatly decreased attention to the subject in all parts of the technical literature. It is true, as I have noted, that the subject would still come up as a standard category.21 But as two California experts commented in 1967, “The prevailing trend is to reject the concept as totally useless for scientific purposes and to banish the problem of accident repeatedness to the realm of scientific limbo . . . It is the authors’ view that the baby has thereby been thrown out with the bath water.” One of the authors in fact had just reported that accident prone behavior in children was not a transient condition but persisted over time, and the two authors pointed out again that people who worked with children continued to see individual youngsters who, over and over, got hurt.22 Clearly, however, statistics provided an entryway for thinkers who found problems with accident proneness as an idea and an instrument for promoting safety. Both the concept and investigative studies came under attack from experts who started out with the doubts raised by statisticians. Psychologists Alexander Mintz and Milton L. Blum in 1949 began by making a widely cited distinction between the personal characteristics of an accident prone person and the person’s “accident liability,” which referred to all of the factors in accidents, such as workplace or other job

171

172

ch a p t er 9

conditions. They concluded that personal proneness was a very minor factor when one finished all of the calculations.23 At almost the same time, in a major essay on accident and safety psychology, two University of California psychologists, Edwin E. Ghiselli and Clarence W. Brown, summarized what became mainstream doubts. They questioned basic assumptions upon which the idea of accident proneness rested. Starting right out with the idea that a few workers cause most of the accidents, “then the major emphasis in a safety program should be given to locating and separating the so-called ‘accident repeaters.’ ” Unfortunately, the investigators continued, applying a test from statistical discussions, over a period of time, the statistical profile did not hold, as repeaters in fact did change. Moreover, tests did not have statistically sufficient predictive power to be of great use even in selecting pilots or workers on the basis of an inclination to have accidents.24 Another factor that began to play a large part in the rising tide of criticism of accident proneness was a shift in ideas about psychological testing for accident proneness. Where, before the war, investigators differed about which tests were effective, divergences in reported results pushed a number of experts in the 1950s to grow increasingly skeptical that any tests had predictive value—and therefore to reject them all.25 As the statistical critiques began to wind down in the mid-1950s, other refinements and reservations appeared in the technical literature, and outright attacks increased.26 Some investigators simply got negative results when trying to find ways of predicting who would have an accident. An aviation psychologist, for example, wrote to express his frustration with attempting to get useful data and acceptable analysis in either other experimenters’ thinking or in the “complexities of the uncontrolled world surrounding operational accidents.” Eventually he concluded that having accidents did not predict having further accidents—a denial of the fundamental idea of proneness.27 For twenty years, from the late 1950s to the late 1970s, investigators repeatedly addressed the problems around the idea of accident proneness through reviews of the literature, sometimes in connection with new data, but often not.28 Such review articles signified experts’ uncertainty amidst the welter of data, argument, and critiques. Experts trying to master the subject found not only disparate individual investigations but too many approaches and, worse, contradictory findings. In this situation, and attempting to be helpful, the literature reviewers imposed demanding standards on previous and current research, and almost always much of a review consisted of finding both research and conceptualization seriously flawed.

Eclipse of the Idea Among Experts

A particularly detailed and damaging review appeared in 1964, written by two experts from Northern Ireland. They concluded that “there was a growing heresy that perhaps the very concept itself should be reappraised.” It might be, they conceded, that people can have transitory “spells” of accident proneness, but it is not a lasting trait. They appealed to the common sense that “tendency to accident is an undoubted hazard of living. But the wisest course might be to discourage the use of any catch phrase; [and] the abuse of ‘accident proneness’ must always serve as a grim example.” 29 Other authors of review articles did not always employ such an astringent tone, but they did tend to reject the validity of parts or all of the previously established idea of accident proneness. 30 In 1971, the Medical Journal of Australia published a symposium on accident proneness. Only one of the participants defended using the idea of accident proneness. She lumped it together with other behavior that disrupted job performance in a factory, and she urged clinical and managerial attention to the individual worker. The other participants in the symposium believed that the concept was not valid and was a distraction. The first speaker reviewed the history of the idea and then concluded that with the 1960s, “the cult of accident proneness to explain and preach cause and effect of events had reached and indeed passed its zenith.” Another added that “accident proneness is a bogy that has appeared before me from time to time during my work in safety. It is a term which has done much to harm the accident-prevention movement, particularly in its effect in limiting research.” And still another contributor concluded that trying to focus on an accident prone driver was not practical. Altogether, the panel drew attention to managerial responsibilities or to engineering solutions for safety work—leaving attention to individual workers to supervisors and clinicians. 31 Indeed, regardless of the tone, the critics of the idea of accident proneness were usually not just negative. Instead, they were advocates of safety strategies other than relying on weeding accident prone people out of dangerous situations. For such critics, accident proneness distracted safety advocates from more effective considerations. An American writer deplored the way in which resources had been diverted to accident proneness research instead of looking at management and engineering. “The first time the accident-prone people theory fell flat on its face in industry was in the thirties when a psychologist found that one department of 200 men was having the majority of accidents in a large railroad company . . . when the superintendent of these men was transferred to another department[,] suddenly the accidents decreased,” and those in

173

174

ch a p t er 9

the superintendent’s new department increased. And in another case, redesign of equipment kept the operators of the equipment from being labeled accident prone. 32 Again and again, practical considerations became the factor that undermined support for using the idea of accident proneness. “The literature in English in the 1970s,” concluded one review author tellingly, “contains several position statements by safety officers . . . decrying the notion of accident proneness, largely because it may allow managements to escape their responsibilities in machine design, in selection and training of personnel in safe operating procedures, and in meticulous attention to the environment where high energy transfer can impinge on the human.” 33 In the last part of the twentieth century, practitioners who believed that accident proneness was invalid used historical accounts—however brief—to try to stamp out any remaining remnants of it. Clearly they believed portraying accident proneness as merely a historical idea was a means of discrediting it. A 1985 expert in his book on injury control devoted a whole chapter to the evidence that “accident proneness” (he even put the term in quotation marks) was not a viable idea in safety work. Instead, he noted, his colleagues and he had moved to considering risk groups (a shift that I note below in this chapter). As a clinical entity, he conceded, accident proneness might still have an existence, but that was not of interest to those attempting to control injury. A 1992 writer was even more vehement, classifying the idea of accident proneness with outdated Freudianism and psychology and the traditional attempts to blame the victim in any accidents by asking “What was wrong with people that caused them to injure themselves?” 34

def ender s of t he ide a In the face of all of these attacks over several decades, a number of investigators, albeit a diminishing number, continued to refine ideas of accident proneness and directly or indirectly to answer the critics. 35 Shaw and Sichel in 1971, for example, were still trying to point out that in carefully formulated terms, accident proneness was a useful and valid idea.36 And enough analysis had occurred that it was possible to organize the concept, as did a California psychologist in 1970, separating short-term from long-term accident proneness and breaking down the factors and characteristics of various sub-types. 37 Those experts who continued to develop and defend the concept after the 1950s of course operated under changing conditions. First of all, as I

Eclipse of the Idea Among Experts

have noted, the rates of industrial accidents declined to such an extent that they ceased to be the huge social problem they once were, and education and management campaigns kept the numbers down to levels that did not set off special social action. The number of vehicle accidents, however, continued to escalate along with increasing amounts of traffic, so that if accident proneness was to be a relevant and important factor, it would most likely be so in the area of traffic accidents. 38 Therefore much of the defense of the idea came from those engaged in projects focused on traffic accidents. In 1958, in a major, long-range study of transport drivers, the Finnish expert, Sauli Häkkinen (whose work was mentioned in the last chapter), met the statistical and logical objections of critics of accident proneness. Häkkinen also produced data from a stable, comparable population over a number of years and in other ways recognized and corrected for the faults critics had found in previous studies. He concluded that tests of a limited number of elements of character and ability—constant factors— were significantly predictive of consistently high and low rates of accidents if one controlled for age and experience. He found for example that fatigue was not a factor, but skill in manipulating the vehicle and regularity in work were. Finally, the critical factor to count was not type or circumstances of an accident but simply the number of accidents a driver had. Häkkinen’s work was particularly impressive because, years later, he published a follow-up study. The same psychological tests discriminated well, and, moreover, the accident proneness pattern persisted in the same group of drivers for more than twenty-five years. He concluded that “with well-planned and controlled studies it is possible to detect relevant human factors which in general accident statistics and larger studies are overshadowed by many changing and uncontrolled factors.” 39 A variety of other psychologists conducted experiments to define and predict an aberrant tendency for a person to have accidents, experiments that were much more carefully formulated and hedged than those of their predecessors. One investigator, reporting from Beirut, showed that environment and type of accident were irrelevant for the accident prone person, if one sets out a broad definition of “mishaps”: “individuals who admit to having accidents report having them in a wide variety of situations, some of which, like losing things . . . and hurting people’s feelings . . . would not normally be regarded as accident situations.” Other investigators found that only accidents involving unsafe behavior should be counted, for they did, in fact, predict more accidents that would result from unsafe acts.40

175

176

ch a p t er 9

Those experts continuing to argue for accident proneness as part of personality used both experimental and clinical data. Mere individual differences, such as psychomotor abilities, were not as successfully linked to accident incidence as were personality tests. Tests for extraversion detected industrial accident repeaters among males, for example, in one British study—even though the subjects had different occupations. Another study, from the University of Michigan, found that, beyond alcohol use, fatal accidents came with chronic psychopathology, social stress, or acute pre-accident emotional disturbances, suggesting that accident behavior was a complex force but one subject to analysis.41 As the years passed, accident proneness survived most clearly in clinical studies or those based on dynamic interpretations, as noted from another point of view in the last chapter. As two Detroit physicians concluded in 1966, Current knowledge indicates that there is an accident prone syndrome . . . The concept must be broadened to include sociologic understanding in addition to the psychological aspects. It is, first of all, not usually a lifelong affair. Although it is undoubtedly true that some individuals go from one accident to another throughout their existence, most members of this group do not. Rather, many people develop self-destructive needs from time to time. At these moments, accidents can provide less devastating satisfaction than other expressions like suicide.42

pr ac t ic a l use a s def ense As suggested by the clinicians, much of the defense and persistence of the idea of accident proneness came directly from the fact that many experts found that it remained a useful idea in practical circumstances. One approach was to justify educational and therapeutic programs because they could target key people. That is, if accident proneness can be corrected, then it is useful to identify the accident prone person who can be treated. A British safety expert used the idea to justify her safety programs: “I am firmly convinced that there are people who are accident-prone. Their mental make-up is such that observation and judgment are not equally matched with awareness of immediate environment or even potential danger. These are the people who are sure to trip over the uneven paving stone, or catch hold of the red-hot pan handle.” 43 A foundry physician told of his group’s experience in identifying work-

Eclipse of the Idea Among Experts

ers who showed distinctly elevated rates of injury—and they did fi nd such workers. In each case, the medical staff identified any problems that could be corrected, with results that fully justified making the identification of the employee and taking action. He offered this case: L.A.—Breaking iron on hot line and sorting into bin. Injuries in the two months before the interview: (1) sand left eye; (2) sand left eye; (3) sand left eye; (4) sand right eye; (5) sand right eye; (6) bruise [of] forehead; (7) burn of left arm; (8) burn of left arm; (9) burn of left arm; (10) burn of left arm; (11) bruise of scalp; (12) blister right fi nger. Cause: The employee had very poorly fitted goggles. Used gloves but wore short sleeves. He had to reach across a belt of hot iron to get casting and deposit it in bin. This caused left arm burns. Solution: Goggles were fitted properly. It was suggested that he wear a grinder’s sleeve on the left arm.

The employee continued to have injuries in the next two months, but only five of them, as opposed to twelve before (a reduction of 58%). The physician was convinced most of the later injuries “were due to failure to use the suggested procedures consistently.” In another case, “his attitude was indicative of a psychological problem.” Simply identifying the source of high accident rates was of great help to everyone involved.44 One major arena in which accident proneness continued to be useful was in selection of employees, particularly those involved in transport. The eminent psychologist H. J. Eysenck was still asserting in the 1960s that psychologists could assist “in the elimination of accidents. Such help may be forthcoming in identifying, eliminating or retraining the accident-prone driver.” And as late as 1973, a team of Israeli psychologists was offering new tests for commercial driver selection.45 On a particularly practical level, a General Motors industrial physician observed that doctors in their physical examinations of applicants could note, not least of all, scars from previous accidents that “may be silent testimonies of an accident-prone personality and should, if verified, influence placement.” 46 The other major arena of practicality in which the idea of accident proneness survived after 1960 was, again, in connection with traffic injuries. In part this concern overlapped the problem of choosing commercial drivers. In 1992, a British safety investigator, E. M. Holroyd, published a study

177

178

ch a p t er 9

of accident rates that called into question many of the arguments that his predecessors had made against the idea of accident proneness. By following for a dozen years or more a very large population of drivers, Holroyd was able to demonstrate, as did Häkkinen a generation earlier, that some drivers did indeed, over many years, show elevated “personal accident rates.” Both evidence and statistical analysis refuted the skepticism of many previous writers, and the approach, Holroyd concluded, could be usefully “applied to studies of the accident rates of professional drivers and factory workers, for whom the exposure is more uniform.” 47 Although some authorities recognized the significance of Holroyd’s work, the fact remains that by the 1990s, no obvious audience appeared to revive expert interest in accident proneness by that name.

ch a nged circ u msta nces a nd a pproaches By the last decades of the twentieth century, the complexity of evidence and analysis and the contradictory nature of the findings therefore encouraged those concerned with accidents to avoid the term if not the idea of accident proneness.48 Children who lacked motor coordination, for example, in one study had fewer accidents than those who had more motor coordination.49 As early as 1954, a British investigator found two different types of accident susceptibility, one in younger workers, one in older workers.50 Some experts continued to believe that everyone had been asking the wrong question: the most important data were the factors that cause some people to remain free of accidents. 51 But investigators in the postwar decades had still another circumstance with which to contend. The standards for research in the field of accident causation in general were rising (parallel with those in statistical studies), and work on accident proneness often fell afoul of those standards—beyond the specific problems that I have already noted. In 1964, a landmark book appeared criticizing shortcomings in research in the accident field, and accident proneness investigations were especially conspicuous as examples of flawed work. 52 The review articles (noted above) that marked the end of the classic work on accident proneness reflected these new standards applied to both theory and methodology in earlier—and current—investigations. 53 Moreover, psychology, the home discipline for accident proneness, was changing. A new emphasis on cognition made the automatic or unconscious behavior of an accident prone person of less interest to investigators. 54 An Iowa State psychologist in 1978 made the trend clear by

Eclipse of the Idea Among Experts

advocating “cognitive accident-avoidance training for beginning drivers.” Although he measured outcome by “accidents per driver year,” he did not use the concept of accident proneness in any form at all. 55 Or investigators could focus on mental processes that were in themselves complex, such as information processing.56 Even when the experts put all of the tests together, the results did not satisfy their colleagues.57 Perhaps the last word was that of an investigator for the Royal Air Force Institute of Aviation Medicine: It will be noted that none of the topics discussed above have an overriding, or even particularly clear-cut, relationship with pilot-error accidents although all these topics have an apparent pertinence to the problem. Closer examination of individual accidents reveals why this is so, for it is very rare for [an] accident to be caused entirely because of deficiencies or other discrete human factors problems; culpability almost invariably lies with an interaction between these and other factors. The possible number of such interactions are virtually limitless; consequently, the determination of the most likely aetiology of a given accident is likely to remain a non-precise “science.” 58

It was in this world of growing complexity that the idea of accident proneness simply faded away among relevant expert communities. Indeed, experts migrated into different specialties in the late twentieth century. Safety engineering and industrial hygiene split. Then other distinct specialists appeared, as one commentator noted: “environmental engineers, product safety engineers, fire protection engineers, and various plant engineering specialists.” All of these experts were working alongside a variety of public health specialists. 59 Many people, however, continued to believe at various levels that some people had more accidents than was their quota. So the noncommittal term, “repeaters,” persisted. “The word, ‘proneness’ has been used to imply a series of personality traits which have not been precisely identified thus far,” wrote McFarland in 1962. “Repeater” as merely descriptive of course left the type of accident analysis open. 60 Wrote the practical Australian safety officer Arthur A. Sampson, one of those who believed accident proneness was a “myth,” “When we find accident repeaters, we do not label them ‘not to be let out alone’, or ‘dangerous, do not touch’, but we try to find out why, and in all cases there is a valid reason. Most common reasons are because of physical handicaps (often not apparent), or because the work is not suited to their temperament, or they are under mental stress.” 61

179

180

ch a p t er 9

But to accommodate this kind of individual difference between human beings, experts came to depend upon other ways of thinking about the social problem of unexpected injuries and errors. These other ways of thinking effectively replaced accident proneness in the technical literature. Two were especially obvious: epidemiology and risk groups.

epidemiolog y Just at the time when accident proneness was reaching a high point in the 1950s, the study of epidemiology was changing rapidly. The classic form was to locate where in a given population there had been illnesses and to infer causal factors, or at least control factors, from those circumstances. Typically, a single bacterial or toxic cause was indicated. But in the 1950s, epidemiology was brought to bear on complex causes, particularly of chronic diseases. In 1957, two Americans pointed out that not only had accidents, especially traffic accidents, “taken on the characteristics of a mass disease of epidemic proportions,” but “accidental injuries do, in fact, provide frequency distributions in time and place that are comparable to those of mass diseases, and show similar biologic constants.” 62 But the new form of epidemiology led practitioners to go further and study populations “prospectively” to find what people came down with which malady so as to determine what causes were operating. In the last half of the twentieth century, experts applied both the retrospective and prospective approaches to the problem of accidents. Indeed, accident proneness was a classic case of a malady discovered through retrospective epidemiological tracing of who in a group had suffered injuries. To later experts, the fact that the concept had a retrospective basis could indicate weakness in evidence supporting the idea. 63 Public health experts also used epidemiology to move away from blaming individuals for injuries. Where traditionally the careless, clumsy, or accident prone person had implicit or explicit responsibility for accidents, public health leaders invoked general, impersonal, social viewpoints. The late twentieth-century epidemiology of accidents was not biological or behavioral. Rather, epidemiologists placed injuries in a very broad environmental setting.64 The epidemiology of accidents went in two different directions. One was an extension of the gathering of statistics. That is, as infectious diseases receded in importance and accidents became more and more conspicuous as causes of injury and disability, experts used numbers to raise the public alarm and study injury incidence more closely. Eventually,

Eclipse of the Idea Among Experts

these numbers produced risk groups—an analysis that was specially related to accident proneness, as I shall indicate below. The second direction in which epidemiology went was to define all accidents in terms of host, agent, and environment. Beginning with a key article in 1949, experts began to apply this special epidemiological approach to accidents. 65 By the mid-1950s, this new accident analysis tended to sideline the idea of accident proneness. The analysis now was in terms of the host, namely, the child or adult; the agent, that is, whatever physically inflicted injury; and the environment, which could be physical or psychosocial or both. The “repeaters” phenomenon just was not necessary any more for those analyzing accidents to achieve safety, although at first the “unusually susceptible” child might be a minor host factor—very much diminished, however, in the new complex context. 66 Thus, insofar as epidemiologists used patterns of occurrence to infer causation, their terms of analysis became host, agent, and environment. In practice, however, most of the epidemiologists’ attention went to the environment. The injuring agent, such as a falling beam or a vehicle, was generally left to safety engineers. The host, meanwhile, either diminished in importance or was translated into a risk group. Or experts could refer to some behavior for which education and discipline would be appropriate correctives. 67 In 1988, for example, the International Labour Organisation/World Health Organization report on the epidemiology of occupational accidents focused on getting sound data and then analyzing the work environment: “The Committee stressed the multifactorial nature of occupational accidents, which were considered as statistically foreseeable consequences of technical-social failure of the work system.” In such a view, individual differences played no part. Only factors that could be dealt with in the aggregate such as age groups or types of industries entered into this kind of epidemiology. 68 Moreover, injuries differed from diseases that had a clear causal factor, like a bacterial agent. Injuries resulted from complex circumstances. The epidemiology of injuries—or accidents more generally—therefore could introduce social viewpoints, as opposed to clinical treatment of the individual. By the end of the century, public health experts were arguing strongly that mass individual actions had powerful impacts on injury statistics. The main example was of course drunk driving, which was a complex act, brought on by complicated social and personal factors, but a large factor in automobile and other accidents the world around. Personal risk factors, in the eyes of epidemiologists, tended to be voluntary actions, rather than unwilled risky behavior. 69

181

182

ch a p t er 9

In the 1960s, led by American William Haddon, Jr., epidemiologists refined the idea of agent to mean an energy carrier, like a train or a fire or chemical, because any accident involved energy. By focusing on the energy and trying to suggest ways of modifying agents and environments, epidemiologists led the way into emphasizing engineering solutions for safety problems. The agents that actually physically inflicted damage typically fell under the category of engineering. Moreover, by invoking complicated energy carriers—agents—and an extremely complex environment with a variety of types of elements contributing to any injury, accident analysis necessarily now included many factors operating at once.70 Indeed, this new outlook spawned a developing special study, ergonomics, the interaction between the biology of the human body and the physical environment (see next chapter). From such studies came child booster seats for automobiles, shock absorbent materials, road layouts, fireproof clothing, and many other technical changes. Michael Guarnieri, reviewing the history of safety expertise, identified a breakthrough at the beginning of the 1960s that “shifted the focus of research from behavioral psychology to engineering and epidemiology. Psychology might explain a rear end collision of an older driver by a teenager in interesting but largely unusable Freudian terms, whereas [it would be better to] examine brake systems, speed limits, bumper design, and alcohol laws.” The new expertise, Guarnieri continued, “led to air bags, head restraints, padded dashboards, and stronger bumpers, doors and fuel tanks,” as well as the removal of “killer poles” that caught drivers who got slightly off the road. Physicians and engineers now produced “an avalanche of work with models systems and computers on cars, gloves, shoes, hard hats, sports equipment, eye wear, even prophylactics.” 71 In this way, epidemiology led directly to the new emphasis on engineering safety that I shall discuss in the next chapter. Epidemiologists also introduced still another shift that minimized concern about any possible accident proneness. By shifting to the term “injury” instead of accident, they included intentional as well as the unintentional actions that by definition constituted accidents. Not only did this shift now exclude errors, such as actions that damaged goods, rather than people, in a factory, but by including murder, assault, and suicide, epidemiologists could bring in even more social factors. And they did, introducing categories such as ethnicity and socioeconomic deprivation. Such categories suggested using education and engineering, plus social reform as means of remediation—in place of individual treatment. Safety expert Guarnieri in 1992 claimed that “the word ‘accident’ has almost

Eclipse of the Idea Among Experts

disappeared from modern scientific and engineering use although, unfortunately, it remains embedded in popular journalism and common use.” In 1997, Judith Green went so far as to report that “through epidemiological mapping even chance events have been made predictable, and the accident no longer has a legitimate place.” In this new context of many factors and shifted viewpoint, possible accident proneness in the “host” in an accident did not attract important consideration.72 Experts did not just abandon the idea that there were, for example, some children who were prone to have accidents, however. In this case, the host, or child, became just part of a “risk group.” Experts increasingly were tending to think not of individuals who were accident prone, but of “special groups of vulnerable people—for example, the elderly, the handicapped, the infant or the toddler,” as the author of a 1965 World Health Organization report put it. Otherwise space, time, weather, and unsafe conditions were of special concern, obscuring any individual inclinations.73

r isk a nd r isk grou ps In the last decades of the twentieth century, then, one of the most notable fates of the idea of accident proneness was to be reconceptualized in terms of risk groups.74 And experts found risk groups to be particularly useful because the risk group was based on relatively objective or neutral characteristics, such as, notoriously, age or obvious disease condition, or, around the turn of the twenty-first century, measurable obesity, rather than on a personality trait or an attitude measured by a mental test.75 In his classic on the late twentieth-century “risk society,” Ulrich Beck pointed out the philosophical and political aspects of risk and the unequal distribution of risk. Others have also placed the idea of risk in a larger social framework, as one analyst suggested drily that one could cut the accident rate most easily by increasing unemployment (which would mean young men, now with less money, would drive less).76 Certainly as Beck and others pointed out, allowing people to take chances was an essential part of freedom. Indeed, one needed to make errors in order to learn. In an open society, everyone had to do a cost-benefit analysis to see if a risk was worth taking (a cognitive approach to which I shall return later).77 Judith Green describes three stages of thinking about accidents, stages that might fit the narrative sequence laid out above. The first, premodern stage, was fate. The second stage was determinism, the modernity phase in which Europeanized thinkers viewed accidents as the result of

183

184

ch a p t er 9

natural causes. And the last phase, out of “high modernity,” was the risk stage, in which everyone had to make calculations of uncertainty—but one in which being part of a grouping according to risk—the insurance model—provided some personal assurance even though the individual as such was not a major factor.78 The ultimate in the insurance model was aggregating risk so that the individual simply disappeared.79 The original idea was, of course, the “risk factor.” 80 William Rothstein has described how the idea of risk factor came in first in life insurance and then, around 1960, into the health sciences. 81 Contrary to one’s expectations, insurance executives, like the British insurers in the mid-1930s, throughout the twentieth century generally ignored the idea of accident proneness. Instead, ironically, it was the insurance idea of risk that came into public thinking about accidents. 82 Insurance executives everywhere for decades were well aware of the concept of accident proneness. 83 Or they were at least ready for it. Even without the name, the repeater phenomenon continued to be obvious in insurance “experience.” In 1942, one expert reported that “the Boston Index Bureau says that of more than 100,000 accident claimants reported to it every year, between 40 and 50 per cent are discovered to have a previous accident history.” 84 The term did finally show up occasionally in relevant insurance literature. An insurance physician in 1946, at the height of the use of the idea, made the equation: Insurability is assessed with awareness of accidental [sic] proneness. Awareness and advance recognition are not the same. Some persons are accident prone. They get more than their share. Worse, they cause more than their share. Reasons are several. Slow reaction. Inattention. Moodiness. Preoccupation. Faulty judgment. Failure to anticipate. Anyone who plays athletic games sees many such individuals. Healthy. Vigorous. But clumsy. Poor timing. Slow coordination. If, additionally, they are impaired in vision, in hearing, in health, or in age, they lack alertness. So it happens to them, and those who insure them.85

The idea therefore did appear incidentally in the insurance literature. The last example I have found is from a USA text of 1978, in which the author noted accident proneness as part of the standard disqualifying category, “moral hazard,” in an insurance applicant: “a person has a mental attitude which causes him to suffer more loss than others, perhaps subconsciously—e.g., persons who are accident prone.” 86

Eclipse of the Idea Among Experts

But in various settings and various ways, as Marbe discovered in the 1920s, insurance experts did not factor in individual workers or drivers. Instead, they worked with overall rates or such factors as occupation and exposure to danger. Indeed, underwriters in the early years of the twentieth century were already rating the insurance risk of occupational groups and charging accordingly. Actuaries and dentists were “select and preferred” or “extra preferred.” Cowboys were “extra hazardous,” lumberjacks were “perilous,” and acrobats were “extra perilous.” 87 At any time, insurers preferred to insure categories and groups for accidents. If one could bury in a larger group accident prone people or those with other personal peculiarities, it would be advantageous economically. There could, in the first place, be less money going to sales agents. But most important, the process of selection of individual risks was costly. Altogether, gathering individuals into groups to take advantage of the averaging in great numbers brought great savings. 88 Thus the idea of risk groups was early on implicit in an insurance viewpoint. That is, insurance effectively led people to think in terms of aggregating people into groups who could gain some security against the effects of unwanted events. And even where some people had to be disaggregated out of a group, it was on the basis of aggregating them into another, smaller group, namely, a more specific risk group. 89 Even in motorcar insurance, as demonstrated in the British incident noted in chapter 6, insurance executives came to rating individual drivers by accident claims only very slowly, in the middle of the twentieth century. Before that time, firms were operating on the basis of other groupings for risk. But eventually company experience showed that young drivers were decisively worse insurance risks than others, and so people under a certain age all became the object of much higher rates as a risk group, without regard to individual records (figure 9:1).90 Among health workers, thinking in terms of risk factors represented a major change in understanding human illnesses. A risk factor such as high blood pressure was found to predict death rates in a given population, and if there were multiple causes or unknown causes, statistical conceptualizations of diseases could be very important. Indeed, insofar as shifts in lifestyle could control risk factors, health advocates had found a powerful weapon to ward off illness: changing people’s lifestyles. The most notorious risk factor was smoking cigarettes, an unambiguous behavior, and indeed it was work on statistics concerning that particular lifestyle that changed the science of epidemiology substantially beginning in the 1950s.91

185

186

ch a p t er 9

figure 9:1. Insurance journals at midcentury were including illustrations to reinforce the industry campaign to justify charging higher insurance rates for young drivers, implicitly confi rming that some drivers had substantially more accidents than other drivers. The caption under this photo was “Young drivers have too many accidents—and very serious ones.” Source: Casualty and Surety Journal, January 1950, 48.

The idea that having accidents was a risk factor for having further accidents was, however, for many late twentieth-century experts, as I have suggested, not a useful idea. The idea of risk factor could, however, translate into another way of thinking: grouping into a risk collectivity everyone who was subject to a particular risk factor. And so it was that, from one point of view, people who had accidents could constitute one or more risk groups, depending on how one classified the accidents. The idea of risk groups as such (as opposed to “risk”) came explicitly into the technical accident discussion only in the 1960s. The first article title in Medline to include the term “risk group” applied to any medical condition appeared in 1955. The second article title, from 1963, actually reported that various groups of people who attempted suicide were more or less likely later to act successfully to kill themselves. This research built

Eclipse of the Idea Among Experts

directly on evidence that followed the accident model, starting with the idea that people who committed suicide often had a record of attempted suicides, just as “accidenters” had records of having accidents. Through the rest of the 1960s and after, the idea of risk groups became familiar in public health publications.92 In 1983, an official report of a committee of the American Medical Association spelled out how the transition had occurred in connection with accidents: Many studies of persons affected by traumatic injuries have made it apparent that the injuries are not distributed randomly, or “accidentally,” among populations, just as lung cancer, coronary artery disease, and infections such as whooping cough and malaria are not random occurrences. Rather, injuries in general, including those related to motor vehicles, are likely to involve specific groups of persons who can be described by age, sex, occupation, personal actions, and other characteristics. In addition to the examination of such host characteristics, studies of injuries also involve scrutiny of the key epidemiologic variables of agent, environment, and time. Thorough analyses of such variables provide a solid foundation for the understanding of any health problem and may lead to the development of effective preventive measures.93

t hink ing in t er ms of r isk grou ps The idea that children, and in particular active boys, were at risk to be hurt came easily. The practical consequence was that one should watch them closely, whether or not one was thinking of them as “accident prone.” 94 In the second half of the twentieth century, another population group based on age, old people, particularly came under the same sort of classification. That is, old people got hurt more often than other people.95 But at that point, experts were thinking less and less often in terms of accident proneness and more and more often that all old people fell into a risk group. And before long, the essentially insurance term, “risk,” came to be applied to the young as well as the old, and to people with disabilities as well as those in certain age groups. This shift was not just a change of classification or a new way of seeing people. As I shall comment further later, when injury prone people became mere demographic segments, those concerned with safety adopted a new strategy. They no longer targeted individual members of the risk groups for special attention and treatment. Instead, they applied

187

188

ch a p t er 9

engineering measures that would protect a person in the risk group, whether or not the person was specially accident prone. That is, it was no longer necessary in the last part of the twentieth century to discriminate between people on the basis of individual proclivities to get hurt. “Accident susceptibility in the elderly” came from physiological decline and illnesses common among all members of the group, not a special personal attribute or personality trait. Each old person had a general accident inclination, like the g factor that I noted earlier. And for all of them, staircases, slippery surfaces, stoves, and automobiles were environmental hazards (figure 9:2).96 Remarkably, this was exactly the same outcome that

figure 9:2. The idea of aged people as a major risk group hardened substantially in the late twentieth century, as illustrated in this cartoon. Source: Garrett E. Snipes, “Accidents in the Elderly,” American Family Physician, July 1982, 117.

Eclipse of the Idea Among Experts

resulted from another safety effort that was contributing to sidelining accident proneness, as I have already noted: accident analysis.97 In the middle of the century, as the longevity of populations in Westernized areas increased dramatically, a number of experts began to pay attention to old people in a different way. In factories, wrote an American official, “the only disadvantage of older workers, the data indicate, is that their disabilities last longer once they are injured. But they are, on the whole, less likely to be absent as frequently and perhaps less likely to be injured than younger workers.” Other investigators, however, began to find age a significant factor in accident frequency. If one tried to separate the factor of age from the factor of experience (workers with more experience had fewer accidents—unless early on the accident prone had already been weeded out), age as a factor in accidents could trump experience.98 By the end of the 1960s, experts were tending to talk, not of accident proneness, but of accidents among people for whom age was the defi ning characteristic for the group.99 Among drivers, everyone knew that young people, and males in particular, had more accidents and were “a distinct high-risk group.” Old people matched the unfortunate accident records of young drivers.100 But old people, in cars or elsewhere, were particularly easy to identify, and their experience in suffering injuries was obvious to most observers.101 But beyond accident analysis (a subject to be examined further in the next chapter), once age was used to define “risk groups” for accidents, it was easy for experts to try to use other external characteristics (as opposed to forces operating internally) to define other groups with high accident rates. Authorities knew that in traffic accidents, alcoholics as well as young people accounted for a substantial disproportion of accidents. So users of mind-altering drugs also caused accidents in undue numbers. But so did some types of hemiplegic patients. And the list of groups at risk could expand indefinitely. Sometimes the label “accident prone” was used, and sometimes not. In a hospital in 1982, for example, administrators developed a “profile” of the “accident prone employee” (“has a lower salary, is older, fails to record his height, weighs more, has a longer length of employment, and lives in a less affluent section of the city”). This profile obviously consisted of factors that were verifiable without mental tests or individual clinical inferences. And the report was part of a program now labeled as “risk management.” 102 One of the obvious risk-group categories was that of sex. Boys in fact had very different rates of injury from girls and so would be a special risk group (figure 9:3). Yet as males aged, manliness was also perceived

189

190

ch a p t er 9

figure 9:3. In the 1960s and 1970s a new hazard appeared in the workplace—a hazard that had previously been restricted to women: “The problem of long hair [in men] is international in scope. Above is the subject of a poster produced by the National Occupational Safety Association of South Africa pointing up the danger.” Source: A. B. Maines, “Is the Job Getting in Your Hair?” ASSE Journal, August 1972, 34, reproduced through the kind permission of NOSA, South Africa.

as skill in avoiding accidents, and carelessness and risk taking could be negative attributes for men.103 The category of sex was problematic from the beginning. Women workers were already singled out as a special group, as noted repeatedly in earlier chapters. But in general the dangerous workplace was perceived as a masculine space, as opposed to the less hazardous areas of female employment.104 From the beginning of the twentieth century, various writers (as well as sexist comedians) discussed women as motor vehicle drivers and connected their supposed characteristics with a tendency to incur auto accidents.105 In 1955, for example, a speaker’s discussion of accident rate differentials on the basis of age brought out in discussion a strong statement about women as a general risk group— without using the term.106 With the idea in the late twentieth century that accident proneness could be temporary, and with the increasing emphasis on temporary medical conditions, those who sought to classify femaleness as a risk fac-

Eclipse of the Idea Among Experts

tor and women as a risk group hit upon attempting to connect menstruation, a more or less measurable condition, with an increased accident rate. In the 1960s, a London practitioner found a connection between menstruation and accidents. A German investigator could not confi rm her findings, but tentatively connected accidents to menopause.107 As in that study and others, the shift from personality trait to measurable physiological change indicated how risk groups were replacing a trait of accident proneness. As late as 2002, a writer from Imperial College, London, invoked the idea of “risk-prone behavior” to note that “the rise in accidental and violent death among men coincides precisely with the onset of puberty.” 108 The equating of risk group with accident proneness was substantial. Sometimes the phrasing was very suggestive, such as an “increased susceptibility to injury” group.109 As early as 1965, a California investigator tested the state law that denied driving privileges to people with certain chronic disease conditions. He found that “drivers with diabetes, epilepsy, cardiovascular disease, alcoholism and mental illness averaged twice as many accidents” as a control group. He also noted that age, attitude, severity of medical condition, “and previous accident experience” also had to be taken into consideration. But it was clear that except for “attitude” and a nod to the repeater factor, chronic medical conditions, plus age, created groups at risk—groups that could be confi rmed by measuring the number of accidents they incurred. And these were chronic groups, not a repeater group into which people could be thrown by temporary upsets or illnesses.110 In the last part of the twentieth century, authorities worked to manage the risk shared by a group of people by creating conditions that prevented accidents among either the general population, accident vulnerable or not, or among the large groups of people who fell into high-risk groups. As an editor commented in 1967, yes, an older worker may have more accidents. Yet “he” is a good worker, and “his productive value is such that it is well worth the effort necessary to adapt the job to him as he grows older.” Yes, both the elderly and their caregivers should have safety training, and, yes, physicians and others should look for physical causes of accidents in each elderly person, wrote a Scottish physician in 1973. But, he continued, many problems could be avoided by simply designing housing and town environments with simple safety designs, such as full banisters on stairs, elimination of steps, and convenient railings. A decade later, a New York gerontologist noted that it was possible to identify elderly accident “repeaters” among hospital patients. Again, staff gave them

191

192

ch a p t er 9

individual attention, but still engineering efforts were called for—safer beds and wheelchairs especially. Indeed, “all repeaters in the hospital were restrained after their fi rst accident, but this did not prevent them from having subsequent accidents.” Clearly by identifying various risk groups, the potential to engineer safety was limitless.111 But so was the potential for social discrimination. In 1992, for example, two accident experts suggested setting different and more rigorous standards (allowing fewer “points”) in order to disqualify aged drivers.112 Risk groups had the potential of implying deviant identities for members of that group. Even at the end of the twentieth century, using the idea of risk groups often brought investigators back to accident repeaters, even if not back to accident proneness. “Male sex and previous injury” were significant predictors of more injuries in children. And the further conclusion was that educational and other preventive programs might better be directed indiscriminately at large populations rather than at targeted risk groups defined by social or other criteria.113 Again, engineering means (“passive” safety as opposed to an “active” educational safety program) also covered large populations and often did not depend upon defi ning risk groups or those who were accident prone.114 As one expert worried about the connection between engineering and risk groups put it in 1998: Should we say about males, ‘let them get injured—it is biological and nothing can be done about it’? Only a social scientist would conclude this. Social wisdom has always known that many people are naturally predisposed to do things we preferred they not do, including injuring themselves and others. We know plenty of ways to reduce injury rates for everyone by creating a safer environment. But this strategy will still leave males with high injury rates.115

Where systemic errors were added to personal accident susceptibility or inclination or proneness, engineering again could cover all people as well as all of the systems in which they encountered technologies that caused injuries. Neither accident proneness nor even risk groups were essential to engineering safety.116 Knowing about risk groups, however, could suggest areas on which to focus engineering efforts.

10.

BYPASSING ACCIDENT PRONENESS WITH ENGINEERING

The last chapter followed experts as accident proneness went through stages of retreat and substantial transformation into risk groups. In this chapter, one last dimension of the history of the idea of accident proneness comes fully into place: how did people concerned with traumatic injury and safety find it so natural, as they were abandoning accident proneness, to turn instead to using engineering to meet the social problem of accidents? In order to explore this great social displacement, it is necessary to go back in time and reframe slightly what was happening in society and in public discourse as the fate of accident proneness played out. As people then were often acutely aware, the idea of accident proneness raised fundamental human problems. The beginning of the story is familiar. Before the idea of accident proneness, injuries and errors came to people because of bad luck or the will of God. Or they occurred because someone could be blamed for carelessness or acting without proper forethought. Or possibly the person was clumsy, which was a physical attribute. But the cause was human beings. Europeans were more sensitive to worker rights than were North Americans, but to people of the early twentieth century, labor was a commodity, and in an employment situation, one could discharge the person who was the source of any kind of trouble or bother. Beyond that, there was at first 193

194

ch a p t er 10

no systematic way except education and discipline to reduce the grim, scandalous accident rate.1 And it is true that still in the late twentieth century, safety education and safety rules and laws continued to constitute a major resource by which societies attempted to reduce injury and other damage. With the idea of accident proneness, experts could associate some or many of the injuries and errors with people who showed a pattern of causing trouble. And with that new conviction, responsible leaders in various societies faced the problem of how to stop those accident prone people from initiating accidents. From the 1920s on, various thinkers tried to figure out how to keep the worst offenders from operating or interacting with machines.

t he probl em of u nsat isfac tory wor k er s a nd dr i v er s It was obviously not a problem to remove an employee who was drunk or who did not follow safety rules (the number of workers who refused to wear inconvenient protective goggles and subsequently suffered eye injuries surpasses imagination; figure 10:1). But when someone was a conscientious worker and still attracted accident, through no apparent willing or conscious action, could that person be separated from employment? By the 1920s, psychologists and managers had devised many means to keep undesirable employees from being hired in the first place—most notably not only individual interviews but aptitude tests. The additional quest for tests to identify accident prone job applicants, as I have noted, was international and intense. Since selecting employees passed for creating a meritocracy, few people objected. During World War i and especially in World War ii, psychologists had worked to devise tests for quick emergency reactions and other desirable traits in military and other aviators. In the end, they came to focus on very specific reactions, not general personality such as might include accident proneness. But investigators working to select pilots did not resolve the confusion between personality and specific reactions any more than did their colleagues in other fields. Pilots likely to have accidents and a personality trait of accident proneness were two different things.2 Screening pilots or employees thus constituted one widely accepted procedure. But when capable people already employed could not help themselves, and had “bad luck,” how could managers justify discriminating against them? In fact, employment conditions were often brutal, be-

Bypassing Accident Proneness with Engineering

figure 10:1. This photo of a British worker welding a pipe joint in 1946 caught a significant event. “At the exact moment of exposure a spark [flew] off from the welding rod . . . towards the right eye of the welder [and] was deflected by the protective goggles, which thus saved the welder’s eye from an injury.” The unspoken message was, what if the worker had not been wearing goggles? Source: “Welders—Wear Goggles!” Industrial Safety Survey, 18 (1942), 97.

yond the fact that workers were often turned out on the slightest excuse. A personnel expert in 1949 asserted that if all other safety measures failed, then (using the customary formulation), high-accident workers “should be either transferred to less hazardous departments where their abilities can be used or eliminated from the force. They constitute an ever-present hazard not only to themselves but to their fellow workmen.” But then he added harshly, harking back to a much earlier formulation, which he now associated with accident proneness: “It is obvious that many of these accidents are caused by sheer stupidity and/or lack of safety know-how, and stand to be repeated by these accident-prone individuals.” 3 In the 1920s and especially the 1930s, a number of psychologists who were spreading the idea of accident proneness were transparently humanitarian in their inclinations.4 As I noted above (chapter 6), experts as different as Morris Viteles and J.-M. Lahy were sympathetic to the point of view of the workers. Even as these experts served management, they insisted repeatedly that their findings of accident proneness should not be used as a reason to deprive a worker of his or her employment, although many did want to exclude them from working with hazardous

195

196

ch a p t er 10

machinery. Farmer, Chambers, and Kirk in 1933 suggested that “to remove a beginner with a high degree of accident proneness from a dangerous department to a less dangerous one might entail no hardship but have only good effects.” 5 After World War ii, in the first wave of technical criticism of the idea of accident proneness, social considerations continued to come up explicitly. A Cambridge psychologist in 1948 noted that his tests showed a predisposition to a “disorder of function” that led to accidents, but he backed hurriedly away from “eliminating the accident-prone from underground employment.” 6 A California public health expert in the same period openly attacked the idea of discrimination against accident prone people. The hazardous environment in which they functioned was equally at fault, he wrote, and society should change the environment (which was a necessary element in any accident) and not depend on finding and changing “hidden psychological motivation” to determine either safety or employment.7 In the second half of the twentieth century, attempts to “eliminate” accident prone people from employment in which they operated machines troubled experts on two counts. One was a simple matter of practicality. If engineering, even costly engineering, could reduce the toll from accidents, it was simply easier to take that route. Discriminating against people was both socially and politically too difficult. 8 As a Swedish neurologist pointed out in 1960, even in the obvious cases of persons with epilepsy, who might be dangerous drivers, attempts at legal means to prevent epileptics from operating motor vehicles had met with notable lack of success. For one thing, he wrote, “It is very unlikely that compelling physicians to report [anyone with epilepsy] would do anything to increase highway safety. If an epileptic has an ordinary sense of responsibility, he can be made to desist from driving anyway.” 9 The second problem that made resort to engineering solutions the easy way out was concern about fairness—reflecting, as in the 1930s, an increasing willingness of people (to varying degrees, depending upon the local standards) to emphasize the rights and privileges of individual citizens against the imperative of the common good. This preoccupation with the fear of injustice appeared, for example, in the 1980 monograph of the U.S. National Academy of Medicine and National Research Council on Injury in America, simply on the basis that the data “on the groups that are particularly susceptible to injuries” were inadequate, except possibly in the area of traffic. In the area of traffic, it was possible, as noted above (chapter 6), for the British to react in the 1930s as if tests for accident

Bypassing Accident Proneness with Engineering

proneness would constitute a “third degree” torture. Thirty years later, two American writers observed, “If stringent tests were used in licensing, the screams of rejected driver’s license applicants would be heard in every state capitol.” 10 Writer after writer pointed out that even in the case of traffic accidents, eliminating a group from driving involved possible injustices, indeed, inevitable instances of unfairness. After many years of experience, “Who,” asked some Canadian investigators in 1991, “is caught in the sieve? ” Others pointed out that test results were not foolproof anyway, and altogether the growing awareness of the complexity of factors in accidents made safety people cautious about restricting opportunities for anyone. Indeed, they could warn explicitly about that new social sin, “blaming the victim,” in this case, the victim of the accident.11 Nevertheless, the one continuing exception was in the area of motor vehicle and transportation operators. For drivers, the North American system of “points” to remove from the roads offenders against driving rules (with or without accidents) was widely copied, on the basis that there was a well-known high correlation between having accidents and driving dangerously.12 The concern about accident prone tram, bus, and train operators on the European Continent that had initiated intense interest in the idea of accident prone individuals in the World War i era continued. It spread geographically around the world, and it extended especially to airplane pilots.13 It spread even in the face of criticism of the idea of an accident prone person, for in this case the public good overrode concerns about fairness and faith in engineering safety—or even doubts that testing and selecting were fair.

t y pes, t r a i ts, a nd her edi t y In the early twentieth century, thinkers and policymakers in Europeanized countries, as I have shown, struggled with the problem of actions that had bad social consequences (like errors and injuries) but were not consciously willed by the person. Accident proneness obviously fitted into the category of unwilled actions. And many thinkers invoked types and traits in standard arguments. Originally, accident prone people were simply by defi nition those who had an abnormal number of accidents—the Unfäller, the accidentistes, etc. But, as I have noted, those like Marbe, who pioneered the idea, described the tendency to have accidents as a special trait—a personality trait. Immediately the idea fell into the classic controversy of that day, heredity

197

198

ch a p t er 10

versus environment. Was the trait of accident proneness inherent, so that people who met the definition did not bear responsibility for their actions? The trait, moreover, as I have suggested, rapidly morphed into the idea of type.14 There was a type of person who had accidents, and there was another type who was free of accidents. Although public discussion was usually not clear, many thinkers believed that traits could be inherent or that a person was inherently of one type or another. The trait argument often was modeled, as I have suggested previously, on discussions of intelligence: one could test many talents, but underlying them all was a g factor of general intelligence. So in accident proneness, people who had occupational accidents were also likely to suffer injuries at home or in traffic. Using that model, many psychologists, especially, attempted to test whether or not a person accident prone in one setting would also be accident prone in another.15 The type argument, that a person who had accidents represented a human type, was more common among Continental scientists. Or one could be of a particular personality type, a type in which accident proneness was only one element (extravert or introvert, for example). In 1966, one psychiatric team divided people who got hurt into the accident type and the suicide type—which in itself was a basic denial of the dynamic interpretation that all of those who kept getting injured were unconsciously but directly self-destructive.16 Also in the second half of the twentieth century, some investigators began to assert that there were other, different types of accident prone people, so that the category of accident prone could be understood further by personality or reaction types. In traffic, for example, there was a type labeled the “high-risk driver.” Two Minnesota psychologists noted that “the concept of a personality type which is accident prone is difficult to defend, but the concept of habitual patterns of attitudes and behaviors which contribute differentially to the occurrence of accidents finds some support.” 17 And by late in the century, under the influence of cognitive psychology, types had become “styles.” 18 Those who believed that some people had more accidents than others (not by chance) therefore offered a wide range of medical, biological, and psychological explanations as to why one person differed from another. Yet if there was an inclination, a proneness, a disposition, a predisposition, a susceptibility, or an Anlage, again and again thinkers came back to the question, what was the origin? 19 And of course in that question was another: can the accident prone person be cured? Many times the hope of controlling or curing accidents conditioned how an expert thought about origins—just as Lahy and Korn-

Bypassing Accident Proneness with Engineering

gold in 1936 rejected the hopelessness in Marbe’s formulations of types of accidenters.20 Twenty years after Lahy’s declaration, another French expert, J. Zurfluh, concluded that any apparent disposition or predisposition to accidents was the result of psychosocial factors, not a biological trait. Anglo-American writers, too, could cite empirical studies that cast doubt on the idea that accident proneness was a hereditary trait (siblings of repeaters, for example, were not repeaters). Altogether, advocates of safety campaigns simply concluded that “although statistically one person may be concerned in more accidents than another, this is not a valid reason for attributing ‘accident proneness’ as an innate quality of that person . . . In any case it would be wrong to think of anyone as incapable of benefiting from a scheme to prevent accidents.” 21 Despite such pressure, even in the second half of the twentieth century, the question of possible hereditary or “constitutional” traits or tendencies to have accidents continued to surface directly or indirectly. A Harvard investigator in 1949 used the new language of host-agent-environment but still concluded, “Accident proneness is a host factor related to social attitudes and psychologic differences, and yet apparently is as inborn and fi xed as any other inherited characteristic.” In 1954, a University College, London psychologist had difficulty in reconciling his findings that although brothers did not demonstrably show similar accident patterns, accident-free fathers produced a notable percentage of accident-free sons. A 1971 study showed that twins in fact tended to share the characteristic of being accident prone when it was present.22 Once again, risk groups raised troubling questions. Left-handedness was an apparent genetic trait that included significantly more “injury proneness,” as an Arkansas group concluded in 1993 (18.1% as opposed to 10.5% in right-handed children). To them, these individuals were members of a group who shared a “risk factor.” 23 Similar conclusions, that left-handedness was a personality trait that led to accident proneness, had come out of Marbe’s department already in 1929.24 As late as 1999, a group of Greek experts found that left-handed children ran a higher risk of upper-limb injury than did controls. Members of this risk group were demonstrably identified by an inherited trait. Faced with the possibility that there exists a hereditary factor in rate of injuries suffered, the Greek experts suggested three types of preventive measures: (1) parental alertness that goes beyond acceptance of left handedness and incorporates guidance and advice on both behavioural and technical issues; (2) notification of teachers and other care givers so that they

199

200

ch a p t er 10

will provide a supportive environment both during educational and sports activities; and (3) ergonomic adjustment of instruments, when feasible, and changes in the layout of furniture and other equipment to increase compatibility of the macroenvironment with the needs and preferences of the left handed.25

This summary indicates transparently the persistent attempts of safety experts in the late twentieth century to take special pains, especially through engineering means, to fit the psychosocial and physical environments to people who might be labeled disadvantaged when in fact they might be endangered.

soci a l de v i a nce Particularly in the United States for several decades after World War i, thinkers often suggested that, among people, differences in patterns of behavior were social constructions, and those who were out of style suffered discrimination because elements in society labeled them “deviant”—not because of any inherent flaw or behavior pattern. There was then the further conclusion that differences—deviance—should not be negatively sanctioned. Of course people at the time sometimes conceptualized accident prone people, who were by defi nition different, as a socially deviant group. In the 1930s, a writer in The Times sensed the social consequences of identifying people who found that they fell into the group of unfortunates who caused accidents: “Give them time, and they will grow proud of the distinction . . . they will found clubs; and to be accident-prone will be to despise the normally capable as the professed plain man despises the intellectual.” By the 1950s, as was noted earlier, investigators were correlating accident records with court and social agency encounters and suggesting seriously that accident repeaters might be lumped and labeled with other social deviants.26 At the high point of interest in accident proneness, people who might initiate injuries presented a social problem: Society cannot wait for the accident prone to reform themselves. Many of them cannot do so. Most of them are unable to recognize, or at any rate to acknowledge their addiction. In any event, we cannot let their self-esteem continue to endanger the lives of others as well as their own . . . we must be on the lookout for accident prone people, and then protect ourselves against them.

Bypassing Accident Proneness with Engineering

They were, according to such thinking, like pyromaniacs or sex offenders, and society needed in a parallel way to identify and restrain the accident prone.27 In the 1960s, investigators were noticing that people who showed the most social conformity—the direct opposite of what would become deviance—suffered the fewest accidents. Moreover, “the injury-repeater tends to feel less inclined toward accepting or complying with rules, standards and social customs.” 28 And within a few years, the term “deviant” appeared frequently in discussions of accidents.29 In the light of continuing studies of violators of traffic regulations and of a variety of young people, there was indeed a high probability that those who broke rules also broke bones and suffered other injuries. Concluded a 1970 investigator in the United States, “There can be little question that, on a behavioral level, accidents are closely associated with various indices of behavioral deviation.” He reasoned further that methods applied to correcting delinquency and inducing more social conformity might help reduce accident rates. 30 Investigators increasingly noted the social class differences that affected both accident rates and deviant behavior. Many earlier accident studies appeared flawed because they were limited to privileged social groups or, as in traffic violations, investigations targeted disadvantaged population segments. It was well know in the postwar decades that those already suffering from various kinds of social discrimination would show higher rates of both physical and mental illness. Accidents showed similar patterns, as in one 1975 study of U.S. hard-core unemployed and “ghetto” hires: “The most striking fi nding in this review was the very high rate of on-duty accidents among hard-core and ghetto hires”—surprising and alarming because, contrary to stereotype, the same population did not in fact show increased health problems. 31

st r at egies to avoid t he ide a of acciden t proneness With all of this baggage of hereditary predisposition, fairness, and a possible new category of social deviant, it was no wonder that, after the 1950s and 1960s, safety and other experts tried simply to bypass the idea of accident proneness. Indeed, by the end of the twentieth century, the pattern of the ultimate fate of the idea of accident proneness had come into sight. As indicated in the preceding chapter, the idea persisted in a few very specific areas, chiefly driver and pilot testing, and to a limited extent in the clinical field. But otherwise safety experts utilized two strategies

201

202

ch a p t er 10

simply to dispense with the idea, even though accidents (if not accident prone people) constituted an enormous social problem throughout the world. One bypassing strategy was to expand the old tactic of engineering safety, the outcome to which I have already alluded repeatedly. The other strategy was to reconceptualize the human factor in accidents. Reconceptualizing the human factor in turn grew out of two trends. One was that of ever more closely analyzing accidents. The other was the rise of cognitive psychology, which turned investigators away from the basic psychological approaches within which psychologists for decades had worked out and advocated the idea of accident proneness. Not only did behavioral and dynamic psychologies become less commanding in psychology, but leading cognitive psychologists and their allies attacked the whole idea of persistent personality and personality traits, one of the foundations of most ideas of accident proneness. 32 Emphasizing cognitive processes rather than emotional processes gave a fundamental rationale to engineering. 33 Emotional elements in accident causation were inherently individual. Cognitive processes were more universal. That distinction had two consequences. First, general engineering for safety, based on cognitive patterns, avoided the individual and could be applied to everyone. Second, the solution for individual emotional differences that caused accidents was to discriminate against the individuals. To deal with cognition, not only was engineering a solution but traditional safety education, possible with cognitive elements, could be invoked. Cognitive psychology thus brought optimism into the safety movement, as well as offering a new approach to understanding why accidents happen.

compl e x i t y l e a ds away f rom acciden t proneness As safety experts from a variety of backgrounds reviewed the possibility of using accident proneness to explain why accidents happen, the experts continued through the twentieth century to apply higher and higher standards of analysis to their empirical data about accidents (figure 10:2). This continual upgrading has been remarked on above, and it seems so natural that the cumulative significance might be underestimated. Clearly authorities were repeatedly disappointed by both the quality of accident data and the analysis of the data. With surprising steadiness, they therefore moved away from considering humans and possible human attributes and examined instead the environment and the specific situation in which individual accidents occurred. As early as 1962, an American psychologist criticized his colleagues: “Instead of studying the

Bypassing Accident Proneness with Engineering

figure 10:2. French safety engineering experts in the 1970s were analyzing the injuries incurred, and how, in this BMW that smashed into another car. The male driver, age 34, 1.65 meters tall, weighing 61 kilograms and wearing a seat belt, had a thoracic contusion. The front seat passenger, female, 34 years old, 1.55 meters tall and weighing 50 kilograms, suffered a sternum fracture and lumbar aches. Source: F. Hartemann et al., “Description of Lateral Impacts,” in Sixth International Technical Conference on Experimental Safety Vehicles (Washington: U.S. Department of Transportation National Highway Safety Administration, 1977), 558.

causes of individual accident, they have been overly preoccupied with why one person has more accidents than his neighbor . . . It seems easier to determine retrospectively the number of accidents an individual has had than to do the careful qualitative analysis of the pre-accident behaviors and other events which contributed to any one of them.” 34 Yet in the end, the experts found baffling complexity in the human and contingent events that were involved in each accident. And so they turned increasingly to seeking engineering and technical solutions to safety problems, solutions that did not depend upon differences between individuals but covered all expectable human frailties and to some extent differences in circumstances. For years, aside from the theories and objections of statisticians, the reliability of data underlying many studies of accident proneness continued to melt under the scrutiny of hard-headed investigators. 35 How, for example, in transportation, can one lump collision and pedestrian or pas-

203

204

ch a p t er 10

senger accidents under one heading? How can one trust an employee or a supervisor in a factory to give an accurate account of what happened so that the accident can be classified for statistical purposes? —especially when each was motivated to show a good safety record. 36 Or, what happened when the data collected by different investigators resulted in contradictory analyses or did not correlate with each other at all? 37 From the earliest efforts, many experts found the collection and processing of accident data unsatisfactory. 38 For a time, as I noted earlier, investigators tried to get at “unsafe actions” as opposed to the people who acted. In that way, accident prone people would be simply those who behaviorally acted in such a way as to bring on an accident. And in the 1970s, experts could report that isolating the actions from the people did indeed predict clusters of accidents. 39 This focus on unsafe actions that caused accidents was still a circular definition, but at least it eliminated the category of accident prone people, a category that included such items as a child passenger injured in a car crash—who should not be included among the accident prone drivers who committed unsafe actions. One pioneering investigator, for example, attempted to use motion picture recordings to put actual individual actions on film so that one could try to see what unsafe acts consisted of. Another team used a simulator in which the experimenters suddenly thrust a dummy into the path of a car and then analyzed the driver’s reactions. This group commented that “such accidents as rear-end collisions, loss of control on turns, accidents during rain or snow, and accidents related to mechanical malfunction are usually grouped together. They may have completely unrelated human characteristics associated with incidence and severity of outcome.” 40 In any event, safety experts increasingly wanted to dissect each accident to find all of the factors that were involved. A Cincinnati team in 1965 summarized the new viewpoint: It may be seen that accidents are multi-factorial in origin, that each employee has a certain accident potential which is variable and dependent upon many internal and external factors. Universal risk, abnormal environment, and personal physical attributes should also be considered. The severity of interpersonal problems, the individual’s satisfactions or dissatisfactions act together with ever-present dangers in his environment, with chance factors, in interlocking fashion, to determine the accident potential of an employee in any given point in time.41

Bypassing Accident Proneness with Engineering

And since the analysis always showed complexity, any actions taken by safety experts to reduce accidents had to be individual, suited to the host, the agent, and the (now very) complex environment.42 This awareness of complexity produced two outcomes over the years. First, experts insisted that one had to manage everything—all of the people, all of the different types of environment, everything from general plant morale to street lighting and machine and furniture design.43 Complexity in the second place led directly to engineering and technological solutions, for changes in the material environment could at least correct physical circumstances that furnished elements that went into any accident. In 1980 in a transitional book on psychological research on accidents and safety, a leading German authority found that in traffic safety analysis, at least, it was possible to maintain that accident proneness existed— perhaps phenomenologically—as a personal factor or as a statistical group factor. But he pictured either type of factor as merely an element in the whole system that produced an accident, a system dominated by environmental factors found through accident situation analysis.44 Even before 1980, many other authors in safety publications were explicitly rejecting the idea of accident proneness in favor of concentrating on how to engineer the environment. Wrote an Australian investigator, “In accident investigations, the desirable technique of identifying the links in the causal chain so that countermeasures can be introduced is neglected in favor of the hunt for a scapegoat . . . This is a thoroughly negative approach. This emphasis on carelessness reached its zenith in the doctrine of accident proneness which, though now discredited, nevertheless retains some adherents.” 45

er ror By the 1990s, analysis of accidents had reached a point that made the whole idea of an accident look very different from earlier conceptualizations. Neither the idea that someone had a proneness to have “accidents” nor the idea that someone was an accident repeater received much attention. Major writers instead were referring to “errors” and systemic flaws.46 Yet human factors persisted and even grew as a concern, as a Swedish expert explained, with perhaps a little irony: (1) . . . technological systems have become more complex, hence more difficult to control; (2) . . . improved models and methods for “human

205

206

ch a p t er 10

error” analysis have made this cause more likely; (3) . . . technological systems have become more reliable, hence raising the relative number of other causes; or (4) . . . it is sometimes cheaper and more convenient to put the blame on a human than to redesign an entire system.47

Conceiving of human “error” grew out of the teachings of cognitive psychologists, whose work was referred to in a different context in the last chapter. Beginning especially in the 1960s, accident psychologists’ views were deeply affected as the behavioral and testing psychology that dominated most of at least American psychology eventually gave way to the new emphasis on cognitive processes (figure 10:3). And since psychologists were the main supporters of the idea of accident proneness, the institutional underpinnings of that idea diminished greatly. Psychologists at first began inquiring into errors in cognition in terms of failures in attention. Even in everyday, habitual tasks, people could be distracted and produce errors—usually considered to be slips or lapses (explained in terms of cognition, not a Freudian unconscious). The people who made slips or lapses at once recognized that they had made an error. Psychologists wrote of different “styles” of cognitive processes that produced errors. Some people were just more absent-minded than others. In-

figure 10:3. An emphasis on complexity and organization in conceptualizing accident processes. Source: James T. Reason, “Understanding Adverse Events: The Human Factor,” in Clinical Risk Management: Enhancing Patient Safety, ed. Charles Vincent, 2nd ed. (London: BMJ Books, 2001), 15, reproduced with the kind permission of Wiley-Blackwell.

Bypassing Accident Proneness with Engineering

vestigators could also by this time show that stress of various kinds could increase the numbers of “attentional slips and memory lapses.” Indeed, James Reason, who became the best-known figure in explaining error, was referring casually in the late 1980s to “proneness to cognitive failure” as “a stable disposition” and people who are “excessively error prone.” 48 The Three Mile Island event in the United States in 1979 particularly helped focus experts on error, as opposed to accident.49 Thinkers began to write particularly about errors that were involved in system failures. Indeed, ultimately (and ironically), experts could ask, “Are some types of systems more prone to accidents? ” as did one group in 2000. And in a parallel to the idea of differences among individual people, these experts answered in part, “Some systems are more prone to accidents than others because of the way the components are tied together.” 50 Reason, in his fundamental survey of 1990, took up the idea that a system, rather than a human being, might have “a[n accident] pathogen” in it, and he used the occasion to distance error theory from accident proneness: The resident pathogen metaphor has a number of attractive features, but it is far from being a workable theory. Its terms are still unacceptably vague. Moreover, it shares a number of features with the now largely discredited accident proneness theory, though it operates at a systemic rather than at an individual level . . . Accident proneness theory failed because it was found that unequal accident liability was, in reality, a “club” with a rapidly changing membership. In addition, attempts to fi nd a clearly defi nable accident-prone personality proved fruitless. 51

Reason advocated fi xing the system, for example to insist on routine maintenance that would counter a memory lapse, rather than using general technological fi xes to cover possible individual errors that might cause accidents. He also distinguished between errors that consisted of “slips, lapses, trips, and fumbles: execution failures,” on the one hand, and errors that were “mistakes: planning or problem solving failures.” Reason was reinforcing the safety experts’ tendency to focus on fi xing mistakes that had produced systems that were not protected against slips and lapses. That is, error analysis reinforced attempts to engineer safety.52 Thinking in terms of systems underlined how policymakers were shifting their concern away from personal and small-scale accidents to

207

208

ch a p t er 10

figure 10:4. A system accident, or disaster, affecting animals, plants, and a local human population: in July 1976, in Seveso, Italy, a runaway reaction in a plant producing trichlorophenol caused a chemical spill over about 700 acres. This victim acquired a painful and unsightly chloracne in the ear. Source: E. Homberger et al., “The Seveso Accident: Its Nature, Extent and Consequences,” Annals of Occupational Hygiene, 22 (1979), 339, reproduced with the kind permission of Oxford University Press.

disasters and environmental damage. Wrote Jobst Conrad from Germany in 1980, “With the increasing importance of complex large-scale technologies, problems relating to safety and the environment have taken on a new significance . . . the development of a research area called risk assessment can be regarded as the response to a societal demand.” And indeed, safety was often being reconceptualized in terms of major hazards threatening many people and often operating slowly (figure 10:4). Beyond the Three Mile Island paradigm were general problems such as chemical pollution, food additives, radiation, and infrastructure soundness (as in dams, bridges, and pipelines). Almost all called for engineering remediation or at least better design and explicit protection against individual cognitive error. 53

erg onomics a nd hu m a n fac tor s ps ycholog y Eventually also playing into the rise of cognitive psychology was a legacy from World War ii that had a substantial impact on applied psychologists who might be interested in accidents. The result was two movements that

Bypassing Accident Proneness with Engineering

ended up much allied but still often distinct: not only ergonomics, but, in addition, human factors psychology. Both dealt directly with the interface of humans with technology. 54 Ergonomics tended to succeed the older time-and-motion studies and traditional psychotechnology, including worker selection. Initially depending heavily on anatomy and physiology, ergonomists worked not to fit the worker to the job but more and more to fit the job to the worker, as their new slogan proclaimed. Ergonomists expanded their mission to designing as much of the work environment as they could. Eventually, to quote a 2001 text, they sought to “maximize safety, efficiency and comfort by shaping the design and operation of the technology to the physical and psychological capabilities and social needs of the user”—a formulation that could apply to consumer products as well as the work situation. 55 Human factors psychologists, who tended to be concentrated at first in North America, traced their ancestry to the development of systems theory during World War ii, in which psychologists had to work with complex interactions between technological systems and human beings. These psychologists found that traditional single-factor stimulusresponse psychology did not meet their needs. In the founding textbook of human factors psychology, published in 1949, the authors, all from Johns Hopkins University, explicitly traced their roots to time-and-motion studies and employee selection techniques. The job now, they wrote, was to outfit machines with dials, instruments, and controls so that humans could operate them safely. The authors included little explicitly on accidents—often, even at this early date, referred to under the more general category of “errors”—because the importance to safety of all of the material in the text was obvious. They were, after all, interested in “the limits of human skill, speed, sensitivity, and endurance,” beyond which of course came error and breakdown. 56 Some of the human factor work was traditional safety thinking. Everyone knew about the tragedies that came of a worker’s defeating a safety switch or device (a problem at one point responsible for half of the machinery accidents in factories) (figure 10:5). Or designers could fail to guard or test equipment, a problem in consumer items as well as workplace machinery. 57 But as continuing transit—and especially airplane— accidents, when analyzed deeply, showed, humans operating machines continued to make errors.58 As the decades went by, however, and cognitive approaches to psychology became more dominant, human factors psychologists more explicitly reflected the cognitive approach. Later in the twentieth century, the

209

210

ch a p t er 10

figure 10:5. Throughout the twentieth century, safety experts had to deal with the serious problem of humans who inactivated or bypassed safeguards on machinery, as exemplified in this Italian safety poster from as early as 1927: “Beware! Do not remove safeguards!” Source: “New Posters,” Industrial Safety Survey, 3 (1927), 112.

increasing use of computers reinforced the idea that examining closely the ways in which humans perceived, reasoned, made decisions, and took action could help improve humans’ complicated interactions with their technological environment. 59 This approach turned out to be exactly the opposite of the approach of looking for accident prone people and getting those people to change or leave. In 1969, the British National Institute of Industrial Psychology had a section on “Industrial Accidents and Ergonomics” and was focusing on job design and explicitly avoiding the idea of accident prone workers. 60 Like the ergonomists, then, the human factors psychologists worked to devise engineering solutions to safety problems. Indeed, there was throughout the second half of the twentieth century substantial overlap between experts trained and working as engineers and other kinds of safety workers. There were times, too, when human factor experts had to work with engineers who were unsympathetic to human factors. For decades, however, human factors psychologists more and more often found

Bypassing Accident Proneness with Engineering

that they had to confront not only individual design problems but whole systems. There the human factor could affect safety, but in the end, system design tended to end up in the hands of engineers. 61

t he f ina l r ise of engineer ing to domina nce The engineering approach to safety problems that ultimately rendered the idea of accident proneness largely irrelevant was, as must now be obvious, not new. Already by the time of World War i, as suggested in chapter 1, guarding machinery, removing physical hazards, and devising safety devices were commonplace measures all over the Europeanized world. 62 In heavily unionized areas of Europe, as I have already noted, this engineering emphasis in occupational safety was particularly strong. In other areas, engineering continued to be important in the safety movement. Beyond laws and driver training for traffic safety, for example, highway and automobile design (not least the brakes!) were effective in modifying the results of human behavior and so reducing injuries and death—just as the ergonomists were beginning to claim. 63 It is true that highway design proceeded faster than vehicle improvements, which the vehicle manufacturers opposed effectively, but the trend to engineering protection was obvious.64 In 1952, a public health pediatrician could list examples of other kinds of engineering that were preventing children’s injuries: “a pair of plastic scissors that cut nothing but paper, a plug to cover exposed wall sockets [electrical outlets], and a plastic drinking container” that would not shatter if dropped. Perhaps the most spectacular safety achievement of the second half of the twentieth century was the placing of safety caps on medicine bottles so that little children could not poison themselves.65 All along, however, many experts perceived that neither the engineering of safety nor the conventional education and management effort (still using the motto of “Engineering, Education, and Enforcement”) 66 would suffice. By focusing on the human factor, they created an opening for identifying accident prone people. And after World War ii, as I have indicated, many experts continued to insist that fi nding the accident prone people offered great potential for social benefits. A management consultant in 1948 for example challenged the engineering approach directly in order to emphasize accident proneness: As long as an organization feels that the approach to accident prevention lies primarily in machine design, machine guards and the physi-

211

212

ch a p t er 10

cal safe-guarding of hazardous areas, allocation of the responsibility to engineering is logical. More and more, management will realize that physical conditions are not the determining factor in accidents but that accidents are essentially a psychological problem; a matter of symptomatic behavior of the individual. Then the need for re-allocation to a department specializing in human relations will be perfectly obvious. 67

Into the 1960s, experts interested in applying the idea of accident proneness to injury prevention continued to emphasize that engineering strategies were inadequate. As two investigators commented sarcastically after an extensive and thorough survey of injuries to children, the data did not suggest that any particular strategy would work. In the case of children’s falls from high chairs, “Mothers can be warned (repeatedly) of the general lack of healthy acrophobia in normal children; a widely publicized campaign to saw the legs off high chairs may be instituted; and more efficient shock-absorbing floor coverings may be designed and their purchase urged. To date, however, control programs of this nature have not been attended by striking success in reducing accidents.” 68 Despite such skepticism, in the end, many streams led to technological and engineering solutions to safety problems. Psychologists and physicians did not hesitate to advocate technological fi xes for individuals whom they identified as “accident prone.” Although faulty eyesight had long been mentioned as a factor in accident proneness, after World War ii, safety experts advocated technological means to diagnose and correct visual problems in both workers and drivers who were having accidents (in part through a program encouraged by optical equipment maker Bausch & Lomb). Simple eye tests turned out to be helpful, but vision defects could be subtly partial or complex—and still a major cause of accidents. Even those defects, however, usually were subject to simple technical remediation—correctly prescribed glasses, which “cured” accident proneness. 69 Visual problems represented just one example. No matter what the bodily defect, experts devised technological devices to prevent the defect from causing an accident—including in 1983 a technical arrangement to keep railway locomotive drivers vigilant and able to pay attention to important cues.70 Experts also used epidemiology of both direct and complex elements to advocate technological and engineering solutions. Officials in a Welsh steel company in 1959 found that a quarter of their accidents were hand and finger injuries. Looking more closely, the injuries were concentrated

Bypassing Accident Proneness with Engineering

figure 10:6. Everyday, simple safety technologies continued at all times to control the effects of accidents. “Above is an actual photograph of the hard hat that saved the life of an employee of Weirton Steel Co. . . . . The employee was working near the base of a dust catcher, while other employees were doing construction work above. A steel wedge which they were using became dislodged and fell 87 feet, striking the workman point down, on the head, penetrating his hard hat to a depth of two inches. Miraculously, the man escaped without a scratch.” Source: Ohio Industrial Monitor, 19 (1946), 158.

on the thumb, index finger, and third fi nger, plus the backs and palms of hands. A new, specially designed glove was introduced, and, despite a growing labor force, the number of hand injuries per month dropped from 200 to 128—a very inexpensive but effective engineering device (figure 10:6).71 When accident analysis and epidemiology created a demand for technological and design measures to prevent injuries, the nature of the “human factor” took on a different look. Wrote the Harvard safety expert, Ross A. McFarland, in 1962, “It is obvious that basic research on the human factors in automotive safety should be expanded so as to provide convincing evidence for additional safety considerations in design . . . The precise measurement of impact forces in motor vehicle crashes, both in regard to the equipment itself and to the occupants, is being carried out experimentally. Additional design criteria should result from these studies.” 72 And as more attention went to risk groups, shown by epidemiology to be vulnerable—“the elderly, the handicapped, the infant or the toddler”—

213

214

ch a p t er 10

so in the safety literature did engineering solutions appear for each one. Sometimes, of course, engineering solutions could themselves constitute a hazard for a special group, as did the automobile glass tinted to increase general safety by reducing glare. That particular improvement could reduce visibility for a risk group—old people—at night and so cause a different kind of accident.73 Another expert offered an argument that concentrating on changing human behavior had delayed some sensible safety measures. The personas-cause theory, “if applied to the widespread thermal and electrical injures associated with early house-wiring systems . . . would have led to concentrating on attempts to influence human behavior rather than the development of the fuse.” 74 Advocates of engineering safety of course were increasingly aware of the work of ergonomists and human factor psychologists as well as other types of investigators. Including psychological and social knowledge about humans therefore became not an alternative to engineering for safety, but a part of the process. Already in 1962 an American safety expert stated baldly that engineering was the way to go: “What engineering revision has done is to make the job relatively safer despite the man . . . The old tried and true principle of building safety into the job, to make it safe despite the man, must always be used.”75 A later writer offered the example of a series of injuries that caused managers to focus on the workers who were having the repeated accidents, only to have an engineer point out that the equipment they were using was inherently unstable and under normal use would tip over!76 Simple technological devices could enhance any safety effort. Painting stripes to separate traffic on roads and later putting in reflective markers to help drivers is an obvious example of engineering that saved countless lives.77 The extent to which safety shifted from attention to humans to attention to physical changes in the immediate environment is remarkable. In a British book on child accidents published in 1977, for example, low-technology changes in the physical environment clearly constituted the best and most direct means to achieve safety. One investigator went to look at a window through which a child had fallen and was seriously hurt. The investigator found that the window was so designed that the investigator himself almost fell out when he checked to see how far it would open. The experts simply ignored efforts to identify the victims and attempt to educate them. Engineering was the obvious solution.78 Moreover, at the end of the twentieth century, advocates could show that engineering had saved untold numbers of lives in an environment more and more dominated by technology.79

Bypassing Accident Proneness with Engineering

In the second half of the twentieth century, as the systems model came to include input from the human factors experts, the end product was still engineering. 80 Systems engineering appeared very advanced and sophisticated not only because it dealt simultaneously with many different kinds of factors, but because it did include input from many types of specialists. Moreover, systems permitted managers to measure tradeoffs between conflicting agendas in the interdependent elements in the system—including the views of safety experts. Safety as a part of systems tended to come into manufacturing first, and only later into traffic, because of the extreme complexity that experts found in traffic. Motor vehicle transport safety, commented an engineer in 1968, has been guided by the “if only” principle. “If only drivers were more careful, accidents wouldn’t happen.” The industrial safety effort is guided by the “since they do happen” principle. Since men and machines do fail, it is rational, economical, and moral to anticipate and provide for their inevitable failure.

The idea, then, was to “anticipate every type of accident that may occur” and build in safeguards. “All areas of safety engineering, such as road design and vehicle design, must deal with the question: ‘How will the drunk react? ’ ” 81 The engineering approach was therefore aimed essentially at overriding the individual differences that were basic in the idea of accident proneness. Indeed, there was an underlying egalitarianism in the engineering approach. As a Michigan analyst saw in 1980, everyone believed that children who suffered socioeconomic deprivation suffered more injuries than privileged children. But engineering protections could affect all children, without regard to class or class-determined parenting skill or neglect. 82 Safety experts, while trying to accommodate a broad range of individual behaviors, all of which could be rendered harmless, nevertheless did take into account the risk groups that were so conspicuous in the literature. But the adjustments were engineering adjustments. As fi rst one risk group and then another appeared in professional discussions, engineering answered safety concerns for each one. A Scottish physician in 1975 showed how the transition from a clinical approach to engineering the physical environment was proceeding: “The old lady with a fractured femur should excite clinical curiosity. Why did she fall? Does she have undiagnosed disease or is her house dangerous? ” 83

215

216

ch a p t er 10

As late as 1995, a French public health pediatrician could recognize that, beyond engineering, epidemiology still raised questions in the quest for safety. There were still “the psychosocial components of the risk approach, with the still unclear and unsolved problem of ‘accident proneness.’ ” 84 But, in general, safety advocates bypassed the problem by engineering safety. 85 As Australian safety expert Colin Cameron wrote in 1975, For the past 20 years or so, reviewers have concluded, without exception, that individual susceptibility to accidents varies to some degree, but that attempts to reduce accident frequency by eliminating from risk those who have a high susceptibility are unlikely to be effective. The reduction which can be achieved in this way represents only a small fraction of the total. Attempts to design safer man-machine systems are likely to be of considerably more value. 86

f rom indi v idua l dif f er ences to ega l i ta r i a n t echnolog y Using engineering and technology to protect people from encounters with technology was therefore not just ironic. It embodied what happened in the last part of the twentieth century. 87 Slowly, almost imperceptibly, but inexorably, societal leaders in developed countries, in a major shift in strategic emphasis, turned away from manipulating people and instead emphasized overwhelmingly the strategy of engineering safety as a way of dealing with the dangers that came from their technological civilization. 88 As a Finnish safety expert explained, “A minor slip, normally totally harmless, may become a critical factor. The protection therefore should not be a change of behaviour, but an error-tolerant, forgiving technical solution.” 89 Such a major cultural change of course, as I have shown, did not occur without support and input from other social streams—epidemiology, ergonomics and human factors psychology, and, one should emphasize, the availability of wealth to create devices and systems that would accommodate great variations in patterns of individual human behavior.90 Indeed, what happened in safety shows how fundamental was the reorientation of Europeanized societies over the decades as all of the streams came together in the effort to engineer safety. In that world, there was no place for the idea of accident proneness. Even as engineering continued to be the dominant approach to con-

Bypassing Accident Proneness with Engineering

trolling accidents, officials in government, business, and nonprofit entities persisted, it is true, in their traditional programs of education and discipline. At the turn of the twenty-fi rst century, moreover, one could hardly avoid noticing that the persistent problem of disparities in accident incidence between individuals continued to surface, particularly in traffic injuries. Road safety experts recognized that some groups contained inept drivers, particularly the aged. But many other individuals had a “disposition” to crash. The disposition was cognitive and reactive, and Anglophone experts, particularly, were writing about trying to give those with a bad record a new self-identity, that of a “safe driver.” The experts believed that working with a negative identity, such as a bad or dangerous driver, was not productive. Certainly they tried to avoid the idea of an accident prone person. Moreover, the idea of accident proneness as part of one’s personality carried with it all of the conceptual baggage of “personality.” By contrast, simple “disposition” was much easier to work with if one was going to change human behavior.91 Nevertheless, even this discussion of differential accident records took place within a discourse in which a context of engineering improvements to avoid injuries was simply assumed. A leading expert, British psychologist Robert West, wrote in 1997, “Individuals are indeed consistent over time in their accident liability . . . there is clear evidence that certain individuals are at considerably greater risk of accident than others.” Yes, West noted, enforcement, education, and screening would help reduce the accident toll. But his concluding recommendation was “engineering modifications to roads and vehicles that take account of the deeply ingrained nature of bad driving practices.” 92

217

CONCLUSION

Throughout the preceding chapters, I have tried to set up several ways of viewing the history of the idea of accident proneness. For closure and conclusion, I emphasize again a number of interpretive contexts, including: the history of accidents; the history of safety movements; the history of psychology and psychological science; the strategy of individual treatment; the great social shift to a technological society; and, by continual allusion, the eternal problem presented by the non-uniformity of human beings. In the end, the history of the idea of accident proneness provides insights into the ways in which people in Europeanized cultures interacted with their built environments, their technology. The leading edge of technology took the form of devices, tools, and machines. The awful, obvious price that they all exacted from human bodies ultimately created a new social problem: accidents. In a society in which labor was a commodity, as in the late nineteenth century, careless or unlucky people could be dealt with arbitrarily. But as the twentieth century unfolded, human encounters with technology became so frequent and so intense that even a slight change in the incidence of human error could have powerful personal, economic, and social effects. The “accident” became a spectacular as well as everyday problem in Europeanized societies. In factories, on the road, in recreational areas, and in the home, identifying “misfits of the machine age” held great promise to reduce the costs of human error.1 Were there people who made more errors and suffered more injuries than others in their societies in the twentieth century? Could a history of having accidents serve to predict that any particular person would have further accidents? The answer is not clear. Perhaps one’s viewpoint determined the answer. Without being certain of the details of how and why, 219

220

Conclusion

one could, however, agree with Ulrich Beck, writing late in the century, that risk was distributed unevenly.

nat u r a l izing acciden ts Virtually all discussions of the problem of accidents started out with a recital of how many people were killed or injured each year, or the economic costs of errors and injuries. Throughout the twentieth century, the figures continued to be horrifying. Between 1925 and 1928, when the idea of accident proneness first appeared formally, almost 1,000 children were killed in street accidents in London alone—and for each fatality, another 35 were injured. Already in 1951, the press in the United States, with reporters watching as the statistics mounted, carried stories about who would be “the Millionth Victim,” the millionth American killed in an auto accident since 1900. As late as 2000, one research group concerned with all kinds of accidents counted in the United States 149,075 fatal injuries for just that one year, 1,869,857 injuries requiring hospitalization, and 48,108,166 medically attended injuries not hospitalized. At all times, worldwide, injuries affected younger people most often, disabling them from productive activities far more than, typically, an illness would.2 And accompanying such figures was a generally unstated further concern: the economic costs of production errors and property damage. Moreover, over the years, each author concerned about accidents wrote with the awareness of the paradox that in the Europeanized world, public attention, concern, and policy were minimal and certainly not proportional to the magnitude of the problem of accidental traumatic injuries. 3 Both ordinary people and opinion leaders perceived accidents as a “natural” part of the cost of industrialism, which they equated with modernity and progress. Later, in the increasingly post-industrial world, thoughtful people naturalized accidents by viewing them as a product of the clash of a natural human body with an also natural (or historically inevitable) built environment. In a workplace, for example, the worker’s body was in an ecological relationship with the workplace environment and with the larger technological system.4 This late twentieth-century environmental thinking could go further. As part of an ecological system, accidents were normal as well as natural phenomena. Moreover, in the fluid, chaotic processes of natural systems, specific events were unpredictable—and the more so in unstructured situations such as the home or recreational settings, where routines and movements were much less scripted than in the workplace, and therefore even more “natural.” 5

Conclusion

As the evidence in this book shows, one type of naturalistic thinking about accidents was commonplace in the late nineteenth century. Every action had consequences. Typical bourgeois thinkers in Westernized countries therefore believed that human beings could, by appropriate behavior, control their own destinies. The model in place in the early twentieth century, however, was that humans could go on and control nature. Altogether, then, as technology intruded on their lives, people thought that accidents took place when someone who should have been able to control events did things wrong instead of doing them right. Yet psychologists in Germany and Britain then saw in statistics another natural phenomenon: some people were suffering more accidents than others. And the accident-free Nichtunfäller were also just natural products, as were the accident prone. The two groups were perhaps like another familiar dualistic category, drunk rather than sober. But, again, and unlike carelessness or drunkenness, moral responsibility was not attached to the natural trait of accident proneness. The whole point of searching for naturalistic causes of errors and injuries was, of course, to try to set up methods to prevent them. When rules, discipline, dismissals, and persuasion in safety movements did not adequately control injuries, the idea of accident proneness provided a resource with which to devise further measures that could diminish the human and economic costs of injuries. In 1980, a California expert listed six types of safety hazards in the human environment: 1. 2. 3. 4. 5. 6.

infections and degenerative disease; “natural” catastrophes; failure of large technological systems; discrete, small-scale accidents; low-level, delayed-effect hazards; and sociopolitical disruptions.

It was of course all very well to think globally and environmentally. Such thinking about safety served the good social purposes of extending the idea of safety to all kinds of risks and systems in the environment, broadly speaking.6 But one might point out that, when aggregated, everyday “discrete, small-scale” unforeseen incidents—that is, accidents in the ordinary use of the word—remained an appalling, costly problem in the twenty-first century. The new post-industrial, post-accident proneness thinkers had still not succeeded in making the accident statistics become insignificant.

221

222

Conclusion

r ise a nd fa l l As we look back, then, from a time much later than the 1920s, it is possible to see that for half a century and more, experts and managers applied the idea of accident proneness in active, practical ways. Yet, eventually, most people working at the junction of person, technology, and society forsook the concept.7 The marked rise and fall of a distinctive scientific idea continues to constitute a historical social phenomenon. Thomas Kuhn made familiar the idea that, in history, scientific concepts depended upon communities of thinkers who would leave one set of ideas and move to another. It was not, Kuhn showed, that the old paradigmatic belief was definitively refuted. Young, or not-so-young, scientists simply moved on to new questions, just as they quietly abandoned old theories of mountain building when they shifted their attention to tectonic plate movements. In the history of ideas, and in the history of scientific ideas specifically, historians often pay particular attention to the origins of a new concept, as I have done in the case of accident proneness. That idea had come in just when circumstances had prepared many people to fi nd an idea of inclination to accident useful and attractive. The factory system, governmental organizations, and insurance schemes had furnished statistics so that clusters of accidents became noticeable. The fearful toll of injuries, plus growing corporate responsibility for transport accidents, intensified concern about any mishap. Moreover, as some perceptive people at the time noticed, accident proneness offered another means by which managers could deflect attention from their own shortcomings and direct it to individual workers or groups of workers. At the same time, mental tests and the idea of personality traits were flourishing in psychology, and psychologists believed behavior was often automatic or unconsciously motivated, such as one would find operating in an accident habit. With the psychologization of the West, the idea of accident proneness was diffused on the basis of common beliefs, whether technical or popular. Further, beyond the theoretical model, specific industrial, technological, and social dilemmas caused more and more people to use the idea. Altogether, it is easy to see why, over decades, so many experts and managers thought that they could decrease accidents by putting in charge of machines only those people who did not habitually have accidents. Yet late in the twentieth century, experts began to make strong statements rejecting the idea of accident proneness, some, as I have noted, even

Conclusion

going so far as to call it “a myth.” 8 Increasingly often, in many venues, the concept no longer appeared. This shows an unmistakable decline in the history of the idea of accident proneness—despite the lasting appearances and fleeting reappearances of versions of the idea into the twenty-first century.9 The late twentieth-century transition, from a focus on the individuals who had accidents to improving technological systems to accommodate individual differences, was reflected in many areas of culture. Before World War ii, writers on the great impact of technology on the world tended to set humans in dramas based on the human-machine interface. At one point, dramatists and artists were exploring the clash of humans with the devices that those humans had created. Late in the twentieth century, the drama was one in which humans were torn between their desire for technology and the attraction of the organic or natural, as underlined by parts of the environmental movement.10 Post-industrial environmentalists brought an additional factor that moved the focus of safety concerns away from individual accidents. Their big picture, of interacting systems, tended to focus more on disasters, including nuclear war, than on aggregated personal injuries. It was catastrophes that fascinated people.11 Media presentations made people conscious of large plane crashes rather than the passenger who tripped on a step or the worker who fell between baggage carts. Moreover, ecological concerns could be long term, such as polluting the general atmosphere with chemicals, rather than a single sudden collision or exposure. It was amidst such new ecological emphases in picturing the world that the gradual turn from accident proneness to engineering took place. Indeed, since everyone in every system possessed the “strong human tendency to make mistakes,” targeting the most accident prone came, in the eyes of experts and policymakers, to appear to be futile.12 In 2005, an announcement appeared on the internet, “Watch America’s most accident prone family in their cartoon shorts.” By that point in time, then, it was possible to portray accident prone people, like many other social deviants, as mere objects of humor and ridicule—albeit it turned out in this case to be in an ad for a health insurance company! So far had the idea—although still being used—departed from the serious and even tragic conceptions of the earnest people who had originated and employed it (figure c:1). When an idea has faded away in the communities where it had once originated and flourished, historians have not been so attentive as they are to innovation. Scholars focus on “new” successor ideas, without paying

223

224

Conclusion

figure c:1. Still a common, perhaps humorous, idea in the twenty-first century. Source: Reproduced with the permission of Creators Syndicate.

much attention to what happened to the older idea.13 Or they typically follow the expansion of innovation, as in “technology transfer,” again without noticing, except by the way, what was displaced.14 Just as in the case of the origins, it is also possible, as suggested in the preceding chapters, to invoke changing circumstances to understand why accident proneness or any other idea faded away. Beyond the social questions of inequalities and discrimination that a trait of accident proneness aroused, when cognitive psychology came in and began to replace the older behavioral and dynamic psychologies, psychologists had new ways to approach accident behavior, ways that not only were based in cognitive processes but worked very well to further the shift to engineering as the means to prevent accidents. Accident proneness therefore appears to fall into the common pattern of historic ideas that once flourished and then declined. In these cases, it is tempting, as I have suggested, to cast an idea in terms of the organic metaphor: birth, growth, maturity, senescence, and, ultimately, death. One of the problems in doing so, of course, is that in the case of ideas, death is often very hard to prove. Old ideas keep showing up in new guises. And obsolete science keeps reappearing. Indeed, one of the main functions of popularizing science has long been to discredit ideas that formerly were quite acceptable science, such as Ptolemaic cosmology or phrenology, but later served as folk wisdom or superstition. Perhaps it was not inappropriate that as late as July 2006 on the U.S. ABC television network, it was an astrologer who declared that Gemini people are “accident prone.” 15

t r e at ing acciden ts on a n indi v idua l ba sis One partial legacy did persist through the stage of decline: treating each person with an accident record by individual, clinical means. Individual

Conclusion

treatment persisted not so much because of the idea of accident proneness as a trait or illness but because individual treatment was a major social strategy applied to all problems even after the middle of the twentieth century. This approach was part of the basic and high point of accident proneness as part of accident prevention, and individual treatment persisted as a part of the afterlife of an idea that most experts thought had died. The social compulsion for clinical use of the idea of accident proneness was deeply cultural. Just before World War i, a new general social strategy was evolving to solve problems that arose in the organizational societies of the West. The strategy was modeled on medical practice. Each time a person had a problem, a professional (often a doctor) would deal with that individual. Because it was the individual who had to be adjusted, the organizational system did not need to be adjusted. Thus, when an accident prone person could be identified, a professional could deal with that aberrant individual and try to make him or her fit in with the system, or, as was more often the literal case, with the machine.16 This clinical, one-on-one counseling model continued to operate through the end of the twentieth century. Sometimes the person who intervened was a manager, who worked with an individual employee. But often if accident proneness appeared as a symptom in a worker or driver, one or more clinicians could try to remedy that symptom. The model was dramatically deployed in the Detroit court’s psychopathic clinic for deviant motorists as early, as I have noted, as 1921. And while psychiatrists did not make accident proneness a significant illness or even an independent syndrome, when a problem pattern appeared in a patient’s history, clinicians of various kinds applied their talents to try to help the affl icted individual. “There are as many accident-proneness patterns as there are individual accident repeaters,” concluded one psychiatrist.17 A clinical model carried a residual belief that accident proneness or something functionally similar did exist, and physicians and other clinicians were obliged to deal with it. As a California clinical psychologist wrote tartly in 1970, “No clinical psychologist or psychiatrist experienced in working with individual patients could ever be convinced by mere theorizing that ‘accident-prone’ people do not exist.” 18 Nor was accident proneness the only tendency to which individual human organisms were subject and which demanded clinical treatment. At the end of the twentieth century, for example, some people appeared to physicians to be “infection prone.” That is, many experts on physiology held that nervous and other neuroendocrinological systems could transform strong external inputs, often conceptualized as stress, into immune

225

226

Conclusion

system vulnerability so that the individual organism would suffer a series of infections, strikingly parallel to a series of accidents.19 In either case, whether accident proneness or infection proneness, still many decades after the 1950s, an individual approach to the problem of a record of repeated accidents or illnesses appeared to clinicians to be appropriate. Where members of the public were endangered, as by mass transit operators or motor vehicle drivers, social standards crystallized so that an accident record could disqualify a person from operating the machines. Requiring physical examinations of each individual person was commonplace, and point systems for individual violations of safe driving spread in the last half of the twentieth century to keep many people off the roads. The rigorous Spanish truck driver testing, for example, lasted effectively through more than eight decades of great political stresses and changes.20 Airplane pilot records and tests were extremely selective everywhere. The individual approach to the accident prone person was therefore continually present in some areas of both private practice and public function.21 There were also two other practical arenas in which the individual approach to accident histories, construed as predictive tendencies, persisted after the apparent death of the idea. One was in the treatment of children with motor dysfunction. As I noted at the beginning, even the English word, “clumsiness,” came back into clinical use. A second was the urgent need of teachers and athletic coaches to deal with the fact that some children and some players were clearly prone to incur injuries (figure c:2).22 Yet in a world in which the very term “accident” was embargoed, much

figure c:2. A poster of the Corporation of Delta, British Columbia, Parks, suggesting the early twenty-fi rst-century technological protections available for a skateboarder who insists on indulging in the risky behavior of that sport. The poster reads: “It’s your body, our rules. Wearing a helmet, wrist, elbow and knee protection is required every time you’re on the Skatepark or in any Delta park.” Source: www.corp.delta.bc.ca.

Conclusion

less accident proneness, the practical and clinical applications of the idea became very limited. Despite, then, the occasional recognition that a series of accidents constituted a personal problem, experts in mainstream safety moved away from a focus on individual flaws or susceptibilities. But just as Newtonian mechanics continued to be useful in teaching and engineering long after theorists in physics had adopted new frameworks, so clinical understandings and regulation of accident repeaters had practical uses, notably in traffic regulation, long after opinion leaders had abandoned accident proneness as such as a basis for safety efforts.

acciden t proneness a nd moder nism In the introduction, I called attention to the “defining moment” at the end of the nineteenth century when, according to Roger Cooter, accidents became a public rather than a private matter. The gradual death of the idea of accident proneness, as I have described it, constituted another defining moment, albeit a very extended moment. At that point, accidents became a problem that engineers were to solve without specially identifying people who were responsible for a large part of the accidents. Machinery now had to be adapted to each individual, who was supposed to have freedom to be as deviant as he or she wished—an expansion of “the age of narcissism.” In such a world, what worked for safety, for all people, was engineering. The earlier idea of accident proneness, however, had resonated with much that defi ned modernity. I commented at the beginning of this book that accident proneness began as part of early twentieth-century modernity, with the factory system and mass labor movements. The rest of the story spells out many ways in which agents of modernity thought and operated as they had to deal with casualties of the machine age. When humans encountered technology, the questions of rationality, irrationality, mass solutions, individual differences, and other areas of psychology and social justice intruded. Many historians have reflected on the contemporary appearance of modern industrialization with the decline of formal thinking and the simultaneous growth of thinkers’ engagement with irrational forces. Again, accident proneness shows concretely how some of the modernist leaders tried to deal with unlucky people in their midst. In an intellectual milieu emphasizing irrational behavior, accident proneness had a special cogency, for thinkers could use the idea of un-

227

228

Conclusion

willed error to address the rapidly intensifying confrontation between humans and machines. In a later era, in which both intellectuals and marketers emphasized personal dignity and choice, it was appropriate that societal attempts to stigmatize, segregate, and control accident prone people would diminish greatly, along with the intellectual underpinnings of such policies.23 What I have characterized as the great shift to engineering came about not only because of the availability of resources but also because of new ways in which opinion leaders in Europeanized societies viewed fundamental questions. Not least of the circumstances facilitating the shift to engineering was the coming of what historian Paul Forman descriptively calls the post-modern age of the late twentieth century, that is, the era after modernity (not “postmodern” philosophy). In that new age, as Forman points out, technology, rather than science, provided a model for conceptualizing the world.24 It was in such a world that an engineering, rather than a human factor, approach to controlling accidents made sense. The idea of accident proneness had come out of an earlier modernist attempt to apply psychological science to a serious social problem. By 1985, however, a group of experts on traffic safety illustrated dramatically how complete was the shift to post-modernity that had taken place. They wrote with disbelief that half a century earlier, an official U.S. government report had “betrayed an overwhelming emphasis on collision prevention, with a special section devoted to identifying the collisionprone driver. It said nothing about improving the design of vehicles or roadways to minimize impact once a crash occurs” (figure c:3).25 One major issue implicit in the idea of accident proneness, however, overshadowed all others: the question of human differences, the inequalities of bodies as well as circumstances.26 Were there “accidenters” who were “industrially unfit”? 27 In late twentieth-century public discussions, labeling some people accident prone was not a welcome contribution. Many people had, after all, shrunk from accepting that nature or society had produced “born losers.” 28 Of course many commentators on risk taking in society held and hold that freedom to injure oneself and others should not be constrained, that risk taking, whether conscious or unconscious, is part of one’s personality and is a significant part of living. Indeed, as a number of writers pointed out, having accidents in a technological, post-modern world could constitute normality, could be just adapting to the culture.29 Or it was possible to make a cost-benefit analysis. As early as the 1930s,

Conclusion

figure c:3. In 1976, an advanced technology, the air bag to protect automobile drivers, careless or not, was being advocated at an international technical conference. Source: Charles Strother, Michael U. Fitzpatrick, and Timothy P. Egbert, “Development of Advanced Restraint Systems for Minicars RSV,” in Sixth International Technical Conference on Experimental Safety Vehicles (Washington: U.S. Department of Transportation National Highway Safety Administration, 1977), 456.

a German administrator bluntly asked how much one wants to slow down the production in a factory just to avoid accidents. An American insurance expert in 1969 could ask, in the face of traffic fatalities, “Does the malum of motorization outweigh the bonum? ” Or, as a number of writers have pointed out, pursuing risk in a variety of machines has been positively attractive in the West. In early twentieth-century Germany and

229

230

Conclusion

Britain, the initial publicity about injuries related to technologies such as the airplane made the victims of accidents into admired martyrs for developing safety for the rest of us—hardly a way to discourage people from taking chances. 30 Later in the twentieth century, a number of economists suggested that too much spent on safety made people feel safe so they would take more chances. Similarly, too high compensation rates, some experts held, would cause workers to hurt themselves to obtain disability benefits (the same idea that I noted was strongly argued just at the beginning of the twentieth century on the Continent). In a slashing attack in 1984, two scholars pointed out that these arguments were another version of the old idea of carelessness and the assumption of worker rationality. The “new” economics advocates were forgetting the element of irrationality that, in the real world, played into having an accident—an irrationality such as was posited by people who advocated and used the idea of accident proneness. 31 Clearly at least one successor idea to accident proneness was not an improvement over identifying employees and drivers who were accident repeaters.

acciden t proneness in histor ic a l per spec t i v e If the experts who in the twentieth century worked to diminish the modern phenomenon of accidents had one great lesson to pass on, it was that accidents are complex. Each accident was complex in itself. 32 Accidents in various aggregates were complex. And all of them were also complex in the specific historical contexts with which they interacted. 33 Much more intimidating complexity, however, lay in the interaction of the idea of accident proneness with general cultural trends. Ulrich Beck, whose work comes up repeatedly, has attempted to uncover the dynamics of the risk society that emerged in the late twentieth century. Members of that society had to calculate the advantages of technology against one’s personal advantage when the costs were unevenly distributed—just as in the wider marketplace standards that dominated most societies in that period. 34 Judith Green has particularly explored for the recent past the interrelationships between group risk and people’s understandings of personal accidents. “The accidental . . . as a category of misfortune,” she holds, cannot be “easily wrenched from the field of discourses that produce it: risk, uncertainty and responsibility.” Yet the fatalism that complexities of the modern risk society produced in statisticians and citizens alike, she points out, were not the same as the fatalism that existed before modernist rationality and attempts at prevention. 35

Conclusion

From all of the complexities, from theorizing or from analysis of actual mishaps, come major hints of the more distant profile of the history of accident proneness. The simultaneous, independent appearance of the idea, as I have emphasized, shows that a wider cultural context, a great social problem of technological societies, impelled thinkers to seek understanding of error and accident in the face of injury and waste. My surprising negative finding that psychiatrists did not adopt accident proneness as a medical entity is, however, unexpectedly significant. The fact that the professional experts most sensitive to cultural trends did not medicalize accident proneness suggests that as an intellectual construction, the idea of accident proneness grew and developed in a restricted, not a general, cultural milieu. It is true that the idea incidentally involved major cultural issues, such as heredity versus environment, but the idea of accident proneness was only secondarily and derivatively driven by such general cultural forces. First came the immediate crisis of mangled bodies and material losses. It was appropriate, therefore, that late in the twentieth century, experts shifted their focus from accident prone people to the technological fi xes with which humans attempted to control those injuries and damages. It may be that, as in other shifts in the history of scientific ideas, that which was new simply diverted leading thinkers’ attention away from the previous focus of attention. The fact remains that, beyond a limited clinical application, accident proneness was effectively displaced by engineering—in the safety movement and as a general social strategy. Moreover, that great shift was emblematic of the technological society of the late twentieth century that Forman has identified. As I noted in chapter 10, historians of technology have made a great contribution by balancing impressive accounts of technological development with thoughtful consideration of the environmental costs of technological devices and policies. 36 Scholars implicitly raise a question: are there ways to control hazardous technology that are better than trying to change people who initiate accidents? Yet, except when they explore ironies of technological expansion, historians in fact seldom take up the direct costs that mechanization exacted from material goods and especially from human bodies. 37 Recent scholars who write about the machinehuman interface typically refer to robotics and engineering psychology, not to injuries. 38 It is true that later, in the beginning years of the twenty-first century, some safety experts were trying to initiate a new phase in injury prevention and control. Their goal was to supplement engineering with strategies to change human behavior, using the cognitive, behavioral, and

231

232

Conclusion

social psychology of the day. Because “epidemiological principles” and environmental health models guided this thinking, these safety experts aimed primarily at community-wide change, which could include targeting risk groups. But they were also explicit that “risk communication” was designed to change individual as well as “community” behavior. The program thus reemphasized the education and enforcement elements of the traditional “Three E” approach, along with engineering. As one formulation put it, “Ultimately the best solutions will combine innovative approaches to modifying individual and community behavior, along with technical and environmental factors that enable and reinforce behavior change.” 39 At the least, looking at the stages in the history of accident proneness, as I maintain in this book, makes it possible to discern major shifts in the ways in which people of the Europeanized world dealt with their collisions with technology. The accident proneness idea did grow out of real conditions across their cultures, which produced the people who appeared to their contemporaries to be tragic “misfits of the machine age.” 40 Over decades, the fundamental types of problems did not change. Circumstances, however, did. In application, focusing on people who repeatedly made errors and got hurt did not solve satisfactorily the terrible problem of accidents with machines and the built environment. It is, as my narrative shows, in the decline of the idea of accident proneness among experts, a shift without the drama of the explicit simultaneity of the origin, that the outlines of the profound displacement from accident prone people to technological fi xes and engineering emerge. To a surprising extent, technological designs and fi xes did enable people of the late twentieth century to avoid questions of how to handle folks who manifested a pattern of making repeated errors. Experts still could not, in the end, translate into satisfactory prediction the common observation that some people had significantly more accidents than others. Or the experts could not turn the observation into action or policy without an intolerable amount of discrimination against some drivers, some workers, and possibly other categories of persons whose involuntary actions and lack of talents made them “different” or “other.” Still, it would be impossible to ignore the fact that there were commonsense and inconsistent exceptions for people operating public transportation machinery or monitoring major technological systems like a power grid. People who temporarily or consistently kept making errors were a symbol of the enduring encounter of humans with the technology they had created. A set of stairs, a kitchen knife, a cleaning fluid, a power saw, an airplane, an oil drilling platform, an automated chemical factory—all de-

Conclusion

manded some degree of human conformity. Persistent human differences therefore continued to produce unevenly distributed error and tragedy. Ignoring or ideologically denying individual differences in “inclination to accident” did not make the costs and tragedies go away. Nor did modifying technology invalidate the insight of a continuum of deeply concerned experts that some people had lots of “bad luck.” In the age of technology at the end of the twentieth century, engineers, psychologists, and other experts continued to try to make both simple and elaborate provision to overcome the inequality of individuals’ responses to technological contrivances. These implicit social perfectionists were, in effect, trying to achieve radical equality in the face of irreducible human differences. In attempting to control the dangerous unwanted consequences of using technology, these idealistic agents of society were not usually consciously egalitarian. Yet they had come to hope that with engineering they could protect all members of their communities, even people with a history of suffering accidents. For the moment, at least, the psychologists, engineers, and other experts would not have to face the problem of individual accident proneness again.

t he acciden t prone m a n See the accident prone man. His body has been stitched and pinned. He was run over by a car or attacked By mad adolescents with hatchets For no apparent reason. He will show you the scars, The ones on the outside. But look at him now. He is not crying or shouting. He is putting a foundling 20 Into his billfold. And this very morning he will meet A woman who will change his life.

But really, this is no way to be, Too many bumps, too many ups and downs. Things should be done, corrective things. Yet if God nods in a heaven of any sort, If the sky really sags with glad souls, The accident prone man May suddenly one day Find himself there, Like one who has dozed At a symphony concert And just now starts awake To the first bars Of his favorite piece.

f r a nz k . b a sk e t t, 19 9 4*

*Franz K. Baskett, The Accident Prone Man (Washington: Orchises, 1994), 12. A typographical error in the original has been corrected on the advice of the author, who has very generously authorized the reprinting of this poem.

233

NOTES

in t roduc t ion 1. As will be taken up in chapter 4, vaguely similar thinking—but without a specific formulation—also arose in the 1910s in the United States and France. French writers were not among the originators of the idea, and later French experts never did accept either the English or German terms and instead continued to speak of a “predisposition” or “disposition” to [have] accidents. As late as 1967, Jean-Marie Faverge, Psychosociologie des accidents du travail (Presses Universitaires de France, 1967), 126, was citing a colleague who introduced the term l’accidentabilité, which Faverge believed would avoid the implication in the English term of determinants that are “innate, constitutional, non-modifiable.” The language itself therefore foretold one important type of controversy the concept aroused for many decades. 2. Much of what follows is taken, by permission of Cambridge University Press, from John C. Burnham, “Accident Proneness (Unfallneigung): A Classic Case of Simultaneous Discovery/Construction in Psychology,” Science in Context, 21 (2008), 99–118, which includes a historiographical discussion. 3. The literature on “multiple discoveries” is summarized and explored in Augustine Brannigan, The Social Basis of Scientific Discoveries (Cambridge: Cambridge University Press, 1981), David Lamb and Susan M. Easton, Multiple Discovery: The Pattern of Scientific Progress (Amersham: Avebury Publishing Company, 1984), and Dean Keith Simonton, Scientific Genius: A Psychology of Science (Cambridge: Cambridge University Press, 1988), chap. 6. Buhm Soon Park, “The Contexts of Simultaneous Discovery: Slater, Pauling, and the Origins of Hybridisation,” Studies in History and Philosophy of Modern Physics, 31B (2000), 451–474, reviews the literature on simultaneous discovery from a recent point of view and offers a case study that deals with a phenomenon that appeared in two different disciplines. 4. See especially Robert K. Merton, The Sociology of Science: Theoretical and Empirical Investigations, ed. Norman W. Storer (Chicago: University of Chicago Press, 1973). 5. Further types of doubts are described in such publications as Ilana Löwy, “Variances in Meaning in Discovery Accounts: The Case of Contemporary Biology,” Historical Studies in the Physical and Biological Sciences, 21 (1990), 87–121. Alan G. Gross, “Do Disputes over Priority Tell Us Anything About Science? ” Science in 235

236

Notes to Pages 4–6

Context, 11 (1998), 161–179, explores the sociological aspects of discovery, with qualifications about time and agent: “A scientific discovery is the public attribution of novelty to a claim regarded by a relevant scientific community as possible and as the consequence of following appropriate methods” (168). 6. Frank A. Haight, “Accident Proneness, The History of an Idea,” Automobilismo e Automobilismo Industriale, 42 (1964), 536–546, draws the sequence in much more black-and-white terms, based solely on English-language research. This is the only freestanding explicit history of accident proneness that I have found, but it is in fact using a historical format to make a polemical attack on the idea. Haight’s article is therefore used below, in an appropriate chronological location, to exemplify growing opposition to the idea in the 1960s. Many other authors in the course of their expositions on the subject of accident proneness presented historical summaries, and Lynette Shaw and Herbert S. Sichel, Accident Proneness: Research in the Occurrence, Causation, and Prevention of Road Accidents (Oxford: Pergamon Press, 1971), turned a detailed and close reading of the literature through the 1960s into a de facto history that will be referred to repeatedly below. A recent contribution is Martin Lengwiler, “Gendering of Industrial Risks: Conceptions and Practices of Male ‘Accident Proneness’, 1880–1920 (Britain, Germany and Switzerland),” unpublished paper generously furnished by the author. 7. Donald R. Kelley, The Descent of Ideas: The History of Intellectual History (Aldershot: Ashgate, 2002), explores the literature on how the history of ideas has become increasingly problematic for philosophical historians. In practice, as my narrative will show, tracing the idea of accident proneness works. 8. The original formulation of “technological fi x” was Alvin M. Weinberg, “Can Technology Replace Social Engineering? ” Bulletin of the Atomic Scientists, 22 (1966), 4–8. 9. In the 1920s and even much later, psychologists often changed focus or pursued more than one specialization. They had in common a background in experimental psychology. I have sometimes tried to identify psychologists who were working consistently in a recognized subfield, particularly in industrial psychology, or on a particular subject such as (later) traffic psychology. 10. Ann M. Dellinger, David A. Sleet, and Bruce H. Jones, “Drivers, Wheels, and Roads: Motor Vehicle Safety in the Twentieth Century,” and Daniel M. Albert, “The Nut Behind the Wheel: Shifting Responsibilities for Traffic Safety Since 1895,” in Silent Victories: The History and Practice of Public Health in Twentieth-Century America, ed. John W. Ward and Christian Warren (New York: Oxford University Press, 2007), 343–378, suggest that just at the end of the twentieth century, Americans, at least, became partially disillusioned with engineering and reverted substantially to education in the area of auto safety. 11. See, for example, items catalogued under “accidents—psychological aspects—juvenile fiction.” 12. “Accident-Prone Lorries”; The Times, 7 March 2003; Beth Daly, “Barge Owner Has History of Oil Spills, Violations,” Boston Globe, 1 May 2003. 13. Claire Martin, “He Spent Accident-Prone Life Picking Himself Up,” Denver Post, 24 September 2006, about a man of many misfortunes killed in an auto accident and quoting his sister, “He was kinda accident prone, I swear to God, even before he was born.” (The term “accident prone” appears in the print version

Notes to Pages 6–8

headline, not the online version.) The Tom Cruise story is from the New Zealand Herald, 2 September 2004. Another story reported accident prone dog and cat pets (Pet Planet, 18 September 2000). Both of the latter items emerged from an online search. 14. John Penn, Accident Prone (New York: Doubleday, 1990 [c. 1987]), 110. 15. An incisive and relevant account is Hermann Loimer and Michael Guarnieri, “Accidents and Acts of God: A History of the Terms,” American Journal of Public Health, 186 (1996), 101–107. In the literature of accident proneness, concern over the defi nition of what an accident is persisted through the late twentieth century. 16. Robert Campbell, “Philosophy and the Accident,” in Accidents in History: Injuries, Fatalities and Social Relations, ed. Roger Cooter and Bill Luckin (Amsterdam: Rodopi, 1997), 17–34. Judith Green, “Accidents: The Remnants of a Modern Classificatory System,” in ibid., 35–58; the quotation is from 51. 17. Judith Green, Risk and Misfortune: A Social Construction of Accidents (London: UCL Press, 1997), provides a sophisticated overall history of thinking about accidents. 18. Events that happened because of malicious intent, as in a fight or a crime, have not been considered accidents, although, as will be noted below, in the late twentieth century, they often were conflated with accidents under the noncommittal heading of “injuries.” 19. Accidents in History, ed. Cooter and Luckin. See, for example, Elizabeth Towner and John Towner, “Developing the History of Unintentional Injury: The Use of Coroners’ Records in Early Modern England,” Injury Prevention, 6 (2000), 102–105. 20. Green, Risk and Misfortune; Andreas Killen, Berlin Electropolis: Shock, Nerves, and German Modernity (Berkeley: University of California Press, 2006). 21. Thomas R. Forbes, “By What Disease or Casualty: The Changing Face of Death in London,” Journal of the History of Medicine and Allied Sciences, 31 (1976), 413; Thomas R. Forbes, “Mortality Books for 1820 to 1849 from the Parish of St. Bride, Fleet Street, London,” Journal of the History of Medicine and Allied Sciences, 27 (1972), 25–29; Loimer and Guarnieri, “Accidents and Acts of God.” 22. See especially Karl Figlio, “What Is an Accident? ” in The Social History of Occupational Health, ed. Paul Weindling (London: Croom Helm, 1985), 180–205. Figlio emphasizes legal and labor issues. 23. Newton Bosworth, The Accidents of Human Life; With Hints for Their Prevention, or the Removal of Their Consequences (London: Lackington, Allen, and Co., 1813), iii–iv. 24. Again, the historical literature on safety movements is scant, although there are a few notable contributions regarding events in the United States. See Mark Aldrich, Safety First: Technology, Labor, and Business in the Building of American Work Safety, 1870–1939 (Baltimore: Johns Hopkins University Press, 1997); Michael Guarnieri, “Landmarks in the History of Safety,” Journal of Safety Research, 23 (1992), 151–157; and Tom Christoffel and Susan Scavo Gallagher, Injury Prevention and Public Health: Practical Knowledge, Skills, and Strategies, 2nd ed. (Sudbury MA: Jones and Bartlett, 2006), chap. 2, which also contains, 39–48, a chronology of landmark events in the United States, 1937–2004. Martin Lengwiler, “Between War Propaganda and Advertising,” in Medicine, the Market and the Mass Media: Producing Public

237

238

Notes to Pages 8–9

Health in the Twentieth Century, ed. Virginia Berridge and Kelly Loughlin (London: Routledge, 2005), 55–75, provides incisive descriptions of European trends, with comparisons to American safety efforts, and Jamie L. Bronstein, Caught in the Machinery: Workplace Accidents and Injured Workers in Nineteenth-Century Britain (Stanford: Stanford University Press, 2008), surveys the record and scant historical literature, pointing out especially the differences between worker perceptions and those of employers and legislators. Green, Risk and Misfortune, especially chap. 5, puts the professionalization of accident prevention into a broad context. Barbara Young Welke, Recasting American Liberty: Gender, Race, Law, and the Railroad Revolution, 1865–1920 (Cambridge: Cambridge University Press, 2001), traces transportation law in the United States to raise the question of who owned the new technologies and who was responsible for consumers (transport passengers), a question with a focus partly different from those raised in the occupational safety movement. H. S. Person, “Safety Movement,” Encyclopaedia of the Social Sciences XIII, 503–506 [1934], provides a contemporary history and analysis. 25. See, for example, the special issue, “Risk,” Daedalus, 119 (1990), no. 4. 26. Martin Lengwiler, “Gendering of Industrial Risks,” 3, quotes Carl Hoyos that “the history of accident psychology can be reduced to the history of the concept of ‘accident proneness.’ ” [“Die Geschichte der Unfallpsychologie ist weitgehend eine Geschichtes des Unfällerbegriffs.”] 27. David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007), describes the confl ict between those who contended that accidents were inevitable and those who contended accidents were avoidable. Aldrich, Safety First, provides a somewhat different framework. See my concluding chapters below. 28. H. M. Vernon, “The Neglect of the Human Factor in the Prevention of Industrial Accidents,” International Labour Review, 33 (1936), 780–781, for example, held the American campaign up as a model. 29. See, for a random example, Cynthia Illingworth, “Playground Equipment Injuries,” in Children, the Environment and Accidents, ed. R. H. Jackson (Tunbridge: Pitman Medical, 1977), 121–125. The issues and the concern of Anglo-American experts even at a very late date can be followed in John D. Langley, “The Need to Discontinue the Use of the Term ‘Accident’ when Referring to Unintentional Injury Events,” Accident Analysis and Prevention, 20 (1988), 1–8; D. C. Girasek, “How Members of the Public Interpret the Word Accident,” Injury Prevention, 5 (1999), 19–25; and S. A. Evans, “Banning the ‘A Word’: Where’s the Evidence? ” Injury Prevention, 7 (2001), 172–175. Other implications in the terminology are discussed below. 30. Roger Cooter, “The Moment of the Accident: Culture, Militarism and Modernity in Late-Victorian Britain,” in Accidents in History, ed. Cooter and Luckin, 107–157. Elsewhere, Roger Cooter, Surgery and Society in Peace and War: Orthopaedics and the Organization of Modern Medicine, 1880–1948 (Manchester: Centre for the History of Science, Technology and Medicine, University of Manchester, 1993), chap. 5; Cooter describes the changing relationship between accidental injuries and health care. Anson Rabinbach, “Social Knowledge, Social Risk, and the Politics of Industrial Accidents in Germany and France,” in States, Social Knowledge, and the Origins of Modern Social Policies, ed. Dietrich Rueschemeyer and Theda Skocpol (Princeton: Princeton University Press, 1996), 51. Bronstein, Caught in

Notes to Pages 9–12

the Machinery, is the most recent historian to present the sequence of events in Britain. 31. Oliver Wendell Holmes, in The Collected Works of Justice Holmes: Complete Public Writings and Selected Judicial Opinions of Oliver Wendell Holmes, ed. Sheldon M. Novick (3 vols., Chicago: University of Chicago Press, 1995), III: 170. 32. Use of the term “Europeanized societies” in this book recognizes that technology and the injuries that came with technology were also found in “developing” countries into which, over time, technology was coming, including rural areas of developed countries. C. J. Romer and M. Manciaux, “Accidents in Childhood and Adolescence: A Priority Problem Worldwide,” in Accidents in Childhood and Adolescence: The Role of Research, ed. M. Manciaux and C. J. Romer (Geneva: World Health Organization, 1991), 1: “It was long believed, and still is believed by some, that accidents occur only in developed countries, being the price that has to be paid for industrialization, technology, urbanization, and motorization. This is not true. In the developing countries, accidents are perhaps just as common, and their consequences are often more serious.” 33. G. H. Miles, “Economy and Safety in Transport,” Journal of the National Institute of Industrial Psychology, 2 (1925), 192. Eric A. Finkelstein et al., The Incidence and Economic Burden of Injuries in the United States (New York: Oxford University Press, 2006). World Report on Road Traffic Injury Prevention, ed. Marjie Peden et al. (Geneva: World Health Organization, 2004), 3. 34. Accidents in History, ed. Cooter and Luckin; Aldrich, Safety First. These have recently been supplemented by specialized studies such as Welke, Recasting American Liberty, and Mark Aldrich, Death Rode the Rails: American Railroad Accidents and Safety, 1828–1965 (Baltimore: Johns Hopkins University Press, 2006). Green, Risk and Misfortune, portrays preventive measures from several different viewpoints. 35. I myself have written about the shift to engineering solutions in connection with ghastly injuries to children. But until I was deep into the abstract idea of a person’s having a tendency, disposition, inclination, or proneness to have accidents and make errors, I did not see how, in the great shift to engineering safety, was embodied a history of modern humanity and human problems more fundamental than I had imagined. John C. Burnham, “Why Did the Infants and Toddlers Die? Shifts in Americans’ Ideas of Responsibility for Accidents—From Blaming Mom to Engineering,” Journal of Social History, 29 (1996), 817–837. 36. For example, Paul R. Kemmeter et al., “Colonic Injury and Intraspinal Penetration from High-Pressure Molton Plastic Injection: Case Report,” Journal of Trauma, Injury, Infection, and Critical Care, 44 (1998), 738–740. 37. “ . . . Accident,” Casualty and Surety Journal, March 1951, 14. 38. A typical dual-purpose defi nition, for example, is in the glossary of Paul A. Alcorn, Social Issues in Technology: A Format for Investigation, 3rd ed. (Upper Saddle River NJ: Prentice Hall, 2000), 282–283: “Technology. That whole collection of ways in which the members of a society provide themselves with the material tools and goods of their society. The collection of artifacts and concepts used to create an advanced sociopolitico-economic structure.” 39. David Edgerton, The Shock of the Old: Technology and Global History Since 1900 (Oxford: Oxford University Press, 2007).

239

240

Notes to Pages 13–14

ch a p t er 1 1. For example, “Epidemiology of Accidents,” Lancet, 11 April 1953, 730. Neil W. Rheiner et al., “Investigation of Compliance in the Severely Injured: A Literature Review,” Rehabilitation Nursing, July–August 1981, 12. Similar figures came from other localities. 2. Anson Rabinbach, The Human Motor: Energy, Fatigue, and the Origins of Modernity (New York: Basic Books, 1990), 228–234, summarizes some Continental developments. Much background about Britain appears in P. W. J. Bartrip and S. B. Burman, The Wounded Soldiers of Industry: Industrial Compensation Policy, 1833–1897 (Oxford: Clarendon Press, 1983), especially chap. 1, and Jamie L. Bronstein, Caught in the Machinery: Workplace Accidents and Injured Workers in Nineteenth-Century Britain (Stanford: Stanford University Press, 2008). Julian Go III, “Inventing Industrial Accidents and Their Insurance,” Social Science History, 20 (1996), 401–438. Some concern about the social problem of accidents had already moved governments to collect a few statistics outside of insurance activities. 3. Albeit not in less serious injuries: Henry J. Harris, “The Increase in Industrial Accidents,” American Statistical Society Publications, 13 (1912), 23–27. In the mid-1920s, the U.S. National Safety Council was still reporting steady increases in the rate of occupational accidents and also, ominously, in the severity of the accidents: Clyde J. Crobaugh and Amos E. Redding, Casualty Insurance (New York: Prentice-Hall, 1929), 543. 4. Harold G. Villard and P. Tecumseh Sherman, German Experience in Accident Prevention (New York: Workmen’s Compensation Publicity Bureau, 1914). 5. A typical discussion was Charles Corfield, “What Is an Accident? ” Bristol Medical-Chirurgical Journal, 31 (1913), 148–153, who in despair suggested it would be better to cover all injuries with medical insurance without worrying about defi nitions of accident. Hermann Loimer and Michael Guarnieri, “Accidents and Acts of God: A History of the Terms,” American Journal of Public Health, 86 (1996), 101–107. Karl Figlio, “What Is an Accident? ” in The Social History of Occupational Health, ed. Paul Weindling (London: Croom Helm, 1985), 180–205, gives a nuanced history of English law. 6. Anson Rabinbach, “Social Knowledge, Social Risk, and the Politics of Industrial Accidents in Germany and France,” in States, Social Knowledge, and the Origins of Modern Social Policies, ed. Dietrich Rueschemeyer and Theda Skocpol (Princeton: Princeton University Press, 1996), 48–89, uses German and French changes and arrangements as the model. The U.S. case is explored in part in John Fabian Witt, The Accidental Republic: Crippled Workingmen, Destitute Widows, and the Remaking of American Law (Cambridge MA: Harvard University Press, 2004), and Price V. Fishback and Shawn Everett Kantor, A Prelude to the Welfare State: The Origins of Workers’ Compensation (Chicago: University of Chicago Press, 2000), who, 3, remark explicitly that mechanization of the workplace undermined the traditional argument that each worker was individually responsible for his or her own safety. Villard and Sherman, German Experience in Accident Prevention. Bronstein, Caught in the Machinery, offers another narrative and analysis. 7. Franz Kafka for years held an appointment to write reports on occupational accident claims, reports now published in Franz Kafka, Amtliche Schriften, ed. Klaus Hermsdorf and Benno Wagner (Frankfurt am Main: S. Fischer, 2004), as was called to my attention by Lydia Marinelli.

Notes to Pages 14–17

8. See note 5. Barbara Young Welke, Recasting American Liberty: Gender, Race, Law, and the Railroad Revolution, 1865–1920 (New York: Cambridge University Press, 2001), shows how the extension of the idea of accident brought in gendered thinking as well as the idea of ownership. Mark Aldrich, Safety First: Technology, Labor, and Business in the Building of American Work Safety, 1870–1939 (Baltimore: Johns Hopkins University Press, 1997). See also chapter 4, below. 9. “A Big Step Toward Elimination of the Accident Hazard,” National Safety News, 29 November 1920, 8. Among Metropolitan Life industrial policyholders, in 1922, 46 percent of the fatal injuries were in public areas like roads and buildings, and most of those (89 percent) were automobile or pedestrian accidents; “Hazard Factors in the Accident Mortality Record,” Safety Engineering, July 1924, 29. 10. Stephen Davies, “ ‘Reckless Walking Must Be Discouraged,’ The Automobile Revolution and the Shaping of Modern Urban Canada to 1930,” Urban History Review, 18 (1989), 123–138; Peter C. Baldwin, Domesticating the Street: The Reform of Public Space in Hartford, 1850–1930 (Columbus: Ohio State University Press, 1999); Peter D. Norton, “Street Rivals: Jaywalking and the Invention of the Motor Age Street,” Technology & Culture, 48 (2007), 331–359. E. S. Clowes, “Street Accidents—New York City,” Publications of the American Statistical Association, 13 (1912), 449–456; the New York figures were for 1912. 11. Marx, “Fliegerverletzungen,” Berliner klinische Wochenschrift, 51 (1914), 53– 54. H. Graeme Anderson, “Aeroplane Accidents,” Journal of the Royal Naval Medical Service, 4 (1918), 51–68. 12. See, for example, a detectable change in the number and in the alarmed tone of articles in the National Safety News as early as 1920. Francis L. Grahlfs, “Read About Yourself, Mr. Motorist,” Safety Engineering, 50 (1925), 157–160. David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007), provides another account. Or see an insurance company experience, “Travelers’ Accident Insurance: Motor Vehicles Largest Cause of Accidental Deaths,” Safety Engineering, 59 (1930), 245. Despite the fact that the acute concern in the United States paralleled the huge number of motorcars there, the same concern also was found all over Europe. See Clay McShane, “The Origins and Globalization of Traffic Control Signals,” Journal of Urban History, 25 (1999), 379–404. 13. W. Moede, “Unfallverhütung auf psychotechnischer Grundlage,” Industrielle Psychotechnik, 3 (1926), 16–22. 14. For example, Gustavus Myers, “A Study of the Causes of Industrial Accidents,” American Statistical Society Publications, 14 (1914), 672–685. The later role of an epidemiological approach is taken up in chapter 9 below. 15. Henry J. Harris, “The Occupation Hazard of Locomotive Firemen,” American Statistical Association Publications, 14 (1914), 177–202. 16. For example, A. F. Stanley Kent, “The ‘Monday Effect’ in Industry,” Journal of Physiology, 50 (1916), lv–lvi; “Danger Hour,” Safety Engineering, 27 (1914), 457. 17. Safety Engineering, 26 (1913), 385. Later studies, particularly of fatigue, produced a somewhat different, but distinctive, profi le. 18. Untitled editorial, Safety Engineering, 28 (1914), 288. A parallel example of a supervisor fatality was in “Elevator, Boiler, and Flywheel Accidents,” Insurance Engineering, 15 (1908), 545. 19. See, for example, Edward D. Jones, Industrial Accidents and Their Prevention (Bulletin no. 47, Employment Management Series no. 7, Washington: Federal

241

242

Notes to Pages 17–21

Board for Vocational Education, 1919), based in part on material from the National Safety Council. A most striking instance of a psychological program, in this case dealing with airplane pilots who had accidents, was described by Otto Selz, “Über den Anteil der individuellen Eigenschaften der Flugzeugführer und Beobachter an Fliegerunfällen,” Zeitschrift für angewandte Psychologie, 15 (1919), 254–300. Pilots were tested for psychophysical characteristics, including distractability and nervous irritability, and despite the detailed statistics, including circumstantial factors such as side winds, the investigators did not see patterns of clumping accidents in individuals that others did in comparable compilations. I am indebted to Alexandre Métraux for this reference. 20. As will be noted below in the last chapters, “human factor” later came to take on a much more specific meaning. 21. This contrast, to which I shall refer again, is described particularly by Martin Lengwiler, “Between War Propaganda and Advertising: The Visual Style of Accident Prevention as a Precursor to Postwar Health Education in Switzerland,” in Medicine, the Market and the Mass Media: Producing Health in the Twentieth Century, ed. Virginia Berridge and Kelly Loughlin (London: Routledge, 2005), 57–65. 22. Aldrich, Safety First, 132, who emphasizes the shift of responsibility for safety to management. I shall return to national differences below. 23. H. B. Rockwell, “Why Accidents Happen,” Street Railway Journal, 26 (1905), 211–214. Figlio, “What Is an Accident? ” notes that the English legal history reflected some level of popular understanding of causal sequences that led to accidents. In one distant era, the law held the material object that inflicted damage to a person to be in some sense culpable. 24. “The Human Factor in Accident Occurrence,” Engineering and Industrial Management, 1 (1919), 261. 25. J. Andrew Mendelsohn, “Medicine and the Making of Bodily Inequality in Twentieth-Century Europe,” in Heredity and Infection: The History of Disease Transmission, ed. Jean-Paul Gaudillière and Ilana Löwy (London: Routledge, 2001), 21–79. See, for example, E. W. Scripture, “Reaction Time in Nervous and Mental Diseases,” Journal of Mental Science, 62 (1916), 698–719. 26. The Book of Accidents: Designed for Young Children (New Haven: S. Babcock, 1830). 27. Mary Martha Sherwood, Duty Is Safety; or, Troublesome Tom (Philadelphia: Geo. S. Appleton, 1847); the quotation is from 64. 28. It is only fair to note that risk taking could be construed favorably. In “Deaths of Aviators,” Boston Medical and Surgical Journal, 169 (1913), 365, an editorial writer noted that 342 aviators had died since 1908, but he [not likely she] went on to comment, “No doubt the spirit which prompts men to undertake a trial of the air differs widely in persons of varying temperament, but the fact of importance is that fl ight by human agency is possible, and therefore there will unquestionably always be found men of courage to undertake its further development.” 29. “Carelessness an Accident Factor,” Safety Engineering, 27 (1914), 456. “Hitting the Mind,” Safety Engineering, 28 (1914), 242–243. 30. “Carelessness of Users the Chief Cause of Electrical Fires,” Electrical World, 75 (1920), 1447. “Due to Heedlessness,” Safety Engineering, 28 (1914), 149. 31. R. C. Richards, “President Richards’ Address Before the Annual Meeting of

Notes to Pages 22–24

Members,” National Safety News, 4 October 1920, 8. Blanke, Hell on Wheels, takes up the “bad driver.” And see chapter 4, below. 32. Denunciations of alcoholic intoxication were commonplace in all Western discussions of accidents; see, for example, Revenstorf, “Ertrinkungsgefahr und Schwimmkunst,” Münchener medizinische Wochenschrift, 54 (1907), 2229–2232. 33. There are histories of other homely types of failures, but all of those that have come to my attention have moral implications that have attracted scholarly attention. One such basic, homely idea is laziness; see, for example, Georg Hirsch, Characterologische Studien über die Faulheit (Halle a. S.: Carl Marhold Verlagsbuchhandlung, 1931), and the currently flourishing literature on acedia, in the literature on the history of mental illnesses, such as the early article by Stanley W. Jackson, “Acedia the Sin and Its Relationship to Sorrow and Melancholia in Medieval Times,” Bulletin of the History of Medicine, 55 (1981), 172–185. 34. Clumsiness per se is not yet among subjects taken up by disability historians. See Catherine J. Kudlich, “Disability History: Why We Need Another ‘Other,’ ” American Historical Review, 108 (2003), 763–793. 35. Judith Mary Peters, “Developmental Coordination Disorder (DCD) as a Distinct Syndrome: A Conceptual and Empirical Investigation” (Ph.D. thesis, Institute of Education, University of London, 2006), chap. 1, especially 16–22. 36. Ibid., 22. 37. Ibid., chap. 1, provides a detailed examination of terminology. 38. Immanuel Kant, Anthropology from a Pragmatic Point of View, trans. Mary J. Gregor (The Hague: Martinus Nijhoff, 1974), 27: “One who accomplishes whatever he does with ease is skillful, and one whose activity betrays his trouble is awkward” [the German word is schwerfällig]. I am grateful to John Forrester for calling this to my attention. 39. Among other descriptions of psychologization (and this one unsympathetic) is Martin L. Gross, The Psychological Society: A Critical Analysis of Psychiatry, Psychotherapy, Psychoanalysis and the Psychological Revolution (New York: Random House, 1978); accident proneness is cited near the beginning of the book (15). 40. Dupré and Merklen, “L’Insuffisance Pyramidale physiologique de la première enfance et le syndrome de débilité motrice,” Revue neurologique, 1909, 1073–1074. 41. J.M. Peters, A.L. Barnett, and S.E. Henderson, “Clumsiness, Dyspraxia and Developmental Co-Ordination Disorder: How Do Health and Educational Professionals in the UK Defi ne the Terms? ” Child: Care, Health and Development, 27 (2001), 399–412. 42. C. M. Fisher, “A Lucunar Stroke: The Dysarthria-Clumsy Hand Syndrome,” Neurology, 17 (1967), 614–617. 43. See especially Peters, “Developmental Coordination Disorder,” 25–32. 44. The foregoing paragraphs have been informed particularly by information furnished by George Paulson and Judith Peters, to both of whom I am very deeply indebted. An example of the literature is Sasson S. Gubbay, The Clumsy Child: A Study of Developmental Apraxic and Agnosic Ataxia (London: W. B. Saunders Company, 1975). 45. See, for example, Daniel D. Arnheim and William A. Sinclair, The Clumsy Child: A Program of Motor Therapy (Saint Louis: C. V. Mosby, 1975); Helping Clumsy Chil-

243

244

Notes to Pages 24–28

dren, ed. Neil Gordon and Ian McKinlay (Edinburgh: Churchill-Livingstone, 1980). Peters, “Developmental Coordination Disorder,” offers additional details and insights, including, 24, the fact that recent authorities may use the term “clumsy” but never use the term “accident prone.” 46. See, for a further early example and the exact phrasing, Daniel D. Arnheim and Robert A. Pestolesi, Developing Motor Behavior in Children: A Balanced Approach to Elementary Physical Education (Saint Louis: C. V. Mosby, 1973), chap. 9. Peters, “Developmental Coordination Disorder,” contains a recent authoritative discussion. 47. These particular examples are from Theodore Toepel, “The Effect of School Life on the Physical Child,” Mind and Body, 20 (1913), 91, and George V. N. Dearborn, “Some Factors in the Development of Voluntary Movement in the Infant,” Mind and Body, 18 (1912), 374, reprinted from New England Medical Monthly. 48. Julia Thomas and Annie Thomas, Thomas Psycho-Physical Culture (New York: Edgar S. Werner, 1892), 23. 49. Luther Halsey Gulick, “Gymnastic Therapeutics,” Physical Training, 2 (1903), 177–178. William Burdick, “A Study of Functional Exercises in Some Nervous Diseases,” Mind and Body, 18 (1911), 61. Emil Rath, “The Theory, Terminology, and Technique of Gymnastic Dancing,” Mind and Body, 18 (1912), 443. 50. R. S. Woodworth, “The Accuracy of Voluntary Movement,” Psychological Monographs, 13 (1899). 51. See, for example, John J. B. Morgan, “The Overcoming of Distraction and Other Resistances,” Archives of Psychology, 35 (1916), no. 4; F. A. C. Perrin, “An Experimental Study of Motor Ability,” Journal of Experimental Psychology, 4 (1921), 24–56. Charles S. Myers, “Contributions to the Study of Shell Shock,” Lancet, 2 (9 September 1916), 466, attributed speech disorders to “disorders of the voluntary muscular system generally.” 52. Frederic Lyman Wells, “On the Psychomotor Mechanisms of Typewriting,” American Journal of Psychology, 27 (1916), 47–70. 53. See, for example, the material, for a long time heavily French, indexed in the Psychological Index. The subject was eventually one of those subject to regular reviews in the Psychological Bulletin, such as Edward K. Strong, Jr., “Fatigue, Work, and Inhibition,” Psychological Bulletin, 12 (1915), 416–419; idem., 13 (1916), 430–433. 54. Rabinbach, The Human Motor, describes these events in a rich cultural and intellectual context. The fundamental ideas were available in a popular as well as technical form on the western edge of the Atlantic as well, as in Horatio W. Dresser, Human Efficiency: A Psychological Study of Modern Problems (New York: G. P. Putnam’s Sons, 1912), especially 36–45. 55. Bernard J. Newman, “Shop Standards and Fatigue,” National Safety News, November 1920, 8. 56. Rabinbach, The Human Motor, especially 231–234. Josephine Goldmark, Fatigue and Efficiency: A Study in Industry (New York: Charities Publication Committee, 1912), translated and republished basic sources as well as analyzing the existing literature. See especially Emory S. Bogardus, “The Relation of Fatigue to Industrial Accidents,” American Journal of Sociology, 17 (1912), 206–222, 351–374, 512–539, who like others pinpointed speed and monotony as factors in muscular inaccuracy that led directly to accidents. 57. Lord Ashley, quoted in Goldmark, Fatigue and Efficiency, 202.

Notes to Pages 28–29

58. Emil Roth and Francesco S. Nitti, both quoted in Goldmark, Fatigue and Efficiency, 205 and 213. 59. Goldmark, Fatigue and Efficiency, 71; she treated accidents explicitly on 71– 89 and 192–220. Rabinbach, “Social Knowledge, Social Risk,” 76–77. 60. Clarence Stone Yoakum, “An Experimental Study of Fatigue,” Psychological Monographs, 11 (1909), no. 3. 61. John J. B. Morgan, “The Effect of Fatigue on Retention,” Journal of Experimental Psychology, 3 (1920), 319–333. 62. Rabinbach, The Human Motor, offers a summary history, and a history directly relevant to the German development of accident proneness is in Annette Mülberger Rogele, La aportación de Karl Marbe a la psicología: un enfoque crítico (tesi doctoral, Universitat Autónoma de Barcelona, Bellaterra, 1995), 299–303. H. Piéron, “L’Association Internationale de Psychotechnique entre les deux guerres mondiales,” Travail humain, 15 (1952), 122–132. José Germain, “Passé, present et avenir de la psychotechnique,” Travail humain, 17 (1954), 149–172, gives his impressions of previous and later development of the field. A scholarly overview of the origins is Siegfried Jaeger and Irmingard Staeuble, “Die Psychotechnik und ihre gesellschaftlichen Entwicklungsbedingungen,” Die Psychologie des 20. Jahrhunderts, 15 (1981), 53–95. Additional background is in Historical Perspectives in Industrial and Organizational Psychology, ed. Laura L. Koppes (Mahwah NJ: Lawrence Erlbaum Associates, 2007), especially Andrew J. Vinchur and Laura L. Koppes, “Early Contributors to the Science and Practice of Industrial Psychology,” 37–58. The classic background study of the United States is Loren Baritz, The Servants of Power: A History of the Use of Social Science in American Industry (Middletown: Wesleyan University Press, 1960). 63. See, for example, the pioneering work, Hugo Münsterberg, Psychology and Industrial Efficiency (Boston: Houghton Miffl in, 1913). In the larger, German version of his work, which was very influential in Europe, Hugo Münsterberg, Grundzüge der Psychotechnik, 2nd ed. (Leipzig: Johann Ambrosius Barth, 1920), he used the term “psychotechnology,” but it did not survive in the works of other authors writing in English. A contemporary British survey is Charles S. Myers, “Psychology and Industry,” British Journal of Psychology, 10 (1919), 177–182. A survey of programs in various countries was Paul Devinat, Scientific Management in Europe (Geneva: International Labour Office, 1927). Judith A. Merkle, Management and Ideology: The Legacy of the International Scientific Management Movement (Berkeley: University of California Press, 1980). Thomas J. Van De Water, “Psychology’s Entrepreneurs and the Marketing of Industrial Psychology,” Journal of Applied Psychology, 82 (1997), 486–499, emphasizes that in the United States, management viewpoints and psychologists’ viewpoints fed into each other. 64. Münsterberg, Psychology and Industrial Efficiency, 130, and Münsterberg, Grundzüge der Psychotechnik, especially 403–419. A full, rich history of his work in the context of psychology is Matthew Hale, Jr., Human Science and Social Order: Hugo Münsterberg and the Origins of Applied Psychology (Philadelphia: Temple University Press, 1980), especially 185–188, chap. 10. Münsterberg enters the story of accident proneness again in later chapters. 65. Franziska Baumgarten, “Wirtschaftspsychologie aus dem Auslande,” Zeitschrift für angewandte Psychologie, 19 (1921), 214–231.

245

246

Notes to Pages 30–32

66. E.g., Pearce Bailey, “Efficiency and Inefficiency—A Problem in Medicine,” Mental Hygiene, 1 (1917), 196–210. Sidney I. Schwab, “Neurasthenia Among Garment Workers,” American Labor Legislation Review, 1 (1911), 27–33. See below, chapter 7. 67. A particularly eloquent example was Alberto del Moral, “Higiene de los accidentes del trabajo: La selección,” Revista Médica de Sevilla, 31 (1912), 97–102, who was concerned that a legal, rather than medical, outlook overlooked the possibility of repeated accidents of the same kind. 68. Bailey, “Efficiency and Inefficiency,” 198. 69. Béla Sándor, “Die Reaktionzeiten des Menschen,” Industrielle Psychotechnik, 8 (1931), 233–248, for example, reviewed the history of reaction time and explicitly linked this inherited characteristic, part of the “psychophysical constitution,” to the new driver’s tests with which governmental authorities were attempting to remove dangerous drivers from the roads. The classic account of reaction time is Edwin G. Boring, A History of Experimental Psychology, 2nd ed. (New York: AppletonCentury-Crofts, 1950), chap. 8. A nuanced account with a different focus is Simon Schaffer, “Astronomers Mark Time: Discipline and the Personal Equation,” Science in Context, 2 (1988), 115–145, who notes that the term “personality” was often used to denote individual differences in reaction times. The international appeal of reaction time studies is emphasized in Jacqueline Carroy and Henning Schmidgen, “Reaktionsversuche in Leipzig, Paris und Würzburg: Die deutsch-französische Geschichte eines psychologischen Experiments, 1890–1910,” Medizinhistorisches Journal, 39 (2004), 27–55. Scripture, “Reaction Time in Nervous and Mental Diseases.” 70. Edwin G. Boring, Sensation and Perception in the History of Experimental Psychology (New York: Appleton-Century-Crofts, 1942), 182–190. 71. George Wilson, Researches on Colour-Blindness. With a Supplement on the Danger Attending the Present System of Railway and Marine Coloured Signals (Edinburgh: Sutherland & Knox, 1855), especially 125, 148. More than a century later, color blindness continued to be a problem, even among pedestrians: Alfred H. Lawton and Gordon J. Azar, “Some Observations of Behavior of Older Pedestrians,” Medical Times, 92 (1964), 69–74. 72. J. Ellis Jenkins, Color-Vision and Color-Blindness: A Practical Manual for Railroad Surgeons, 2nd ed. (Philadelphia: F. A. Davis Company, 1905), 4–6, 50–53. 73. David G. Winter and Nicole B. Barenbaum, “History of Modern Personality Theory and Research,” in Handbook of Personality Theory and Research (New York: Guilford Press, 1999), 3–27, suggest that in the twentieth century, personality, beyond traits, consisted of motives, cognitions, and context, with traits as the basic measurable element. James D. A. Parker, “In Search of the Person: The Historical Development of American Personality Psychology” (doctoral diss., York University, 1991), especially the summary on 140. 74. Myers, “Psychology and Industry,” 181. 75. Leonard W. Ferguson, “Employment Management (1): Origins and Development,” Heritage of Industrial Psychology, 13 (1964), 249–272. 76. Henri Ellenberger, Discovery of the Unconscious: The History and Evolution of Dynamic Psychiatry (New York: Basic Books, 1970), 495–496, 463n, reviews this evidence, citing specific examples. Europeans and Americans often knew the work of Hans Gross, Criminal Psychology: A Manual for Judges, Practitioners, and Students, trans.

Notes to Pages 32–36

Horace M. Kallen (Boston: Little, Brown, 1911), in which Gross discussed deceit and errors in testimony—sometimes in the light of bias and other psychological influences. 77. Sigmund Freud, The Psychopathology of Everyday Life (vol. 6, The Standard Edition of the Complete Psychological Works of Sigmund Freud), especially 162–163. 78. An English-language summary of the Freudian idea that accidents can be motivated was Samuel A. Tannenbaum, “Accidents and Symptomatic Actions,” Psyche and Eros, 3 (1922), 129–150. See below. 79. Major Greenwood and Hilda M. Woods, A Report on the Incidence of Industrial Accidents upon Individuals, with Special Reference to Multiple Accidents (Medical Research Council, Industrial Fatigue Research Board, Report no. 4, London: His Majesty’s Stationery Office, 1919), 9. 80. Chs. Widmer, “Über Unfalldisposition,” Die Therapie der Gegenwart, 60 (1919), 441–447. Widmer’s article is devoted mostly to speculating about the evolutionary failure involved in accident neurotics, and from a Lamarckian viewpoint, and I have not noticed that other writers cited his article. That his observations were published, however, suggests that the problem was available in the culture, and he himself suggested that his fi ndings might have important applications in factory safety. 81. Martin J. Flyzik, “Psychology of the Causes and Prevention of Accidents,” Safety Engineering, 39 (1920), 113; sentence construction sic. 82. William H. Tolman and Leonard B. Kendall, Safety: Methods for Preventing Occupational and Other Accidents and Diseases (New York: Harper & Brothers, 1913), 29. 83. C. L. Ferguson, “The Hospital’s Chronic Customers,” National Safety News, March 1933, 23. 84. Boyd Fisher, Mental Causes of Accidents (Boston: Houghton Miffl in Company, 1922), 5–6, 9, 10, 22, 284. There was no preparation for the terms and no explicit follow-up. Fisher wrote and consulted widely on management and had been involved in manager training in World War i and was later a major figure in setting up the U.S. Rural Electrification Administration; see, for example, Ronald R. Kline, Consumers in the Country: Technology and Social Change in Rural America (Baltimore: Johns Hopkins University Press, 2000), chaps. 5 and 6. Boyd’s book and his terms appear to be isolated, and yet the managing director of the U.S. National Safety Council wrote a foreword to the book. An instance of a psychologist, Max S. Henig, who cited Boyd’s book, is noted below in chapter 4. 85. It is possible that Farmer or someone else had seen Fisher’s book before 1925 and adopted the phrase from him. I have no evidence for such possibilities other than the fact that the Wellcome Library copy of the book is from the Myers Library collection, and Myers might or might not have made the book available to his former student, Farmer.

ch a p t er 2 1. Anne Cutler and David Fay, “Introduction,” in Rudolf Meringer and Carl Mayer, Versprechen und Verlesen: Eine psychologisch-linguistische Studie (Amsterdam: John Benjamins, 1978), xxvi. A. Thumb and K. Marbe, Experimentelle Untersuchungen über die psychologischen Grundlagen der sprachlichen Analogiebildung (Leipzig: Wilhelm Engelmann, 1901), especially chaps. 2 and 3.

247

248

Notes to Pages 36–39

2. Karl Marbe, “Über das Gedankenlesen und die Gleichförmigkeit des psychischen Geschehenes,” Zeitschrift für Psychologie, 56 (1910), 241–263. 3. Karl Marbe, in Murchison, A History of Psychology in Autobiography, III, 181–213. [Karl Marbe,] Selbstbiographie des Psychologen: Geheimrat Prof. Dr. Karl Marbe in Würzburg, ed. Emil Abderhalden (Halle [Saale]: Deutsche Akademie der Naturforscher Leopoldino, 1945), is a later account, emphasizing Marbe’s research. 4. Karl Marbe, Die Gleichförmigkeit in der Welt: Untersuchungen zur Philosophie und positive Wissenschaft (Munich: C. H. Beck’sche Verlagsbuchhandlung Oskar Beck, 1916), 384. The translation is mine. Marbe later, in his written expositions, did not connect the tendency to become ill or have bad luck with the inclination to have accidents. Roger Cooter has commented to me on the striking parallel of Marbe’s account to the account of how the founder of phrenology, Franz Josef Gall, as a boy noticed the correlation between character traits and conformations of the head in his schoolmates. For example, J. G. Spurzheim, Phrenology, In Connexion with the Study of Physiognomy, 2nd ed. (Boston: Marsh, Capen & Lyon, 1834), 13, writes of Gall: “From an early age he was given to observation, and was struck with the fact, that each of his brothers and sisters, companions in play, and schoolfellows, possessed some peculiarity of talent or disposition.” 5. The defi nitive work on Marbe is Annette Mülberger Rogele, La aportación de Karl Marbe a la psicología: un enfoque crítico (Tesi Doctoral, Bellaterra: Publicaciones de la Universitat Autónoma de Barcelona, 1995); see especially 305–369. I am extremely grateful to Annette Mülberger Rogele for making her work available to me and for advising me on Marbe’s work. On 310, Mülberger Rogele notes that Marbe had a lawyer friend who got him involved in the court cases. A particular account of Marbe’s philosophical psychology is in English in Martin Kusch, Psychological Knowledge: A Social History and Philosophy (London: Routledge, 1999). 6. Karl Marbe, “Psychologische Gutachten zum Prozess wegen des Müllheimer Eisenbahnunglücks,” Fortschritte der Psychologie und ihrer Anwendungen, 1 (1913), 339–374. Karl Marbe, “The Psychology of Accidents,” The Human Factor, 9 (1935), 102–103. 7. Mülberger Rogele, La aportación de Karl Marbe, especially 311–317. Marbe’s account is Karl Marbe, “Ueber Psychologie und Eisenbahnwesen,” Archiv für Eisenbahnwesen, 47 (1924), 729–744. W. Peters, “Karl Marbe: 1869–1953,” American Journal of Psychology, 66 (1953), 645–647, connects Marbe’s dissatisfaction with later developments in the Würzburg school of psychology to his turning his attention to applied psychology. 8. Marbe, Die Gleichförmigkeit in der Welt. Marbe’s statistics were in the stage of “collective regularities and frequencies” such as was standard in the pre–World War i world; Gerd Gigerenzer et al., The Empire of Chance: How Probability Changed Science and Everyday Life (Cambridge: Cambridge University Press, 1989), especially 62. Mülberger Rogele, La aportación, found that although Marbe used statistics to fi nd uniformities in natural phenomena, he did not contribute to the development of statistics, and, outside of a small group of mostly his students, his philosophy had only a limited impact on other scholars or scientists. 9. Marbe, Die Gleichförmigkeit in der Welt, especially 382–387. See J. B. Maller, “Studies in Character and Personality in German Psychological Literature,” Psychological Bulletin, 30 (1933), 209–232, and, for example, H. Gaudig, Die Idee der Persönlichkeit und ihre Bedeutung für die Pädagogik (Leipzig: Quelle & Meyer, 1923).

Notes to Pages 39–42

10. Relevant studies appeared in the organ of Marbe’s institute, Fortschritte der Psychologie und ihrer Anwendungen, published between 1912 and 1917; see, for example, Michael Bauch, “Psychologische Untersuchungen über Beobachtungsfehler,” 1 (1913), 169–226; Hermann Klugmann, “Über Fehler bei der Reproduktion von Zahlen,” 4 (1917), 327–367. 11. Jakob Stoll, “Zur Psychologie der Schreibfehler,” Fortschritte der Psychologie und ihrer Anwendungen 2 (1914), 1–133 12. See especially Karl Marbe, “Die Bedeutung der Psychologie für die übrigen Wissenschaften und die Praxis,” Fortschritte der Psychologie und ihrer Anwendungen, 1 (1912), 5–82, especially 18–26. Marbe also mentioned work on the fatigue factor in his discussion. 13. Marbe, Die Gleichförmigkeit in der Welt, 384–385. Marbe, “Ueber Psychologie und Eisenbahnwesen,” 741–742, where he incidentally also used the phrase “Unfällen neigend.” 14. K. Marbe, “Zur praktischen Psychologie der Unfälle und Betriebsunglücke,” Verhandlungen der phys.-med. Gesellschaft zu Würzburg, 1925, 172–175, especially 172– 173. Marbe there also used the term “disposition,” which is taken up in the next paragraph. 15. Karl Marbe, “Über Unfallversicherung und Psychotechnik,” Praktische Psychologie, 4 (1923), 257–264. The idea of disposition or predisposition was vague and troubled at the turn of the twentieth century, but very generally in use. Predisposition was a major legal issue in Germany and France, where insurance claims could depend upon whether in an accident the worker had been predisposed to hernia or lumbago or some other malfunction presumably brought about by an industrial accident. See Olivier Lenoir, Conferences sur les accidents du travail et les affections traumatiques (Paris: Vigot Frères, 1909). 16. Karl Marbe, Praktische Psychologie der Unfälle und Betriebsschäden (Munich: R. Oldenbourg, 1926), especially 14, 54. 17. Ibid., especially 15–25, 56–59, 78–83, 106–109. Marbe believed that some layers of personality traits might change, but other layers continued to be stable. 18. Ibid., unpaginated “Vorwort” and 107–110. Insurance executives, as will be noted below, did not respond to the idea of risk-laden individuals because actuaries were operating on the basis of insuring large groups, not individuals. 19. Karl Marbe, “Psychologie und Versicherungswesen,” Zeitschrift für die gesamte Versicherungs-Wissenschaft, 25 (1925), 337–346, for example, was, despite the title, largely a summary of his 1926 monograph to which he referred explicitly (and despite the official 1925 date of volume 25). I return to Marbe’s publicizing below. 20. See, for example, Otto Lipmann, Wirtschaftspsychologie und psychologische Berufsberatung (Leipzig: Johann Ambrosius Barth, 1921). 21. See, for example, the contents of Industrielle Psychotechnik or Walther Moede, “Ergebnisse der industriellen Psychotechnik,” Praktische Psychologie, 2 (1921), 289– 328. Hugh Quigley, “Industrial Psychology in Germany,” Psyche, 3 (1923), 229–234, especially 233. Quigley contrasted the discipline in the German system with the degrading automatism of American taylorization, which was becoming less relevant in industry as factories became automated. 22. See, for example, Heinrich Kuhn, Arbeitsleistung und Beleuchtung: Beschreibung und Ergebnisse psychotechnischer Wirtlichkeitsversuche (Deutsche Psychologie, VI, Halle: Carl Marhold, 1927).

249

250

Notes to Pages 42–46

23. [Dr. Glasel], “Selection Tests on the German Railways,” Journal of the National Institute of Industrial Psychology, 3 (1926), 201–204. W. Stern, “Über eine psychologische Eignungsprüfung für Strassenbahnfahrerinnen,” Zeitschrift für angewandte Psychologie, 13 (1918), 91–104, who tested for attention and reaction times. Other, earlier testing in other German mass transit venues is taken up below, chapter 4. 24. See, for example, in addition to material already cited, C. Heydt, “Eignungsprüfungen für den Rangierdienst,” Industrielle Psychotechnik, 1 (1924), 140–147; Osw. Heller, “Eignungsprüfung und Unfallvorbeugung in der Holzindustrie,” Industrielle Psychotechnik, 1 (1924), 99–118. 25. Marbe, “Über Unfallversicherung und Psychotechnik,” 259, was already in 1923 clear about his special approach: “Es gibt offenbar gewisse physiologische und psychologische Qualitäten, die ganz abgesehen von offensichtlichen körperlichen und seelischen Defekten, zu Unfällen disponieren. Wir dürfen daher von einem persönlichen Moment oder einem persönlichen Faktor reden, der beim Zustandekommen der Unfälle mitwirken kann.” 26. Marbe, “Psychologie und Versicherungswesen.” 27. Marbe, “Über Unfallversicherung und Psychotechnik,” 258. Baumann, review of Marbe, Praktische Psychologie der Unfälle, in Industrielle Psychotechnik, 4 (1927), 319. 28. Maria Schorn, “Unfallaffi nität und Psychotechnik,” Industrielle Psychotechnik, 1 (1924), 156–160. Marbe, as suggested above, had in 1926 mentioned investigating schoolchildren, making an empirical confi rmation of his own boyhood impression that some kids tended to have all the accidents; Marbe, Praktische Psychologie der Unfälle, 33–39. 29. E. Schmitt, “Unfallaffi nität und Psychotechnik im Eisenbahndienst,” Industrielle Psychotechnik, 3 (1926), 144–153, 363–366. Marbe’s 1924 presentation is noted further, below, in chapter 4. 30. Anton Grüb, “Zur Psychologie der Eisenbahnunglücke und Eisenbahnunfälle,” Archiv für die Gesamtepsychologie, 69 (1929), 207–282. 31. Fritz Giese, Methoden der Wirtschaftspsychologie (Berlin: Urban & Schwarzenberg, 1927), 564. 32. See, for example, Erich A. Klockenberg, “Beiträge zur Psychotechnik der Schreibmaschine und ihrer Bedienung,” Industrielle Psychotechnik, 1 (1924), 209–246. 33. David Meskill, “Characterological Psychology and the German Political Economy in the Weimar Period (1919–1933),” History of Psychology, 7 (2004), 3–19. The decline of psychotechnology is a special focus of Andreas Killen, “Weimar Psychotechnics Between Americanism and Fascism,” Osiris 22 (2007), 48–71. 34. For example, the article preceding Schorn’s: K. A. Tramm, “Die Verhütung der Unfälle durch Propaganda,” Industrielle Psychotechnik, 1 (1924), 148–156. Tramm is discussed below. 35. Examples include Kurt Seesemann, “Psychotechnical Studies on Industrial Safety Propaganda,” Industrial Safety Survey, 3 (1927), 63–67; Bertold Buxbaum, “Psychologische Betrachtungen zum Unfallschutz,” Maschinenbau, 8 (1929), 618–620. 36. Helene Bapp, Ueber Unfälle bei Kindern (Bonn: L. Neuendorff, 1927). 37. Kurt Schoch, Der Unfallbegriff in der schweizerischen privaten Einzel-UnfallVersicherung (Schleitheim: Buchdruckerei J. G. Stamm, 1930), especially 152, 173–

Notes to Pages 46–49

177, 183–184. Schoch drew especially on philosophical and legal literature. I allude again to the lack of effects of ideas of accident proneness in insurance organizations in chapters 4 and 9. 38. H. Piéron, “L’Association Internationale de Psychotechnique entre les deux guerres mondiales,” Travail humain, 15 (1952), 124–127. The association did not flourish in the 1930s. See, for example, A. Gemelli and M. Ponzo, “Les Facteurs psychophysiques qui prédisposent aux accidents de la rue et les perspectives d’organisation psychotechnique preventive,” Journal de psychologie normale et pathologique, 30 (1933), 781–782. 39. H. Hildebrandt, “Zur Psychologie der Unfallgefahrdten,” Psychotechnische Zeitschrift, 3 (1928), 1–8. 40. Ibid. Hildebrandt himself favored standard American-style propaganda and persuasion techniques to change worker behavior; in “Unfallpsychologie,” Wärme- und Kälte-Technik, 31 (Heft 5, 1929), 8–9, dealing with such techniques, he did note incidentally, 9, and with restraint, that further research was needed to test Marbe’s theory. H. Hildebrandt, “Unfallpsychologie,” in Der Mensch im Fabrikbetrieb; Beiträge zur Arbeitskunde, ed. Friedrich Ludwig (Berlin: J. Springer, 1930), 38–72, in his survey gave a very prominent place to personal accident affi nity (a term he preferred to Neigung, inclination, although he also used that term). He gave a full explanation of Marbe’s expositions, perhaps more sympathetic than in his previous publication, but he indicated clearly he wanted more testing and analysis not only of the Unfäller but of each accident itself and the predisposing circumstances of the accident, not just of the person. 41. N. Ach, “Psychologie und Technik bei Bekämpfung von Auto-Unfällen,” Industrielle Psychotechnik, 6 (1929), 92. 42. Schürmann, abstract of Karl Schwandtke, “Unfallverhütung,” Stahl und Eisen, in Zentralblatt für Gewerbehygiene, 7 (1930), 91. See, for example, A. Eiselsberg, “Alkohol und Unfall,” Wiener klinische Wochenschrift, 43 (1930), 6–7. 43. W. Moede, Lehrbuch der Psychotechnik (Berlin: Julius Springer, 1930), 443. 44. Walther Moede, “Unfäller und Nichtunfäller im Lichte der eignungstechnischen Untersuchung,” Industrielle Psychotechnik, 11 (1934), 1–10. 45. See, for example, E. E. Brakeman and C. S. Slocombe, “A Review of Recent Experimental Results Relevant to the Study of Individual Accident Susceptibility,” Psychological Bulletin, 26 (1929), 15–38, which contains no German references at all (!) and even omits work by the British pioneers in accident proneness. Morris S. Viteles of the University of Pennsylvania, “Psychology in Industry,” Psychological Bulletin, 27 (1930), 605–606, commented on what he considered this glaring omission. 46. H. M. Vernon et al., “Accidents in Industry and the Human Factor,” in Occupation and Health: Encyclopaedia of Hygiene, Pathology and Social Welfare (Geneva: International Labour Office, 1930), no. 109. The European approach to industrial psychology as such did not often include accidents. See, for example, Arthur W. Kornhauser, “Industrial Psychology in England, Germany, and the United States,” Personnel Journal, 8 (1930), 421–434, and Walter J. Heller, “Industrial Psychology and its Development in Switzerland,” ibid., 435–441. 47. Arnulf Rüssel, Arbeitspsychologie (Bern: Verlag Hans Huber, 1961), 238–259. 48. W. Hergt, “Über Unfallpsychologie,” Monatsschrift für Unfallheilkunde, 58

251

252

Notes to Pages 49–53

(suppl. 48) (1955), 66. There was even an East German who translated the term as “Unfall-Affi nität”: H. Redetzky, “Neue Gesichtspunkte bei der Betrachtung Ursachen der Betriebsunfälle,” Zentralblatt für Chirurgie, 43a (1952), 2094. Ragnar Berfenstam and Bo Vahlquist, “Die Bedeutung des Unglücksfalles als Todesursache bei Kindern,” Zeitschrift für Kinderheilkunde, 76 (1955), 499. Franziska BaumgartenTramer, “Typologie der Unfäller,” Zeitschrift für Präventivmedizin, 4 (1959), 228. Theodor Ehlers, “Über persönlichkeitsbedingte Unfallgefährdung,” Archiv für die gesamte Psychologie, 117 (1965), 252–279, for example resorted largely to a neutral and descriptive “individuellen Unfallhäufigkeiten.” [individual accident frequency]. 49. Review of Marbe, Praktische Psychologie der Unfälle und Betriebsschäden, in Journal of the National Institute of Industrial Psychology, 3 (1927), 278–279. 50. The copy of Marbe’s book now in the Wellcome Library belonged to Myers. Clearly he knew about the work. See next chapter. 51. E. Farmer, “The Study of Personal Differences in Accident Liability,” Journal of the National Institute of Industrial Psychology, 3 (1927), 436n. 52. Marbe, Praktische Psychologie der Unfälle, 65, had noted Vernon’s 1918 monograph citing fatigue, alcohol, haste, temperature, lighting, and the like as factors in accident rates, but he showed no awareness of the British work that identified workers who tended repeatedly to have accidents.

ch a p t er 3 1. General background is in Hugh Armstrong Clegg, A History of British Trade Unions Since 1889 (3 vols., Oxford: Clarendon Press, 1964–1985), II, chaps. 4 and 5. 2. [Illegible] to H. F. Heath, 26 October 1916, Industrial Fatigue Research Board Papers, FD 5/48, The National Archives, Kew. 3. Later the Industrial Health Research Board. See L. S. Hearnshaw, A Short History of British Psychology, 1840–1940 (New York: Barnes & Noble, 1964), 247–248; Major Greenwood and Hilda M. Woods, A Report on the Incidence of Industrial Accidents upon Individuals, with Special Reference to Multiple Accidents (Medical Research Council, Industrial Fatigue Research Board, Report no. 4, London: His Majesty’s Stationery Office, 1919); the quotation is from 2. Another account is R. S. F. Schilling, “Industrial Health Research: The Work of the Industrial Health Research Board, 1918–44,” British Journal of Industrial Medicine, 1 (1944), 145–152. The Board came to depend on private funding. The complex political and social context of these events is taken up in Deborah Thom, “ ‘A Revolution in the Workplace’ ? Women’s Work in Munitions Factories and Technological Change 1914–1918,” in Women Workers and Technological Change in Europe in the Nineteenth and Twentieth Centuries, ed. Gertjan de Groot and Marlou Schrover (London: Taylor & Francis, 1995), 97–118. 4. Schilling, “Industrial Health Research.” 5. H. M. Vernon, An Investigation of the Factors Concerned in the Causation of Industrial Accidents (Ministry of Munitions, Health of Munitions Workers Committee, Memorandum no. 21, London: HMSO, 1918), especially 4–7, 22–23. Vernon was greatly troubled by the inexactitude of accident statistics; whether or not a worker would report an injury, he found, depended very much on the individual worker. 6. Ibid., especially 45–46. Correspondence of the Board between 1918 and 1920 showed that Board members were concerned about Vernon’s going beyond the

Notes to Pages 53–55

evidence and adding his own interpretations. And Board members were aware that the focus might better be “output in relation to wages.” Papers of the Industrial Fatigue Research Board, FD1–122, The National Archives. George Newman to Walter Fletcher, 5 February 1918, referred disrespectfully to “Vernon’s screed on Accidents.” Ibid. Board members were also aware of a need to avoid appearing to be acting “in the interest of employers” as opposed to employees; L. S. Lloyd to Walter Fletcher, 5 September 1919. Ibid. 7. Greenwood’s given name was “Major”; it was not an indication of rank. Details of his life appear in “Major Greenwood, D.Sc, F. R. C. P., F. R. S.,” British Medical Journal, 2 (15 October 1949), 877–879, and Lancelot Hogben, “Major Greenwood, 1880–1949,” Obituary Notices of Fellows of the Royal Society, 7 (1950), 139–154. For his place in medical statistics, see J. Rosser Matthews, “Major Greenwood versus Almond Wright: Contrasting Visions of ‘Scientific’ Medicine in Edwardian Britain,” Bulletin of the History of Medicine, 69 (1995), 175–197, and Anne Hardy and M. Eileen Magnello, “Statistical Methods in Epidemiology: Karl Pearson, Ronald Ross, Major Greenwood, and Austin Bradford Hill, 1900–1945,” in A History of Epidemiologic Methods and Concepts, ed. Alfredo Marabia (Basel: Birkhäuser Verlag, 2004), 207–221. 8. Greenwood and Woods, A Report on the Incidence of Industrial Accidents upon Individuals, 2. 9. Ibid. The authors clearly were aware of the great dangers of accidents specifically in munitions works, as well as all of the factors that Vernon had described. Moreover, unlike many later investigators, they found that workers who took sick leave did not necessarily have matching accident records. 10. Ibid., especially 5, 7, 9, 11. 11. Major Greenwood and G. Udny Yule, “An Inquiry into the Nature of Frequency Distributions Representative of Multiple Happenings with Particular Reference to the Occurrence of Multiple Attacks of Disease or of Repeated Accidents,” Journal of the Royal Statistical Society, 83 (1920), 255–279. The original model was distribution of disease incidence in different houses. 12. Major Greenwood, quoted in Royal Statistical Society Journal, 90 (1927), 536– 537. E. Farmer, “The Study of Personal Differences in Accident Liability,” Journal of the National Institute of Industrial Psychology, 3 (1927), 433, wrote that Greenwood and Yule, “prompted mainly by the desire to advance statistical theory, . . . examined the nature of accident distribution” in the records of the Ministry of Munitions. Farmer (see below) later seemed to forget Yule. E. M. Newbold, A Contribution to the Study of the Human Factor in the Causation of Accidents (Medical Research Council, Industrial Fatigue Research Board, Report no. 34, London: His Majesty’s Stationery Office, 1926), 53, noted ambiguously that she was using “some additional figures relating to women munition workers obtained during the war by Miss C. Allen and myself,” but she did not directly claim that they noticed the distribution of accidents among workers at that time. 13. Major Greenwood, “Accident Proneness,” Biometrika, 37 (1950), 24. 14. See particularly ibid. and Eric Farmer, quoted in ibid., 544–545, who emphasized how closely the later group, notably Newbold, Farmer, and Chambers, worked together. Over the years, Yule tended to get dropped out of the narrative. See below.

253

254

Notes to Pages 56–58

15. Edgar L. Collis and Major Greenwood, The Health of the Industrial Worker (Philadelphia: P. Blakison’s Son, 1921), chap. 8, especially 195–201, 203. The book was also published in London. Because so much of the Greenwood and Woods paper was quoted in the text, I have inferred that the relevant passages were written by Greenwood. 16. As late as 1951, P. L. McKinlay, “Major Greenwood, 1880–1949,” Biometrika, 38 (1951), 2, in an obituary called this element in Greenwood’s work “among his most important contributions to medical statistics.” 17. Medical Research Council and Department of Scientific and Industrial Research, First Annual Report of the Industrial Fatigue Board ([London:] HMSO, 1920), 12. 18. I have not found any correspondence revealing Board members’ specific reactions to the publication of Greenwood and Woods or investigations of accident repeaters as I did remarks concerning Vernon’s work. 19. Medical Research Council, Industrial Fatigue Research Board Annual Report, 1922, 8. The hard times of 1920–1921 apparently slowed down at least this line of research. 20. Chambers moved to the University of Cambridge when Farmer did in 1935 and served as “assistant director of research in industrial psychology” until 1962. 21. Industrial Fatigue Research Board Annual Report, 1923, 9–10; 1924, 8–10. The phrase “a worker specially prone to accident,” using the word “prone,” occurred in the preface to Ethel E. Osborne et al., Two Contributions to the Study of Accident Causation (Medical Research Council, Industrial Research Board Report no. 19, London: His Majesty’s Stationery Office, 1922), iv, but there is no evidence that this use of the word was other than casual, and “accident proneness” did not appear until 1924 [1925], as recounted above. 22. Charles Samuel Myers, in A History of Psychology in Autobiography, ed. Carl Murchison (Worcester MA: Clark University Press, 1936), III, 222. Information from the Psychological Register and the University of Cambridge University Archives. The dearth of records concerning Farmer is remarkable. I have not been able to locate even an obituary, despite extensive efforts, and various archives and records of the university contain only odd bits of information. The annual reports of the Board and the papers of the Board in The National Archives in Kew furnish details of personnel appointments and projects, but they add little substantive to the published annual reports. I am deeply indebted to Professor John Mollon, who has collected personal material about Farmer that he was kind enough to share with me. Farmer apparently ended his days in a rural setting running a care institution for troubled adolescent children from wealthy families. A little additional personal information appears in “E. H.,” in The Times, 19 August 1976, 14h, published shortly after Farmer’s death. 23. E. Farmer, “The Method of Grouping by Differential Tests in Relation to Accident Proneness,” in Industrial Fatigue Research Board, Annual Report, 1924, 43–45. The annual reports were essentially public relations publications making a case for funding of the Board, and so the circulation of this paper was very limited. P. Froggatt, “Short-Term Absence from Industry. III. The Inference of ‘Proneness’ and a Search for Causes,” British Journal of Industrial Medicine, 27 (1970), 305n, notes the use of the term “prone” in earlier medical works, as in “prone to inflammation” (1804) or “prone to migraine or neuralgia” (1899).

Notes to Pages 58–60

24. Greenwood and Woods, A Report on the Incidence of Industrial Accidents; the quotation is from 11. 25. See, for example, Newbold, A Contribution to the Study of the Human Factor. A variation, “personal tendency,” in Ethel M. Newbold, “Practical Applications of the Statistics of Repeated Events, Particularly to Industrial Accidents,” Royal Statistical Society Journal, 90 (1927), 489, was not taken up by safety workers. H. M. Vernon, Accidents and Their Prevention (Cambridge: Cambridge University Press, 1936), 28, stated that “accident-prone” was an “expression fi rst used by Eric Farmer.” 26. Eric Farmer and E. G. Chambers, A Psychological Study of Individual Differences in Accident Rates (Medical Research Council, Industrial Fatigue Research Board Report no. 38, London: His Majesty’s Stationery Office, 1926), 3. 27. Medical Research Council and Department of Scientific and Industrial Research, First Annual Report of the Industrial Fatigue Research Board, 1920, 12. It seems clear that both Farmer and Marbe were speaking of an attribute of otherwise normal people, thus excluding medical conditions such as organic ataxias, epilepsies, etc. 28. Greenwood and Woods, A Report on the Incidence of Industrial Accidents, 4. 29. Newbold, A Contribution to the Study of the Human Factor. 30. H. D. Harrison, Industrial Psychology and the Production of Wealth (London: Methuen, 1924), for example, based his book largely on reports of the Fatigue Board and NIIP but, probably because of his conventional mindset, totally ignored the early mentions of individuals predisposed to accident. 31. See, for example, H. M. Vernon, “The Causation and Prevention of Industrial Accidents,” Lancet, 196 (5 April 1919), 549–550; H. M. Vernon, Industrial Fatigue and Efficiency (London: George Routledge & Sons, 1921), especially 203. 32. See, for example, “Factors Concerned in the Causation of Industrial Accidents,” Automotive Industries, 39 (1918), 916–918; “Industrial Accidents,” Engineering, 17 May 1918, 551–552; “The Human Factor in Accident Occurrence,” Engineering and Industrial Management, ns 1 (1919), 261–262; Frank Watts, “The Present Condition of Industrial Psychology,” Psyche, 2 (1921), 154–164; and especially the summary in the preface in Osborne et al., Two Contributions to the Study of Accident Causation, i–x; the usual Board summary of Greenwood’s and Woods’s work is in a paragraph in a context emphasizing personal factors, on vii. 33. Hearnshaw, A Short History of British Psychology, covers the work of the NIIP and its relationship to the Fatigue Board, especially 275–282. Myers left several accounts, including Charles S. Myers, Industrial Psychology in Great Britain (London: Jonathan Cape, 1926), chap. 1, including, 17: “The relations between the Industrial Fatigue Research Board and the National Institute of Industrial Psychology are intimate and harmonious. Each body is mutually (though not officially) represented on the other; investigators have been loaned or transferred from one body to the other according to need”; and they of course collaborated. 34. For example, Journal of the National Institute of Industrial Psychology, 1 (1923), 203, 347; 3 (1926), 176. 35. G. H. Miles and A. B. B. Eyre, “An Investigation into Breakage Problems,” Journal of the National Institute of Industrial Psychology, 1 (1922), 132–140, 2 (1924), 150–154. 36. See particularly S. V. Keeling, “Recent Tests for Competence in Tram Driving,” Journal of the National Institute of Industrial Psychology, 3 (1926), 86–93.

255

256

Notes to Pages 60–65

37. Leonard P. Lockhart, A Short Manual of Industrial Hygiene for Managers, Foremen, Forewomen and Industrial Supervisors Generally (London: John Murray, 1927), 67. The only literature cited was the 1919 report of Greenwood and Woods. 38. H. M. Vernon, “The Human Factor and Industrial Accidents,” International Labour Review, 13 (1926), 673–683. W. H. Cameron, “The Personal Factor in Accident Prevention,” Safety Engineering, 51 (1926), 38–40. 39. A. Stephenson, “Accidents in Industry,” Journal of the National Institute of Industrial Psychology, 3 (1926), 196–200. 40. Newbold, A Contribution to the Study of the Human Factor, especially 3, 5, 13. Farmer and Chambers, A Psychological Study of Individual Differences, iii n, note the number as thirteen. 41. Newbold, A Contribution, especially 60–61. 42. Farmer and Chambers, A Psychological Study of Individual Differences, especially 2. Newbold, “Practical Applications of the Statistics of Repeated Events,” 487–547, especially 517; in the article, she explained her statistical reasoning and summarized her previous fi ndings. 43. Farmer and Chambers, A Psychological Study of Individual Differences, especially 4 and 36. In 1925, E. Farmer and E. G. Chambers, “Concerning the Use of the Psycho-Galvanic Reflex in Psychological Experiments,” British Journal of Psychology, 15 (1925), 237–254, concluded that the tests did not distinguish between individuals, yet they could differentiate groups who were of “the same mental type.” 44. For example, although limited to “special susceptibility to accident,” the author who abstracted the Industrial Fatigue Research Board already had the idea: “England Makes Discoveries to Lessen Industrial Fatigue,” Industrial Psychology Monthly, 2 (1927), 205–206. The term “proneness to accident” did fi nally show up in the journal in “Recent Progress,” ibid., 487. 45. Morris S. Viteles, “Psychology in Industry,” Psychological Bulletin, 25 (1928), 326–327. Morris S. Viteles and Helen M. Gardner, “Women Taxicab Drivers: Sex Differences in Proneness to Motor Vehicle Accidents,” Personnel Journal, 7 (1929), 349–355. 46. C. S. Slocombe and W. V. Bingham, “Men Who Have Accidents: Individual Differences Among Motormen and Bus Operators,” Personnel Journal, 6 (1927), 251– 257; some account, noting the origin in management, is given in W. V. Bingham, “Individual Differences in Industrial Personnel. A Study of Accident-Prone Motormen,” Eugenical News, February 1930, 19–20. Clinton D. Smith, “Discovering Traits That Make Unsafe Trainmen,” National Safety News, July 1929, 31–33; this study was apparently inspired by one at the Milwaukee Railway and Light Company. J. S. Baker, “Do Traffic Accidents Happen By Chance? ” National Safety News, September 1929, 12–14. 47. Eric Farmer, “Psychological Study of Accident Proneness,” Personnel Journal, 9 (1930), 115–120, especially 116–117. 48. E. G. Chambers, “Personal Qualities in Accident Causation,” Journal of Industrial Hygiene, 12 (1930), 223–232. 49. G. H. Miles, “The Psychology of Accidents,” Journal of the National Institute of Industrial Psychology, 5 (1930), 183–192. 50. Ibid., 188. The British had no domestic examples of transport screening to report.

Notes to Pages 66–70

51. In 1929, Eric Farmer and E. G. Chambers, A Study of Personal Qualities in Accident Proneness and Proficiency (Medical Research Council, Industrial Health Research Board Report no. 55, London: His Majesty’s Stationery Office, 1929), reported the progress of their search for psychophysiological tests to detect accident proneness and confi rmed the complexity of the phenomenon; they insisted that it existed and correlated with industrial efficiency. 52. E. L. Collins, review of Farmer and Chambers, A Study of Personal Qualities, in Journal of Industrial Hygiene, 12 (1930), 84–85. A. Stephenson, “Industrial Accidents,” in Industrial Psychology, ed. Charles S. Myers (London: Thornton Butterworth, 1929), 122–140; the quotation is from 132. Another example is “Fitness for Individual Vocations,” Engineering, 129 (1930), 24.

ch a p t er 4 1. See, for example, Friedrich Dorsch, Geschichte und Probleme der angewandten Psychologie (Bern: Verlag Hans Huber, 1963), 141–153; and Hartmut Häcker and Wilfried Echterhoff, “Traffic Psychology,” in A Pictorial History of Psychology, ed. Wolfgang G. Bringmann (Chicago: Quintessence Publishing, 1997), 503–505, and Horst U. K. Gundlach, “The Mobile Psychologist: Psychology and the Railroads,” in ibid., 506–509. 2. See the rating schedule worked out by Sadie Myers Shellow and Walter J. McCarter, “Who Is a Good Motorman? ” Personnel Journal, 6 (1927), 338–343, making explicit the subjective and implicit standards of judgment. James R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society (Cambridge MA: Harvard University Press, 1986), 231, noted that in the late nineteenth century, the Erie Railroad already was rating train engineers by how efficiently they operated the locomotives. 3. See, for example, Peter van Drunen, Pieter J. van Strien, and Eric Haas, “Work and Organization,” in A Social History of Psychology, ed. Jeroen Jansz and Peter van Drunen (Malden MA: Blackwell, 2004), 142–151. Paul Devinat, Scientific Management in Europe (Geneva: International Labor Office, 1927), 118, 250–251, for example, placed the transport operator work of Berlin (which is discussed below) and other centers in a context of vocational aptitude testing. 4. J.-M. Lahy, La Selection psychophysiologique des travailleurs: Conducteurs de tramways et d’autobus (Paris: Dunod, 1927), 2. L. Bacquerisse, preface in ibid., v–xiii. Lahy’s career is described in S. Pacaud, “J.-M. Lahy (1872–1943),” Travail humain, 15 (1952), 338–343, and William H. Schneider, “The Scientific Study of Labor in Interwar France,” French Historical Studies, 17 (1991), 410–446. 5. J.-M. Lahy, “La Supériorité professionnelle chez les conducteurs de tramways dans ses rapports avec la consommation d’énergie électrique,” La Technique moderne, 7 (1913), 388–390. 6. Hugo Münsterberg, Psychology and Industrial Efficiency (Boston: Houghton Miffl in Company, 1913), 63–64. A German edition was also issued at this time. Matthew Hale, Jr., Human Science and Social Order: Hugo Münsterberg and the Origins of Applied Psychology (Philadelphia: Temple University Press, 1980), chap. 10. 7. Martin J. Insull, quoted in “Psychological Tests for Accident Prevention,” Electric Railway Journal, 39 (1912), 394–395. 8. Münsterberg, Psychology and Industrial Efficiency, 63–96.

257

258

Notes to Pages 70–74

9. Morris S. Viteles, “Research in Selection of Motormen. Part I. Survey of the Literature,” Journal of Personnel Research, 4 (1925), 107. 10. P. W. Gerhardt, “Scientific Selection of Employees,” Electric Railway Journal, 47 (1916), 943–945. 11. Lahy, La Selection psychophysiologique, especially 129–148,173–197. 12. One early example is Gustav Kafka, “Zwei neue Apparate zur Eignungsprüfung für Strassenbahner,” Beihefte zur Zeitschrift für angewandte Psychologie, 29 (1922), 95–101. The extant literature was reviewed in 1925 by Viteles, “Research in Selection of Motormen. Part I,” 100–115, and Albert P. Weiss and Alvhh R. Lauer, Psychological Principles in Automotive Driving (Columbus: Ohio State University, 1930), 11–16, give a short history of tests devised. 13. A. Schreiber, “Das Prüflaboratorium für Berufseignung bei den Königlich Sächsischen Staatseisenbahnen,” Zeitschrift des Vereines deutscher Ingenieure, 6 (1918), 446–451. A. Schreiber, “Mitteilungen aus dem Prüflaboratorium für Berufseignung bei den Sächsischen Staatseisenbahnen,” ibid., 63 (1919), 653–657. 14. F. Strauss, “The Selection and Training of Locomotive Staff as Factors in Accident Prevention,” Industrial Safety Survey, 13 (1937), 159. 15. See, for example, “Psychologische Veranstaltungen der Eisenbahnbehörde,” Zeitschrift für angewandte Psychologie, 18 (1921), 199–200; Riedel, “Bemerkungen zur Eignungsprüfung bei Fahrzeugführerberufen,” Zeitschrift für angewandte Psychologie, 19 (1921), 196–213; [A.] Schreiber, “Das Prüflaboratorium für Berufseignung bei der Eisenbahn-Generaldirektion Dresden,” Praktische Psychologie, 2 (1921), 232–239; Alfred Gradenwitz, “Psychological Tests for Motormen,” Electric Railway Journal, 59 (1922), 143–146. 16. K. A. Tramm, “Die Auswahl und Ausbildung des Fahrpersonals auf psychotechnischer Grundlage,” Verkehrstechnik, 15 September 1915, 25–28, and, for example, K. A. Tramm, “Über psychotechnische Bewegungsstudien an Strassenbahnnotbremsen,” Praktische Psychologie, 1 (1920), 252–256, and K. A. Tramm, “Die Bewährung des psychotechnischen Prüfverfahrens für Strassenbahnführer,” Industrielle Psychotechnik, 1 (1924), 36–42. See the notice in J. L. Prak, “How Germany and Holland Fit the Worker to the Job,” Industrial Psychology Monthly, 2 (1927), 422–423. 17. The tie to pilot testing was explicit in one streetcar driver testing group: Hildegard Sachs, “Studien zu Eignungsprüfung der Strassenbahnführer,” Zeitschrift für angewandte Psychologie, 17 (1920), 198–225 18. Strauss, “The Selection and Training,” 157. Gradenwitz, “Psychological Tests for Motormen”; the quotation is from 143. 19. G. H. Miles, “Economy and Safety in Transport,” Journal of the National Institute of Industrial Psychology, 2 (1925), 192–193. 20. W. Moede, “Unfallverhütung auf psychotechnischer Grundlage,” Industrielle Psychotechnik, 3 (1926), 16–22; the quotation is on 20. Lahy, La Selection psychophysiologique, 175. Lahy in his bibliography did not list Marbe or any of his students. 21. “Psychotechnischer Ausschuss der Deutschen Reichsbahn,” Industrielle Psychotechnik, 1 (1924), 87–88. The anonymous reporter noted that the subject was set for further discussion. 22. There was probably also some similar preparation among the general, ed-

Notes to Pages 74–76

ucated public; see, for example, “How Paris Picks Its Motormen,” Literary Digest, 30 May 1925, 22–23, in which the classification of candidates for the job of motorman was “good,” “bad,” and “very bad.” 23. Strauss, “The Selection and Training of Locomotive Staff,” made this point explicitly. Lahy, La Selection psychophysiologique, 227–231, lists in his bibliography reports in various languages from psychotechnologists all over the Europeanized world who were testing transport operators. Not all of the studies of transit workers were productive, even for occupational screening; see, for example, Ernst Brezina and Josef Wasti, “Anthropologische, konstitutions- und gewerbehygienische Untersuchungen an Wiener Strassenbahnbediensteten,” Mitteilungen der anthropologischen Gesellschaft in Wien, 59 (1929), 19–38. 24. W. V. Bingham, “Achievements of Industrial Psychology,” Mental Hygiene, 14 (1930), 375, lists American projects that “paralleled in this country elaborate developments in the techno-psychological laboratories of the Paris tramways and German railways,” suggesting that he believed that efforts in the USA were independent. 25. The management of that company was already developing a system of grading employees “on the basis of proficiency” into three classes, which would be the basis for wage increases. Clearly the grouping of employees was common industrial practice. Morris S. Viteles, Job Specifications and Diagnostic Tests of Job Competency Designed for the Auditing Division of a Street Railway Company (Philadelphia: University of Pennsylvania, 1922), 6. 26. Morris S. Viteles, “Research in the Selection of Motormen. Part II. Methods Devised for the Milwaukee Electric Railway and Light Company,” Journal of Personnel Research, 4 (1925), 173. 27. Ibid., 173–199. 28. During his stay abroad, Viteles became acquainted with the work of Tramm in Berlin in reducing accident records. See Morris S. Viteles, “Psychology in Business—in England, France and Germany,” Annals of the American Academy of Political and Social Science, 110 (1923), 207–220, especially 217–218. 29. Sadie Myers Shellow, “Research in Selection of Motormen in Milwaukee,” Journal of Personnel Research, 4 (1925), 223–237. Viteles, “Research in Selection of Motormen. Part I,” 104–105. M. McCants, “Tests Used in Selecting Employees,” Electric Railway Journal, 60 (1922), 710–715, showed the determination to fi nd screening tests even when many suggested tests and standards were not demonstrably useful. Perhaps the most important positive article was Gradenwitz, “Psychological Tests for Motormen,” which was heavily illustrated with the Berlin tests in action. 30. C. S. Slocombe and W. V. Bingham, “Men Who Have Accidents: Individual Differences Among Motormen and Bus Operators,” Personnel Journal, 6 (1927), 251– 257. E. E. Brakeman and C. S. Slocombe, “A Review of Recent Experimental Results Relevant to the Study of Individual Accident Susceptibility,” Psychological Bulletin, 26 (1929), 15–38; “Psychological Tests and Accident Proneness,” British Journal of Psychology, 21 (1930), 29–38. The authors did cite the work of Lahy on Paris. For example, “Tram Drivers,” abstract in Journal of the National Institute of Industrial Psychology, 4 (1929), 475. W. V. Bingham, “Individual Differences in Industrial Personnel: A Study of Accident-Prone Motormen,” Eugenical News, 15 (1930), 19–26; this

259

260

Notes to Pages 76–80

was, as the discussion showed, an odd place for this paper to appear. Walter V. Bingham, “Management’s Concern with Research in Industrial Psychology,” Harvard Business Review, 10 (1931), 45–48. 31. See previous note; the quotation is from Bingham, “Individual Differences,” 23. Bingham, “Achievements of Industrial Psychology.” 32. The Accident-Prone Employee ([New York: Metropolitan Life Insurance Company, 1930], 5–6. See below, chapter 7, on psychiatrists. 33. Ibid., especially 8, 10. 34. A significant article is K. A. Tramm, “Unser Verhalten bei Verkehrsgefahren,” Industrielle Psychotechnik, 8 (1931), 53–57. The chiefly American literature, although including major European writings, is described in C. H. Lawsche, Jr., Psychological Studies of Some Factors Related to Driving Speed on the Highway (Lafayette IN: Purdue University, 1940), 7–56. The U.S. history in general is covered incisively by Ann M. Dellinger, David A. Sleet, and Bruce H. Jones, “Drivers, Wheels, and Roads: Motor Vehicle Safety in the Twentieth Century,” in Silent Victories: The History and Practice of Public Health in Twentieth-Century America, ed. John W. Ward and Christian Warren (New York: Oxford University Press, 2007), 343–362, and Daniel M. Albert, “The Nut Behind the Wheel: Shifting Responsibilities for Traffic Safety Since 1895,” in ibid., 363–378. 35. N. Ach, “Psychologie und Technik bei Bekämpfung von Auto-Unfällen,” Industrielle Psychotechnik, 6 (1929), 88. Miles, “Economy and Safety in Transport,” 192. Miller McClintock, “Traffic Regulation,” Encyclopaedia of the Social Sciences, XV, 70–75 [1935], an incisive and authoritative summary. 36. David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007), provides a general history of events in the United States, including efforts to demonize the bad driver. 37. Harry R. DeSilva, “Age and Highway Accidents,” Scientific Monthly, 47 (1938), 536. Background information is available in such accounts as Uwe Fraunholz, Motorphobia: Anti-Automobiler Protest in Kaiserreich und Weimar Republik (Göttingen: Vandenhoeck & Ruprecht, 2002), 55–62, 70–74; Dietmar Fack, Automobil, Verkehr und Erziehung: Motorisierung und Sozialisation zwischen Beschleunigung und Anpassung 1885–1945 (Opladen: Leske & Budrich, 2000), with the example from 107; and James J. Flink, America Adopts the Automobile, 1895–1910 (Cambridge MA: MIT Press, 1970), especially chap. 6; the example is from 175. 38. Daniel M. Albert, “Psychotechnology and Insanity at the Wheel,” Journal of the History of the Behavioral Sciences, 35 (1999), 291–305; the quotation is from 294. Theophile Raphael et al., “One Hundred Traffic Offenders,” Mental Hygiene, 13 (1929), 809–824. 39. F. A. Moss and H. H. Allen, “The Personal Equation in Automobile Driving,” Journal of the Society of Automotive Engineers, 16 (1925), 415–420; discussion was reported in “Psychology of Automobile Driving,” ibid., 269–271. These pre-1926 experts wanted to make driver testing a requirement for a license but limited themselves to thinking in terms of classical personal reaction time—a constituent of most people’s ideas of accident proneness as it became defi ned by psychological tests. 40. See especially A. J. Snow, “Reduction of Automobile Accidents by Use of Psychological Tests,” Journal of the Society of Automotive Engineers, 17 (1925), 163–166;

Notes to Pages 80–82

A. J. Snow, “Tests for Transportation Pilots,” Journal of Applied Psychology, 10 (1926), 37–51; A. J. Snow, “Tests for Chauffeurs,” Industrial Psychology, 1 (1926), 30–45. 41. See previous note and A. J. Snow, Psychology in Business Relations (Chicago: A. W. Shaw Company, 1925), especially 524. Snow’s figures were utilized later by writers dealing explicitly with the idea of accident proneness. 42. A. J. Snow, “It Will Take More than Education to Stem Traffic Accidents,” Industrial Psychology Monthly, 2 (1927), 360. Blanke, Hell on Wheels, notes many similar formulations at that time. 43. Snow, “Tests for Chauffeurs,” 45n. 44. Morris S. Viteles, “Transportation Safety by Selection and Training,” Industrial Psychology Monthly, 2 (1927), 119–128. Morris S. Viteles and Helen M. Gardner, “Women Taxicab Drivers: Sex Differences in Proneness to Motor Vehicle Accidents,” Personnel Journal, 7 (1929), 349–355. Viteles’ work as an implicit publicist for the idea of accident proneness is taken up below. 45. Walter V. Bingham, “Psychology and Highway Safety,” Scientific Monthly, 31 (1930), 552–556, quotation from 553–554. Walter V. Bingham, “The Accident-Prone Driver,” Human Factor, 6 (1932), 158–169. 46. “Individuals Susceptible to Accidents,” Industrial Psychology Monthly, 2 (1927), 265. Examples of notes are untitled items in ibid., 1 (1926), 604; 2 (1927), 158, 483; “Making Highways Safe,” ibid., 1 (1926), 377–378; “Auto Drivers’ Eyes Need Attention,” ibid., 501–502. I assume that Laird wrote the editorials and notes. 47. Douglas Fryer, “Progress in Great Britain by the Industrial Fatigue Research Board,” Industrial Psychology Monthly, 3 (1928), 40–41. Theodore R. Johnson, “Mental Maturity Necessary for Safe Auto Operation? ” ibid., 77–80. “Tests to Detect the Accident Prone,” ibid., 380. 48. A. R. Lauer, “Travaux du Driving Research Laboratory,” Travail humain, 17 (1954), 41–52, described his career. In fact, the two-year grant under which he worked on the problem, from the National Research Council, began in 1927. 49. David Wechsler, “Tests for Taxicab Drivers,” Journal of Personnel Research, 5 (1926), 24–30. 50. Sadie M. Shellow, “The Accident Clinic: How It Functions and What It Accomplishes,” Personnel Journal, 9 (1930), 207–215. Earlier work was summarized in Sadie Myers Shellow, “Selection of Motormen: Further Data on Value of Tests in Milwaukee,” Journal of Personnel Research, 5 (1927), 183–188. 51. “Weeding Out Accident-Prone Improves Driving,” National Safety News, July 1931, 28. The story was anonymous, and since the journal was using the term, it could well have been introduced by the writer. Within a year, another writer, this time named, James Stannard Baker, “What Can We Do for High-Accident Drivers? ” ibid., June 1932, 19–20, 63–64, popularized the idea but used the term, “highaccident driver,” consistently. 52. Weiss and Lauer, Psychological Principles in Automotive Driving, especially 3–4 and chap. 6. The investigators found over a hundred “prone-to-accident drivers” and had comparable control groups. 53. See especially W. Moede et al., “Fahrerprüfungen,” Industrielle Psychotechnik, 3 (1926), 23–34, an international survey, and, for example, William Forster, “A Test for Drivers,” Personnel Journal, 7 (1928), 161–171, who was still thinking in terms of employees but included all drivers.

261

262

Notes to Pages 83–88

54. [Probably Charles S. Slocombe], “Job Analysis of Highway Safety,” Personnel Journal, 17 (1938), 30. 55. G. F. Michelbacher, “Casualty Insurance for Automobile Owners,” Proceedings of the Casualty Actuarial and Statistical Society of America, 5 (1919), 233 and 213– 242 in general. Peter C. Baldwin, Domesticating the Street: The Reform of Public Space in Hartford, 1850–1930 (Columbus: Ohio State University Press, 1999), describes the class and ethnic elements in struggles of pedestrians with growing automobile and truck traffic in the United States to control the streets. 56. Thomas N. Boate, “Auto Inspections Gain Ground,” Casualty and Surety Journal, September 1951, especially 53, 57–58. Glenn L. Wood et al., Personal Risk Management and Insurance, 4th ed. (2 vols., Malvern PA: American Institute for Property and Liability Underwriters, 1989), I, 78–81. 57. A general brief history of Anglo-American casualty and auto insurance is in C. A. Kulp, Casualty Insurance: Workmen’s Compensation, Public Liability, Automobile, Accident and Health, Burglary, Plate Glass, Power Plant, Forgery, Credit, and Other Lines (New York: The Ronald Press, 1928), 21–24. Another is H. W. Dingman, Insurability: Prognosis and Selection: Life—Health—Accident (Chicago: Spectator Company, 1927), especially chaps. 2 and 3. Harry Dingman, Risk Appraisal (Cincinnati: The National Underwriter Company, 1946), 50–58. Contemporary examples include Kulp, Casualty Insurance, 262–266; Winfield W. Greene, “Should the Compensation Premium Reflect the Experience of the Individual Risk? ” Proceedings of the Casualty Actuarial and Statistical Society of America, 2 (1916), 347–355. 58. W. M. Johnson, quoted in Proceedings of the Casualty Actuarial Society, 7 (1920), 354. The basis of cancellation of some insurance is spelled out in J. W. Laird, “NonCancellable Accident and Health Insurance Underwriting Problems,” Proceedings of the Casualty Actuarial Society, 7 (1920), 305–307 59. Clyde J. Crobaugh and Amos E. Redding, Casualty Insurance (New York: Prentice-Hall, 1929), 33–34. 60. G. F. Michelbacher, Casualty Insurance Principles (New York: McGraw-Hill Book Company, 1930), 352. 61. As will be noted below, statistical sophistication before the 1930s was surprisingly little developed even where one might expect it, and the actual statistics in casualty insurance outside of worker group insurance were not adequate for advanced treatment. See, for example, B. D. Flynn, “Review of the Actuarial and Statistical Work in the Various Branches of the Casualty Insurance Business,” Proceedings of the Casualty Actuarial Society, 7 (1920), 1–9. 62. See Blanke, Hell on Wheels, especially chap. 6.

ch a p t er 5 1. See, for example, a review in an American journal, “The Course of Accident Prevention Development,” Safety Engineering, 59 (1930), 293–294, 346; and, similarly, Arthur C. Carruthers, “Safeguarding First Gets Safety First,” Safety Engineering, 58 (1929), 289–290. A contemporary international account is H. S. Person, “Safety Movement,” Encyclopaedia of the Social Sciences XIII, 503–506 [1934]. 2. H. W. Heinrich, “The Origin of Accidents,” Travelers Standard, 16 (1928), 121–137. 3. Clarence Otis Sappington, “A Five Years’ Sickness and Accident Experience

Notes to Pages 88–90

in the Edison Electric Illuminating Company of Boston,” Journal of Industrial Hygiene, 6 (1924), 81–101, revealed that the company had started gathering statistics in 1913. To their surprise, the figures for the 1920s revealed that sickness was much more important economically than accidents. Nevertheless, accidents continued to be a major concern. 4. “Failures—Not Accidents,” Safety Engineering, 58 (1929), 178. 5. See, for example, “A Defi nite Plan of Accident Cause Analysis,” Industrial Safety Survey, 6 (1930), 103–104; J. E. Bathurst, “Getting at the Hidden Causes,” National Safety News, May 1932, 31; or “75 Separate Causes of Motor Accidents Listed,” Weekly Underwriter, 123 (1930), 820. 6. See, for example, R. L. Forney, “What We Do Not Know about Accidents,” American Journal of Public Health, 19 (1929), 46–50; W. Dean Keefer, “What’s the Use of Accident Statistics,” Chemical Markets, 29 (1931), 161–167; W. H. Cameron, “A Key to Accurate Fleet Accident Record Keeping,” Gas Age-Record, 68 (1931), 307–308; “Aetna Accident Claims Show Hazards of Sport,” Weekly Underwriter, 124 (1931), 872; E. C. Sleeth, “When the Accident Isn’t in the Records,” National Safety News, 22 (1930), 38, 114. 7. Johnson O’Connor, Born That Way (Baltimore: Williams & Wilkins Company, 1928), 35. 8. Abstract in Journal of the National Institute of Industrial Psychology, 3 (1927), 451. 9. Many examples have been given above. See, in addition, [H. Hildebrandt,] “Unfall-Psychologie,” Maschinenbau, 8 (1929), 265–266; Chas. I. Saks, “The Importance of Eyesight Surveys in Industry,” Safety Engineering, 58 (1929), 7–8; Kurt Seesemann, “Psychotechnical Aptitude Tests in the Service of Accident Prevention,” Industrial Safety Survey, 5 (1929), 157–161; Alvhh R. Lauer, “Disabilities that Cause Accidents, “ National Safety News, October 1932, 48–49, 86–87; Max S. Henig, “Intelligence and Shop Accidents,” Industrial Arts Magazine, 17 (1928), 265–266; and Max S. Henig, “Intelligence and Safety,” Journal of Educational Research, 16 (1927), 81–87, which was summarized in R. Duthil, “Accidents et Intelligence,” Psychologie et la vie, 2 (1928), 10. 10. See, for example, H. M. Vernon et al., “Accidents in Industry and the Human Factor,” in Occupation and Health: Encyclopaedia of Hygiene, Pathology and Social Welfare (2 vols., Geneva: International Labour Office, 1930), I, no. 109. 11. A. Gemelli and M. Ponzo, “Les Facteurs psychophysiques qui prédisposent aux accidents de la rue et les perspectives d’organisation psychotechnique preventive,” Journal de psychologie normale et pathologique, 30 (1933), 781–811. 12. Harold S. Hulbert, “Mental Mechanics in Accident Prevention,” National Safety News, November 1929, 12; another version appeared as Harold S. Hulbert, “Aids in Handling Oneself and Others, Especially in Relation to Accident Prevention and Safety,” Rock Products, 11 October 1930, 92. 13. G. P. Crowden, “Industrial Efficiency and Fatigue,” Proceedings of the Royal Society of Medicine, 23 (1930), 479–485. M. Greenwood, reported in ibid., 485, and Major Greenwood, Epidemics and Crowd Diseases: An Introduction to the Study of Epidemiology (New York: Arno Press, 1977, reprint of 1935), 130–134, spells out what he meant by psychological, including accident proneness, 132. 14. For example, Tapio Voionmaa, “Alcoholism and Industrial Accidents,” In-

263

264

Notes to Pages 91–94

ternational Labour Review, 11 (1925), 200–228; A. Eiselsberg, “Alkohol und Unfall,” Wiener klinische Wochenschrift, 43 (1930), 6–7. 15. See, for example, C. H. Murray, “Measuring Illness Experience by the Accident Rule,” National Safety News, July 1930, 21–22; Karl Mayer, “Die Beurteilung der Fahrdienstfähigkeit bei isolierten Pupillenstörungen (Spätneurosyphilis),” Deutsche medizinische Wochenschrift, 57 (1931), 547–549; Henry H. Kessler, “Is ‘Poor Health’ Blamed for Too Much? ” National Safety News, June 1933, 23–24,48. “The Course of Accident Prevention Development,” 346. 16. Arthur C. Carruthers, “Correct Determination of Accident Causes,” Safety Engineering, 60 (1930), 219–220. 17. As was done by Max S. Henig, “Mental States Other than Intelligence that Cause Work Accidents,” Industrial Arts and Vocational Education. 60 (1931), 7–9. Of course there was much concern about chronic mental illnesses that could involve dangerous behavior, as in “Mental Condition in Accident Causation,” Safety Engineering, 59 (1930), 230, 280. 18. Prof. Nippe, “Über die Beschäftigung von Epileptikern in Maschinenbetrieben,” Archiv für Psychiatrie und Nervenkrankheilen, 9 (1931), 234–236. 19. See, for example, Grace Overmyer, “Children Who Get Hurt,” Survey, 63 (1929), 337–338; Albert Frederick Stevens, Jr., “Accidents of Older Workers: Relation of Age to Extent of Disability,” Personnel Journal, 8 (1929), 138–145; “Many over 45 Victims of Speeding Motor Cars,” Weekly Underwriter, 124 (1931), 440; C. S. Slocombe, “The Dangerous Age in Industry,” National Safety News, July 1930, 68–69. Slocombe is especially interesting, as he was just moving into becoming a major user of the idea of accident proneness; see below. These groupings show up again in chapter 9 in a much stronger form as “risk groups.” 20. American examples include Marie Cornell, Industrial Injuries to Women in 1928 and 1929 Compared with Injuries to Men (United States Department of Labor, Women’s Bureau Bulletin no. 102, Washington: Government Printing Office, 1933); Mary Anderson, “Safety for Women Workers: A Challenge to Industry,” National Safety News, February 1933, 25–27. For general background, see Joel A. Tarr and Mark Tebeau, “Housewives as Home Safety Managers: The Changing Perception of the Home as a Place of Hazard and Risk,” in Accidents in History: Injuries, Fatalities and Social Relations, ed. Roger Cooter and Bill Luckin (Amsterdam: Rodopi, 1997), 201–207. See the further discussion of gendered thinking in chapter 9. 21. Robert E. Martin, “Are You Fit to Drive Your Car? ” Popular Science Monthly, April 1930, 55–56. The writer was informed by reports of work by psychologist Knight Dunlap, which appear not to have been published. 22. Heinrich, “The Origin of Accidents.” H. W. Heinrich, “A New Theory of Safety,” Safety Engineering, 57 (1929), 3–5. H. W. Heinrich, “You Will Probably Have 30 Injuries for Every 300 Accidents,” ibid., 56, 76–77. H. W. Heinrich, Industrial Accident Prevention (New York: McGraw-Hill Book Company, 1931), chap. 7, 315–326. 23. The Journal of Industrial Hygiene, for example, and Monatsschrift für Unfallheilkunde. 24. Morris S. Viteles, “Psychology in Industry,” Psychological Bulletin, 25 (1928), 326–327, 27 (1930), 604–606. 25. Adelbert Ford, A Scientific Approach to Labor Problems (New York: McGraw-Hill Book Company, 1931), 159, 161–162.

Notes to Pages 94–99

26. Harold Ernest Burtt, Psychology and Industrial Efficiency (New York: D. Appleton, 1931), especially chap. 8. Henry J. Welch and George H. Miles, Industrial Psychology in Practice (London: Sir Isaac Pitman & Sons, 1932), 139–145. 27. Charles S. Myers, Industrial Psychology in Great Britain (London: Jonathan Cape, 1926), 98. Myers did cite Greenwood’s and Woods’s paper. The book appeared in an American printing under a different title. Charles S. Myers, Industrial Psychology in Great Britain, 2nd ed. (London: Jonathan Cape, 1933), 94–95. A. Stephenson, “Industrial Accidents,” in Industrial Psychology, ed. Charles S. Myers (London: Thornton Butterworth, 1929), 122–140. C. S. Myers, “Some Recent Researches in Great Britain on the Psychology of Work,” Journal of the National Institute of Industrial Psychology, 5 (1931), 432–433, had almost the same wording as his earlier survey. Myers of course was probably also the author of anonymous abstracts of the accident proneness literature. 28. “Industrial Psychology,” Engineering, 129 (1930), 321. 29. “Fitness for Individual Vocations,” Engineering, 129 (1930), 24. 30. W. Dean Keefer, “Is Accident Proneness Curable? ” American Machinist, 74 (1931), 679–681. 31. See the general background in Loren Baritz, The Servants of Power: A History of the Use of Social Science in American Industry (Middletown CT: Wesleyan University Press, 1960), especially chap. 3. 32. Covert’s upward spiral model of popularizing technical ideas is described in John C. Burnham, How Superstition Won and Science Lost: Popularizing Science and Health in the United States (New Brunswick: Rutgers University Press, 1987), especially 39–40. 33. W. D. Ryan, “The Paramount Issue” (Information Circular 6428, United States Bureau of Mines, Washington: United States Bureau of Mines, 1931). 34. G. R. Petrie, “To Reduce Accidents—Be ‘Hard-Boiled,’ ” Factory and Industrial Management, 79 (1930), 326–328. And, for example, C. K. Thomas, “When Men Are ‘Careless,’ Look Higher Up,” National Safety News, December 1933, 27. 35. N. Ach, “Psychologie und Technik bei Bekämpfung von Auto-Unfällen,” Industrielle Psychotechnik, 6 (1929), 87–97, especially 92. 36. David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007), 124 and passim, emphasizes the domination of education along with engineering in the 1920s and 1930s safety efforts. 37. See, for example, C. E. Pettibone, “Mental Causes of Accidents,” Safety Engineering, 61 (1931), 9–11; Robert Clair, “Let Managerial Control Reduce Fleet Accidents,” National Safety News, December 1933, 20. Arthur S. McArthur, “Horseplay Prone,” National Safety News, July 1933, 23.

ch a p t er 6 1. For example, Henry H. Kessler, Accidental Injuries: The Medico-Legal Aspects of Workmen’s Compensation and Public Liability (Philadelphia: Lea & Febiger, 1931), contained no mention of accident proneness, but a second edition in 1941, 55, did in one place use the idea and the term. 2. See, for example, the Swiss publication, Progrès de la psychotechnique/Progress of Psychotechnics/Fortschritte der Psychotechnik, 1 (1939–1945).

265

266

Notes to Pages 99–102

3. See, for example, the summary in Arnold Wilson and Hermann Levy, Workmen’s Compensation (2 vols., Oxford: Oxford University Press, 1941), II, 30–33. 4. The curious history of physicians’ interest and lack of interest in accident proneness is taken up in the next chapter. 5. The foregoing generalizations are based upon a very extensive collection of material on the subject. Since the project was begun using British libraries and sources, and since rich runs of European journals exist in the United States, the sampling was probably not unduly skewed by the USA base of the author. Additional impressionistic evidence from the citations and bibliographies in European publications confi rms the probable validity of the sample. And see Arthur W. Kornhauser, “Industrial Psychology in England, Germany, and the United States,” Personnel Journal, 8 (1930), 421–434. 6. C. A. Oakley, “Accident Prevention in Industry,” Occupational Psychology, 16 (1942), 112. 7. A. E. Sinclair, “Our Accident Record—What and How,” National Safety Congress Transactions, 25 (1936), 289. 8. See, for example, B. H. Kirkbride, “Driving Errors,” Travelers Standard, 24 (1936), 21–27, especially 23–24. 9. H. M. Vernon, Accidents and Their Prevention (Cambridge: Cambridge University Press, 1936), especially chaps. 2 and 3. 10. And see, for example, Schürmann, abstract of Karl Schwandtke, “Unfallverhütung,” Stahl und Eisen, in Zentralblatt für Gewerbehygiene, 7 (1930), 91, who referred to Germany as “the motherland of technical accident prevention.” 11. The spread of the American model of educating managers and workers is particularly striking in the central international publication, Industrial Safety Survey. In Spain, the limitations of legal measures drove safety authorities to education and propaganda. Javier Silvestre, “Workplace Accidents and Early Safety Policies in Spain, 1900–1932,” Social History of Medicine, 21 (2008), 67–86. 12. H. W. Heinrich, “Twenty Years After,” National Safety News, August 1937, 14. 13. The Hawthorne experiments to try to improve factory worker productivity were carried out between 1924 and 1933, and they soon became familiar and symbolic to midcentury thinkers in many fields; see Richard Gillespie, Manufacturing Knowledge: A History of the Hawthorne Experiments (Cambridge: Cambridge University Press, 1991). Accidents were not a consideration in these studies; see L. J. Henderson and Elton Mayo, “The Effects of Social Environment,” Journal of Industrial Hygiene, 18 (1936), 401–416. 14. Charles S. Myers, “The Human Factor in Accidents,” Human Factor, 8 (1934), 266–279, for example, was unusual in grouping together traffic accidents, factory accidents, and breakages in restaurant work. 15. State of Ohio, Department of Highways, Summary of Traffic Accidents on State Highways Outside Municipalities (Columbus: Ohio Department of Highways, 1940), 25–27. “Accident-Prone Areas Studied on Ohio Highways,” Public Works, November 1940, 24. “Accident Prone Mileage: State Highways Outside Municipalities, 1947” (Columbus: Ohio Department of Highways, Division of Traffic and Safety, [1948]), carries a similar report with a similar map, for a later year. Erich Mittenecker, “Wie gross is die ‘Unfallneigung’ der Strasse? ” Psychologische Beiträge, 9 (1966), 288–293, apparently independently made a parallel application of Marbe’s original term to roadway areas in which high numbers of accidents occurred.

Notes to Pages 103–105

16. Examples include Robert E. Martin, “Are You Fit to Drive Your Car? ” Popular Science Monthly, April 1930, 55–56; Curtis Billings, “Science Measures the Driver’s Defects,” National Safety News, July 1934, 9–11, 48. 17. The investigators reporting in Benjamin F. Jones et al., Fatigue and Hours of Service of Interstate Truck Drivers (U.S. Public Health Service, Public Health Bulletin no. 265, Washington: United States Government Printing Office, 1941), for example, simply built accident proneness into their investigation of fatigue as individual variation. Not everyone had heard of or used the term, of course. I have not found any practical means of assembling quantitative data to confi rm my impressionistic report. See, for example, Du Pont safety official Harold L. Miner, “Attacking Personal Injury,” Safety Engineering, May 1939, 15–18, 26, who contended that management and engineering could handle all safety problems and that (16) “a few years ago we heard a great deal about the ‘accident prone’ employee. Personally I have always felt that it was rather begging the issue. A study of minor injuries made in an effort to determine trends, showed several apparent ‘repeaters’ but it was usually found that these repetitions of specific types of injuries to certain employees could be eliminated if conditions under which they worked were improved . . . Personally I do not know of a single employee who could rightfully be classified as ‘accident prone.’ ” I have not discovered the origin of the neutral, merely descriptive English term “repeater.” The fi rst journal article I found to use the term technically in a title was George E. Sanford, “Study the Accident Repeaters,” National Safety News, January 1931, 32. The distinction was still strong in Andrew Dakoski, “Putting Psychology to Work,” National Safety News, September 1955, 72–74. 18. See the previous chapter, where the Covert model is described. 19. “Purely Co-Accidental,” Time, 19 January 1942, 68. 20. Medical Research Council, Industrial Health Research Board, The Personal Factor in Accidents (London: His Majesty’s Stationery Office, 1942), especially 3, 16. 21. “The Personal Factor in Accidents,” Engineer, 154 (1942), 201–204. “Accident Proneness,” ibid., 212. May Smith, “A Study of Accidents,” Nature, 150 (1942), 396–397. 22. “Accident Proneness in Factories, Great Britain,” Monthly Labor Review, 55 (1942), 1199–1202. “Accident Proneness,” Industrial Medicine, 12 (1943), 109–111, where the writer maintained that tests for accident proneness could be applied to those in skilled occupations but that they were not applicable to unskilled workers. “The Personal Factor in Accidents,” Medical Journal of Australia, 9 January 1943, 33–34. 23. An obscure article, for example, on the subject of accident proneness, Flanders Dunbar, “Medical Aspects of Accidents and Mistakes in the Industrial Army and in the Armed Forces,” War Medicine, 4 (1943), 161–175, gained an extraordinary amount of attention in other publications, including popular publications. See next chapter. 24. Eric Farmer and E. G. Chambers, A Study of Personal Qualities in Accident Proneness and Proficiency (Industrial Health Research Board, Report No. 55, London: His Majesty’s Stationery Office, 1929), especially vi–vii, 41. Other factors in accident incidence, such as temperature and lighting, Farmer and Chambers left to their colleague, Vernon, whose work is noted above. The division of labor was made explicit in annual reports of the Board and elsewhere.

267

268

Notes to Pages 105–107

25. See, for example, Gustav Ichheiser, “Subjective Accident Causes and Psychological Methods of Accident Prevention,” Industrial Safety Survey, 15 (1939), 125–130; G. Herdan, “The Logical and Analytical Relationship between the Theory of Accidents and Factor Analysis,” Journal of the Royal Statistical Society, 106 (1943), 125–142; and, from a later period, Gerald F. King and James A. Clark, “Perceptual-Motor Speed Discrepancy and Deviant Driving,” Journal of Applied Psychology, 46 (1962), 119. The issue comes up again, here and there, below. 26. Eric Farmer, “A Note on the Relation of Certain Aspects of Character to Industrial Proficiency,” British Journal of Psychology, 21 (1930), 46–49. 27. Eric Farmer, The Causes of Accidents (London: Sir Isaac Pitman & Sons, 1932), especially 60–61; he also published a summary in an American journal, Eric Farmer, “Recent Research into the Causes of Industrial Accidents,” Journal of Industrial Hygiene, 14 (1932), 84–86. His much-noted distinction between proneness and liability he elaborated at the beginning of World War ii, “Accident Proneness and Accident Liability,” Occupational Psychology, 14 (1940), 121–131, incidentally summarizing his other contentions about accident proneness. 28. E. Farmer, E. G. Chambers, and F. J. Kirk, Tests for Accident Proneness (Industrial Health Research Board, Report no. 68, London: His Majesty’s Stationery Office, 1933). 29. Eric Farmer, “Accident Proneness,” Safety Engineering, June 1934, 247–248; July 1934, 23. 30. E. Farmer and E. G. Chambers, A Study of Accident Proneness Among Motor Drivers (Industrial Health Research Board, Report no. 84, London: His Majesty’s Stationery Office, 1939), especially iii. Eric Farmer, “Accident-Proneness on the Road,” Practitioner, 154 (1945), 221–226, quotation from 224. 31. The “trait” problem I discuss below. The tendency to have accidents as a part of one’s personality was another matter; a Berlin graphologist, L. KroeberKeneth, “Unfallneigung und Handschrift,” Industrielle Psychotechnik, 12 (1935), 321–334, for example, used Marbe’s classifications and idea of accident proneness as part of personality to try to set up ways to identify the accident prone by their handwriting. 32. Walther Moede, “Unfäller und Nichtunfäller im Licht der eignungstechnischen Untersuchung,” Industrielle Psychotechnik, 11 (1934), 1–10, for example, still attempted to analyze Marbe’s categories into nonexistence by translating them into conventional vocational testing or training. Rich background is found in David Meskill, “Characterological Psychology and the German Political Economy in the Weimar Period,” History of Psychology, 7 (2004), 3–19. 33. Peter Glück, “Psychologische Analyse und Prüfung der Unfallaffinität von sieben- bis neunjährigen Kindern,” Archiv für die Gesamte Psychologie, 93 (1935), 1–41. Wilhelm Grundler, “Untersuchungen zur Typologie des Unfällers in der Eisenund Stahlindustrie,” Arbeitsphysiologie, 8 (1935), 97–133. Hans Dombrowsky, Warum Unfälle? (Braunschweig: Friedr. Vieweg & Sohn, 1935), especially 7. 34. Karl Marbe, “The Psychology of Accidents,” Human Factor, 9 (1935), 100–104. 35. Annette Mülberger Rogele, “La aportación de Karl Marbe a la psicología: un enfoque crítico (tesi doctoral, Universitat Autónoma de Barcelona, 1995), especially 417–420. Anson Rabinbach, The Human Motor: Energy, Fatigue, and the Origins of Modernity (New York: Basic Books, 1990), 278–282, and William H. Schneider, “The

Notes to Pages 107–109

Scientific Study of Labor in Interwar France,” French Historical Studies, 17 (1991), 410– 446. The Zentralblatt für Gewerbehygiene, for example, by 1936 was, except for a couple of book notices, not carrying any material relevant to accident proneness but only material about engineering and education to prevent accidents. One might add that since so much of the accident publication shifted to the United States, the fact that Marbe’s German-language work was not accessible to many Anglophone experts also inhibited his further influence; K. C. Garrison, “The Human Element in Accident Prevention,” Safety Engineering, 71 (1936), 258–260, for example, cited the recent paper that Marbe published in English, not any of his German works. 36. Herbert Heymann, “Die Bedeutung der Unfälle und ihrer Bekämpfung,” Wiener medizinische Wochenschrift, 93 (1943), 453–459. Walther Moede, Eignungsprüfung und Arbeitseinsatz (Stuttgart: Ferdinand Enke, 1943). And see, for example, Ichheiser, “Subjective Accident Causes and Psychological Methods.” 37. K. Marbe, “Ueber Unfäller, Schadenstifter und den Wiederholungssatz,” Zentralblatt für Arbeitswissenschaft, 2 (1948), 76–77. 38. See, for example, Karl Marbe, “Theorie der motorischen Einstellung und Persönlichkeit,” Zeitschrift für Psychologie, 129 (1933), 305–322. 39. J.-M. Lahy and S. Korngold, “Recherches expérimentales sur les causes psychologiques des accidents du travail,” Travail humain, 4 (1936), 1–64, especially 7– 15. J.-M. Lahy and S. Korngold, Recherches expérimentales sur les causes psychologiques des accidents du travail (Paris: Publications du Travail Humain, 1936), was substantially a reprint of the article but included more data. 40. J.-M. Lahy, La Selection psychophysiologique des travailleurs: Conducteurs de tramways et d’autobus (Paris: Dunod, 1927), did not reference Marbe, Greenwood, or Farmer. By 1936, Lahy was well aware of German and British publications. 41. Lahy and Korngold, “Recherches expérimentales.” J.-M. Lahy and S. Korngold, “La Fatigabilité est-elle une cause des accidents? ” Travail humain, 4 (1936), 153–162. 42. Lahy and Korngold, “Recherches expérimentales,” especially 58–63. 43. Ibid., especially 11–13. J. Girault, “Lahy, Jean-Maurice,” Dictionnaire biographique de movement ouvrier français, 33 (1988), 140–141. Schneider, “The Scientific Study of Labor in Interwar France,” especially 426–428. Lahy had earlier written opposing American taylorization of labor on humanitarian grounds. 44. See, for example, Guy Palmade, La Psychotechnique (Paris: Presses Universitaires de France, 1948), 15–19. 45. Eric Farmer, “Psychological Causes of Accidents: A Critical Notice,” Human Factor, 11 (1937), 415–416. 46. [Probably Charles S. Slocombe], “Job Analysis of Highway Safety,” Personnel Journal, 17 (1938), 25. 47. “J.K.,” abstract of A. Kolodnaia, “Analyse psychologique des accidents de chemin de fer,” Travail humain, 2 (1934), 135–136. Lahy in his work cited similar material from Soviet writers. 48. Volker Roelcke, “Programm und Praxis der psychiatrischen Genetik an der Deutschen Forschunganstalt für Psychiatrie unter Ernst Rüdin: Zum Verhältnis von Wissenschaft, Politik und Rasse-Begriff vor und nach 1933,” Medizinhistorisches Journal, 37 (2002), 21–55, reminds us that the idea of inherited psychological traits was, even at that time, not simple.

269

270

Notes to Pages 109–112

49. H. Laugier, J. Monnin, and D. Weinberg, “Contribution a l’étude du facteur individuel dans les accidents du travail,” Travail humain, 5 (1937), 392–407, especially 407. 50. Alvhh R. Lauer, “A Plea for the Negative,” National Safety News, April 1936, 12. 51. Ichheiser, “Subjective Accident Causes and Psychological Methods,” 127. Charles A. Drake, “Detecting the Accident-Prone Worker,” Personnel, 18 (1942), 281. 52. A. Wyn Williams, “The Accident-Prone Employee,” Textile World, May 1943, 68. 53. Garrison, “The Human Element in Accident Prevention,” 260. 54. Fritz Giese, Methoden der Wirtschaftspsychologie (Berlin: Urban & Schwarzenberg, 1927), 564, for example, had immediately after the initial publications interpreted Marbe’s groupings as inherited personality traits. Carsten Timmerman, “Constitutional Medicine, Neoromanticism, and the Politics of Antimechanism in Interwar Germany,” Bulletin of the History of Medicine, 75 (2001), 717–739, notes the distinction at that time between inherited traits and holistic inheritance. 55. Eug. Schreider, Facteurs physiologiques et psychologiques de la predisposition aux accidents (Paris: Conservatoire National des Arts et Métiers, [1937]). 56. Grundler, “Untersuchungen zur Typologie des Unfällers in der Eisen- und Stahlindustrie.” The search for such a g was explicit in Herdan, “The Logical and Analytical Relationship between the Theory of Accidents and Factor Analysis,” especially 136. 57. J. J. Logie, “Outside Troubles Make Unsafe Workers,” National Safety News, February 1934, 7, 32. Rex B. Hersey, “Emotions and Accidents,” National Safety News, August 1934, 13–15, 62. N. E. Eckelberry, “Unsuspected Physical Defects,” National Safety News, May 1940, 36. Edward R. Graniss, “Accident Investigation and Cause Finding,” Mechanical Engineering, 63 (1941), 209. 58. “Accidents: The Human Factor Is Discussed,” Mine and Quarry Engineering, 5 (1940), 261. Other examples include C. A. Oakley, “Psychological Factors in Accident Causation,” Occupational Psychology, 14 (1940), 181–183; A. W. Meinke, “Small Fleet Safety,” Safety Engineering, June 1940, 22–25; Verne K. Harvey and E. Parker Luongo, “Industrial Medicine and Accident Prevention,” Industrial Medicine, 14 (1945), 377–381. 59. Lauer, “A Plea for the Negative,” 13. 60. There were of course enthusiasts for one type of testing or another. An upper New York eye expert, for example, found that thorough vision testing could identify accident prone employees; N. Frank Stump, “Spotting Accident-Prone Workers by Vision Tests,” Factory Management and Maintenance, 103 (1945), 109–112. 61. Dombrowsky, Warum Unfälle?, especially 90. Hans Dombrowsky, “Eignungsprüfungen und Unfallhäufigkeit,” Industrielle Psychotechnik, 13 (1936), 143–148. 62. R. Bonnardel, “Recherches expérimentales sur la prevention des accidents au moyen des methods psychotechniques,” L’Année psychologique, 1942, 84–93. 63. See, for example, Lowell S. Selling, “Psychiatry in Industrial Accidents,” Advance Management, 10 (1945), 70–75. 64. For example, Mary May Wyman, “Accident Repeaters,” Safety Education, October 1945, 6–9; Harry R. DeSilva, “Normal Versus Accident Drivers,” Journal of Psychology, 7 (1939), 337–342.

Notes to Pages 113–118

65. Charles A. Drake, “Accident-Proneness: A Hypothesis,” Character and Personality, 8 (1940), 335–341. 66. Charles S. Slocombe, “The Psychology of Safety,” Personnel Journal, 20 (1941), 42–50. 67. E. G. Martin, “Shotgun or Rifle? ” National Safety News, June 1939, 38. 68. “Accident Proneness Shows, Even in Cinders in Eyes,” National Safety News, December 1935, 71. Homer L. Humke, “First Month Found Most Dangerous,” Personnel Journal, 14 (1936), 337. Examples include W. V. Bingham, “Those Accident Addicts,” National Safety News, October 1934, 47, 83; Charles S. Slocombe, “How to Cut Accident Costs,” Personnel Journal, 16 (1937), 134–141; Heinrich, “Twenty Years After,” 15. The concern about minor incidents appeared as well, for example, in the wartime British report, The Personal Factor in Accidents. Minor accidents could of course also indicate dangerous conditions as well as accident prone people; Hilton E. Wright, “The ‘Acorn’ of a Major Accident,” National Safety News, January 1934, 25–26. See below on consequences to employees. 69. See, for example, Charles S. Myers, “The Psychological Approach to the Problem of Road Accidents,” Nature, 136 (1935), 740–742. 70. “Traffic’s Worst Offenders,” Safety Engineering, 68 (1934), 229–232. 71. Motor-Vehicle Traffic Conditions in the United States, Part 6, The Accident-Prone Driver (Washington: United States Government Printing Office, 1938). Harry R. DeSilva, Why We Have Automobile Accidents (New York: John Wiley & Sons, 1942). 72. DeSilva, Why We Have Automobile Accidents, especially chap. 9; the quotation is from 198. 73. Motor-Vehicle Traffic Conditions in the United States, Part 6, The Accident-Prone Driver, 8. 74. “Driving Tests,” The Times, 18 May 1934. “Memorandum on Accident Proneness Prepared under the Direction of the Medical Research Council,” April 1934, Public Record Office FD1/4017, 11798, The National Archives, Kew. 75. David Munro to Ernest Bevin, 27 April 1934, Public Record Office FD1/4017. Munro was former director of medical services of the British air forces. 76. Ernest Bevin to David Munro, 2 May 1934, Public Record Office FD1/4017. This was the Bevin who later held important government appointments, including that of foreign minister. 77. E. M. Tylor [Excess Insurance Company Limited] to Minister of Transport, 1 March 1935, Public Record Office. [Unreadable] to David Munro, 21 March 1935, Public Record Office, FD 4017 111798. “E.M.” to Secretary, Ministry of Transport, 28 March 1935, Public Record Office FD1/4017 111798. See also chapter 4, above. 78. “Accident Proneness, Note of a Meeting Held at the Ministry of Transport on Friday, 5th April, 1935, Public Record Office FD1/4018 111798. Major Greenwood to David Munro, 6 April 1935, Public Record Office FD1/4018 111798. 79. “Accident-Prone Test—A New Third Degree,” Daily Herald, 29 March 1935. 80. “Road Accident Proneness,” extract from The Morning Post, 21 March 1935, and other clippings in Public Record Office FD1/4017 111798 and 4018 111798. 81. Alexander Chleusebairgue, “The Selection of Drivers in Barcelona,” Occupational Psychology, 14 (1940), 146–161. G. H. Miles and D. F. Vincent, “The Institute’s Tests for Motor Drivers,” Human Factor, 8 (1934), 245–257. Report by the Select Committee of the House of Lords on the Prevention of Road Accidents (London: His Majesty’s

271

272

Notes to Pages 118–122

Stationery Office, 1939), 28. The continuing British struggle with “fitness” was described, for example, in D. W. Elliott and Harry Street, Road Accidents (Baltimore: Penguin Books, 1968), 52–59. 82. There were already two brief historical accounts of the U.S. driver’s testing efforts, with different conclusions: A. R. Lauer, “Fact and Fancy Regarding Driver Testing Procedures,” Journal of Applied Psychology, 21 (1937), 173–184, and Lowell S. Selling, “The Psychological Approach to the Traffic Problem,” Scientific Monthly, 44 (1937), 547–554. H. M. Johnson, “Born to Crash,” Colliers, 25 July 1936, 28, 58, 60. Hugh A. Dunlop, “Medical Standards of Fitness for Driving,” Practitioner, 154 (1945), 201–204. For example, Marion M. Kalez and Rieber C. Hovde, “Pilots with Repeated ‘Pilot-Inaptitude’ Accidents,” Aviation Medicine, 16 (1945), 370–375. 83. See, for example, Dombrowsky, “Eignungsprüfungen und Unfallhäufigkeit.” 84. Adolf Maria Manoil, “Vocational Guidance in Prevention of Occupational Accidents,” Occupations, 21 (1943), 382. The use of such terms as “industrially fit” suggests the common mindset of social Darwinism common in some social arenas at that time. 85. Chleusebairgue, “The Selection of Drivers in Barcelona,” 147. David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007), provides an American context for the problem of disciplining bad drivers. 86. Gollasch, “Aptitude Tests,” 153; the wording of course might be that of the translator (possibly Gollasch himself); or see Williams, “The Accident-Prone Employee.” “Is He Really Accident-Prone? ” Safety Engineering, 90 (1945), 11. “Accident Proneness,” Aircraft Production, 5 (1943), 328–329. Leverett D. Bristol, “Medical Aspects of Accident Control,” Journal of the American Medical Association, 107 (1936), 654, a typical statement. 87. Johnson, “Born to Crash,” 60. 88. Bingham, quoted above in chapter 4, was one who was outspoken. W. H. Cameron, “Reducing Plant Accidents under High-Speed Defense Production,” Personnel, 18 (1941), 5. 89. Slocombe, “How to Cut Accident Costs.” 90. C. L. Sankey, “Safety from an Industrial Viewpoint,” Journal of Educational Sociology, 11 (1937), 52. 91. Morris S. Viteles, “The Scientific Approach in Selecting the Right Man for the Job,” Edison Electric Institute Bulletin, 10 (1942), 264–268. Morris S. Viteles, “The Application of Psychology to Industry and Business,” Occupational Psychology, 16 (1942), 55–64. Clifford E. Jurgensen, “How Much Do We Know about Accident Causes? ” National Safety News, November 1945, 84. 92. Charles S. Slocombe, “The Psychology of Safety, Part ii,” Personnel Journal, 20 (1941), 105–112; quotation is on 112. 93. Clifford E. Jurgensen, “Getting Personal,” National Safety News, August 1945, 90.

ch a p t er 7 1. Much of this chapter is taken, through the kind permission of German Berrios, from “The Syndrome of Accident Proneness (Unfallneigung): Why Psychiatrists Did Not Adopt and Medicalize It,” History of Psychiatry, 19 (2008), 251–274.

Notes to Pages 122–123

2. See, for example, Lester S. King, Medical Thinking: A Historical Preface (Princeton: Princeton University Press, 1982), 162–163. Robley Dunglison, A Dictionary of Medical Science, revised ed. (Philadelphia: Blanchard and Lea, 1860 [c. 1857]), 890: “A name given by the Empirical Sect to the union of symptoms which takes place in diseases.” Robert A. Aronowitz, “When Do Symptoms Become a Disease? ” Annals of Internal Medicine, 134 (2001), 803–808, concludes that sociocultural factors determine when symptom clusters turn into disease, which is implicit in the history that follows below. A recent major historical treatment of disease defi nition is Jacalyn Duffi n, Lovers and Livers: Disease Concepts in History (Toronto: University of Toronto Press, 2005). Charles Rosenberg, “Contested Boundaries: Psychiatry, Disease, and Diagnosis,” Perspectives in Biology and Medicine, 49 (2006), 407–424, describes psychiatry as a “boundary manager.” 3. See, for example, Marion M. Kalez and Rieber C. Hovde, “Pilots with Repeated ‘Pilot-Inaptitude’ Accidents,” Aviation Medicine, 16 (1945), 370–375. While by the World War ii era, the idea of accident proneness was a common one in various types of medical literature, it often was alluded to in only general terms; see, for example, Hermann Levy, “Medicine and Industrial Safety,” Lancet, 2 (6 July 1946), 20–22. 4. The classic discussion is Peter Conrad and Joseph W. Schneider, Deviance and Medicalization: From Badness to Sickness, 2nd ed. (Philadelphia: Temple University Press, 1992), but see Peter Conrad, The Medicalization of Society: On the Transformation of Human Conditions into Treatable Disorders (Baltimore: Johns Hopkins University Press, 2007). Another recent summary of the literature and idea of medicalization is in Stephen Petrina, “The Medicalization of Education: A Historiographic Synthesis,” History of Education Quarterly, 46 (2006), 503–531. Adele Clarke et al., “Biomedicalization: Technoscientific Transformations of Health, Illness, and U.S. Biomedicine,” American Sociological Review, 68 (2003), 161–194, distinguish between a later version of medicalization and that alluded to here. Rosenberg, “Contested Boundaries,” 408n, cautions that medicalization was not an agent but only a generalization of historical processes. 5. Simon Wessely, “Chronic Fatigue: Symptom and Syndrome,” Annals of Internal Medicine, 134 (2001), 838–843. 6. Probably the most scholarly historical explication is Elizabeth Lunbeck, The Psychiatric Persuasion: Knowledge, Gender, and Power in Modern America (Princeton: Princeton University Press, 1994). 7. See, for example, the work of a Viennese psychologist, Rudolf Nejedilk, “Psychische Ursachen von Fussgangerunfallen,” Wiener Archiv für Psychologie, Psychiatrie und Neurologie, 2 (1952), 193–207, who, following Marbe and Tramm, compared those who had accidents with those who did not and concluded that in injured pedestrians there were strong psychophysiological differences such as sensory perception and a tendency to turn the head the wrong way, but unconscious confl icts were not a differentiating factor. 8. Injuries that befell patients and workers in psychiatric hospitals did not, remarkably, stimulate any substantial attention to accident proneness as such. See, for example, the notable absence of the idea in William H. Vicary, “An Analysis of Accidents at a Hospital for Psychiatric Patients,” Medical Bulletin of the Veterans’ Administration, 18 (1941), 292–296.

273

274

Notes to Pages 124–126

9. An example of how far early psychoanalytic ideas were from the accident problem is Fritz Giese, “Psychoanalyse im Fabrikbetriebe,” Zeitschrift für psychoanalytische Pädagogik, 1 (1926), 97–107, who merely invoked some psychoanalytic terms in an attack on capitalist exploitation of labor, particularly in the form of vocational testing. Giese did not include the subject of accidents at all. 10. Karl Marbe, Praktische Psychologie der Unfälle und Betriebsschäden (Munich: Druck und Verlag von R. Oldenbourg, 1926), 63. In his footnote, Marbe noted the possibility suggested by Freud’s evidence that an accident could eventuate from guilt factors outside of consciousness. 11. See, for example, E. W. Scripture, “Reaction Time in Nervous and Mental Diseases,” Journal of Mental Science, 62 (1916), 698–719. K. Marbe, Grundzüge der forensichen Psychologie (Munich: Oskar Beck, 1913), 64–65. This research tradition linked unintentional errors in thinking with psychiatric and psychoanalytic diagnoses. 12. There is a substantial literature on this entire subject, which extends into the subject of shell shock during World War i. See, for example, Traumatic Pasts: History, Psychiatry, and Trauma in the Modern Age, 1870–1930, ed. Mark S. Micale and Paul Lerner (Cambridge: Cambridge University Press, 2001). 13. See, for example, F. W. Dershimer, “Practical Mental Hygiene in Industry,” Journal of Industrial Hygiene, 5 (1923), 5–6. The classic account is Esther FischerHomberger, Die traumatische Neurose: vom somatischen zum socialen Leiden (Bern: Hans Huber, 1975). Even physicians focused on sequelae often expressed their concern about the personal factor that made an accident victim different from those who did not suffer accidents; see, for example, H. Graeme Anderson, “Aeroplane Accidents,” Journal of the Royal Naval Medical Service, 4 (1918), 51–68. 14. Even, as Fischer-Homberger, Die traumatische Neurose, 169–170, points out, to the point of implicitly blaming the victim. 15. Heinrich Meng, “Kind und Unfallneurose,” Die medizinische Welt, 3 (1927), 230–233. James H. Huddleson, Accidents, Neuroses and Compensation (Baltimore: Williams & Wilkins Company, 1932), chap. 2, 206–207. H. Flanders Dunbar, whose work is discussed below, for example at the beginning of her discovery of accident proneness as a part of psychosomatic medicine, cited the College of Physicians and Surgeons psychiatrist, F. W. Dershimer, “The Prevention of Traumatic Neuroses,” Journal of Industrial Hygiene, 16 (1934), 40–51, who clearly described the emotional factors that could dynamically produce a traumatic neurosis after an injury. 16. S. A. Spilman, “The Psychic Element in Accident Cases,” Iowa Medical Journal, 10 (1904), 440. Spilman did not mention the Ganser syndrome of simulated psychosis or somaticization, as George Paulson suggests to me. G. Ledderhose, “Ueber Simulation und ihre Entlarvung in der Unfallchirurgie,” Deutsche medizinische Wochenschrift, 33 (1907), 965–968, for example, wrote that true simulation was very rare, and he or she (only the initial appears in the literature) emphasized the uncertain state of medical knowledge. 17. A number of physicians were drawn into thinking about accidents through the public health concern with preventing accidents. This concern drew them into the safety movement. See, for example, “Accident Prevention: A Public Health Function,” Canadian Public Health Journal, 28 (1937), 394. 18. See, for example, Ermin L. Ray, “The Psychological Versus the Psychiatric

Notes to Pages 126–128

Method in Industry,” Mental Hygiene, 11 (1927), 140–147. Kenneth E. Appel, “Psychiatry in Industry,” Occupational Therapy and Rehabilitation, 10 (1931), 213. 19. I return to this mode below in the conclusion. 20. See, for example, Augusta Scott, “Neuropsychiatric Work in Industry,” Mental Hygiene, 7 (1923), 521–537; Henry Weinstein, “The Neurotic Factor as It Affects the Police Department of the City of New York,” Medical Journal and Record, 126 (1927), 727–729. 21. C. Macfie Campbell, “Personal Factors in Relation to the Health of the Individual Worker,” Mental Hygiene, 13 (1929), 483–484. The phrasing suggests the adaptive orientation of Adolf Meyer and the famous book by William Healy, The Individual Delinquent, in which he applied dynamic as well as commonsense interpretations. 22. Millais Culpin, “A Study of the Incidence of the Minor Psychoses—Their Clinical and Industrial Importance,” Proceedings of the Royal Society of Medicine, 21 (1927), 419–430, especially 424. In subsequent psychiatric writings, Culpin did not mention accident proneness, even though for years he continued to collaborate with members of the team who had worked on the subject at some point. Edward E. Mayer, “Accidents and the Mind,” Safety Engineering, 60 (1930), 45–50, especially 46, 49. 23. Richard H. Price, quoted in Seventeenth Annual Meeting of the International Association of Industrial Accident Boards and Commissions (USBLS Bulletin 536, Washington: U. S. Bureau of Labor Statistics, 1931), 279. 24. Morris S. Viteles, “Psychology and Psychiatry in Industry: The Point of View of a Psychologist,” Mental Hygiene, 13 (1929), 361–377. 25. J.-M. Lahy, “La Psychotechnique et la psychiatrie,” Prophylaxie mentale, 9 (1934), 87–102, viewed the relationship as psychiatrists’ pragmatically using mental tests—in this case especially for detecting psychiatric patients with organic brain damage. 26. V. V. Anderson, Psychiatry in Industry (New York: Harper & Brothers, 1929), especially chap. 9 and 278 and 292. 27. Anderson, Psychiatry in Industry, 3. Pearce Bailey, “Efficiency and Inefficiency— A Problem in Medicine,” Mental Hygiene, 1 (1917), 197, characterized psychology as “one of the medical sciences” in the context of industrial psychology. 28. Eugen Kahn, Psychopathic Personalities, trans. H. Flanders Dunbar (New Haven: Yale University Press, 1931). When Kahn did write about accidents, Eugen Kahn, “Unfallereignis und Unfallerlebnis,” Münchener medizinische Wochenschrift, 72 (1925), 1458–1459, he wrote about psychiatric problems that followed the accident, and, like others, he believed that neuroses after an injury depended on personality and preexisting. underlying psychopathology. 29. Herman M. Adler, “Unemployment and Personality—A Study of Psychopathic Cases,” Mental Hygiene, 1 (1917), 16–24. 30. The connection with the formal idea of deviance is taken up below in chapter 10. 31. Esther Loring Richards, “All Men Are Not Equal,” National Safety News, March 1934, 13–16. 32. Stanley Cobb, “Applications of Psychiatry to Industrial Hygiene,” Journal of Industrial Hygiene, 1 (1919), 343–347.

275

276

Notes to Pages 128–134

33. Jau Don Ball, “Industrial Psychiatry,” American Journal of Psychiatry, ns 1 (1922), 639–677. 34. As late as 1945, the subject was remarkably absent in Aubrey Lewis, E. H. Capel, and D. Elizabeth Bunbury, “Psychiatric Advice in Industry,” British Journal of Industrial Medicine, 2 (1945), 41–45. 35. A. Rendle Short, “Self-Infl icted Injuries in Civilian Practice,” British Medical Journal, 1 (26 February 1921), 298. Alexander Nicoll, “A Remarkable Case of Persistent Ingestion of Needles and Their Removal from Different Parts of the Body, Especially from the Organs of the Abdominal Cavity, by Laparotomy Several Times Repeated,” Lancet, 14 March 1908, 772–778. 36. Thomas H. Kellock, “A Form of Self-Mutilation of the Penis in Young Boys,” British Journal of Children’s Diseases, 12 (1915), 209–212. Hagedorn, “Abnorme Selbstbeschädigungen,” Zeitschrift für Chirurgie, 137 (1916), 125–132. 37. Alvhh R. Lauer, “Why They Drive That Way,” National Safety News, April 1934, 15–18. 38. Otto Löwenstein, “The Nature of Individual Accident Proneness,” Industrial Safety Survey, 10 (1934), 1–9. In 1933, Löwenstein was director of the Rheinischen Provinzial-Kinderheilanstalt für Seelisch Abnorme and the Pathopsychologischen Institut der Universität Bonn. 39. Karl A. Menninger, “Psychoanalytic Aspects of Suicide,” International Journal of Psycho-Analysis, 14 (1933), 376–390. 40. Karl A. Menninger, “A Psychoanalytic Study of the Significance of SelfMutilation,” Psychoanalytic Quarterly, 4 (1934), 408–466. This very extensive article drew substantially upon the mainstream psychoanalytic literature of that day. 41. Ibid. 42. Karl A. Menninger, “Purposive Accidents as an Expression of SelfDestructive Tendencies,” International Journal of Psycho-Analysis, 17 (1936), 6–16. Italics in original. 43. Ibid., 8n, 15–16. 44. Sigmund Freud, The Psychopathology of Everyday Life (vol. 6 of The Standard Edition of the Complete Psychological Works of Sigmund Freud, London: Hogarth Press, 1960), especially 174, 180. Not all physicians, even dynamic psychiatrists, of course, agreed with Freud’s approach; see Samuel A. Tannenbaum, “Accidents and Symptomatic Actions,” Psyche and Eros, 3 (1922), 129–150, who denied the plausibility of the motives that Freud attributed to patients who had accidents. 45. Freud, The Psychopathology of Everyday Life, 162 and chapter 8 in general as well as chapters 5 and 9. Rudolf Meringer and Karl Mayer, Versprechen und Verlesen: Eine psychologisch-linguistische Studie (Stuttgart: G. J. Göschen’sche Verlagshandlung, 1895 [reprinted, ed. Anne Cutler and David Fay, Amsterdam: John Benjamins, 1978]). Mayer was a professor of psychiatry and nervous pathology at the University of Innsbruck. 46. Menninger cited one case from “Fragment of an Analysis of a Case of Hysteria,” in the original edition of Sigmund Freud, Collected Papers (5 vols., New York: Basic Books, 1959 reprint), III, 145. The use of either German or English concordances to Freud’s works does not yield relevant material beyond that cited by Menninger and other writers. 47. Karl Abraham, Selected Papers, trans. Douglas Bryan and Alix Strachey (New

Notes to Pages 134–136

York: Brunner/Mazel, 1979 reprint of 1927 edition), 56–62, especially 61. Menninger also cited a case of a motivated accident, showing that the unconscious motivation was primary and the accident a merely secondary phenomenon, from Franz Alexander, The Psychoanalysis of the Total Personality: The Application of Freud’s Theory of the Ego to the Neuroses, trans. Bernard Glueck and Bertram D. Lewin (New York: Nervous and Mental Disease Publishing Co., 1930), especially 30. 48. Menninger, “Purposive Accidents,” especially 6. 49. Abraham, Selected Papers, 62. Ibid., 59: “When . . . a person has lost all pleasure in life, and the thought is obviously present in his mind that it would be better to die than go on living under such conditions; and when that person meets with an accident under circumstances which suggest that it might have been avoided, then I consider that we are justified in assuming there is an unconscious intention of suicide.” 50. “Planned Accidents,” Safety Engineering, August 1936, 54, reported that psychoanalyst Gregory Zilboorg had “declared that many fatal automobile accidents are deliberate suicides . . . The practicing psychiatrist is only too familiar with the neurotic and ostensibly normal individual who labors under the pressure of a violent but unconscious trend of self-destruction and who either runs his automobile into a telegraph pole or lets himself be run over by an approaching car.” 51. Conceivably the idea of a “repetition compulsion” (from Freud) could have attached itself to the “accident habit.” In some thinkers, as Lawrence S. Kubie, “A Critical Analysis of a Repetition Compulsion,” International Journal of PsychoAnalysis, 20 (1939), 390–402, pointed out, a compulsion to repeat was just another version of the idea of habit. In any event, psychoanalytic thinkers did not make the connection to accidents. 52. See, for example, M. Tramer, “Kausistiches zur Frage der psychogenen Unfallkonstellation,” Schweizerische medizinische Wochenschrift, 59 (1929), 1398. 53. Lydia G. Giberson, “Emotional First Aid,” National Safety News, February 1940, 10–11, 56, 80–84. 54. H. Flanders Dunbar et al., “The Psychic Component of the Disease Process (Including Convalescence), in Cardiac, Diabetic, and Fracture Patients,” American Journal of Psychiatry, 93 (1936), 649–672. H. Flanders Dunbar et al., “The Psychic Component of the Disease Process (Including Convalescence), in Cardiac, Diabetic, and Fracture Patients. Part II,” American Journal of Psychiatry, 95 (1939), 1319–1342. She summarized the fi ndings in Flanders Dunbar, “The Relationship between Anxiety States and Organic Disease,” Clinics, 1 (1942), 888: Of the fracture patients, “eighty per cent two or more accidents, majority three or more. Accident habit especially frequent in decade 15–25. Mainly the result of falls and traffic accidents. Many childhood accidents.” 55. Dunbar, “The Psychic Component . . . Part II.” Flanders Dunbar, “Medical Aspects of Accidents and Mistakes in the Industrial Army and in the Armed Forces,” War Medicine, 4 (1943), 161–175; quotation from 164. The idea that the phrase “accident habit” was in common use in medicine years before does not fi nd confi rmation in the medical literature, although Dunbar used it herself in 1939. It is entirely possible that she projected her own use onto the past or, more intriguingly, she was quoting some medical folklore that recognized as common sense that there were patients who repeatedly had accidents.

277

278

Notes to Pages 136–138

56. For an understanding of psychosomatic medicine, I am drawing heavily upon Robert C. Powell, “Helen Flanders Dunbar (1902–1959) and a Holistic Approach to Psychosomatic Problems. I. The Rise and Fall of a Medical Philosophy,” Psychiatric Quarterly, 49 (1977), 133–152, and Robert C. Powell, “Helen Flanders Dunbar (1902–1959) and a Holistic Approach to Psychosomatic Problems. II. The Role of Dunbar’s Nonmedical Background,” ibid., 50 (1978), 144–157. An incisive description of the general framework is in Theodore M. Brown, “The Rise and Fall of American Psychosomatic Medicine,” http://human-nature.com/freeassociations/riseandfall.html. 57. Flanders Dunbar, Psychosomatic Diagnosis (New York: Paul B. Hoeber, 1943), especially chap. 4. Following Dunbar’s lead, for example, was Arnold J. Rawson, “Accident Proneness,” Psychosomatic Medicine. 6 (1944), 88–94. 58. Dunbar, Psychosomatic Diagnosis, especially 179, 189. 59. Ibid., 188–190. 60. Ibid., especially 225, 246–247. 61. Flanders Dunbar, Mind and Body: Psychosomatic Medicine (New York: Random House, 1947). Flanders Dunbar, Psychiatry in the Medical Specialties (New York: McGraw-Hill Book Company, 1959), chap. 4. Flanders Dunbar, “Do You Have the Accident Habit? ” Science Digest, February 1948, 7–10. 62. See, for example, the centrality of Dunbar in the major editorial, “The Accident-Prone Individual,” American Journal of Public Health, 39 (1949), 1036–1038. 63. Otto Fenichel, The Psychoanalytic Theory of Neurosis (New York: W. W. Norton, 1945), 506. One other possibility for the idea of type to carry accident proneness into psychiatry in the 1930s and 1940s came through the interest of Alexandra Adler in the subject. Adler was a chief advocate of the variety of dynamic psychiatry originated by her father, Alfred Adler. In Austria and the United States, she studied workers who suffered repeated accidents, initially using terms from Marbe. Her work made no impression on the literature, however, beyond popularizing the idea of accident proneness. Alexandra Adler, “Beitrag zum Problem der Unfallshäufung,” Wiener medizinische Wochenschrift, 84 (1934), 293–295; Alexandra Adler, “The Psychology of Repeated Accidents in Industry,” American Journal of Psychiatry, 98 (1941), 99–100. 64. For example, Joseph L. Fetterman, “Neuropsychiatric Aspects of Industrial Accidents,” Industrial Medicine, 15 (1946), 96–100. 65. Powell, “Helen Flanders Dunbar,” parts 1 and 2. 66. See, for example, Robert M. Goldenson, “Accidents Don’t Always Happen by Accident,” Good Housekeeping, May 1945, 36–37, 212–213. 67. For example, E. Eduardo Krapf, “Accidentes y operaciones como expresión de tendencias auto-destructivas,” Revista médica del Hospital Británico, 1 (1944), 36– 40, who pointed out that there are many forms that self-destructive actions can take, including “disposición al accidente.” 68. John Leuba, “Women Who Fall,” trans. Cecily de Monchaux, International Journal of Psycho-Analysis, 31 (1950), 6–7. Another example is Margaret E. Fries and Paul J. Woolf, “Some Hypotheses on the Role of the Congenital Activity Type in Personality Development,” Psychoanalytic Study of the Child, 8 (1953), 57–58. 69. Norman Tabachnick and Robert E. Litman, “Character and Life Circumstances in Fatal Accident,” Psychoanalytic Forum, 1 (1966), 66–70, with discussion

Notes to Pages 138–140

70–74, 239–240. Robert E. Litman and Norman Tabachnick, “Fatal One-Car Accidents,” Psychoanalytic Quarterly, 36 (1967), 248–259. 70. Jules H. Masserman, The Practice of Dynamic Psychiatry (Philadelphia: W. B. Saunders Company, 1955), for example, indexes accident proneness in several places, but in a very inconsistent and confusing way. Two cases under that heading are listed, but in each case only one accident was involved, and one of those was a case of deliberate self-mutilation! Dunbar is cited in support of the idea, and in a footnote a critic is cited. 71. For example, psychiatrist Philip J. Moorad, “Human Factors in Accident Liability,” Industrial Medicine, 16 (1947), 494–498, has a subtitle, “With Special Reference to Accident Repeaters.” Yet the text covers accident prone workers in only one place, while most of the exposition is devoted to general factors such as age and physical disability. Or Jerome M. Kummer, “A Psychiatrist Looks at Problem Drivers,” Medical Times, 91 (1963), 160–164, who did not use the term or the idea of accident proneness at all. 72. American Psychiatric Association, Diagnostic and Statistical Manual: Mental Disorders (Washington: American Psychiatric Association, Mental Hospital Service, 1952), 13. There is a substantial historical literature on changes in the DSM, including Gerald N. Grob, “Origins of DSM-I: A Study in Appearances and Reality,” American Journal of Psychiatry, 148 (1991), 421–431; Lloyd H. Rogler, “Making Sense of Historical Changes in the Diagnostic and Statistical Manual of Mental Disorders: Five Propositions,” Journal of Health and Social Behavior, 38 (1997), 9–20. 73. See especially Rick Mayes and Allan V. Horowitz, “DSM-III and the Revolution in the Classification of Mental Illness,” Journal of the History of the Behavioral Sciences, 41 (2005), 249–267, and Mitchell Wilson, “DSM-III and the Transformation of American Psychiatry: A History,” American Journal of Psychiatry, 150 (1993), 399–410. A random negative example from later years, with not a hint of an approach to accident proneness is David P. Moore and James W. Jefferson, Handbook of Medical Psychiatry (St. Louis: Mosby, 1996). 74. See, for example, Joseph L. Fetterman, Practical Lessons in Psychiatry (Springfield: Charles C. Thomas, 1949), 209–210, 316–317, who used accident proneness both as a presenting syndrome to be explained dynamically and, in another illustrative case, as a symptom of a more general neurosis. Compare Comprehensive Textbook of Psychiatry/III, ed. Harold I. Kaplan, Alfred M. Freedman, and Benjamin J. Sadock, 3rd ed. (2 vols., Baltimore: Williams & Wilkins, 1980), 1953–1957, with Comprehensive Textbook of Psychiatry/IV, ed. Harold I. Kaplan and Benjamin J. Sadock, 4th ed. (2 vols., Baltimore: Williams & Wilkins, 1985), where the subject does not appear. The subject was, however, alive and well in a textbook on child psychiatry: Raymond Sobel, “Accident Proneness,” in Basic Handbook of Child Psychiatry, ed. Joseph D. Noshpitz (2 vols., New York: Basic Books, 1979), II, 637–643. 75. Alexander H. Hirschfeld and Robert C. Behan, “The Accident Process,” Journal of the American Medical Association, 186 (1963), 193–199, 300–306, especially 193, 196. 76. Examples include Cicero Christiano de Sousa, “Acidentes múltiplos como sintoma neurotico,” Arquivos de neuro-psiquiatria, 1947, 155–166; A. Hedri, “Eine Indikationserweiterung der Psychotherapie: Die Unfallbereitschaft,” Fortschritte der psychosomatische Medizin, 3 (1963), 154–159.

279

280

Notes to Pages 140–143

77. An incisive description as of 1966 is Robert C. Behan and A. H. Hirschfeld, “Current Concepts of Accident Proneness,” Archives of Environmental Health, 13 (1966), 462–466. 78. See, for example, Peter Husband and Pat E. Hinton, “Families of Children with Repeated Accidents,” Archives of Disease in Childhood, 47 (1972), 396–400. 79. For example, Gerald Gordon, “The People Who Get Hurt,” National Safety News, February 1954, 20–21, 113–120, who portrayed accident proneness as a treatable mental illness. I. Csillag and E. Hedri, Jun., “Personal Factors of Accident Proneness,” Industrial Medicine, 18 (1949), 29–30. Alan A. McLean and Russell De Reamer, “The Accident Repeater: Psychiatric and Statistical Considerations,” Psychosomatic Medicine, 29 (1961), 564. Karl Menninger, “Mental Attitudes and Safety,” Menninger Quarterly, 13 (1959), 7. 80. See, for example, Johannes Grunert, “Unfallkeim—Unfalldisposition— Unfall: Entwurf einer psychodyamischen Unfalltheorie,” Zeitschrift für experimentelle und angewandte Psychologie, 8 (1961), 519–539. One broad survey that includes some Continental literature is J. H. Schultz, “Unfall und Psychoanalyse,” Zentralblatt für Verkehrs-Medizin, 1 (1956), 259–266. 81. As noted earlier, psychologization is taken up in many general cultural histories. Some general treatments appear in such works as Ellen Herman, The Romance of American Psychology: Political Culture in the Age of Experts (Berkeley: University of California Press, 1995); James H. Capshew, Psychologists on the March: Science, Practice, and Professional Identity in America, 1929–1969 (New York: Cambridge University Press, 1999), especially chap. 11; Philip Cushman, Constructing the Self, Constructing America: A Cultural History of Psychotherapy (Reading MA: AddisonWesley Publishing Company, 1995), which covers more than the United States; and Inventing the Psychological: Toward a Cultural History of Emotional Life in America, ed. Joel Pfister and Nancy Schnog (New Haven: Yale University Press, 1997). 82. Harold Brandaleone and Richard H. Daily, “A Method of Reducing the Accident Rate in Bus Operation—Selection of the Bus Operator,” Industrial Medicine and Surgery, 19 (1950), 233. Fred P. Kiefer, “Help for the Accident Prone,” Safety Maintenance and Production, October 1951, 21. 83. Joseph Connolly, “Accident Proneness,” British Journal of Hospital Medicine, 26 (1981), 470. My own survey confi rms an obvious decline beginning by the 1970s. 84. Sebastiano Timpanaro, The Freudian Slip: Psychoanalysis and Textual Criticism, trans. Kate Soper (London: NLB, 1976 [1974]). See the discussion below of the rise of cognitive psychology. 85. Allan Young, The Harmony of Illusions: Inventing Post-Traumatic Stress Disorder (Princeton: Princeton University Press, 1995), 104–105, in a discussion that more generally offered many instructive parallels to, and comparisons with, accident proneness. 86. John MacIver, “Psychological Aspects of Accident Causation,” Industrial Medicine and Surgery, 28 (1959), 231. 87. Franz Alexander, “The Accident-Prone Individual,” Public Health Reports, 64 (1949), 362. 88. Julian A. Waller and Howard W. Mitchell, “Is ‘Accident Proneness’ a Useful Concept for Medical Practice? ” Medical Times, 93 (1965), 36–39. “Accident susceptibility” in fact did not appear in the psychiatric or medical literature.

Notes to Pages 143–145

89. Other competing areas in medicine also existed. Peter Husband, “The Accident-Prone Child,” Practitioner, 211 (1973), 335–344, gave a history of how accident proneness entered the field of pediatrics, 336–338, a subject to which I return in the next chapter. He confi rmed the existence of accident repeaters, but like the psychiatrists he concluded that treatment should be on an individual basis for each patient. 90. John J. Brennan and Adolph G. Ekdahl, “Are Psychotics Accident Prone? ” American Journal of Psychiatry, 120 (1963), 175–176; Matthew W. Buttiglieri and Marie Guenette, “Driving Record of Neuropsychiatric Patients,” Journal of Applied Psychology, 51 (1967), 96–100. Gunnel Mellbin, Traffic Accidents and Mental Health, trans. Helen Frey (Stockholm: P.A. Norstedt & Söner, 1971). 91. Stephen Patnode, “ ‘Their Lack of Masculine Security and Aggression Was Obvious’: Gender and the Medicalization of Inebriety in the United States, 1930–50,” Canadian Bulletin of the History of Medicine, 24 (1907), 67–92, suggests that for a time—coincident with the flourishing of the idea of accident proneness— alcoholism also was not medicalized. 92. German E. Berrios, “The History of Descriptive Psychopathology,” in Psychiatric Epidemiology: Assessment Concepts and Methods, ed. Juan E. Mezzich, Miguel R. Jorge, and Ihsan M. Salloum (Baltimore: Johns Hopkins University Press, 1994), 47–68, especially 53–55. H. Sass and S. Herpertz, “Personality Disorders: Clinical Section,” and Waltraud Ernst, “Personality Disorders: Social Section,” in A History of Clinical Psychiatry: The Origin and History of Psychiatric Disorders, ed. German E. Berrios and Roy Porter (London: Athlone Press, 1995), 633–655. 93. See particularly German E. Berrios, The History of Mental Symptoms: Descriptive Psychopathology Since the Nineteenth Century (Cambridge: Cambridge University Press, 1996), 424–436. Trying to fit accident proneness into the category of personality disorder led medical writers to think in terms of types and traits. Accident prone people could of course be conceptualized in terms of types, as Dunbar did—once one accepted the phenomenon of accident repeater. The most promising category for accident proneness was as a personality trait. But in psychiatric discourse, personality disorders had a framework different from a single behavioral deviation, much less one established by statistical observation. Personality disorders— particularly the mid-twentieth-century diagnosis of psychopathic personality— were based on a social perception, that the patient had antisocial motives. The whole point of accident proneness was that if the victim had a detectable motive, it was to harm oneself, not to act out antisocial “psychopathic” tendencies. In fact, as I have noted, the broad category of personality disorder in DSM-III did not include accident proneness among the possible signs, symptoms, and categories.

ch a p t er 8 1. Of course, all along many experts appeared not to need or want the idea; see, for example, Injuries and Accident Causes in the Manufacture of Clay Construction Products (United States Department of Labor Bulletin no. 1023, Washington: U.S. Government Printing Office, 1951). 2. Joseph E. Barmack and Donald E. Payne, “Injury-Producing Private Motor Vehicle Accidents among Airmen: Psychological Models of Accident-Generating Processes,” Journal of Psychology, 52 (1961), 3. In 1964, Frank A. Haight, “Accident

281

282

Notes to Pages 146–150

Proneness, The History of an Idea,” Automobilismo e Automobilismo Industriale, 4 (1964), 45, concluded that “it would probably be no exaggeration to say that a thousand papers on the subject have appeared in the past forty years.” The basic post-1945 profi le of items specifically on accident proneness is based on my own survey of many hundreds of articles and books from that period. 3. See, for example, the largely American postwar literature cited by J. Zurfluh, Accidents du travail et formation-sécurité (Paris: Dunod, 1957). Lynette Shaw and Herbert S. Sichel, Accident Proneness: Research in the Occurrence, Causation, and Prevention of Road Accidents (Oxford: Pergamon Press, 1971), 38; chap. 9, especially 135–136. 4. Milton D. Kramer et al., The Motor-Vehicle Driver: His Nature and Improvement (Saugatuck CT: Eno Foundation for Highway Traffic Control, 1949), offered a thorough summary of thinking at that point. 5. Jean Surry, Industrial Accident Research: A Human Engineering Appraisal (Toronto: Labour Safety Council, Ontario Department of Labour, 1971), 149–153. Some background is in Joel A. Tarr and Mark Tebeau, “Housewives as Home Safety Managers: The Changing Perception of the Home as a Place of Hazard and Risk, 1870–1940,” in Accidents in History: Injuries, Fatalities and Social Relations, ed. Roger Cooter and Bill Luckin (Amsterdam: Rodopi, 1997), 196–233. 6. C. A. Oakley, “The Prevention of Factory Accidents,” Occupational Psychology, 26 (1952), 50. 7. Once again I refer to the Catherine Covert model of the way in which, with time, popularization becomes more and more technical and less and less elementary. 8. The only previous such book beyond Marbe’s was a short monograph, H. Hildebrandt and K. Ross, Individuelle Unfallaffinität (Berlin: Richard Schoetz, 1932). 9. I have been unable to identify Norman Roberts Lykes, despite an intensive search. He wrote in the manner of a journalist. 10. Norman Roberts Lykes, A Psychological Approach to Accidents (New York: Vantage Press, 1954), 49–50. 11. Ibid., especially 68; some italics removed. 12. Ross A. McFarland, Roland C. Moore, and A. Bertrand Warren, Human Variables in Motor Vehicle Accidents: A Review of the Literature (Boston: Harvard School of Public Health, 1955), 19 and passim. 13. For example, Charles A. Drake, “Detecting the Accident-Prone Worker,” Personnel, 18 (1942), 276–281. 14. Morris S. Schulzinger, The Accident Syndrome: The Genesis of Accidental Injury, A Clinical Approach (Springfield IL: Charles C. Thomas, 1956), especially ix, 59. Only one chapter is explicitly on accident proneness, but the subject pervades the whole book. 15. See chapter 5, above. 16. M. S. Schulzinger, “Accident Proneness,” Industrial Medicine and Surgery, 23 (1954), 151–152, quotation from 151, and he made the point repeatedly in several other publications in this period. 17. See, for example, R. F. Tredgold, “Emotional Factors and Accident Causation,” Community Health, 2 (1970), 7–11, especially 8. 18. Sauli Häkkinen, Traffic Accidents and Driver Characteristics: A Statistical and Psychological Study (Helsinki: Institute of Technology, 1958), 79. David K. Trites,

Notes to Pages 150–151

Albert L. Kubala, Jr., and Saul B. Sells, “Aircraft Accidents vs. Characteristics of Pilots,” Journal of Aviation Medicine, 26 (1955), 494. 19. Robert L. Thorndike, The Human Factor in Accidents with Special Reference to Aircraft Accidents (Randolph Field TX: USAF School of Aviation Medicine, 1951). Häkkinen, Traffic Accidents and Driver Characteristics, especially 52–71. B. J. Speroff, “Empathic Ability and Accident Rate Among Steel Workers,” Personnel Psychology, 6 (1953), 297–300. 20. Seiichi Kuraishi, Masahide Kato, and Bien Tsujioka, “Development of the ‘Uchida-Kraepelin Psychodiagnostic Test’ in Japan,” Psychologia, 1 (1957), 104–109. 21. I. D. Brown, “Studies of Component Movements, Consistency and Spare Capacity of Car Drivers,” Annals of Occupational Hygiene, 5 (1962), 131–143. “Attitude Test Predicts Teenage Auto Accidents,” Safety Education, January 1963, 28. 22. William N. McBain, “Arousal, Monotony, and Accidents in Line Driving,” Journal of Applied Psychology, 54 (1970), 509–519. William L. Mihal and Gerald V. Barrett, “Individual Differences in Perceptual Information Processing and Their Relation to Automobile Accident Involvement,” Journal of Applied Psychology, 61 (1976), 229–233. 23. See, for example, Erich Langer, “Kann die Lues einen Grund für Verweigerung oder Entziehung des Führerscheins darstellen? ” Ärztliche Wochenschrift, 7 (1952), 490–492; A. E. Bennett, “Accident Proneness in Multiple Sclerosis,” Journal of Nervous and Mental Disease, 113 (1951), 198–210. 24. See, for example, Medical Aspects of Fitness to Drive Vehicles: A Guide for Medical Practitioners, ed. Leslie G. Norman (London: Royal College of Surgeons, Report by the Medical Commission on Accident Prevention, 1968, and subsequent editions through 1995); Joseph L. Fetterman and Victor M. Victoroff, “Industrial and Forensic Applications of Electroencephalography,” Archives of Industrial Hygiene and Occupational Medicine, 4 (1951), 10–24; Lars Ysander, “The Safety of Drivers with Chronic Disease,” British Journal of Industrial Medicine, 23 (1966), 28–36; B. Herner, B. Smedby, and L. Ysander, “Sudden Illness as a Cause of Motor-Vehicle Accidents,” ibid., 37–41; M. Woldert and G. Ritter, “Hirntumorkranke im Strassenverkehr,” Der Nervenarzt, 54 (1983), 304–310. 25. This remarkable phenomenon of growing alcohol awareness, with an emphasis on driving while drunk, was obvious in the second half of the twentieth century, but it was incidental and possibly irrelevant to the main subject I am pursuing and will not be discussed in detail. It did include “recreational” drug users, and it affected groupings of people at risk for accidents. See, for example, Melvin L. Selzer, “Alcoholism, Mental Illness, and Stress in 96 Drivers Causing Fatal Accidents,” Behavioral Science, 14 (1969), 1–10; David Sohn, “The Drug Abuser as a Safety Hazard,” ASSE Journal, November 1971, 12–16. In this context, accident proneness continued to appear under such headings as “risky driving”; see, for example, Young Drivers Impaired by Alcohol and Other Drugs, ed. Timothy Benjamin (London: Royal Society of Medicine Services, 1987). 26. Hans Selye, “The Stress of Life—New Focal Point for Understanding Accidents,” Industrial Medicine and Surgery, 33 (1964), 621–625. Leon Brody, “Methodology and Patterns of Research in Industrial Accidents,” Annals of the New York Academy of Sciences, 107 (1963), 659–663, believed that accident prone people “as such” did not exist, but he thought that investigating the role of stress in accidents should have

283

284

Notes to Pages 151–153

“the highest priority.” The idea of stress had of course been around in accident research for some time; see, for example, Willard Kerr, “Complementary Theories of Safety Psychology,” Journal of Social Psychology, 45 (1957), 3–9. Cary L. Cooper and Philip Dewe, Stress: A Brief History (Oxford: Blackwell, 2004). Eventually stress was in part reconfigured into more specific and complex physiological components; see, for example, Robert T. Rubin, “Biochemical and Neuroendocrine Responses to Severe Psychological Stress,” in Life Stress and Illness, ed. E. K. Eric Gunderson and Richard H. Rahe (Springfield IL: Charles C. Thomas, 1974), especially 236–239. 27. Alan A. McLean, “Accident Proneness—A Clinical Approach to Injury Liability,” Industrial Medicine and Surgery, 24 (1955), 123, 124. A similar formulation of stress and temporary conditions was still in use in R. A. Alkov, “Psychological Stress, Health and Human Error,” Professional Safety, August 1981, 12–14, and in Steven R. Furney, “Life Change Events as a Predictor of Accident Incidence in a College Population,” Health Education, November/December 1983, 22–24. R. L. Summers, “Stress Management: A New Dimension in Safety,” Professional Safety, January 1982, 38–39. Bruce P. Dohrenwend, “Social Stress and Psychopathology,” in A Decade of Progress in Primary Prevention, ed. Marc Kessler and Stephen E. Goldston (Hanover NH: University Press of New England, 1986), 87–114. 28. For example, Rollo H. Britten, “Illness and Accidents among Persons Living Under Different Housing Conditions,” Public Health Reports, 56 (1941), 609–639. 29. W. A. Tillmann and G. E. Hobbs, “The Accident-Prone Automobile Driver,” American Journal of Psychiatry, 106 (1949), 321–331. W. A. Wong and G. E. Hobbs, “Personal Factors in Industrial Accidents—A Study of Accident Proneness in an Industrial Group,” Industrial Medicine, 18 (1949), 291–294; these Canadian investigators were convinced that “the accident tendency is a lifelong characteristic, and it appears to invade all aspects of life” (292). 30. Boris Speroff and Willard Kerr, “Steel Mill ‘Hot Strip’ Accidents and Interpersonal Desirability Values,” Journal of Clinical Psychology, 8 (1952), 89–91. T. T. Paterson and F. J. Willett, “An Anthropological Experiment in a British Colliery,” Human Organization, 10 (1951), 19–25. 31. A. G. Arbous and J. E. Kerrich, “Accident Statistics and the Concept of Accident-Proneness,” Biometrics, 7 (1951), 390. 32. Schulzinger, The Accident Syndrome, 160. So, for example, Armand M. Nicholi II, “The Motorcycle Syndrome,” American Journal of Psychiatry, 126 (1970), 1588–1595, made this syndrome equivalent to a syndrome of accident proneness in vehicular accident cases. 33. Thomas J. Ruane, “Repeated Trauma as the Presenting Symptom of a Pathological Grief Reaction,” Journal of Family Practice, 10 (1980), 327–328. 34. Logan Wright, Sally Flagler, and Alice G. Friedman, “Assessment for Accident Prevention,” in Handbook of Child Health Assessment: Biopsychosocial Perspectives, ed. Paul Karoly (New York: John Wiley & Sons, 1988), 513. 35. Julian A. Waller and Howard W. Mitchell, “Is ‘Accident Proneness’ a Useful Concept for Medical Practice? ” Medical Times, 93 (1965), 39. Waller and Mitchell’s argument was summarized above, in chapter 7. 36. E. James Anthony, “The Treatment of the Latency Child,” Canadian Psychiatric Association Journal, 21 (1926), 207. The child turned out, in the physician’s eyes, to be suffering from nuclear family confl icts of late latency.

Notes to Pages 154–157

37. Schulzinger, The Accident Syndrome, ix, 215. 38. Harold Marcus, “The Accident Repeater: A Comparative Psychiatric Study,” Industrial Medicine and Surgery, 37 (1968), 768–772. 39. L. G. Norman, “Medical Aspects of Road Safety,” Lancet, 1 (1960), 992, noted that Portugal required an independent medical examination of each applicant for a driver’s license. 40. Joshua A. Mott, “Personal and Family Predictors of Children’s Medically Attended Injuries That Occurred in the Home,” Injury Prevention, 5 (1999), 189–193. For the risk context, see the next chapter. 41. Marion Norris Gleason, “The Accident-Prone Patient,” New York State Journal of Medicine, 48 (1948), 2168–2170. 42. See the discussion of epidemiology in the next chapter. 43. W. von Freytag-Loringhoren et al., Accidents in Childhood as a Public Health Problem (Strasbourg: Council of Europe, European Health Committee, 1972), quotation on 63. 44. E. Richard Weinerman, “Accident-Proneness: A Critique,” American Journal of Public Health, 39 (1949), 1527–1530. 45. Edwin Lipinski, et al., “Occupational Accidents: Some Psychosocial Factors in the Accident Syndrome,” Canadian Psychiatric Association Journal, 10 (1965), 305. 46. Sydney A. Halpern, American Pediatrics: The Social Dynamics of Professionalism, 1880–1980 (Berkeley: University of California Press, 1988), especially 133–143, suggests some of the context of the psychologization of childhood injuries. 47. In a medical dictionary that still included the term in 1997, the note focused on children in stipulating, “Most common patients are males under age 21 . . . More frequent in summer.” Sergio I. Magalini and Sabina C. Magalini, Dictionary of Medical Syndromes, 4th ed. (Philadelphia: Lippincott-Raven, 1997), 4. 48. Vita Krall, “Personality Factors in Accident Prone and Accident Free Children,” Ph.D. dissertation, University of Rochester, 1951. Later investigators had not followed up on the early publications by Marbe and his students noted in chapter 2. The inevitable exception was Nathan W. Ackerman and Leona Chidester, “ ‘Accidental’ Self-Injury in Children,” Archives of Pediatrics, 53 (1936), 711–721. As late as M. W. Partington, “The Importance of Accident-Proneness in the Aetiology of Head Injuries in Childhood,” Archives of Disease in Childhood, 35 (1960), 215, the author was still claiming little research had been carried out on children. 49. A. A. Fabian and Lauretta Bender, “Head Injury in Children: Predisposing Factors,” American Journal of Orthopsychiatry, 17 (1947), 68–79 50. Ruth Morris Bakwin and Harry Bakwin, “Accident Proneness,” Journal of Pediatrics, 32 (1948), 749–752, especially 751. 51. Helen W. Bellhouse and Roy J. Boston, “The Physician and Childhood Accidents,” Journal of the Medical Association of Georgia, 43 (1954), 142–145. 52. Some account of these events is in John C. Burnham, “How the Discovery of Accidental Childhood Poisoning Contributed to the Development of Environmentalism in the United States,” Environmental History Review, 19 (1995), 57–81. The typical figures are from Bruno Gans, “Childhood Accidents,” Home Safety Journal, April 1969, 13. One of the curious aspects of childhood accidents was how they differed in different geographical areas: Ross G. Mitchell, “Accidents in Childhood,” Developmental and Medical Child Neurology, 9 (1967), 767, noted, “British chil-

285

286

Notes to Pages 157–160

dren are most commonly poisoned by household cleaning agents while aspirins head the list in New York, and in Sweden the ingestion of cigarettes is the largest single cause . . . Drowning is an especially important cause of accidental childhood deaths in countries where there are many lakes and canals, as in Finland and the Netherlands.” 53. Elizabeth Mechem Fuller, “Injury-Prone Children,” American Journal of Orthopsychiatry, 18 (1948), 723. 54. William S. Langford et al., “Pilot Study of Childhood Accidents: Preliminary Report,” Pediatrics, 11 (1953), 405. Harold Jacobziner and Herbert Rich, “Home Safety and Accident Prevention in a Child Health Conference,” American Journal of Public Health, 44 (1954), especially 92. 55. George M. Wheatley and Stephen A. Richardson, “Some Approaches to Research in Childhood Accidents,” Pediatrics, 25 (1960), 343–347, quotation from 346. 56. Elizabeth Mechem Fuller and Helen B. Baune, “Injury Proneness and Adjustment in a Second Grade,” Sociometry, 14 (1951), 210–225. Gordon P. Liddle and Robert E. Rockwell, “The Kid with Two Strikes Against Him,” Safety Education, December 1963, 3–5. David Klein, “Some Methodological Problems in Research on Childhood Accidents,” in Childhood Accidental Injury Symposium Proceedings (Charlottesville: University of Virginia School of Medicine, 1966), 112–121. Roger J. Meyer and Joanne Bluestone, “Accidental Injury to the Preschool Child,” Journal of Pediatrics, 63 (1963), 95–105. Hildegard Marcusson, Wilhelm Ochmisch, and Wolfgang Pechmann, Der Unfall im Kindes- und Jugendalter. 1. Teil: Mortalität und Prophylaxie. Ein socialhygienische Studie (Berlin: VEB Verlag Volk und Gesundheit, 1970). Later studies occasionally were not necessarily positive for social factors; see, for example, Jean Golding, “Accidents in the Under Fives,” Health Visitor, 56 (1983), 293–294. 57. Gans, “Childhood Accidents,” 13–14. 58. Raymond Sobel, “The Psychiatric Implications of Accidental Poisoning in Childhood,” Pediatric Clinics of North America, 17 (1970), 653–685, quotation from 655. 59. Helen Ross, reported in Pediatrics, 11 (1953), 413–414. Her camp was an allgirl camp. 60. Peter Husband, “The Accident-Prone Child,” Practitioner, 211 (1973), 335; Peter Husband, “The Child with Repeated Injuries—A Family Problem,” Journal of the Royal College of General Practitioners, 25 (1975), 419. A Florida physician, H. J. Roberts, in a comprehensive book, The Causes, Ecology and Prevention of Traffic Accidents, With Emphasis upon Traffic Medicine, Epidemiology, Sociology and Logistics (Springfield IL: Charles C. Thomas, 1971), still showed throughout his massive survey how pervasive the concept of accident proneness continued to be at that late date, even as he used the term often in a generic and inclusive way. 61. See, for example, from the 1980s, Jerry G. Jones, “The Child Accident Repeater,” Clinical Pediatrics, 19 (1980), 284–288; Melanie A. Suhr, “Trauma in Pediatric Populations,” Advances in Psychosomatic Medicine, 16 (1986), 31–47. 62. Again I am drawing on the Covert model of the popularizing process. 63. For example, William B. McGrath, “The Accident-Prone Child,” Arizona Medicine, March 1963, 31a-33a; “Accident Proneness,” Canadian Medical Association Journal, 90 (1964), 646–647.

Notes to Pages 160–162

64. [Peter Husband and Pat E. Hinton,] “The Psychology of Accidents,” Care in the Home, January 1973, 13–15. The original article is Peter Husband and Pat E. Hinton, “Families of Children with Repeated Accidents,” Archives of Disease in Childhood, 47 (1972), 396–400. 65. Lawrence Galton, “Some Folks Have All the Accidents,” Nation’s Business, April 1947, 36–38, 64–67. “Discovering the Accident-Prone Employee,” Management Review, 36 (1947), 551–552. W. F. Hubbard, “Who Are Your Accident-Prones, and Why,” Safety Engineering, February 1947, 14–15, 31–33. “Psychologists at Denver,” Newsweek, 19 September 1941, 56–57. A. Wyn Williams, “Why Workers Get Hurt,” Mill & Factory, February 1948, 99. 66. For example, “Accident Repeaters,” Safety Engineering, September 1950, 30. 67. See, for example, Simon S. Olshansky, “Accident Proneness and Its Implications for Vocational Counseling,” Journal of Rehabilitation, 14 (1948), 27–28; “The Accident-Prone Individual,” American Journal of Public Health, 39 (1949), 1036–1038. Flanders Dunbar, “Do You Have the Accident Habit? ” Science Digest, February 1948, 7–10. Frank Howard Richardson, “Are You Accident Prone? ” Today’s Health, April 1951, 14–15, 50. 68. Constance J. Foster, “Some Accidents Don’t Just Happen,” Parents’ Magazine, May 1946, 38–39, 161–163, quotation from 39. The recognition by insurance companies was only implicit, that is, through the cancellation provision. See below. 69. “The Unmet Needs . . . of Accident Prone Children,” Safety Education, March 1958, 19. Lawrence and Eda LeShan, “Is Your Child Accident Prone? ” National Parent-Teacher, March 1960, 17–19. 70. Jane W. Kessler, Psychopathology of Childhood (Englewood Cliffs NJ: PrenticeHall, 1966), 349–350, for example, gave a powerful psychological emphasis to psychosomatic behavior and “failure of the ego function of self preservation.” Glen D. Mellinger, “Family Studies of Accidents,” in Childhood Accidental Injury Symposium Proceedings, 121–129. 71. Lee Forrest Hill, “Accidents in Children,” Journal of the Iowa State Medical Society, 50 (1960), 254. See also, for example, Vita Krall, “Personality Characteristics of Accident Repeating Children,” Journal of Abnormal and Social Psychology, 48 (1953), 99–107. 72. After World War ii, a few U.S. tort law specialists, for example, began to write about possible legal applications of the idea of accident proneness, and then the idea, as in other areas of popularization, just faded out. Fleming James, Jr., and John J. Dickinson, “Accident Proneness and Accident Law,” Harvard Law Review, 63 (1950), 769–795. Frank E. Maloney and William J. Rish, “The Accident-Prone Driver: The Automotive Age’s Biggest Unsolved Problem,” University of Florida Law Review, 14 (1962), 364–384. Robert Henley Woody, “Accident Proneness: A Psychological Framework,” Medical Trial Techniques Annual, 1981, 74–82. 73. Benjamin Spock, The Common Sense Book of Baby and Child Care (New York: Duell, Sloan and Pearce, 1946). Subsequent editions were not substantially altered in these areas. 74. Examples include M. E. Steiner, The Psychologist in Industry (Springfield: Charles C. Thomas, 1949), 81–82; R. Bromley-Harker, “Prevention of Accidents,” in Modern Trends in Occupational Health, ed. R. S. F. Schilling (London: Butterworth, 1960), 168; James C. Coleman, Abnormal Psychology and Modern Life (Chicago: Scott,

287

288

Notes to Pages 163–164

Foresman and Company, 1956), 241; Kessler, Psychopathology of Childhood, 349–350. Anna M. Baetjer, in Rosenau’s Preventive Medicine and Public Health, 8th ed., ed. Kenneth F. Maxcy (New York: Appleton-Century-Crofts, 1956), 1106; ibid., 10th ed., 1973, 961: “In industry the majority of accidents occur among a relatively small group of employees, indicating that some persons are more likely to have accidents than others.” 75. My sample includes a total of more than twenty editions of eight textbooks issued between 1946 and 1975. Later textbooks do not extend the profi le significantly, largely because accident proneness simply dropped out. For a European example, see Friedrich Dorsch, Geschichte und Probleme der angewandten Psychologie (Bern: Verlag Hans Huber, 1963), especially 140, 145. European textbooks were so few in number that they do not provide significant additional information. 76. Joseph Tiffi n, Industrial Psychology, 2nd ed. (New York: Prentice-Hall, 1947), especially 421–451. The fi rst edition pages are 282–309. 77. Norman R. F. Maier, Psychology in Industry (Boston: Houghton Miffl in Company, 1946), 346–358, quotation from 349. 78. Thomas Willard Harrell, Industrial Psychology (New York: Rinehart & Company, 1949). 79. See, for example, how a section that started out entirely on accident proneness essentially disappeared in Readings in Industrial and Business Psychology, ed. Harry W. Karn and B. von Haller Gilmer (New York: McGraw-Hill Book Company, 1952), part 6; B. von Haller Gilmer et al., Industrial Psychology (New York: McGrawHill Book Company, 1961), chap. 15; B. von Haller Gilmer et al., Industrial and Organizational Psychology (New York: McGraw-Hill, 1971), chap. 18. 80. See especially Rollin H. Simonds and John V. Grimaldi, Safety Management: Accident Cost and Control (Homewood IL: Richard D. Irwin, 1956), especially 387–389. 81. See, for example, Marland K. Strasser et al., Fundamentals of Safety Education (New York: Macmillan Company, 1964), 83–91; Norman R. F. Maier, Psychology in Industry, 3rd ed. (Boston: Houghton Miffl in Company, 1965), 569–587 (almost all negative); Studies in Personnel and Industrial Psychology, ed. Edwin A. Fleishman, 2nd ed. (Homewood IL: Dorsey Press, 1967), section 8; Lawrence Siegel, Industrial Psychology, 2nd ed. (Homewood IL: Richard D. Irwin, 1969), 296–299. 82. In the 1980 edition, Ernest J. McCormick and Daniel R. Ilgen, Industrial Psychology, 7th ed. (Englewood Cliffs NJ: Prentice-Hall, 1980), did not even list the subject in the index, although there were still a few allusions to the idea in the text. There were always exceptions, like Alton L. Thygerson, Accidents and Disasters: Causes and Countermeasures (Englewood Cliffs NJ: Prentice-Hall, 1977), 75–77. 83. As late as 1989, John V. Grimaldi and Rollin H. Simonds, Safety Management, 5th ed. (Homewood IL: Irwin, 1989), chap. 16, were still trying to stamp the idea out. 84. See, for example, “Accident Epidemiology,” Scientific American, August 1957, 58. M. S. Schulzinger, “A Closer Look at ‘Accident Proneness,’ ” National Safety News, June 1954, 32–33, 194–195. John MacIver, “Psychological Aspects of Accident Causation,” Industrial Medicine and Surgery, 28 (1959), 230–231. “Study Shows Tired Children Have Accidents,” Safety Education, March 1963, 29. 85. Marian Rolen, “Science Says—What Do We Really Know about Accident-

Notes to Pages 164–167

Proneness? ” Child Study, 33 (1955), 27–29. The author noted that there were, nevertheless, a few accident prone children who needed special watching. 86. Andrew Hale, “Who Is Accident Prone? ” Occupational Health, 24 (1972), 445. 87. Frank Longford, Lord Longford’s Book of Accidents & Misfortunes or—Hello Michael Foot, ed. Frank Longford (London: Sidgwick & Jackson, 1982). 88. Franziska Baumgarten-Tramer, “Typologie der Unfäller,” Zeitschrift für Präventivmedizin, 4 (1959), 228. 89. K. Biener, “Forschungsmethoden zur Unfallprävention,” Monatsschrift für Unfallheilkunde, 76 (1973), 166. See similarly M. Schachter, “Contribution à l’étude clinique et psychodiagnostique d’un groupe de poly-accidentés,” Ospedale psichiatrico, 4 (1952), 301. 90. “List of Illnesses and Disabilities Associated with Smoking,” Medical Journal of Australia, 2 (1975), supp. 3–4. 91. John D. MacArthur and Francis D. Moore, “Epidemiology of Burns: The Burn-Prone Patient,” JAMA, 231 (1975), 259–263; George L. Engel, “ ‘Psychogenic’ Pains and the Pain-Prone Patient,” American Journal of Medicine, 26 (1959), 899–918; Paul F. Wehrle et al., “The Repeater Problem in Accidental Poisoning,” Pediatrics, 27 (1961), 619; R. H. Blum, The Management of the Doctor-Patient Relationship (New York: McGraw-Hill, 1960). 92. Paul R. Schnurrenberger et al., “The Zoonosis-Prone Veterinarian,” Journal of the American Veterinary Medical Association, 15 (1978), 373–376. A zoonosis is a disease in a human that is directly transmitted from animals, and such events were an occupational hazard for veterinarians. 93. P. M. van Wulfften, abstracted in Aerospace Medicine, June 1960, A-88–89. 94. Richard J. Contrada and Richard D. Ashmore, Self, Social Identity, and Physical Health: Interdisciplinary Explorations (New York: Oxford University Press, 1999), xiii. P. Froggatt, “Short-Term Absence from Industry. III. The Inference of ‘Proneness’ and a Search for Causes,” British Journal of Industrial Medicine, 27 (1970), 298. 95. R. Sunderland, “Dying Young in Traffic,” Archives of Disease in Childhood, 59 (1984), 754–757. The terminology of standard analysis is taken up below. 96. Susan Powell, “An Accident-Prone Anglo Saxon,” Notes and Queries, n.s. 35 (1988), 154–157. A. J. P. Taylor, From Napoleon to the Second International: Essays on Nineteenth-Century Europe, ed. Chris Wrigley (New York: Allen Lane, 1993), xxiii.

ch a p t er 9 1. Lynette Shaw and Herbert S. Sichel, Accident Proneness: Research in the Occurrence, Causation, and Prevention of Road Accidents (Oxford: Pergamon Press, 1971), 72. 2. Historians, as I noted in the introduction and shall note again in the conclusion, have spent much time on the origins of ideas but relatively little on decline as such. Religious thought in the nineteenth century, for example, is usually conceptualized as the rise of secularism, although in recent historiography some scholars have written about the specific decline of religious ideas and thinking. Evidence of the phenomenon of just fading away is impressionistic and fugitive. For example, in Accident Prevention: The Role of Physicians and Public Health Workers, ed. Maxwell N. Halsey (New York: McGraw-Hill Book Company, 1961), the term “accident proneness” appeared in a very few places as a generic term meaning in-

289

290

Notes to Pages 167–169

cidence of accidents (1, 382) and was taken up technically only in the contribution on epidemiology of accidents. 3. Edward A. Suchman, “A Conceptual Analysis of the Accident Phenomenon,” in Herbert H. Jacobs et al., Behavioral Approaches to Accident Research (New York: Association for the Aid of Crippled Children, 1961), 44; Urie Bronfenbrenner, “Discussion,” in ibid., 139n. 4. A. G. Arbous and J. E. Kerrich, “The Phenomenon of Accident Proneness,” Industrial Medicine and Surgery, 22 (1953), 141–148; the editorial comment is on 141. “Accident proneness” ceased to be a separate entry in Psychological Abstracts in 1956 but continued as an important subhead under “accidents.” 5. While negative evidence is hard to accumulate, one can for example note that accident proneness did not appear in the British National Industrial Safety Study Conference Proceedings, 1969–1974. 6. Eric C. Wigglesworth, “The Fault Doctrine and Injury Control,” Journal of Trauma, 18 (1978), 792. John Langley, “A Study of the Relationship of Ninety Background, Developmental, Behavioural and Medical Factors to Childhood Accidents,” Australian Paediatric Journal, 16 (1980), 244–247. R. A. Alkov, “Psychological Stress, Health and Human Error,” Professional Safety, August 1981, 12. A suggestive sign of the fading-away process was how E. A. Sand, “Psychosocial Factors in Childhood and Adolescence,” in Accidents in Childhood and Adolescence: The Role of Research, ed. M. Manciaux and C. J. Romer (Geneva: World Health Organization, 1991), 82–83, in introducing briefly the idea of accident proneness in 1991 had to cite literature from the 1960s or, at the latest, 1976. 7. See Gerd Gigerenzer et al., The Empire of Chance: How Probability Changed Science and Everyday Life (Cambridge: Cambridge University Press, 1989); and The Road to Medical Statistics, ed. Eileen Magnello and Anne Hardy (Amsterdam: Rodopi, 2002). The special issue, “Statistics and Causation,” Synthese, 121 (nos. 1–2, 1999), implicitly describes technical and philosophical changes in the last half of the twentieth century. 8. See, for example, Ove Lundberg, On Random Processes and Their Application to Sickness and Accident Statistics (Uppsala: Almqvist & Wiksells, 1940, reprinted 1964); F. N. David, Probability Theory for Statistical Methods (Cambridge: Cambridge University Press, 1949), especially chap. 6. C. D. Kemp, “ ‘Accident Proneness’ and Discrete Distribution Theory,” in Random Counts in Biomedical and Social Sciences, ed. G. P. Patil (University Park: Pennsylvania State University Press, 1970), 41–65. 9. Percy W. Cobb, “The Limit of Usefulness of an Accident Rate as a Measure of Accident Proneness,” Journal of Applied Psychology, 24 (1940), 154–159. H. M. Johnson, “The Detection and Treatment of Accident-Prone Drivers,” Psychological Bulletin, 43 (1946), 489–532, especially 495–498. Validity and reliability were the two basic tests that psychologists used for test results in that period. 10. Major Greenwood, “Accident Proneness,” Biometrika, 37 (1950), 24–29. 11. Alexander Mintz and Milton L. Blum, “A Re-examination of the Accident Proneness Concept,” Journal of Applied Psychology, 33 (1949), 195–211; quotation from 210. 12. J. S. Maritz, “On the Validity of Inferences Drawn from the Fitting of Poisson and Negative Binomial Distributions to Observed Accident Data,” Psychological Bulletin, 47 (1950), 434–443. A. G. Arbous and J. E. Kerrich, “Accident Statistics

Notes to Pages 169–171

and the Concept of Accident-Proneness,” Biometrics, 7 (1951), 340–432, quotation from 424. 13. See, for example, Milton L. Blum and Alexander Mintz, “Correlation Versus Curve Fitting in Research on Accident Proneness: Reply to Maritz,” Psychological Bulletin. 48 (1951), 413–418; A. M. Adelstein, “Accident Proneness: A Criticism of the Concept Based upon an Analysis of Shunters’ Accidents,” Journal of the Royal Statistical Society, 115 (1952), 354–400; E. G. Chambers, quoted in Journal of the Royal Statistical Society, 115 (1952), 408. 14. Examples include Wilse B. Webb, “Some Relations Between Two Statistical Approaches to Accident Proneness,” Psychological Bulletin, 50 (1953), 133–139; Grace E. Bates, “Joint Distributions of Time Intervals for the Occurrence of Successive Accidents in a Generalized Polya Scheme,” Annals of Mathematical Statistics, 26 (1955), 705–720; Robert Fitzpatrick, “The Detection of Individual Differences in Accident Susceptibility,” Biometrics, 14 (1958), 50–68. Herbert H. Jacobs, “Mathematical, Psychological, and Engineering Aspects of Accident Phenomena. I. Operational Aspects of the Accident Proneness Concept,” Transactions of the New York Academy of Sciences, 18 (1956), 261–265. 15. See, for example, Herbert S. Sichel, “The Statistical Estimation of Individual Accident Liability,” Research Review, March 1965, 8–15; Frank A. Haight, “On the Effect of Removing Persons with N or More Accidents from an Accident Prone Population,” Biometrika, 52 (1965), 298–300; Erich Mittenecker, Methoden und Ergebnisse der psychologischen Unfallforschung (Vienna: Franz Deuticke, 1962). J. E. Ager and L. G. Norman, review of Cresswell and Froggatt, The Causation of Bus Driver Accidents, in British Journal of Industrial Medicine, 21 (1964), 248. 16. See, for example, Alexander Mintz, “Time Intervals Between Accidents,” Journal of Applied Psychology, 38 (1954), 401–406. Another concept in accident epidemiology, contagion, appeared later; see, for example, C. D. Kemp, “On a Contagious Distribution Suggested for Accident Data,” Biometrics, 23 (1967), 241–255; Evdokia Xekalaki, “The Univariate Generalized Waring Distribution in Relation to Accident Theory: Proneness, Spells or Contagion? ” Biometrics, 39 (1983), 887–895. The applicability of the concept of accident proneness to individuals was still being argued years later; Lindsay Prior, “Chance and Modernity: Accidents as a Public Health Problem,” in The Sociology of Health Promotion, ed. Robin Bunton, Sarah Nettleton, and Roger Burroughs (London: Routledge, 1995), 135–138. 17. The studies are summarized in P. Froggatt, “Short-Term Absence from Industry. III. The Inference of ‘Proneness’ and a Search for Causes,” British Journal of Industrial Medicine, 27 (1970), 309–310. 18. Robert S. McInnes and Dwight M. Bissell, “Accidental Poisoning as an Indication of High Accident Frequency,” Public Health Reports, 75 (1960), 853–858. Paul F. Wehrle et al., “The Repeater Problem in Accidental Poisoning,” Pediatrics, 27 (1961), 619. 19. See, for example, Poul Kølle-Jørgensen, Child Accidents: A Medico-Social Study of 4820 Accidents (Copenhagen: Munksgaard, 1971), who thought that child repeaters were those with inadequate caregiving and (223) “those very active children who always are present where things happen, or perhaps those who always see to it that something is happening where they are.” 20. M. W. Partington, “The Importance of Accident-Proneness in the Aetiology

291

292

Notes to Pages 171–174

of Head Injuries in Childhood,” Archives of Disease in Childhood, 35 (1960), 223. See, similarly, Reuel A. Stallones, “Epidemiology of Childhood Accidents in Two California Counties,” Public Health Reports, 76 (1961), 25–35. 21. For example, Stuart M. Finch, “Psychophysiologic Disorders in Children and Adolescents,” International Journal of Psychiatry in Medicine, 6 (1975), 223–224. 22. Dean I. Manheimer and Glen D. Mellinger, “Personality Characteristics of the Child Accident Repeater,” Child Development, 38 (1967), 492; Glen D. Mellinger, “Family Studies of Accidents,” in Childhood Accidental Injury Symposium Proceedings (Charlottesville: University of Virginia School of Medicine, 1966), 121–129. 23. Mintz and Blum, “A Re-examination of the Accident Proneness Concept.” This was only a slight shift from Farmer’s 1930s defi nition of accident liability. 24. Edwin E. Ghiselli and Clarence W. Brown, Personnel and Industrial Psychology (New York: McGraw-Hill Book Company, 1948), chap 14, especially 377. 25. Frank J. Harris, “Can Personality Tests Identify Accident-Prone Employees? ” Personnel Psychology, 3 (1950), 455–459, quotation from 459. And see the widely cited work of Robert Thorndike noted in the previous chapter. 26. See, for example, Arbous and Kerrich, “Accident Statistics and the Concept of Accident-Proneness”; Earl R. Rice, “The Human Element in Accidents,” Mississippi Valley Medical Journal, 74 (1952), 54–57, 65. 27. D. Russell David and Patricia A. Coiley, “Accident Proneness in MotorVehicle Drivers,” Ergonomics, 2 (1959), 239–246. Wilse B. Webb, “The Illusive Phenomena in Accident Proneness,” Public Health Reports, 70 (1955), 951–956, quotation from 955. Wilse B. Webb, “The Prediction of Aircraft Accidents from PilotCentered Measures,” Journal of Aviation Medicine, 27 (1956), 141–147. 28. See, for example, Leon G. Goldstein, “Human Variables in Traffic Accidents: A Digest of Research and Selected Bibliography” (Highway Research Board Bibliography 31, Washington: National Academy of Sciences-National Research Council, 1962). Review articles were incidentally a telling sign that elementary expositions had decreased substantially in favor of advanced discussions. 29. Peter Froggatt and James A. Smiley, “The Concept of Accident Proneness: A Review,” British Journal of Industrial Medicine, 21 (1964), 1–12, quotations from 6, 11. 30. Jean Surry, Industrial Accident Research: A Human Engineering Approach (Toronto: Labour Safety Council, Ontario Department of Labor, 1971), summarized much of this literature and offered still more criticism. 31. Linda Viney, “Accident Proneness: Some Psychological Research,” Medical Journal of Australia, 30 October 1971, 916–918. Allan Crawford, “Accident Proneness: An Unaffordable Philosophy,” ibid., 905–909. Michael Henderson, “The Accident-Prone Car Driver—Does He Exist? ” ibid., 909–912. Arthur A. Sampson, “The Myth of Accident Proneness,” ibid., 913–916. 32. William W. Allison, “ ‘Accident-Prone’ Theory Can Affect Safety Performance,” ASSE Journal, 14 (1969), 17–19. 33. Joseph Connolly, “Accident Proneness,” British Journal of Hospital Medicine, 26 (1981), 474. 34. Julian A. Waller, Injury Control: A Guide to Causes and Prevention of Trauma (Lexington MA: Lexington Books, 1985), chap. 28. Michael Guarnieri, “Landmarks in the History of Safety,” Journal of Safety Research, 23 (1992), 152–153. Other examples of attempting to kill off the idea include D. L. Mohr and D. I. Clemmer, “The

Notes to Pages 174–178

‘Accident Prone’ Worker: An Example from Heavy Industry,” Accident Analysis and Prevention, 20 (1988), 123–127; and Mark D. Rodgers and Robert E. Blanchard, Accident Proneness: A Research Review (Technical Report DOT/FAA/AM-93/9, Washington: Office of Aviation Medicine, Federal Aviation Administration, 1993), who referred to the concept as “folklore.” 35. E.g., J. D. Keehn, “Accident Tendency, Avoidance Learning and Perceptual Defence,” Australian Journal of Psychology, 13 (1961), 157–169. 36. Shaw and Sichel, Accident Proneness. 37. Frederick L. McGuire, “A Typology of Accident Proneness,” Behavioral Research in Highway Safety, 1 (1970), 26–32. 38. This point is obvious in the relatively very large volume of writings on traffic accidents and is made also explicitly in Sauli Häkkinen, Traffic Accidents and Driver Characteristics: A Statistical and Psychological Study (Helsinki: Institute of Technology, 1958), 7. 39. Ibid. Sauli Häkkinen, “Traffic Accidents and Professional Driver Characteristics: A Follow-Up Study,” Accident Analysis and Prevention, 11 (1979), 7–18, quotation from 18. 40. J. D. Keehn, “Factor Analysis of Reported Minor Personal Mishaps,” Journal of Applied Psychology, 43 (1959), 311–314, quotation from 313. This was therefore another confi rmation of a g, general factor for accident proneness as assumed by many earlier workers. Gerald H. Whitlock, Robert J. Clouse, and Woodford F. Spencer, “Predicting Accident Proneness,” Personnel Psychology, 16 (1963), 35–44. Joan S. Guilford, “Prediction of Accidents in a Standardized Home Environment,” Journal of Applied Psychology, 57 (1973), 306–313, explicitly answered many objections raised by previous research. 41. S. Craske, “A Study of the Relation Between Personality and Accident History,” British Journal of Medical Psychology, 41 (1968), 399–404. Melvin L. Selzer, Joseph E. Rogers, and Sue Kern, “Fatal Accidents: The Role of Psychopathology, Social Stress, and Acute Disturbance,” American Journal of Psychiatry, 124 (1968), 1028–1036. 42. Robert C. Behan, “Current Concepts of Accident Proneness,” Archives of Environmental Health, 13 (1966), 463. 43. Helen M. Nicholson, “Built-In Safety,” Home Safety Journal, April 1968, 3. 44. K. T. Johnstone, “The Identification of Accident Proneness,” Industrial Medicine and Surgery, 24 (1955), 293–295. 45. H. J. Eysenck, “The Personality of Drivers and Pedestrians,” Medicine, Science and the Law, 3 (1963), 416–423, quotation from 417. Daniel Kahneman and Rachel Ben-Ishai, “Relation of a Test of Attention to Road Accidents,” Journal of Applied Psychology, 58 (1973), 113–115. 46. Ernest Wolff, “Accident Proneness: A Serious Industrial Problem—What Can the Industrial Physician Do About It? ” Industrial Medicine and Surgery, 19 (1950), 419–426, quotation from 425. 47. E. M. Holroyd, “The Variation of Drivers’ Accident Rates Between Drivers and Over Time,” Accident Analysis and Prevention, 24 (1992), 275–305, quotation from 304. 48. See, for example, William N. McBain, “Arousal, Monotony, and Accidents in Line Driving,” Journal of Applied Psychology, 54 (1970), 509–519, in which the term

293

294

Notes to Pages 178–180

did not appear, but even in this extreme complexity of analysis, accident history was relevant. 49. J. D. Langley, P. A. Silva, and S. M. Williams, “Motor Coordination and Childhood Accidents,” Journal of Safety Research, 12 (1980), 175–178 50. J. W. Whitfield, “Individual Differences in Accident Susceptibility among Coal Miners,” British Journal of Industrial Medicine, 11 (1954), 126–139. 51. See, for example, David K. Trites, Albert L. Kubala, Jr., and Saul B. Sells, “Aircraft Accidents vs. Characteristics of Pilots,” Journal of Aviation Medicine, 26 (1955), 486–494. 52. William Haddon, Jr., Edward A. Suchman, and David Klein, Accident Research: Methods and Approaches (New York: Harper & Row, 1964). 53. See, for example, Robert Sass and Glen Crook, “Accident Proneness: Science or Non-Science? ” International Journal of Health Services, 11 (1981), 175–190; Frank McKenna, “Accident Proneness: A Conceptual Analysis,” Accident Analysis and Prevention, 15 (1983), 65–71. 54. I return to this subject below in connection with the major shift in safety strategy. 55. Donald H. Schuster, “Cognitive Accident-Avoidance Training for Beginning Drivers,” Journal of Applied Psychology, 63 (1978), 377–379. 56. William L. Mihal and Gerald V. Barrett, “Individual Differences in Perceptual Information Processing and Their Relation to Automobile Accident Involvement,” Journal of Applied Psychology, 61 (1976), 229–233. The shift in psychology is taken up directly in the next chapter. 57. See, for example, D. H. Schuster and J. P. Guilford, “The Psychometric Prediction of Problem Drivers,” Human Factors, 6 (1964), 393–421. 58. R. G. Green, “The Psychologist and Flying Accidents,” Aviation, Space, and Environmental Medicine, 48 (1977), 923. 59. Bryant W. Walker, “Occupational Safety in Industry,” in Occupational Medicine: Principles and Practical Applications, ed. Carl Zenz, 2nd ed. (Chicago: Year Book Medical Publishers, 1988), 3–4. 60. Ross A. McFarland, “Human Factors in Highway-Transport Safety,” in RoadUser Characteristics (Highway Research Board Bulletin 60, Washington: Highway Research Board, 1962), 36. 61. Sampson, “The Myth of Accident Proneness,” 914. 62. Ross A. McFarland and Roland C. Moore, “Human Factors in Highway Safety,” New England Journal of Medicine, 256 (1957), 793. 63. See, for example, Connolly, “Accident Proneness,” 470–481. An example of epidemiology in transition, using classical epidemiology but also the pursuit of groups and environments is Ross A. McFarland and Roland C. Moore, “The Epidemiology of Accidents,” in Accident Prevention: The Role of Physicians and Public Health Workers, ed. Maxwell N. Halsey (New York: McGraw-Hill Book Company, 1961), 17– 45. General background is in works cited elsewhere and particularly in Mark Parascandola, “Epidemiology in Transition: Tobacco and Lung Cancer in the 1950s,” in Body Counts: Medical Quantification in Historical and Sociological Perspective, ed. Gérard Jorland, Annick Opinel, and George Weisz (Montreal: McGill-Queen’s University Press, 2005), 226–248. 64. See, for example, Dalton A. Boot, “Epidemiology of Accidents,” in Trauma:

Notes to Pages 181–183

A Scientific Basis for Care, ed. Emin Kaya Alpar and Peter Gosling (London: Arnold, 1999), 1–15. Tom Christoffel and Susan Scavo Gallagher, Injury Prevention and Public Health: Practical Knowledge, Skills, and Strategies, 2nd ed. (Sudbury MA: Jones and Bartlett, 2006), especially 26, point out that blaming the individual is essentially a politically “conservative” strategy. Judith Green, Risk and Misfortune: A Social Construction of Accidents (London: UCL Press, 1997), especially chaps. 4 and 5, places the epidemiology of accidents in other contexts. 65. John E. Gordon, “The Epidemiology of Accidents,” American Journal of Public Health, 39 (1949), 504–515. This approach was heavily publicized in Ross A. McFarland, Roland C. Moore, and A. Bertrand Warren, Human Variables in Motor Vehicle Accidents: A Review of the Literature (Boston: Harvard School of Public Health, 1955). The original model was based on diseases caused by microbes. 66. Ross A. McFarland, “Epidemiologic Principles Applicable to the Study and Prevention of Child Accidents,” American Journal of Public Health, 45 (1955), 1302–1309. 67. As early as “Epidemiology of Accidents,” Lancet, 1 (1953), 730–731, the move away from the host was noted: “theories of host variability . . . should not blind us to the importance of environmental conditions.” An illustrative transitional statement is John Simpson, “An Epidemiological Approach to Road Accidents,” Practitioner, 188 (1962), 515–523. 68. Joint ILO/WHO Committee on Occupational Health, Epidemiology of WorkRelated Accidents (Technical Report Series 777, Geneva: World Health Organization, 1989), especially 37 69. Robert W. Jeffrey, “Risk Behaviors and Health: Contrasting Individual and Population Perspectives,” American Psychologist, 44 (1989), 1194–1202. See, for example, Julian A. Waller, “Injury as Disease,” Accident Analysis and Prevention, 19 (1987), 13–19. This author, like so many others, ended up advocating engineering solutions as well as social action. 70. See especially Julian A. Waller, “Reflections on a Half Century of Injury Control,” American Journal of Public Health, 84 (1994), 664–670, especially 664–665. 71. Guarnieri, “Landmarks in the History of Safety,” 151–158, quotation from 156. 72. See the literature of the 1990s in Injury Prevention and similar types of publications. Mary D. Overpeck and Elizabeth McLoughlin, “Did That Injury Happen on Purpose? Does Intent Really Matter? ” Injury Prevention, 5 (1999), 11–12, argued that improving the social situation of unfortunate people would reduce both intentional and unintentional injuries, and so making a distinction was irrelevant. Guarnieri, “Landmarks in the History of Safety.” Green, Risk and Misfortune, 129. 73. Henry M. Parrish et al., “Epidemiological Approach to Preventing School Accidents,” Journal of School Health, 37 (1967), 236–240. E. Maurice Backett, Domestic Accidents (Geneva: World Health Organization, 1965), 110. 74. Or “high-risk group” or, still another designation, “high-risk population.” Risk group is a concept derivative, obviously, from the idea of risk and risk factor, but discussions of risk, including histories, have focuses other than risk groups, and the term may not even, for example, be indexed in works on “risk.” Such sources as the OED are not useful in tracing the origins of the term, and current English-language medical dictionaries that I sampled do not include the phrase.

295

296

Notes to Pages 183–185

75. See, for example, Huiyun Xiang et al., “Obesity and Risk of Nonfatal Unintentional Injuries,” American Journal of Preventive Medicine, 29 (2005), 41–45. 76. Ulrich Beck, Risk Society: Towards a New Modernity, trans. Mark Ritter (London: Sage Publications, 1992); I return to this work in the conclusion. G. William Mercer, “Influences on Passenger Vehicle Casualty Accident Frequency and Severity: Unemployment, Driver Gender, Driver Age, Drinking Driving and Restraint Device Use,” Accident Analysis and Prevention, 19 (1987), 231–236. 77. Beck, Risk Society. Examples include Lewis H. Margolis, “Accidental Policy: An Analysis of the Problem of Unintended Injuries of Childhood,” American Journal of Orthopsychiatry, 53 (1983), 629–644; and Baruch Fischhoff, Lita Furby, and Robin Gregory, “Evaluating Voluntary Risks of Injury,” Accident Analysis and Prevention, 19 (1987), 51–62. Ramifications of thinking of a risk society are explored in The Risk Society and Beyond: Critical Issues for Social Theory, ed. Barbara Adam, Ulrich Beck, and Joost van Loon (London: Sage, 2000). 78. Green, Risk and Misfortune, especially 144–155. 79. Jonathan Simon, “The Emergence of a Risk Society: Insurance, Law, and the State,” Socialist Review, September–October 1987, 61–89. 80. “Risk” of course was a term that could be used in many ways, including as a synonym for danger, applied to a situation or environment. See, for example, Philip Powell et al., 2,000 Accidents: A Shop Floor Study of Their Causes Based on 42 Months’ Continuous Observation (London: National Institute of Industrial Psychology, 1971), especially 22–23, in which detailed analysis of “risks” dissipated any possible attention to personal accident proneness. 81. William G. Rothstein, Public Health and the Risk Factor: A History of an Uneven Medical Revolution (Rochester: University of Rochester Press, 2003). 82. See, for example, Allan Mazur, “Societal and Scientific Causes of the Historical Development of Risk Assessment,” in Society, Technology and Risk Assessment, ed. J. Conrad (London: Academic Press, 1980), 151–157, and other comments in that volume. 83. See, for example, P. Stewart Macauley, “Traffic’s Worst Offenders Constitute Only Small Group—Research on Accidents Intrigues Underwriters,” Weekly Underwriter, 132 (1935), 160–162; Randall R. Howard, “Psychology Tackles the Accident-Prone Driver,” Journal of American Insurance, January 1941, 11–13; William H. Seymour, “Man the Most Unpredictable Element,” Weekly Underwriter, 167 (1952), 1728. 84. C. A. Kulp, Casualty Insurance: An Analysis of Hazards, Policies, Companies and Rates (New York: The Ronald Press, 1942), 438. 85. Harry Dingman, Risk Appraisal (Cincinnati: National Underwriter Company, 1946), 58. 86. Mark R. Greene and Oscar N. Serbein, Risk Management: Text and Cases (Reston VA: Reston Publishing Company, 1978), 29, 92. See also, for example, “Accident Prone,” Casualty and Insurance Journal, November 1951, 19, a supposedly humorous item about a U.S. marine who suffered a series of mishaps and then tripped on his way out of the hospital. 87. See, for example, Louis I. Dublin, A Family of Thirty Million: The Story of the Metropolitan Life Insurance Company (New York: Metropolitan Life Insurance Company, 1943), who explains the reasoning, especially 193–194, 201–204, 407–408;

Notes to Pages 185–189

Jesse F. Gelders, “How Dangerous Is Your Job? ” Popular Science Monthly, February 1936, 34–35, 130; Arnold Wilson and Hermann Levy, Workmen’s Compensation (2 vols., Oxford: Oxford University Press, 1941), II, 31. Dingman, Risk Appraisal, 52– 53. Albert H. Mowbray, quoted in Proceedings of the Casualty Actuarial and Statistical Society of America, 2 (1915), 29, suggested that German mutual insurance on workers set a model of dividing industries into standard groups and grades. 88. See, for example, Dingman, Risk Appraisal, 772–773. 89. Simon, “The Emergence of a Risk Society,” generalizes the principle of social management by aggregation. 90. “Reward Plan for Safe Driver,” Safety Engineering, January 1938, 16; Mintz and Blum, “A Re-examination of the Accident Proneness Concept,” 207; N. L. Johnson and F. Garwood, “An Analysis of the Claim Records of a Motor Insurance Company,” Journal of the Institute of Actuaries, 83 (1957), 277–294; Franziska BaumgartenTramer, “Typologie der Unfäller,” Zeitschrift für Präventivmedizin, 4 (1959), 228. An exception was noted by Selzer, Rogers, and Kern, “Fatal Accidents: The Role of Psychopathology, Social Stress, and Acute Disturbance,” 1030. See, for example, William H. Brewster, “Rating the Youthful Driver,” Casualty and Surety Journal, May 1950, 9–13; and James M. Cahill, “Casualty Rating in 1950,” Casualty and Surety Journal, January 1950, 44–49. Age discrimination in life and other insurance (one could renew only to age 60, for example) had long been taken for granted. 91. Rothstein, Public Health and the Risk Factor. An example of the intersection between insurance risk and accident behavior is Paul Slovic, “The Psychology of Protective Behavior,” Journal of Safety Research, 10 (1978), 58–68. 92. Jeremiah A. Barondess et al., “Epidemic of Infectious Hepatitis: Some Notes on Delineation of High Risk Groups and Protection of Exposed Susceptibles by Gamma Globulin,” AMA Annals of Internal Medicine, 95 (1955), 633–645; the authors’ use of the term was casual (age, sex, “race,” geography, family membership) and not analytic. Jacob Tuckman and William F. Youngman, “Identifying Suicide Risk Groups Among Attempted Suicides,” Public Health Reports, 78 (1963), 763–766. Jacob Tuckman and William F. Youngman, “Suicide Risk Among Persons Attempting Suicide,” ibid., 585–587. 93. American Medical Association Council on Scientific Affairs, “AutomobileRelated Injuries: Components, Trends, Prevention,” JAMA, 249 (1983), 3216. 94. Stina Sandels, “An Overall View of Children in Traffic,” in Children, the Environment and Accidents (London: Pitman Medical, 1977), 20–27. 95. J. H. Sheldon, “On the Natural History of Falls in Old Age,” British Medical Journal, 10 December 1960, 1685–1690. 96. Garrett E. Snipes, “Accidents in the Elderly,” American Family Physician, July 1982, 117–122, especially 118. 97. For example, Sheldon, “On the Natural History of Falls in Old Age.” Accident analysis itself of course continued to generate controversy; see, for example, L. S. Robertson, “Groundless Attack on an Uncommon Man: William Haddon, Jr.,” Injury Prevention, 7 (2001), 260–262, and the article that set Robertson off, Malcolm Gladwell, “Wrong Turn: How the Fight to Make America’s Highways Safer Went Off Course,” New Yorker, 11 June 2001, 50–61. 98. Max D. Kossoris, “Absenteeism and Injury Experience of Older Workers,” Monthly Labor Review, July 1948, 16. See, for example, H. F. King and D. Speakman,

297

298

Notes to Pages 189–190

“Age and Industrial Accident Rates,” British Journal of Industrial Medicine, 10 (1953), 51–58. R. H. Van Zelst, “The Effect of Age and Experience upon Accident Rate,” Journal of Applied Psychology, 38 (1954), 313–317. Ross A. McFarland, Alfred L. Moseley, and M. Bruce Fisher, “Age and the Problems of Professional Truck Drivers in Highway Transportation,” Journal of Gerontology, 9 (1954), 338–348. Stephen Griew, “A Study of Accidents in Relation to Occupation and Age,” Ergonomics, 2 (1958), 17–23. 99. See, for example, Elizabeth Repath, “Home Accidents—A Socio-Medical Problem,” Community Health, July/August 1970, 12–17. The general problem of age as a category is addressed historically by John Macnicol, Age Discrimination: An Historical and Contemporary Analysis (Cambridge: Cambridge University Press, 2006). 100. Stanley H. Schuman et al., “Young Male Drivers,” JAMA, 200 (1967), 1026–1030, especially 1030. Stanley R. Mohler et al., “Aircraft Accidents by Older Persons,” Aerospace Medicine 40 (1969), 554–556, reported that in the already selected group of airplane pilots, age did not make any detectable difference in accident rate. 101. John Agate, “Accidents to Old People,” Community Health. 2 (1970), 29–35. And see, for example, Manuel Rodstein, “Effects of Aging on Injury Frequency and Severity,” Safety Maintenance, May 1967, 37–40, 44. Franz Josef Mathey of the University of Bonn, “Attitudes and Behavior of Elderly Pedestrians,” International Journal of Aging and Human Development, 17 (1983), 25–28, hoped to provide special training for the elderly. Accidents in the elderly were particularly conspicuous because physicians used accidents as indicators of physical problems and illnesses. Even one fall, for example, would raise questions, while multiple falls would attract intense attention. See, for example, “Accidents in the Aged May Be First Clue to Onset of Acute Disease,” Geriatrics, September 1969, 34–38; Trevor H. Howell, “Premonitory Falls,” Practitioner, 206 (1971), 666–667; N. A. Hood, “Diseases of the Central Nervous System,” British Medical Journal, 15 November 1975, 398–400. 102. David Sohn, “The Drug Abuser as a Safety Hazard,” ASSE Journal, November 1971, 12–16. Leonard Diller and Joseph Weinberg, “Evidence for Accident-Prone Behavior in Hemiplegic Patients,” Archives of Physical Medicine and Rehabilitation, 51 (1970), 358–363. Joseph R. Horton, “Study Determines Profi le of the Accident-Prone Hospital Employee,” Hospitals, 56 (1982), 105–107. The article was listed under the heading, “Risk Management.” A history of late twentieth-century growth of clinical risk management in Britain, following earlier American examples, is Kieran Walshe, “The Development of Clinical Risk Management,” in Clinical Risk Management: Enhancing Patient Safety, 2nd ed. (London: BMJ Books, 2001), 46–53. 103. John Edward P. Williams-Searle, “Broken Brothers and Sisters of Capital: Disability, Manliness, and Safety on the Rails, 1863–1908” (doctoral diss., University of Iowa, 2004), especially chap. 1. 104. This subject is explored in Martin Lengwiler, “Gendering of Industrial Risks: Conceptions and Practices of Male ‘Accident Proneness’, 1880–1920 (Britain, Germany and Switzerland),” unpublished paper generously furnished by the author. 105. See, for example, “Why Women Are, Or Are Not, Good Chauffeuses,” Outing, 44 (1904), 154–159. The inconclusive early fi ndings are noted above in chapter 4. A. R. Lauer, “Age and Sex in Relation to Accidents,” in Road-User Characteristics (Bul-

Notes to Pages 190–192

letin 60, Washington: Highway Research Board, 1952), 25–35, found sex an insignificant factor unless combined with age to highlight young male drivers. 106. A. Mayer, in a reported discussion of W. Hergt, “Über Unfallpsychologie,” Monatsschrift für Unfallheilkunde, 58, suppl. 48 (1955),63–73, especially 69–70. See the earlier A. Mayer, “Die Sonderstellung der Frau im Unfallgeschehen,” Münchener Medizinische Wochenschrift, 95 (1953), 1169–1172, 1198–1202. Fred L. Mannering, “Male/Female Driver Characteristics and Accident Risk: Some New Evidence,” Accident Analysis and Prevention, 25 (1993), 77–84. 107. Katharina Dawson, “Menstruation and Accidents,” British Medical Journal, 12 November 1960, 1425–1426. A. Mayer, “Menstruationspsyche und Unfall,” Monatsschrift für Unfallheilkunde und Versicherungsmedizin, 62 (1962), 337–338. H.-J. Weist, “Unfall und Menstruationszyklus,” Zeitschrift für die gesamte Hygiene, 12 (1966), 408–420. 108. Ian P. F. Owens, “Sex Differences in Mortality Rates,” Science, 297 (2002), 2008. Mannering, “Male/Female Driver Characteristics and Accident Risk,” found that males’ profi les changed in a way females’ did not. 109. Scott Spreat and Joan C. Baker-Potts, “Patterns of Injury in Institutionalized Mentally Retarded Residents,” Mental Retardation, 21 (1983), 29. 110. Julian A. Waller, “Chronic Medical Conditions and Traffic Safety: Review of the California Experience,” New England Journal of Medicine, 273 (1965), 1413–1420. Waller has appeared previously as a major figure who wrote—mostly negatively—about the idea of accident proneness. Risk groups could of course be broken down into subgroups on the basis of risk for different kinds of accidents; see, for example, Robert C. Eelkema et al., “A Statistical Study on the Relationship Between Mental Illness and Traffic Accidents—A Pilot Study,” American Journal of Public Health, 60 (1970), 459–468. 111. “Editor’s Note,” Safety Maintenance, May 1967, 37. W. Ferguson Anderson, “Getting Old,” Care in the Home, January 1973, 4–7. Harvey Catchen, “Repeaters: Inpatient Accidents Among the Hospitalized Elderly,” Gerontologist, 23 (1983), 273– 276, quotation from 275. 112. Michael A. Gebers and Raymond C. Peck, “The Identification of High-Risk Older Drivers Through Age-Mediated Point Systems,” Journal of Safety Research, 23 (1992), 81–93. 113. Denise Kendrick and Patricia Marsh, “Injury Prevention Programmes in Primary Care: A High Risk Group or a Whole Population Approach? ” Injury Prevention, 3 (1997), 170–175, especially 174. 114. See, for example, Robert A. Dershewitz and Edward R. Christophersen, “Childhood Household Safety,” American Journal of Diseases of Children, 138 (1984), 85–88; J. F. Kraus, “Effectiveness of Measures to Prevent Unintentional Deaths of Infants and Children from Suffocation and Strangulation,” Public Health Reports, 100 (1985), 231–240. 115. J. Richard Udry, “Why Are Males Injured More than Females? ” Injury Prevention, 4 (1998), 95. 116. Indeed, epidemiology could lead experts to thinking about more general factors in accident circumstances than risk groups; see, for example, Katherine K. Christoffel et al., “Childhood Pedestrian Injury: A Pilot Study Concerning Etiology,” Accident Analysis and Prevention, 18 (1986), 25–35.

299

300

Notes to Pages 194–197

ch a p t er 10 1. See, for example, Mark Aldrich, Safety First: Technology, Labor, and Business in the Building of American Work Safety, 1870–1939 (Baltimore: Johns Hopkins University Press, 1997). Barbara Young Welke, Recasting American Liberty: Gender, Race, Law, and the Railroad Revolution, 1865–1920 (New York: Cambridge University Press, 2001), provides another insightful account. 2. Kenton Kroker, “Washouts: Electroencephalography, Epilepsy and Emotions in the Selection of American Aviators During the Second World War,” in Instrumental in War: Science, Research, and Instruments Between Knowledge and the World, ed. Steven A. Walton (Leiden: Brill, 2005), 301–338. 3. Jack A. Holmes, “Industrial Accident Proneness,” Personnel Psychology, 2 (1949), 374. 4. I have not found public statements from labor or socialist groups reflecting any opposition to or, indeed, any interest in, accident proneness. Labor groups of course had safety on the job as a major organizational goal. Workers as individuals, however, often resisted safety rules. Miners, for example, at one point went on strike to avoid wearing a new lamp that was safer but heavier, according to Harold Ernest Burtt, Psychology and Industrial Efficiency (New York: D. Appleton and Company, 1931), 337. But in the large industrial organizations that called forth large labor organizations, as Aldrich, Safety First, 7, notes, “By the 1930s employee involvement in work safety was becoming less important, . . . for safety was becoming increasingly proactive—a matter of factory layout, machine design, hazard assessment, and job evaluation, all of which were the duties of engineers and other managers.” 5. See, for example, Kurt Seesemann, “Psychotechnical Aptitude Tests in the Service of Accident Prevention,” Industrial Safety Survey, 5 (1929), 160. E. Farmer, E. G. Chambers, and F. J. Kirk, Tests for Accident Proneness (Industrial Health Research Board, Report no. 68, London: His Majesty’s Stationery Office, 1933), 27. 6. D. Russell Davis, “The Disorder of Skill Responsible for Accidents,” Quarterly Journal of Experimental Psychology, 1 (1949), 138–139. The investigation grew out of a study of pilot error. 7. E. Richard Weinerman, “Accident-Proneness: A Critique,” American Journal of Public Health, 39 (1949), 1527–1530. Fleming James, Jr., and John J. Dickinson, “Accident Proneness and Accident Law,” Harvard Law Review, 63 (1950), 790–791, pointed out that the courts even then were indulgent of “the handicapped.” 8. Examples include Wilse B. Webb, “The Prediction of Aircraft Accidents from Pilot-Centered Measures,” Journal of Aviation Medicine, April 1956, 141–147; Frank E. Maloney, “Accident Proneness in the Automobile Field,” Journal of the Florida Medical Association, 47 (1960), 31–35. 9. O. Steinwall, “Epilepsy and the Driver’s License,” Acta Psychiatrica et Neurologica Scandinavica, 36 suppl. (1960), 179–186, quotation from 184. Allan Krumholz, “Driving and Epilepsy: A Historical Perspective and Review of Current Regulations,” Epilepsia 35 (1994), 668–674. 10. Injury in America: A Continuing Public Health Problem (Washington: National Academy Press, 1985), 6. J. E. Uhlaner and A. J. Drucker, “Selection Tests— Dubious Aid in Driver Licensing,” in Road-User Characteristics 1963 (Highway Research Record, no. 84, Washington: Highway Research Board, National Academy of Sciences—National Research Council, 1965), 42.

Notes to Pages 197–199

11. E. Hauer et al., “Estimating the Accident Potential of an Ontario Driver,” Accident Analysis and Prevention, 23 (1991), 133–152. Steven Whitman and John L. McNight, “Ideology and Injury Prevention,” International Journal of Health Services, 15 (1985), 35–46. The point was made earlier, in a different formulation, by Erich Mittenecker, Methoden und Ergebnisse der psychologischen Unfallforschung (Vienna: Franz Deuticke, 1962), iii–iv. 12. See, for example, Michel Roche, L’Homme et la route (Paris: Hachette, 1961), especially 226–228. 13. See, for example, S. K. Bose, “A Study of Accident-Prone Personnel Amongst Tram and Bus Drivers in Calcutta,” Indian Journal of Medical Research, 40 (1952), 569– 583. Material on airplane pilots is overwhelming in quantity for the decades after World War ii. 14. Some of the uncertainty of conceptualization is explored in Volker Roelcke, “Programm und Praxis der psychiatrischen Genetik an der Deutschen Forschunganstalt für Psychiatrie unter Ernst Rüdin: Zum Verhältnis von Wissenschaft, Politik und Rasse-Begriff vor und nach 1933,” Medizinhistorisches Journal, 37 (2002), 21–55. 15. Clarence W. Brown and Edwin E. Ghiselli, “Accident Proneness Among Street Car Motormen and Motor Coach Operators,” Journal of Applied Psychology, 32 (1948), 20–23, for example, early identified the assumption, but their fi ndings were negative. 16. Norman Tabachnick et al., “Comparative Psychiatric Study of Accidental and Suicidal Death,” Archives of General Psychiatry, 14 (1966), 60–68. 17. See especially Franziska Baumgarten-Tramer, “Typologie der Unfäller,” Zeitschrift für Präventivmedizin, 4 (1959), 228–237. A. S. Henderson, C. L. Williams, and Janet M. Mills, “An Epidemiological Study of Serious Traffic Offenders,” in Cultures in Collision, ed. Issy Pilowsky (Adelaide: Australian National Association for Mental Health, 1975), 158–162. Paul L. Brown and Ralph F. Berdie, “Driver Behavior and Scores on the MMPI,” Journal of Applied Psychology, 44 (1960), 21. 18. Joseph T. Kunce and Blayne Brewer, “Neuropsychiatric Patients, Accident Proneness, and Interest Patterns,” Journal of Psychology, 63 (1966), 287–290. J. Bristow, B. Kirwan, and D. H. Taylor, “Cognition and Affect in Measures of Driving Style,” Ergonomics, 25 (1982), 939–940. Another formulation, in terms of personality structure, was Johannes Grunert, “Strukturspezifische Unfallkeime des Kraftfahrers,” Zeitschrift für experimentelle und angewandte Psychologie, 8 (1961), 42–53. See below. 19. See the post–World War ii formulations in E. Grafe, “Anlage und Unfall,” Deutsche medizinische Wochenschrift, 18 (1954), 1245–1249, and Johannes Grunert, “Unfallkeim—Unfalldisposition—Unfall: Entwurf einer psychodynamischen Unfalltheorie,” Zeitschrift für experimentelle und angewandte Psychologie, 8 (1961), 519–539. 20. M. Roche, “Les Accidents de la circulation: recherches sur leurs causes, critique de la notion de ‘prédisposition,’ ” Revue de psychologie appliquée, 2 (1952), 263–271, went beyond Lahy in denying a predetermining force—predisposition— in accident repeaters. 21. J. Zurfluh, Accidents du travail et formation-sécurité (Paris: Dunod, 1957), especially 103. M. W. Partington, “The Importance of Accident-Proneness in the Aetiology of Head Injuries in Childhood,” Archives of Disease in Childhood, 35 (1960), 222.

301

302

Notes to Pages 199–201

R. Bramley-Harker, “Prevention of Accidents,” in Modern Trends in Occupational Health, ed. R. S. F. Schilling (London: Butterworth, 1960), 168. 22. John E. Gordon, “The Epidemiology of Accidents,” American Journal of Public Health, 39 (1949), 508. J. W. Whitfield, “Individual Differences in Accident Susceptibility among Coal Miners,” British Journal of Industrial Medicine, 11 (1954), 131. Adam P. Matheny, Jr., Anne M. Brown, and Ronald S. Wilson, “Behavioral Antecedents of Accidental Injuries in Early Childhood: A Study of Twins,” Journal of Pediatrics, 79 (1971), 122–124. 23. Charles J. Graham et al., ‘Left-Handedness as a Risk Factor for Unintentional Injury in Children,” Pediatrics, 92 (1993), 823–826. 24. Maria Schorn, “Untersuchungen über die Handgeschicklichheit,” Zeitschrift für Psychology, 112 (1929), 328. See the apparent independent conclusion of J. Kastner, “Über die Berufseignung Linkshänder,” Psychotechnische Zeitschrift, 10 (1935), 105. 25. Alkistis Skalkidou et al., “Risk of Upper Limb Injury in Left Handed Children: A Study in Greece,” Injury Prevention, 5 (1999), 68–71. Peter Wright, “Is Saying NO to ‘Accident Proneness’ Throwing the Baby Out with the Bathwater? ” Injury Prevention, 3 (1997), 79–81, used left-handers as an example of why one should retain the idea of accident proneness. 26. “Accident-Proneness,” The Times, 12 November 1935. See, for example, Ross A. McFarland and Alfred L. Moseley, Human Factors in Highway Transport Safety (Boston: Harvard School of Public Health, 1954). 27. Lowell S. Selling, “Psychiatry in Industrial Accidents,” Industrial Medicine, 13 (1944), 510; on 508, he referred to the customary “culling . . . along psychophysical lines by trial and error.” Frank Howard Richardson, “Are You Accident Prone? ” Today’s Health, April 1951, 15, 50. Perhaps a better parallel is with developmentally disabled persons, who for much of the nineteenth century were not identified in terms of a social problem, but as individuals. At the beginning of the twentieth century, however, they were, like accident prone people, suddenly made into a social problem and labeled as well as identified as deviant (a parallel suggested to me by Deborah Thom). 28. John L. Benton et al., “Auto Driver Fitness: An Evaluation of Useful Criteria,” Journal of the American Medical Association, 176 (1961), 419–423, especially 421. Thomas W. Jenkins, “Who Will Be Your Injury-Repeaters? ” Safety Maintenance, April 1961, 9. Marland K. Strasser et al., Fundamentals of Safety Education (New York: Macmillan Company, 1964), 89. 29. Edward A. Suchman, “Accidents and Social Deviance,” Journal of Health and Social Behavior, 11 (1970), 4–15, especially 5, reviewed the literature on the connection. See, for example, Michael Henderson, “The Accident-Prone Car Driver—Does He Exist? ” Medical Journal of Australia, 30 October 1971, 912. 30. Suchman, “Accidents and Social Deviance,” 10. Years later, a Swedish expert, Urban Kjellén, “The Deviation Concept in Occupational Accident Control,” Accident Analysis and Prevention, 16 (1984), 289–323, translated into systems theory the idea of accident incidence as deviation. 31. George P. Bisgeier, “Ghetto Hiring,” Archives of Environmental Health, 30 (1975), 440–441. An earlier study was “Good Neighborhoods Have Low Accident Rates,” American City, August 1959, 23.

Notes to Pages 202–205

32. Jack Block and Jeanne H. Block, “Venturing a 30-Year Longitudinal Study,” American Psychologist, 61 (2006), 316–317, from their eyewitness viewpoint offer a striking summary of the change. See Bernard J. Baars, The Cognitive Revolution in Psychology (New York: Guilford Press, 1986); Robert R. Hoffman, “American Cognitive Psychology,” in A Pictorial History of Psychology, ed. Wolfgang G. Bringmann et al. (Chicago: Quintessence Publishing, 1997), 594–598. Recent scholarship appears in several essays in The Life Cycle of Psychological Ideas: Understanding Prominence and the Dynamics of Intellectual Change, ed. Thomas C. Dalton and Rand B. Evans (New York: Kluwer Academic/Plenum Publishers, 2004). A comprehensive critical review is John D. Greenwood, “Understanding the ‘Cognitive Revolution’ in Psychology,” Journal of the History of the Behavioral Sciences, 35 (1999), 1–22. 33. For example, David M. DeJoy, “Attributional Processes and Hazard Control Management in Industry,” Journal of Safety Research, 16 (1985), 61–71. Two Finnish safety experts, J. T. Saari and J. Lahtela, “Job Enrichment: Cause of Increased Accidents? ” Professional Safety, December 1979, 28–32, suggested that, ironically, reducing monotony and routine instead of reducing accidents by keeping workers interested in fact led to more accidents because routines were varied. 34. Joseph E. Barmack, “Methodological Problems in the Design of Motor Vehicle Accident Research,” American Journal of Public Health, 52 (1962), 1868. 35. H. A. Paul, “Unfälle in epidemiologischer Sicht,” Münchener medizinische Wochenschrift, 37 (1967), 2003–2008, on 2006 tried to save the factor of accident proneness by calling “Unfallneigung” the last link in a chain of causal factors. 36. Early examples include Brown and Ghiselli, “Accident Proneness Among Street Car Motormen”; Lawrence LeShan and Jim B. Brame, “A Note on Techniques in the Investigation of Accident Prone Behavior,” Journal of Applied Psychology, 37 (1953), 79–81. 37. Philip Powell et al., 2,000 Accidents: A Shop Floor Study of Their Causes Based on 42 Months’ Continuous Observation (London: National Institute of Industrial Psychology, 1971), especially 22–23. 38. John C. Larson, “Industrial Accident Research: How Can It Be Improved? ” Personnel, 32 (1955), 135–138. Gerald H. Whitlock et al., “The Relation of Accident Proneness and the Tendency to Report Injuries,” Personnel Psychology, 16 (1963), 163–169. 39. See, for example, as early as Henry A. Hepburn, “Some Theoretical Aspects of Industrial Accident Causation—the Accident Sequence,” Occupational Safety and Health, July–September 1953, 113–118; George V. Nichols, Jr., “An Exploratory Study of Some of the Psychological Factors Related to Safety,” ASSE Journal, 17 (1972), 12– 18. Ernest J. McCormick and Mark S. Sanders, Human Factors in Engineering and Design, 5th ed. (New York: McGraw-Hill Book Company, 1982), 577–581. 40. Joan S. Guilford, “Prediction of Accidents in a Standardized Home Environment,” Journal of Applied Psychology, 57 (1973), 306–313. George V. Barrett and Carl L. Thornton, “Relationship Between Perceptual Style and Driver Reaction to an Emergency Situation,” Journal of Applied Psychology, 52 (1968), 169–176, quotation from 173. 41. Edwin Lipinski et al., “Occupational Accidents: Some Psychosocial Factors in the Accident Syndrome,” Canadian Psychiatric Association Journal, 10 (1965), 300. 42. See, for example, Leon Brody, “Methodology and Patterns of Research in

303

304

Notes to Pages 205–207

Industrial Accidents,” Annals of the New York Academy of Sciences, 107 (1963), 659–663; K. H. Link, “Zum Kapitel: Ursachen und Verhütung von Autounfällen,” Monatsschrift für Unfallheilkunde, 71 (1968), 263–265. 43. See, for example, Ernest Levens, “Some Current Trends in Safety Management,” ASSE Journal, 19 (1974), 21–26. 44. Carl Graf Hoyos, Psychologische Unfall- und Sicherheitsforschung (Stuttgart: Verlag W. Kohlhammer, 1980), chap. 6. 45. Eric C. Wigglesworth, “The Fault Doctrine and Injury Control,” Journal of Trauma, 18 (1978), 792, and, for example, Edward J. Bernacki, “Accident Proneness or Accident Liability: Which Model for Industry? ” Connecticut Medicine, 40 (1976), 535–538; John Langley and Phil A. Silva, “A Study of the Relationship of Ninety Background, Developmental, Behavioural and Medical Factors to Childhood Accidents,” Australian Paediatric Journal, 16 (1980), 244–247, 247: “Given the lack of utility and the inhibitory nature of attempts to identify those more liable to accidents, future child accident prevention research should focus on identifying unsafe aspects of the environment with a view to removing or modifying them so that all children may benefit.” See below. 46. Very widely read and cited was Charles Perrow, Normal Accidents: Living with High-Risk Technologies (New York: Basic Books, 1984). See, for example, James T. Reason, “Understanding Adverse Events: The Human Factor,” in Clinical Risk Management: Enhancing Patient Safety, ed. Charles Vincent, 2nd ed. (London: BMJ Books, 2001), 9–30. Frank P. McKenna, “The Human Factor in Driving Accidents: An Overview of Approaches and Problems,” Ergonomics, 25 (1982), 867–877, for example, reviewed human factor approaches, starting with accident proneness, and concluded that accidents occurred in complex systems and urged the further study of human error. 47. E. Hollnagel, “Accident Analysis and ‘Human Error,’ ” in International Encyclopedia of Ergonomics and Human Factors, 2nd ed. (3 vols., London: Taylor & Francis, 2001), II, 1889–1892, quotation from 1889. John Dennis Chasse and David A. LeSourd, “Rational Decisions and Occupational Health: A Critical View,” International Journal of Health Services, 14 (1984), 433–435, for example, launched a slashing attack on economists’ rational decision theory in part on the basis that it, like accident proneness, blamed the victim. 48. See, for example, Donald E. Broadbent, Margaret H. P. Broadbent, and Julian L. Jones, “Performance Correlates of Self-Reported Cognitive Failure and of Obsessionality,” British Journal of Clinical Psychology, 25 (1986), 285–289. James Reason, “Stress and Cognitive Failure,” in Handbook of Life Stress, Cognition and Health, ed. Shirley Fisher and James Reason (New York: John Wiley & Sons, 1988), 405–421. James Reason, Human Error (Cambridge: Cambridge University Press, 1990), was his more defi nitive statement. 49. See especially the crucial report, John W. Senders and Neville P. Moray, Human Error: Cause, Prediction, and Reduction (Hillsdale NJ: Lawrence Erlbaum Associates, 1991), especially 3–10; on 104 they distinguish error from accident; errors may have no consequence, and accidents can happen without human agency. 50. To Err Is Human: Building a Safer Health System, ed. Linda T. Kohn, Janet M. Corrigan, and Molla S. Donaldson (Washington: National Academy Press, 2000), especially 58, 65.

Notes to Pages 207–210

51. Reason, Human Error, 198–199. 52. Ibid., especially 239. Reason, “Understanding Adverse Events: The Human Factor,” especially 11. Lloyd J. Dumas, Lethal Arrogance: Human Fallibility and Dangerous Technologies (New York: St. Martin’s Press, 1999), especially chaps. 6 and 7, took up the consequences of human error in general, including error from illness, drug use, and stress, but did not utilize the concept of one person’s making more than an expected or normal number of errors. In his view, just one error could be lethal. 53. Jobst Conrad, “Society and Risk Assessment: An Attempt at Interpretation,” in Society, Technology and Risk Assessment, ed. J. Conrad (London: Academic Press, 1980), 241. Branden B. Johnson, “Tales of Woe: A Literature Survey,” in Perilous Progress: Managing the Hazards of Technology, ed. Robert W. Kates, Christoph Hohenemser, and Jeanne X. Kasperson (Boulder: Westview Press, 1985), 291–293. 54. Historical perspectives can be found in David Meister, The History of Human Factors and Ergonomics (Mahwah NJ: Lawrence Erlbaum Associates, 1999), and Patrick Waterson and Reg Sell, “Recurrent Themes and Developments in the History of the Ergonomics Society,” Ergonomics, 49 (2006), 743–799. In Britain, the two movements tended to merge; see The MRC Applied Psychology Unit, ed. L. A. Reynolds and E. M. Tansey (Wellcome Witnesses to Twentieth Century Medicine, volume 16, London: Wellcome Trust Centre for the History of Medicine at University College, 2003), especially 31–41. 55. Waterson and Sell, “Recurrent Themes and Developments,” 784–786. See, for example, the classic text, K. F. H. Murrell, Ergonomics: Man in His Working Environment (London: Chapman and Hall, 1965). Jan Noyes, Designing for Humans (Hove: Psychology Press, 2001), xi. 56. Noyes, Designing for Humans, 6–8. Alphonse Chapanis, Wendell R. Garner, and Clifford T. Morgan, Applied Experimental Psychology: Human Factors in Engineering Design (New York: John Wiley & Sons, 1949), 2–5, 7–9, 371–377; quotation from 7. A pioneer, Harry L. Wolbers, quoted in Meister, The History of Human Factors, 231, recalled that in the early days “the idea of designing jobs for people rather than selecting people for jobs was still relatively new.” Most of the pioneers, he continued, came out of industrial psychology and had been working on selecting and training workers. Meister offers a detailed history of the human factors field. See in general Sharolyn Converse Lane, “A Historical View of Human Factors in the United States,” in Historical Perspectives in Industrial and Organizational Psychology, ed. Laura L. Koppes (Mahwah NJ: Lawrence Erlbaum Associates, 2007), 243–263. 57. Richard T. Booth, “Too Important to Be Left to the Engineers,” Professional Safety, November 1980, 35–40. 58. D. Russell Davis, “Human Error and Transport Accidents,” Ergonomics, 2 (1958), 24–33. 59. See, for example, Alex Kirlik, “Cognitive Engineering: Toward a Workable Concept of Mind,” in Human-Technology Interaction: Methods and Models for Cognitive Engineering and Human-Computer Interaction, ed. Alex Kirlik (New York: Oxford University Press, 2006), 3–9, as well as Meister, The History of Human Factors. I am not discussing at length the physical anthropology and “human engineering” element in human factors and ergonomics; see, for example, Albert Damon, Howard W. Stoudt, and Ross A. McFarland, The Human Body in Equipment Design

305

306

Notes to Pages 210–212

(Cambridge MA: Harvard University Press, 1966), dealing with “biomechanical features of man-machine integration” (1–3). 60. See National Institute of Industrial Psychology, Annual Report, especially 1969 and 1972, 7. 61. See Meister, The History of Human Factors, especially 6–8, 202–205; Noyes, Designing for Humans, especially chap. 7. Late twentieth-century theory, including chaos theory applied to error and systems, is summarized by Jop Groeneweg, Controlling the Controllable: The Management of Safety, 4th ed. (Leiden: DSWO Press, Leiden University, 1988). 62. See, for example, American Society of Safety Engineers publications over many years in the early twentieth century. 63. Jameson Michael Wetmore, “Systems of Restraint: Redistributing Responsibilities for Automobile Safety in the United States Since the 1960s” (doctoral diss., Cornell University, 2003), provides an illuminating, detailed case study. 64. See, for example, Christoph Hohenemser and Thomas Bick, “Automobile Accidents,” in Perilous Progress, 179–197; Wetmore, “Systems of Restraint”; and the detailed story of how accident analysis did, and did not, lead to engineering auto safety in Joel W. Eastman, Styling vs. Safety: The American Automobile Industry and the Development of Automotive Safety, 1900–1966 (Lanham MD: University Press of America, 1984). 65. See, for example, Thomas H. MacDonald, “Driver Behavior—Key to Safe Highway Design,” SAE Journal, 1949, 33–43. “How Safe Is Junior? ” Standardization, February 1952, 63. “Childhood Drug Poisonings Reduced,” JAMA, 206 (1968), 1428. 66. See, for example, Earl R. Rice, “The Human Element in Accidents,” Mississippi Valley Medical Journal, March 1952, 56, who incidentally found accident proneness too confused an idea to substitute for the tried and true three E’s. And as Robert J. Schreiber, “The Development of Engineering Techniques for the Evaluation of Safety Programs,” Transactions of the New York Academy of Sciences, 18 (1956), 266–271, observed, education and enforcement could be considered engineering when applied to making workers part of a system. 67. Alfred J. Cardall, “Psychological Factors in Accident Prevention,” Personnel Journal, 26 (1948), 288. George Kienzle, “Will He Have an Accident Today? ” Safety Engineering, November 1948, 9, 52–53, 56–57, likewise deprecated engineering approaches. 68. Reuel A. Stallones and Leslie Corsa, “Epidemiology of Childhood Accidents in Two California Counties,” Public Health Reports, 76 (1961), 25–36, quotation from 32. Ralph L. Barnett and Dennis B. Brickman, “Safety Hierarchy,” Journal of Safety Research, 17 (1986), 49–55, showed that management and knowledge of human factors as well as engineering were necessary for safety, citing for example police revolvers accidentally discharged. 69. See, for example, Bernard T. Parker, “Keeping the Accident-Prone Worker Safe,” Mill & Factory, December 1948, 123–125, who reported a 23% reduction in accidents after a four-year vision testing program; Fred W. Jobe, “An Analysis of Visual Performance in Relation to Safety,” American Journal of Optometry, 25 (1948), 107–116; Joseph Tiffi n, B. T. Parker, and R. W. Habersat, “Visual Performance and Accident Frequency,” Journal of Applied Psychology, 33 (1949), 499–502; H. D. Darcus, “Faulty Vision and the Driver,” Traffic Quarterly, 7 (1953), 198–211. Eric Bateman,

Notes to Pages 212–215

“Aspects of Binocular Vision and Their Effect on Accident Proneness,” Optometric Weekly, 53 (1962), 2019–2023. 70. Gerald J. S. Wilde and John F. Stinson, “The Monitoring of Vigilance in Locomotive Engineers,” Accident Analysis and Prevention, 15 (1983), 87–93. 71. “How Far Can the Accident Rate Be Cut? ” Engineering, 2 February 1962, 172. 72. Ross A. McFarland, “The Epidemiology of Motor Vehicle Accidents,” Journal of the American Medical Association, 180 (1962), 297. 73. Harvey Catchen, “Repeaters: Inpatient Accidents Among the Hospitalized Elderly,” Gerontologist, 23 (1983), 273–276. N. A. Hood, “Diseases of the Central Nervous System,” British Medical Journal, 4 (15 November 1975), 400. E. Maurice Backett, Domestic Accidents (Geneva: World Health Organization, 1965), 110. McFarland, “The Epidemiology of Motor Vehicle Accidents,” 299. 74. William Haddon, Jr., “The Changing Approach to the Epidemiology, Prevention, and Amelioration of Trauma: The Transition to Approaches Etiologically Rather than Descriptively Based,” American Journal of Public Health, 58 (1968), 1431– 1438, quotation from1436. The idea of individual responsibility for injuries never died, and the controversy over bad effects of engineering persisted (seat belts caused people to take more risks, for example); see D. C. Thompson, R. S. Thompson, and F. P. Rivara, “Risk Compensation Theory Should Be Subject to Systematic Reviews of the Scientific Evidence,” Injury Prevention, 7 (2001), 86–91. 75. Theodore Hatch, “Human-Factors Engineering and Safety Research,” Journal of Occupational Medicine, 4 (1962), 2–3. 76. William W. Allison, “ ‘Accident-Prone’ Theory Can Affect Safety Performance,” ASSE Journal, 14 (1969), 17–19. 77. Gordon M. Sessions, Traffic Devices: Historical Aspects Thereof (Washington: The Institute of Traffic Engineers, 1971), chap. 7. Or see Lynn D. Larson et al., “Reduction of Police Vehicle Accidents Through Mechanically Aided Supervision,” Journal of Applied Behavior Analysis, 13 (1980), 571–581. 78. Children, the Environment and Accidents, ed. R. H. Jackson (Tunbridge: Pitman Medical, 1977); the example is from Jackson, “Setting the Scene,” 5. The psychology in the essays was all illustrative of human factors in design. 79. See, for example, “Achievements in Public Health, 1900–1999, Motor-Vehicle Safety: A 20th Century Public Health Achievement,” MMWR, 48 (1999), 369–374. 80. See, for example, William H. Forster, “The Systems Approach: A Tool for Reducing Vehicle/Highway Accidents,” Archives of Environmental Health, 13 (1966), 537–542. Insofar as the system involved using technology to avoid injuries, the technology could of course introduce new conditions in which humans could make errors and cause injury and damage; see, for example, John Gosbee and Laura Lin, “The Role of Human Factors Engineering in Medical Device and Medical System Errors,” in Clinical Risk Management, ed. Vincent, 304–306. 81. Murray Blumenthal, “Dimensions of the Traffic Safety Problem,” Traffic Safety Research Review, March 1968, 7–10, quotations from 10. Or see D. Keith Denton, “The Unsafe Act: Exploring the Dark Side of Accident Control,” Professional Safety, July 1979, 34–37. 82. David Klein, “Societal Influences on Childhood Accidents,” Accident Analysis and Prevention, 12 (1980), 275–281. Jacques Leplat and Xavier Cuny, Les Accidents du travail (Paris: Presses universitaires de France, 1974), especially 49, observed that

307

308

Notes to Pages 215–220

the systems approach accommodated all kinds of individual variations, including predisposition. 83. N. A. Hood, “Diseases of the Central Nervous System,” British Medical Journal, 4 (15 November 1975), 400. 84. M. Manciaux, “Epidemiology, and What Then? ” Injury Prevention, 1 (1995), 211. 85. An example of a collective work in which accident proneness is completely bypassed and engineering emphasized is Occupational Health: Recognizing and Preventing Work-Related Disease and Injury, ed. Barry S. Levy and David H. Wegman, 4th ed. (Philadelphia: Lippincott Williams & Wilkins, 2000). 86. Colin Cameron, “Accident Proneness,” Accident Analysis and Prevention, 7 (1975), 49. 87. See, for example, the contents of safety journals such as the American Society of Safety Engineers Journal. 88. I would emphasize again that very few accidents took place that did not involve human artifacts, or technology. Falls, for example, could at fi rst glance appear not to involve technology, but in fact the statistics showed clearly that a very large percentage occurred on stairs—a technological device—and most of the rest involved tripping or slipping in a built, technological environment, even at home. 89. Jorma Saari, “Risk Assessment and Risk Evaluation and the Training of OHS Professionals,” Safety Science, 20 (1995), 185. 90. Bernhard Rieger, Technology and the Culture of Modernity in Britain and Germany, 1890–1915 (Cambridge: Cambridge University Press, 2005), 73, suggests how, as early as the mid-twentieth century, advertising became a factor emphasizing the advantages of engineering in safety. 91. Robert West, “Application of PRIME Theory of Motivation to Promote Safe Driving,” unpublished paper, 2007, and Robert West, personal communication, 4 July 2007; I am most grateful to Robert West for his assistance. 92. Robert West, “Individual Differences in Traffic Accident Liability: Headlines from a Programme of Research,” Behavioural Research in Road Safety, 7 (1997).

conclusion 1. The phrase was introduced in chapter 6. 2. Richard Shelton Kirby, “Motor Vehicle Accidents,” Encyclopaedia of the Social Sciences, 11, 70 [1933]. Walter E. Schneider, “Victim X—1951’s Traffic Drama,” Casualty and Surety Journal, July 1951, 56–57. Eric A. Finkelstein et al., The Incidence and Economic Burden of Injuries in the United States (New York: Oxford University Press, 2006), 5. Yet these shocking 2000 figures came after substantial decreases in the accident rates. Medically treated injuries per person, for example, had declined by 15 percent over the previous fi fteen years (163). By the end of the twentieth century, the greatly reduced accident toll obviously made examining the idea of accident proneness superficially less urgent. See Daniel M. Albert, “The Nut Behind the Wheel: Shifting Responsibilities for Traffic Safety Since 1895,” in Silent Victories: The History and Practice of Public Health in Twentieth-Century America, ed. John W. Ward and Christian Warren (New York: Oxford University Press, 2007), 368–375. Anthony A. Meyer, “Death and Disability from Injury: A Global Challenge,” Journal of Trauma, Injury, Infection, and Critical Care, 44 (1998), 1–5.

Notes to Pages 220–225

3. Simon Carley, Kevin Mackway-Jones, and Stuart Donnan, “Major Incidents in Britain over the Past 28 Years: The Case for the Centralised Reporting of Major Incidents,” Journal of Epidemiology and Community Health, 52 (1998), 392–398, showed that even medical sources neglected the subject of major accidents. 4. Arthur F. McAvoy, “Working Environments: An Ecological Approach to Industrial Health and Safety,” Technology and Culture, 36 (1995), S145-S172. 5. Ibid. 6. William W. Lowrance, “The Nature of Risk,” in Societal Risk Assessment: How Safe is Safe Enough? ed. Richard C. Schwing and Walter A. Albers, Jr. (New York: Plenum Press, 1980), especially 8. The whole book, including Howard Raiffa, “Concluding Remarks,” 339–341, suggests experts’ lack of specific, local concern about individual injuries and damages, as if the traditional safety movement was withering as “risk management” became the dominating theme. 7. Sharon Clarke and Ivan Robertson, “An Examination of the Role of Personality in Work Accidents Using Meta-Analysis,” Applied Psychology, 57 (2008), 87–108, traced the persistence of the tradition of trying to identify individual accident prone workers but noted, 95, that this line of investigation had for years been only weakly sustained. I am indebted to Ivan Robertson for calling this recent literature to my attention. 8. John Langley, “The ‘Accident Prone’ Child—The Perpetuation of a Myth,” Australian Paediatric Journal, 18 (1982), 243–246. 9. For example, Roderick D. Iverson and Peter J. Erwin, “Predicting Occupational Injury: The Role of Affectivity,” Journal of Occupational and Organizational Psychology, 70 (1997), 113–128. 10. See, for example, Thomas P. Hughes, Human-Built World: How to Think About Technology and Culture (Chicago: University of Chicago Press, 2004). 11. See, for example, Donald J. Zeigler, James H. Johnson, Jr., and Stanley D. Brunn, Technological Hazards (Washington: American Association of Geographers, 1983); Lloyd J. Dumas, Lethal Arrogance: Human Fallibility and Dangerous Technologies (New York: St. Martin’s Press, 1999), especially chap. 4. Kevin Rosario, The Culture of Calamity: Disaster and the Making of Modern America (Chicago: University of Chicago Press, 2007). 12. Dumas, Lethal Arrogance, chap. 7, especially 167, 169. 13. The Life Cycle of Psychological Ideas: Understanding Prominence and the Dynamics of Intellectual Change, ed. Thomas C. Dalton and Rand B. Evans (New York: Kluwer Academic/Plenum Publishers, 2004), although still focused on innovation does contain in some of the essays enlightening passages on the fading away of old ideas. 14. I am describing a general tilt in the scholarship, and I do not mean to do injustice to intelligent colleagues; see, for example, Science and the Quest for Reality, ed. Alfred I. Tauber (New York: New York University Press, 1997), which includes excerpts from historians and philosophers, especially Robert Richards, “Theories of Scientific Change,” 203–227. 15. Good Morning, America, 5 July 2006, 7:35 a.m. E.D.T. 16. Some U.S. origins of this social strategy are described in John C. Burnham, “Medical Specialists and Movements Toward Social Control,” in Building the Organizational Society: Essays on Associational Activities in Modern America, ed. Jerry Israel (New York: Free Press, 1972), 19–30, 249–251.

309

310

Notes to Pages 225–230

17. Alan A. McLean and Russell De Reamer, “The Accident Repeater: Psychiatric and Statistical Considerations,” Psychosomatic Medicine, 29 (1961), 564. 18. Frederick L. McGuire, “A Typology of Accident Proneness,” Behavioral Research in Highway Safety, 1 (1970), 27. McGuire added, “The author’s fi les, for example, contain many cases of individuals who have been singularly devoted to the task of systematically fl irting with physical disaster.” 19. See, for example, Psychoneuroimmunology, Stress, and Infection, ed. Herman Friedman, Thomas W. Klein, and Andrea Friedman (Boca Raton: CRC Press, 1996). 20. Marta Ozcoidi Val et al., “History of Applied Psychology to Public Road Safety and Other Risk Activities in Spain,” an unpublished paper, generously furnished by the authors. 21. It is striking, for example, how, without using the term often, the idea of accident prone individuals was still pervading the U.S. government-funded inquiry, Maxine Esta Lubner, “Aviation Accidents, Incidents, and Violations: Psychological Predictors among U.S. Pilots” (doctoral diss., Columbia University, 1992). 22. See for example Maurice Yaffé, “Sports Injuries: Psychological Aspects,” British Journal of Hospital Medicine, 29 (1983), 224–232; Roeland J. Lysens et al., “The Accident-Prone and Overuse-Prone Profi les of the Young Athlete,” American Journal of Sports Medicine, 17 (1989), 612–619; Michael G. Ehrlich, Michael Hulstyn, and Charles d’Amato, “Sports Injuries in Children and the Clumsy Child,” Pediatric Clinics of North America, 39 (1992), 433–449. 23. One late twentieth-century social attitude was that of some disability activists: “Change society, not people.” See Allen Buchanan et al., From Chance to Choice: Genetics and Justice (Cambridge: Cambridge University Press, 2000). 24. Paul Forman, “The Primacy of Science in Modernity, of Technology in Postmodernity, and of Ideology in the History of Technology,” History and Technology, 23 (2007), 1–152. 25. Thomas Bick, Christoph Hohenemser, and Robert W. Kates, “Regulating Automobile Safety,” in Perilous Progress: Managing the Hazards of Technology, ed. Robert W. Kates, Christoph Hohenemser, and Jeanne X. Kasperson (Boulder: Westview Press, 1985), 314. 26. J. Andrew Mendelsohn, “Medicine and the Making of Bodily Inequality in Twentieth-Century Europe,” in Heredity and Infection: The History of Disease Transmission, ed. Jean-Paul Gaudillière and Ilana Löwy (London: Routledge, 2001), 21–79. 27. Harold S. Hulbert, “Aids in Handling Oneself and Others, Especially in Relation to Accident Prevention and Safety,” Rock Products, 11 October 1930, 92. 28. Scott A. Sandage, Born Losers: A History of Failure in America (Cambridge MA: Harvard University Press, 2005), although in this case referring to economic failure. 29. Car Crash Culture, ed. Mikita Brottman (New York: Palgrave, 2001). 30. Thomas Schlich, “Trauma Surgery and Traffic Policy in Germany in the 1930s: A Case Study in the Coevolution of Modern Surgery and Society,” Bulletin of the History of Medicine, 80 (2006), 86. Leon H. Doman, The Motor Vehicle Traffic Accident Problem: A National Solution (New York: International Insurance Monitor, 1969), 1. Bernhard Rieger, Technology and the Culture of Modernity in Britain and Germany, 1890–1945 (Cambridge: Cambridge University Press, 2005), 74–76. 31. John Dennis Chasse and David A. LeSourd, “Rational Decisions and Occu-

Notes to Pages 230–232

pational Health: A Critical View,” International Journal of Health Services, 14 (1984), 433–445. 32. And the complexity increased greatly when safety experts turned their attention to extremely diverse environments such as “the home,” where activities were not routine as in a factory or traffic behavior; see, for example, I. Jay Brightman, Isabel McCaffrey, and Leonard P. Cook, “Mortality Statistics as a Direction Finder in Home Accident Prevention,” American Journal of Public Health, 42 (1952), 840–848. 33. Judith Green, Risk and Misfortune: A Social Construction of Accidents (London: UCL Press, 1997), with a focus very different from mine, holds that accident itself is a social construction in many senses. 34. Ulrich Beck, Risk Society: Toward a New Modernity, trans. M. Ritter (London: Sage, 1992). The Risk Society and Beyond: Critical Issues for Social Theory, ed. Barbara Adam, Ulrich Beck, and Joost van Loon (London: Sage, 2000). 35. Green, Risk and Misfortune, 203 and especially chap. 6. 36. See, for example, Nathan Rosenberg, Perspectives on Technology (Cambridge: Cambridge University Press, 1976), and many much more recent general works such as Carroll Pursell, The Machine in America: A Social History of Technology (Baltimore: Johns Hopkins University Press, 1995), chap. 13. 37. Historians of disability and of surgery, for example, are concerned with the aftereffects of encounters with technology rather than the prevention of injuries. 38. See, for example, David A. Mindell, Between Human and Machine: Feedback, Control, and Computing before Cybernetics (Baltimore: Johns Hopkins University Press, 2002). One possible exception is the gloomy evaluation of technology, David Edgerton, The Shock of the Old: Technology and Global History Since 1900 (New York: Oxford University Press, 2007), who actually mentions costly accidents; see especially chap. 4. 39. Injury and Violence Prevention: Behavioral Science Theories, Methods, and Applications, ed. Andrea Carlson Gielen, David A. Sleet, and Ralph J. DiClemente (San Francisco: Jossey-Bass, 2006); quotation from David C. Grossman, foreword, xv. I am indebted to Les Fisher for calling this work to my attention. I. B. Pless, “A Brief History of Injury and Accident Prevention Publications,” Injury Prevention, 12 (2006), 65–66, dates the general use of the phrase “injury or accident prevention” to the 1990s and notes that much of that literature focused on children. 40. Helen Altman Klein has pointed out to me that Western thinkers have tended to use dispositional explanations for change—the smart person, the person inclined to one behavior or another. Many Eastern thinkers have tended, by contrast, to emphasize environmental explanations, including physical arrangements, organization, or even other people. Certainly accident proneness is exquisitely dispositional, and the reception of the idea in other cultures would be an interesting inquiry that I have not pursued.

311

BRIEF BIBLIOGRAPHICAL NOTE

This bibliographical note is brief for two reasons. First, surprisingly little existing scholarship touches directly on the subject of this book. My work has turned out to be pioneering in a number of ways. Second, the literature on specific topics is mentioned and discussed in the notations to the text, so that only some very general items need to be mentioned here to suggest, for any reader who may be interested, the scholarly environment in which this book came into being. The literature on simultaneous discovery is covered in Robert K. Merton, The Sociology of Science: Theoretical and Empirical Investigations, ed. Norman W. Storer (Chicago: University of Chicago Press, 1973), and in Augustine Brannigan, The Social Basis of Scientific Discoveries (Cambridge: Cambridge University Press, 1981), supplemented by Buhm Soon Park, “The Contexts of Simultaneous Discovery: Slater, Pauling, and the Origins of Hybridisation,” Studies in History and Philosophy of Modern Physics, 31B (2000), 451–474, and John C. Burnham, “Accident Proneness (Unfallneigung): A Classic Case of Simultaneous Discovery/Construction in Psychology,” Science in Context, 21 (2008), 99–118. Histories of the category of accidents as such are very scarce. There is one book, Accidents in History: Injuries, Fatalities and Social Relations, ed. Roger Cooter and Bill Luckin (Amsterdam: Rodopi, 1997). American incidents in the nineteenth century are featured in volumes with special focuses: Mark Aldrich, Death Rode the Rails: American Railroad Accidents and Safety, 1828–1965 (Baltimore: Johns Hopkins University Press, 2006), and Barbara Young Welke, Recasting American Liberty: Gender, Race, Law, and the Railroad Revolution, 1865–1920 (Cambridge: Cambridge University Press, 2001). Jamie L. Bronstein, Caught in the Machinery: Workplace Accidents and Injured Workers in Nineteenth-Century Britain (Stanford: Stanford University Press, 2008), summarizes and extends the work of British labor and legal 313

314

Brief Bibliographical Note

historians like P. W. J. Bartrip and S. B. Burman, The Wounded Soldiers of Industry: Industrial Compensation Policy, 1833–1897 (Oxford: Clarendon Press, 1983). Julian Go III, “Inventing Industrial Accidents and Their Insurance,” Social Science History, 20 (1996), 401–438, provides background, and I myself have ventured into the history of accidents in the twentieth century: John C. Burnham, “Why Did the Infants and Toddlers Die? Shifts in Americans’ Ideas of Responsibility for Accidents—From Blaming Mom to Engineering,” Journal of Social History, 29 (1996), 817–837. The general background factor of the safety movement is surprisingly no better served. There is one book on the United States, which does not venture far into the twentieth century: Mark Aldrich, Safety First: Technology, Labor, and Business in the Building of American Work Safety, 1870–1939 (Baltimore: Johns Hopkins University Press, 1997). One recent volume covers U.S. traffic safety: David Blanke, Hell on Wheels: The Promise and Peril of America’s Car Culture, 1900–1940 (Lawrence: University Press of Kansas, 2007). Monographic articles include Michael Guarnieri, “Landmarks in the History of Safety,” Journal of Safety Research, 23 (1992), 151–158; Julian A. Waller, “Reflections on a Half Century of Injury Control,” American Journal of Public Health, 84 (1994), 664–670; and Martin Lengwiler, “Between War Propaganda and Advertising: The Visual Style of Accident Prevention as a Precursor to Postwar Health Education in Switzerland,” in Medicine, the Market and the Mass Media: Producing Health in the Twentieth Century, ed. Virginia Berridge and Kelly Loughlin (London: Routledge, 2005), 57–65. One account of traffic safety is Daniel M. Albert, “The Nut Behind the Wheel: Shifting Responsibilities for Traffic Safety Since 1895,” in Silent Victories: The History and Practice of Public Health in Twentieth-Century America, ed. John W. Ward and Christian Warren (New York: Oxford University Press, 2007), 363–378. Recent contexts are covered particularly in David Meister, The History of Human Factors and Ergonomics (Mahwah NJ: Lawrence Erlbaum Associates, 1999), and late twentieth-century theory, error, and systems is summarized by Jop Groeneweg, Controlling the Controllable: The Management of Safety, 4th ed. (Leiden: DSWO Press, Leiden University, 1988). Judith Green, Risk and Misfortune: A Social Construction of Accidents (London: UCL Press, 1997), touches both accidents and the safety movement and, further, includes later insights concerning the risk society than were introduced in the classic, Ulrich Beck, Risk Society: Toward a New Modernity, trans. Mark Ritter (London: Sage Publications, 1992). Accident proneness per se is not well served. As I noted in the introduction, there is one free-standing article, Frank A. Haight, “Accident

Brief Bibliographical Note

Proneness, The History of an Idea,” Automobilismo e Automobilismo Industriale, 4 (1964), 536–546. In 1955, Ross A. McFarland, Roland C. Moore, and A. Bertrand Warren, Human Variables in Motor Vehicle Accidents: A Review of the Literature (Boston: Harvard School of Public Health, 1955), did indeed review the literature, over a thousand titles, including a great deal on accident proneness. A number of writers provided historical background for their articles, especially in the 1960s, and in 1971 Lynette Shaw and Herbert S. Sichel, Accident Proneness: Research in the Occurrence, Causation, and Prevention of Road Accidents (Oxford: Pergamon Press, 1971), fi nally provided a much more detailed and comprehensive historical review, along with some sharply critical commentary, of much of the literature—again focusing on traffic accidents. Material on the English founders of the idea of accident proneness is almost nonexistent except for some obituaries and specialized articles about Major Greenwood. Eric Farmer is almost unknown (there are no identifiable obituaries, for example). Marbe in Germany is fortunately covered masterfully by Annette Mülberger Rogele, La aportación de Karl Marbe a la psychología: un enfoque crítico (tesi doctoral, Universitat Autónoma de Barcelona, Bellaterra, 1995). Most of the other main actors or groups appear today to be minor figures who have not attracted interest from historians, who tend to be interested in theory rather than applied psychology. Robert Powell has written about Helen Flanders Dunbar: “Helen Flanders Dunbar (1902–1959) and a Holistic Approach to Psychosomatic Problems. I. The Rise and Fall of a Medical Philosophy,” Psychiatric Quarterly, 49 (1977), 133–152, and “Helen Flanders Dunbar (1902–1959) and a Holistic Approach to Psychosomatic Problems. II. The Role of Dunbar’s Nonmedical Background,” ibid., 50 (1978), 144–157. An incisive description of the general framework is in Theodore M. Brown, “The Rise and Fall of American Psychosomatic Medicine,” http://human-nature.com/freeassociations/riseandfall.html. For the most part, general background materials are extremely scarce or superficial. Historical treatments of the subject of accident control appear in virtually none of the histories of psychology. The history of both applied psychology and psychotechnology is either not well developed or is just getting under way. The outstanding exception is Anson Rabinbach, The Human Motor: Energy, Fatigue, and the Origins of Modernity (New York: Basic Books, 1990). An introduction to the German literature is incidentally found in Martin Lengwiler, “Gendering of Industrial Risks: Conceptions and Practices of Male ‘Accident Proneness’, 1880–1920 (Britain, Germany and Switzerland),” unpublished paper, who particularly calls attention

315

316

Brief Bibliographical Note

to the work of Carl Hoyos. More historical background is in William H. Schneider, “The Scientific Study of Labor in Interwar France,” French Historical Studies, 17 (1991), 410–446, and David Meskill, “Characterological Psychology and the German Political Economy in the Weimar Period,” History of Psychology, 7 (2004), 3–19. Some framing for psychiatrists’ views of the accident proneness syndrome is found in A History of Clinical Psychiatry: The Origin and History of Psychiatric Disorders, ed. German E. Berrios and Roy Porter (London: Athlone Press, 1995), 633–655, and German E. Berrios, The History of Mental Symptoms: Descriptive Psychopathology Since the Nineteenth Century (Cambridge: Cambridge University Press, 1996), 424–436. A beginning of an exploration of the impact of the insurance model is in William G. Rothstein, Public Health and the Risk Factor: A History of an Uneven Medical Revolution (Rochester: University of Rochester Press, 2003), and, indirectly, Gerd Gigerenzer et al., The Empire of Chance: How Probability Changed Science and Everyday Life (Cambridge: Cambridge University Press, 1989). Jonathan Simon, “The Emergence of a Risk Society: Insurance, Law, and the State,” Socialist Review, September–October 1987, 61–89, is unusually insightful. Epidemiology is explored especially in Mark Parascandola, “Epidemiology in Transition: Tobacco and Lung Cancer in the 1950s,” in Body Counts: Medical Quantification in Historical and Sociological Perspective, ed. Gérard Jorland, Annick Opinel, and George Weisz (Montreal: McGillQueen’s University Press, 2005), 226–248. Scholarship on the history of technology such as would illuminate the history of accidents is disappointingly scarce, although I have drawn on insights in Bernhard Rieger, Technology and the Culture of Modernity in Britain and Germany, 1890–1915 (Cambridge: Cambridge University Press, 2005). Two works do provide specially enlightening framing for the story of the encounter of humans with technology: David Edgerton, The Shock of the Old: Technology and Global History Since 1900 (Oxford: Oxford University Press, 2007), and Paul Forman, “The Primacy of Science in Modernity, of Technology in Postmodernity, and of Ideology in the History of Technology,” History and Technology, 23 (2007), 1–152. This present work therefore rests fundamentally upon journals and books published during the past. Fortunately, print during most of the twentieth century was the central means of communication, and therefore people at that time wrote openly about their work and their thinking. Primary materials can be located through Index Medicus, Psychological Abstracts and Psychological Index, Engineering Index, Industrial Arts Index, Education Index, and the like—plus further references that are listed in the primary sources. There is one outstanding bibliography, Jean Surry, In-

Brief Bibliographical Note

dustrial Accident Research: A Human Engineering Appraisal (Toronto: Labour Safety Council, Ontario Department of Labour, 1971). In addition, just leafing through obvious textbooks, monographs, and runs of (often unindexed) journals has yielded much relevant as well as background material, particularly for various years of Industrielle Psychotechnik, Safety Engineering, National Safety News, Industrial Safety Survey, Accident Analysis and Prevention, Human Factor, and the like. It is also informative that in the late twentieth century, accidents and/or accident proneness disappeared from standard journals such as Industrial Medicine or French and German psychotechnology and medical journals. I found the archival records of the early British groups in The National Archives, Kew, to be mostly uninformative, as was a sample of psychologist organization archives. Mülberger Rogele of course has fully covered relevant German archives regarding Marbe. For later years, the overwhelming amount of published material available rendered archival research superfluous for my broad, general exploration.

317

INDEX

Abraham, Karl, 133–135 absence rates and accidents, 112, 165, 170 accident: defi nition, 6–7, 14; historians and, 7; idea of, 6–9, 183–184, 221, 311n40; non-technological, 7; rationalized, 7–8 accident analysis, 16–17, 146, 149, 176–177, 178–179, 181–182, 202–205; and engineering safety, 188–189, 202–205; and human factors psychology, 209–210. See also error; host-agent-environment; system failure accident complexity, 202–205, 230 accident environment, 181, 182, 311n32. See also environment, workplace; host-agent-environment; streets accident liability, 54; Farmer, 58 accident neurotics, 33, 124–125 accident personal history, 38, 39, 42, 44, 45, 47, 49, 53–57, 70, 75–77, 136, 175, 186, 226. See also repeaters; statistics; Unfäller-Nichtunfäller (high accident-no accident people) division accident proneness: British originators, 50, 51–66; British originators, spread to United States, 60–61, 64–66; bypassing strategies, 201–202, 207–208, 214–216; clinical view, 152, 153–154, 160; death of

idea, 1–2, 4, 207, 210, 223–224, 227; death of idea, and epidemiology, 166; death of idea, and risk group, 189; decline of idea, 166–168, 171– 174, 178, 201–208, 216, 219–220, 222–223; defenders, late twentieth century, 174–178; defi nition, 1, 23, 64–66, 140, 175, 198; defi nition, actions, not people, 204; defi nition, statistical, 152; discrimination against, 5, 91–92, 118–121, 126–127, 173, 188, 195–196, 197, 200–201, 205, 224, 228, 232; early 1930s, 86–87; and error, 206–208, 215; folk observation, 33–35, 125; high point in 1950s, 4, 146–148, 180; historical profi le, 4–5, 10–11, 98, 99, 141, 145–146, 148–149, 152, 166, 193– 194, 201–202, 223–224, 231–232; historiography of, 64, 106, 174; and hysteria, 124–125, 137; industrial accident-traffic accident balance, 102, 146; insurance “moral hazard,” 184; internationalization of idea, 65, 82, 95, 107; late twentieth century, 205–208; levels of discussion, 96–97, 145, 148–152, 159–160; and malingering, 124–125, 130; multifactoral individual, 153–154, 155, 160, 176, 178; 1950s, 147–148; origins, 2, 10–11, 18, 31–32, 33, 222; origins, British, 51–59; origins, German, 36–43; persistence of 319

320

Index accident proneness (continued) idea, 4, 174–178, 201, 205, 211–212, 223–224, 225; personal problem, 160–161, 198–199; popular use of idea, 6, 46, 81–85, 92–97, 99–104, 137, 141, 144, 145, 147, 223, 224; popular use of idea, unfavorable, 117; popularization, post–World War ii, 159–165; popularization, post–World War ii, last example, 160; popularization, post–World War ii, negative evidence, 162; practicality emphasis, 41, 50, 65, 102, 104, 174, 176–178, 222, 225, 227 (see also mass transport safety); prediction failure, 142, 169, 172, 232 (see also insurance, casualty/accident; statistics); public health problem, 155; social analysis, 200–201, 232; and social deviation, 113, 128–129, 151; stages, 167–168; standardization of idea, 67, 96; standardization of idea, 1950s, 147–148, 149, 157; term, 2, 3–4, 6, 58–59, 82–83; term, generic status, 101–102, 164–165; term, spread, 82, 98, 106; tests for, 59, 62, 64–65, 94, 95, 96, 104–105, 110–112, 116, 117, 127, 150, 172, 175–176, 194; tests for, interpretation, 113; tests for, psychological, 112; tests for, psychophysical, 112; time interval in statistics, 169–170; U.S.-centered 1930s on, 99–100, 145–146. See also clinical approach to accident proneness; temporary accident proneness accident sequelae, 124–125. See also pension neurosis accident statistics, 9, 10, 13, 15, 17, 77, 88–89, 220 accidents, minor, 88–89, 92–93, 100, 114 accidents, motivated, 32–33, 132–135, 136–138, 156, 158–159, 161, 197, 227–228, 230; subconscious/unconscious, 78, 130, 132–135, 136–138, 148, 194, 196; unwilled, 10

accidents, time of, 16, 27–28, 44, 69 accidents as a social problem, 9, 10–11, 13, 14–15, 77, 155, 175, 193, 219, 231; costs of, 10, 11, 13, 77; defi ning moment, 9, 227; legal responsibility, 17–18, 20–21 accidents ignored by public, 220, 230–231 affi nity, 39, 44, 49. See also accident proneness: term; accident proneness: term, generic status; accident proneness: term, spread age, 92, 178, 185, 187, 188–189, 191–192, 213–214. See also children; risk groups airplane accidents, 15, 229–230. See also pilot screening alcoholic beverages, 22, 90, 151, 181, 283n25 American Psychiatric Association, 139 Anlage. See traits antibiotics, 157 aptitude testing, 29, 32, 42, 48; in mass transit, 68, 71–76. See also personality tests; psychologists; psychophysiological testing and accident proneness; psychotechnics Asperger’s syndrome, 24 association experiments, 36, 39 ataxia. See clumsiness automobile. See motor vehicle safety; traffic accidents balance of attention between occupational and traffic accidents, 102, 146 Baskett, Franz K., 234 Beck, Ulrich, 183, 219–220, 230 Berlin, 15, 71–72, 77 Bevin, Ernest, 116 Bingham, W. V., 76, 80–81 blaming the victim, 109, 197, 205. See also accident proneness: discrimination against Book of Accidents (1830), 19 Boston, 15, 71–72, 77 Brown, Clarence W., 172

Index carelessness, 5, 10, 17, 19–21, 53, 59, 62–63, 91, 97, 193, 205 categories of people who have accidents, 91–92. See also Marbe, Karl causal chain, 7–8, 15–16, 205, 221. See also accident analysis; determinism Chambers, E. G., 57–59, 64–66, 74, 81, 93, 94, 95, 104–106, 112, 196 character. See personality Chekhov, Anton, 33 children: accident proneness in, 37, 43, 146, 155, 156–159, 161 (see also pediatricians); and decline of idea of accident proneness, 171; education of, 8, 19–20 (see also safety education); injury statistics, 45, 157. See also age chronic fatigue syndrome, 123 Cleveland Railway Company, 76; accident prone factors list, 86–87, 91, 126–127 clinical approach to accident proneness: compatibility with engineering, 215, 224–227; early Detroit, 78, 225; early Milwaukee, 82; factor in preserving idea, 176, 227; post– World War ii, 152–154; sidelining idea of accident proneness, 176, 227; social strategy, 224–225. See also accident proneness: clinical view clinical presentation, 144 clumsiness, 9, 21–24, 33, 193, 230; management of, 24–25, 226 Cobb, Percy W., 168 cognitive psychology, 141, 178–179, 206; and reconceptualizing accidents, 202, 206–208, 209–210, 224 color blindness, 31. See also psychophysiological testing and accident proneness; vision complexity. See accident analysis; accident complexity Connecticut, 1930s traffic accident study, 114 Conrad, Jobst, 208 consumer items, 209, 211. See also home accidents; technology: definition

Cooter, Roger, 9, 227 cost-benefit analysis, 228–230 Covert ascending spiral. See popularization process Cruise, Tom, 6 Culpin, Millais, 126 cultural changes, accident proneness and, 2–5, 31–32, 223–224, 227–230, 231–232. See also Zeitgeist dangerous drivers, 77–83, 84, 118, 146, 160, 197, 198; carryover to accident proneness, 79–81; late twentiethcentury new identity, 221 Dayton Power and Light Company, 82 determinism, 2–4, 7–8, 9, 15, 183– 184, 221. See also causal chain; rationalism Detroit, psychopathic clinic, 78, 225 developmental coordination disorder. See clumsiness deviance. See social deviance and accident proneness diabetes. See diseases as causes of accident proneness disasters, 207–208, 223. See also system failure discipline as accident management, 1, 8, 19–20, 21, 98, 181, 193–194, 209, 216–217, 221. See also managers; traffic accidents discrimination against accident prone people. See accident proneness: discrimination against disease syndrome, 122, 136, 138–139, 140, 142, 144, 158–159 diseases as causes of accident proneness, 30, 90–91, 150–151, 185, 191 disposition, 217. See also predisposition driver licensing. See driver screening driver responsibility, 1930s, 114. See also screening for accident proneness driver screening, 64, 77–81, 114, 177– 178; Britain, 116–117; Spain, 117; transition to accident proneness,

321

322

Index driver screening (continued) 80–81; United States, 118. See also screening for accident proneness; traffic accidents DSM, 139 Dunbar, Helen Flanders, 104, 135–138, 156, 158 Edgerton, David, 12 education. See safety education efficiency, 68–69, 118. See also production errors; productivity; taylorization egalitarianism of engineering solution, 214, 215, 216–217, 233 Electric Railway Journal, 75 emotions as cause of accidents, 148, 150, 155, 161, 164. See also accidents, motivated employee experience and accidents, 18, 59, 64, 103, 189, 195–197 employee rating, basis for accident proneness classification, 67–70, 73–76 employee screening, 29, 31, 41–42, 48, 54, 60, 66, 68–69, 70–77, 105, 106, 108, 114, 116, 118–121, 177–178; 1920s change, 45. See also screening for accident proneness enforcement. See discipline as accident management engineering as a social strategy: bypassing accident proneness, 201–202, 232; major cultural shift, 1, 182, 216, 223; sociopolitical advantages, 196 engineering for individual safety, 4–5, 8, 146, 182, 193, 211–217, 227; criticism of, 211–212. See also safety devices engineers, 5, 95; and human factors psychologists, 210–211 environment, accident, 8, 11, 15, 168, 176, 180, 191–192, 199–200, 202–203, 205, 209–210, 214. See also accident analysis; accident complexity; host-agent-environment; streets

environment, workplace, 16–17, 18, 53, 90, 102, 168, 193–199, 196, 204. See also accident; environment, accident; managers environmental thinking, 208, 221, 223, 231 epidemiology, 155, 166, 168, 180–183, 185; feeds into engineering, 212–214; retrospective versus prospective, 180; social viewpoints, 181, 185 epilepsy, 30, 91, 150, 196. See also diseases as causes of accident proneness ergonomics, 182, 208–209, 210, 211, 214 error, 11, 17–18, 26, 28, 32–33, 36, 93, 205–208, 209, 216, 223, 304n49; cognitive psychology and, 206–208; Marbe and, 39, 40; psychoanalysis and, 124, 133, 141 ethnicity. See social class and accidents experts, 147, 148–152, 163; doubts, 164. See also safety movement; names of professions Eysenck, H. J., 177 Farmer, Eric, 3, 50, 51, 57–59, 64, 74, 79, 81, 93, 94, 95, 100, 109–110, 112, 116, 117, 195, 254n22; later years, 104–106 fatigue, 18, 26–28, 29, 51–53, 69; decreasing importance, 102 Fenichel, Otto, 137 Fisher, Boyd, 34–35 Forman, Paul, 228, 231 Freud, Sigmund, 32–33, 124, 133, 141 Froggatt, Peter, 170 functional diseases, 124 g factor, 105, 110–111, 188, 198 gender-based categories, 51–52, 92, 155, 189–190, 192. See also age Ghiselli, Edwin E., 172 God, acts of. See accident: defi nition Goldmark, Josephine, 28 government safety regulation, 14. See also discipline as accident management

Index Green, Judith, 6, 183–184, 230–231 Greenwood, Major, 33, 49, 53–56, 62, 64, 66, 90, 104, 117, 168, 169 Guarnieri, Michael, 182–183 Gulick, Luther Halsey, 25 habit, 39, 40, 77; accident habit, 136, 277n55 Haddon, William, Jr., 182 Häkkinen, Sauli, 175, 178 Harrell, Thomas Willard, 163 Hawthorne Studies, 102, 152, 154 Health of Munitions Workers Committee, 51 Heinrich, H. W., 92–93, 99, 100 heredity-environment controversy, 39, 40, 44, 50, 92, 108–111, 128, 197–200, 231, 235n1 highway safety. See streets; safety movement; traffic accidents Hildebrandt, H., 47 history of ideas, 1–2, 4, 11, 22, 42, 167– 168, 222–224, 236n7, 289–290n2. See also popularization of accident proneness; simultaneous discovery Holmes, Oliver Wendell, 9 Holroyd, E. M., 177–178 home accidents, 15, 114, 146, 219, 221 host-agent-environment, 181, 182–183, 199, 204–205. See also accident analysis human factor in accidents, 5, 10, 17–19, 28, 29, 48–49, 55–56, 69, 100, 108, 193; post–World War ii refi ning, 202, 213–214; reconceptualizing as error, 205–208, 215 human factors psychology, 208–211, 215; and shift to engineering, 210–211, 214 human nature. See heredityenvironment controversy; individual differences; personality individual differences, 1, 18, 21–22, 29– 32, 36, 48–49, 53, 58, 81, 89–90, 109, 122, 131, 180, 202, 216–217, 219, 228, 233; in accident prone people, 1, 166, 198, 225. See also clinical

approach to accident proneness; habit; personality individual responsibility, 180, 240n6. See also blaming the victim individuality obscured: epidemiology, 183; insurance, 185; risk group, 185 individualized treatment, 130, 140, 142–143, 176, 224–227; distinctive for psychiatrists, 125–126. See also clinical approach to accident proneness individuals, prediction of accident proneness in, 169, 172 industrial accidents, 7, 13–14, 15, 16–17, 27–28, 42, 44, 60, 146, 181, 220; decline after World War ii, 146, 175. See also balance of attention between occupational and traffic accidents Industrial Fatigue Research Board, 51, 52, 55–57, 58–60, 64 Industrial Health Research Board, accident prone driver testing, 116–117; wartime report, 103–104 industrial medicine, 60, 124 industrial mental health, 126–129, 141 industrial psychology, 94–95 industrialization, 13–14, 146, 227. See also industrial accidents; technology Industrielle Psychotechnik, 45 inequality. See egalitarianism of engineering solution; individual differences; risk infection proneness, 226 injury: different from disease, 181; introduction of term, 8–9, 11, 237n18; replaces “accident,” 182–183, 226; sequelae, 42 insurance, casualty/accident, 13, 14, 45–46, 83–85; Britain, 1930s, 116–117; Germany, 13–14, 124; physicians as arbiters of claims, 124; resistance of industry to accident proneness, 45–46, 184–185; source of statistics, 13–14, 38, 117 insurance model for society, 184. See also risk

323

intelligence, 18, 26, 30, 41; relation to accident proneness, 63, 89. See also g factor Johnson, H. M., 168–169 Jung, C. G., 39 Kahn, Eugen, 128 Korngold, Suzanne, 108–109, 198–199 Kraepelin, Emil, 39 Kuhn, Thomas, 222 labor, industrial: commodity, 73, 193; conditions, 194–195; shift from commodity to personnel problem, 102; union concern about accident proneness idea, 116, 300n4 Lahy, J.-M., 46, 68–69, 70–71, 73, 107–109, 112, 168; sociopolitical views, 103–109, 195, 198–199 Laird, Donald, 81 Lauer, Alvhh, 81, 82, 109, 112, 130 left-handedness, 199–200 London, 15, 220 Löwenstein, Otto, 130–131 Lykes, Norman Roberts, 147–148, 161 machine guards. See safety devices machines, speed of, as accident factor, 28, 53, 103, 228–229 malingering. See accident proneness: and malingering managers, 15, 18–19, 34, 42, 67, 71, 97, 99, 111, 128, 173–174, 175; contributions to idea of accident proneness, 67, 70; discharging and transferring workers, 119–120; and safety movement, 9; using idea of accident proneness to escape responsibility, 174, 222 Marbe, Karl, 3, 36–50, 55, 58, 74, 79, 81, 90, 94, 104, 106–107, 108, 110, 112, 133, 159, 168, 185, 197; court experiences, 37–38; criticism of, 47; forming idea of accident proneness, 37–41, 42–43; impact, 43–50, 81, 98; philosophical orientation, 38, 39;

theory, 42–43, 47–50, 51, 106–107; withdrawal 1930s, 106–107 mass transport safety, 10–11, 14, 28, 42, 44, 60, 63–65, 67–77, 86, 209, 222, 232; absorption of idea of accident proneness, 75–77; screening tests, 1920s, 70–73; shift to watching accident records, 74 McFarland, Ross A., 179, 213 mechanization. See technology medical examinations, 30 Medical Journal of Australia, 172 Medical Research Council, 3, 18, 22, 52, 56, 57–58, 116, 117 medicalization, 5, 67, 122–123, 156; accident proneness a negative case, 123, 143–144, 231; psychiatrists and, 123, 131 medicine. See physicians Meng, Heinrich, 124–125 Menninger, Karl, 131–135, 136, 156 menopause. See gender-based categories menstruation. See gender-based categories Meringer, Rudolf, 133 Meskill, David, 45 Milwaukee, 74–75, 82 minor accidents. See accidents, minor misfits, industrial, 90, 105, 110, 219, 228, 232 Mitchell, Margaret, 148 MMPI (Minnesota Multiphasic Personality Inventory), 150 modernist society, 7, 9, 227–230; irrationality, 227–228. See also accidents, motivated; rationalism Moede, Walther, 15, 42, 48, 74, 107, 112 Monday effect, 16. See also accidents, time of; fatigue motor skills, 26, 89, 226. See also clumsiness motor vehicle accidents. See traffic accidents motor vehicle safety, 182, 215. See also traffic accidents

Index multiple discovery. See simultaneous discovery munitions workers, 51–54, 58 Munro, David, 116, 117 Münsterberg, Hugo, 29, 69–70, 75 Muscio, Bernard, 56, 57 Myers, Charles S., 32, 49–50, 57, 60 narcissism, age of, 227 National Institute of Industrial Psychology, 50, 60–61, 65, 107, 210 National Parent-Teacher, 161 National Safety Council, 95, 120 naturalizing accidents, 7–8, 220–221. See also accidents ignored by public; determinism negligence. See carelessness New York City, 15 Newbold, Ethel, 59, 62 Nichtunfäller. See Unfäller-Nichtunfäller (high accident-no accident people) division occupational injuries. See industrial accidents occupations, dangerous, 16, 59, 185. See also insurance, casualty/accident Ohio Division of Traffic and Safety, 102 organizational society, 9 Paris, 15, 68–69, 70–71, 73 pediatricians: and accident proneness, 140, 147, 156–159; and accidents post–World War ii, 156–157; and complex causation, 157. See also children pension neurosis, 230. See also accident proneness: and malingering; accident sequelae percentages method, 169 personal characteristics. See individual differences; personality; traits personality, 30–32, 44, 112, 127–128, 136, 149, 150, 156, 158, 161, 217, 222, 281n93; Marbe, 39–40, 50, 107–108, 110, 128 personality tests, 86, 107, 112, 150, 178

Peters, Judith, 22 physical educators: and motor skills, 24–25; and sports accidents, 226 physicians, 10, 80, 124, 147. See also medicalization; pediatricians; psychiatrists physiological factors. See psychophysiological testing and accident proneness pilot screening, 73, 118, 150, 172, 179, 194, 197; step toward accident proneness idea, 54–55. See also screening for accident proneness Poisson curve. See statistical chance popularization of accident proneness, 5–6; Marbe, 41–43; negative, 1930s, 117; 1920s, 147. See also popularization process popularization process, 96–97, 159–160, 164 Powell, Robert C., 136, 138 predictability, 169, 172 predisposition, 5, 40, 49, 106, 108, 109, 110, 154, 199, 249n15. See also heredity-environment controversy probability. See statistical chance production errors, 10, 60. See also error productivity, 26, 27, 29, 105. See also efficiency; taylorization professional groups and accident proneness, 5, 10, 95. See also engineers; managers; physicians; psychologists; safety movement professional treatment as social strategy, 225. See also clinical approach to accident proneness psychiatric disease classification, 139, 140, 142 psychiatrists, 29–30, 122–129; interest in accident proneness, 123–124, 130–131, 137–138, 140–144, 156, 231; publications, 138–141. See also American Psychiatric Association; industrial mental health; psychosomatic medicine psychiatry, industrial, 125–129, 140–141

325

326

Index psychoanalysts, 124, 131–135, 137–138. See also accidents, motivated psychological testing. See personality tests; psychologists; psychophysiological testing and accident proneness psychologists, 5, 11, 18, 25–28, 30–31, 71, 79–82, 86, 104, 206–207, 222; claim to subject of accident proneness, 5, 40, 41–42, 68–70, 95, 96, 99, 119–120, 122, 123, 125, 127–128, 143, 144, 156, 210–211, 221, 228, 236n9; humanitarian, 195–196; and statistics, 168; twenty-fi rst century, 232. See also cognitive psychology; psychophysiological testing and accident proneness psychologization of society, 121, 141, 156, 222 psychology, applied, 28–31, 208–209; Marbe, 41–42. See also ergonomics; human factors psychology; psychologists; psychotechnics Psychopathic Clinic (Detroit), 78 psychophysiological testing and accident proneness, 30, 31, 51, 63, 71, 76, 82, 86, 89–90, 92, 112, 136, 190–191. See also diseases as causes of accident proneness; environment, workplace psychosomatic medicine, 104, 135–138, 144 psychotechnics, 28, 29, 45, 46, 48, 72–73; Marbe and, 41–43, 46, 74, 106; shift 1920s–1930s, 107 public health. See epidemiology purposive accidents. See accidents, motivated Rabinbach, Anson, 9 railways, 38, 69–70, 71, 75. See also mass transport safety rationalism, 7–8, 148. See also causal chain reaction time, 18, 30–31, 48, 76, 107, 112, 124, 246n69, 260n39. See also psychophysiological testing and accident proneness

Reason, James, 206–207 recklessness. See carelessness recreational accidents, 15. See also sports accidents repeaters, 35, 104, 111, 113, 115, 125, 134–135, 141, 149, 154, 159, 161, 171, 172, 191; clinical presentation, 144; distinguishing repeaters from accident prone, 149; persistence of term, 179. See also accident proneness repetition. See habit; repeaters risk, 219–220; cost-benefit calculation, 228–230; idea of, 8, 183–184; increase with mechanization, 15; a right, 228. See also accident liability risk factor, 184–185, 189. See also insurance model for society; risk groups risk groups, 166, 174, 180–181, 183–192, 199–200, 264n19, 295n74; accident proneness as model, 186–187; basis of grouping, 183–187; communication, 231–232; history, 186–187; leading to engineering, 187–188, 191–192, 200, 213–215 risky behavior, 5, 19–28, 98, 181, 228–229, 242n28 road safety. See traffic accidents Rothstein, William, 184 safety devices, 1, 8, 14, 97, 100, 212; Germany, 48; post–World War ii, 211–214 safety education, 1, 8, 20, 21, 87, 98, 100, 175, 181, 192, 193–194, 216–217, 221, 266n11. See also safety movement safety movement, 1, 8, 9, 10, 11, 13, 19–20, 45, 47–49, 59, 155, 166, 174, 175, 193–194, 221; European profi le, 100; fractured late twentieth century, 179; historiography, 10; home to idea of accident proneness, 100, 111, 113, 147, 171; interwar, 87–89, 99; laws and, 14, 20, 79, 193, 194, 197 (see also discipline as accident management; traffic accidents);

Index twenty-fi rst century, 231–232; U.S., 59. See also engineering headings safety rules. See discipline as accident management Schmitt, E., 44, 74 Schorn, Maria, 43, 46 Schulzinger, Morris, 139–150, 154 screening for accident proneness, 43, 71, 86, 105, 108, 114, 118, 142, 177, 194–197, 201, 226. See also psychophysiological testing and accident proneness selection. See screening for accident proneness self-destructive behavior, 129–130, 131–132, 133, 134, 138, 198. See also accidents, motivated Shaw, Lynette, 146, 166, 174 Sherrington, Charles, 56 shift to technology/engineering, 223, 228–231, 239n35 ship navigators, 69–70 Sichel, Herbert J., 146, 166, 174 simultaneous discovery, 2–4, 11, 50, 231, 235n3; tests for, 3–4 slips. See error Slocombe, C. S., 76, 113, 121 Snow, Adolph J., 79–80, 81 social class and accidents, 112, 201, 215. See also social deviance and accident proneness social deviance and accident proneness, 77, 200–201, 223, 232 social factors and analysis of accidents, 146, 151–152, 154, 158–160 social strategy, 1–2, 4–5, 193–194, 211–217, 223. See also shift to technology/engineering Spock, Benjamin, 162 sports accidents, 88, 226 statistical chance, 8, 38, 54–55, 62, 168, 169, 221 statistics, 168–170, 248n8, 262n61; on accidents, 13, 17, 38–39, 62, 168–170; and British discoverers, 53–56; and decline of accident proneness idea, 142, 168–170; and epidemiology, 180–181; faulty data

in, 203–204; in spreading accident proneness, 95; standards, 168–170 street railways, 14, 15, 68–72, 75–76. See also mass transport safety streets, 15, 146; World War ii, 103 stress, 18, 151, 206–207, 225–226 style, shift from types, 198 styles of thinking and behaving. See cognitive psychology subconsciously driven accidents. See accidents, motivated supervisors. See managers susceptibility, 53, 56, 58–59, 191. See also predisposition syndrome, 122, 273n2; accident proneness as, 152–153 system failure, 207–208, 210–211, 223; and accident proneness, 307n80; safety measures, 215 systems model, 209, 215 systems theory. See human factors psychology Taylor, A. J. P., 165 taylorization, 29; succeeded by ergonomics, 209 technological fi xes, 1, 5, 212, 232. See also engineering as a social strategy; engineering for individual safety; safety devices technology: defi nition, 11–12, 219, 239n38, 308n88; human interface with, 1, 4–5, 7, 10–12, 209–219, 231; implications, 4–5; reconceptualizing, 202–208; shift to environment, 207–208; shift to systems, 12, 207–208, 209. See also shift to technology/engineering temporary accident proneness, 91, 111, 167–168; and gender, 190–191; post–World War ii, 138, 149–150, 170, 172, 173, 111 textbooks, accident proneness in, 44, 48, 55–56, 60, 94, 163–164 three E’s, 155, 212, 232 Three Mile Island event, 207, 208 Time, publicity of accident proneness, 103

327

328

Index traffic accidents, 10, 14–15, 63, 77–83, 86, 114–115, 185, 189, 197, 212, 215–217, 241n12; and accident proneness idea, 175; as epidemiology, 180; Farmer and, 105–106; post–World War ii attention, 146, 148, 175. See also driver screening traits, 30, 31–32, 42, 68, 86, 105, 107, 109–110, 197–198; accident proneness as a trait, 158, 197–198. See also personality Tramm, K. A., 42, 71–72 “Troublesome Tom,” 19 trucks, 77; driver tests in Spain, 226. See also traffic accidents types, 105, 108, 110, 112, 128, 142, 197–198; accident types, 137, 198; personality types, 136, 158, 161; subdivisions, 198. See also personality; style, shift from types Unfäller-Nichtunfäller (high accident-no accident people) division, 38, 40, 48, 53–54, 65–66, 74, 98, 106, 150, 159, 164, 178, 221, 228 Unfallneigung term, 6, 35, 39–40, 44, 49, 107 unsafe drivers. See dangerous drivers unwilled actions. See accidents, motivated Vernon, H. M., 52–53, 56, 57, 59, 100 vision: as accident factor, 89; connected with engineering, 212. See also psychophysiological testing and accident proneness; safety devices

Viteles, Morris, 46, 74–75, 80, 93–94, 120, 195 wealth to enable technology, 216 Wechsler, David, 81 Weiss, A. P., 82 Wells, Frederic Lyman, 26 West, Robert, 217 Widmer, Chs., 33 Wilson, George, 31 women. See gender-based categories Woods, Hilda M., 33, 53–39, 62, 64, 66, 104, 168 Woodworth, R. S., 26 workers, discharge and transfer, 65–66, 74, 77, 114, 118–120, 170, 193–196. See also accident proneness: discrimination against; labor, industrial; managers World Health Organization, International Statistical Classification, 139 World War ii, accident proneness idea impacted, 103–104, 168; increase in accidents, 103 Würzburg School in psychology. See Marbe, Karl Young, Allan, 142 Yule, G. Udney, 54, 55 Zeitgeist, 3–4, 10–11, 67. See also cultural changes, accident proneness and Zurfluh, J., 199