Taming Prometheus Talk About Safety Culture

ANRV381-SO35-17 ARI 5 June 2009 9:28
Taming Prometheus: Talk

About Safety and Culture
by MASSACHUSETTS INSTITUTE OF TECHNOLOGY on 09/10/09. For personal use only.
Susan S. Silbey
Annu. Rev. Sociol. 2009.35:341-369. Downloaded from arjournals.annualreviews.org
School of Humanities, Arts and Social Sciences, Massachusetts Institute of Technology,

Cambridge, Massachusetts 02139; email: ssilbey@mit.edu
Annu. Rev. Sociol. 2009. 35:34169 Key Words

First published online as a Review in Advance on technology, accidents, disasters, systems, management
April 16, 2009
The Annual Review of Sociology is online at Abstract

soc.annualreviews.org
Talk of safety culture has emerged as a common trope in contempo-
This articles doi: rary scholarship and popular media as an explanation for accidents and
10.1146/annurev.soc.34.040507.134707
as a recipe for improvement in complex sociotechnical systems. Three
Copyright ! c 2009 by Annual Reviews. conceptions of culture appear in talk about safety: culture as causal at-
All rights reserved
titude, culture as engineered organization, and culture as emergent and
0360-0572/09/0811-0341$20.00 indeterminate. If we understand culture as sociologists and anthropol-
ogists theorize as an indissoluble dialectic of system and practice, as
both the product and context of social action, the first two perspectives
deploying standard causal logics fail to provide persuasive accounts.
Displaying affinities with individualist and reductionist epistemologies,
safety culture is frequently operationalized in terms of the attitudes and
behaviors of individual actors, often the lowest-level actors, with the
least authority, in the organizational hierarchy. Sociological critiques
claim that culture is emergent and indeterminate and cannot be in-
strumentalized to prevent technological accidents. Research should ex-
plore the features of complex systems that have been elided in the talk
of safety culture: normative heterogeneity and conflict, inequalities in
power and authority, and competing sets of legitimate interests within
organizations.
341
ANRV381-SO35-17 ARI 5 June 2009 9:28
. . .the darkest and most treacherous of all the little of the theoretical edifice sociologists and
countries . . . lie in the tropic between inten- anthropologists have built for cultural analy-
tions and actions. . . . sis. Decades after the social sciences reconcep-
Chabon (2008, p. 29) tualized culture as the medium of lived ex-
perience ( Jacobs & Hanrahan 2005, p. 1), a
Rescuing Prometheus, by the venerable histo- normatively plural system of symbols and
rian of technology Thomas Hughes (1998), meanings that both enables and constrains so-
describes how four large postWorld War II cial practice and action (Sewell 2005, pp. 152
projects revolutionized the aerospace, com- 75; Silbey 2001; 2005a, p. 343), the cultural turn
puting, and communication industries by has taken root in the military and engineer-
transforming bureaucratic organizations into ing professions, and for similar reasons: human
postmodern technological systems. In place action and culture getting in the way of tech-
of centralized hierarchies of tightly coupled nological efficiency. However, unlike the mil-
homogeneous units typical of traditional itarys embrace of culture where critique con-
corporate and military organizations, Hughes fronts its every move (Gusterson 2007), efforts
describes the invention of loosely coupled to propagate safety culture in complex techno-
networks of heterogeneously distributed and logical systems proceed with scant attention to
often collegially connected communities of its ideological implications. Despite the appro-
diverse participants. By the 1980s and 1990s, priation of the term culture, many advocates
new modes of management and designpublic and scholars of technological innovation and
participation coupled with commitments to en- management deploy distinctly instrumental and
vironmental repair and protectionovercame reductionist epistemologies antithetical to cul-
what had been intensifying resistance to tural analysis. We can be protected from the
large-scale, often government sponsored, tech- consequences of our very effective instrumen-
nologies. Prometheus the creator, Hughes tal rationalist logics and safety can be achieved,
(1998, p. 14) writes, once restrained by defense they seem to suggest, by attending to what ad-
projects sharply focused upon technical and vocates of safety culture treat as an ephemeral
economic problems, is now free to embrace yet manageable residue of human intercourse
the messy environmental, political, and social something akin to noise in the system. How are
complexity of the postindustrial world. we to understand this unexpected and unusual
If the engineering accomplishments of the appropriation of the central term of the soft sci-
past 40 years signify a resuscitated capacity to ences by the experts of the hard, engineering
mobilize natural and human resources to pro- sciences?
duce, distribute, and accumulate on historically This article reviews popular talk and schol-
unprecedented scales, proliferating interest in arship about safety culture. Since the 1990s,
safety culture may signal renewed efforts to identifying broken or otherwise damaged safety
tame Prometheus. In the past 20 years, a new culture has become a familiar explanation for
way of talking about the consequences of com- organizational and technological failures. Al-
plex organizations and sociotechnical systems1 though the term safety culture has been de-
has developed. Although culture is a common ployed across institutional sites and scholarly
sociological subject, those talking about safety fields, it is largely absent from sociological
culture often invoke the iconic concept with scholarship. Sociologists studying accidents and
disasters provide a more critical and skep-
tical view of safety culture, if they address
1
The notion of a sociotechnical system stresses the close it at all (e.g., Beamish 2002; Clarke 1989,
interdependence of both the technological artifacts and be- 1999, 2006; Gieryn & Figert 1990; Hilgartner
havioral resources (individual, group, and organizational)
necessary for the operation of any large-scale technology 1992; Perin 2005; Perrow 1999 [1984], 2007;
(Pidgeon 1991, p. 131). Vaughan 1996). However, in engineering and
342 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
management scholarship, the term safety cul- and the interdependencies that are essential to
ture is invoked with increasing frequency and cultural and organizational performances and
seems to refer to a commonly shared, stable set analyses.
of practices in which all members of an orga- This review first provides a historical fram-
nization learn from errors to minimize risk and ing for talk about safety culture because that
maximize safety in the performance of organi- perspective is most clearly missing in much of
zational tasks and the achievement of produc- the research. I suggest that talk about safety
tion goals. culture emerges alongside market discourse
In this review, I argue that the endorsement that successfully challenged the previous cen-
of safety culture can be usefully understood as turies mechanisms for distributing and mitigat-
a way of encouraging and allocating responsi- ing technological risks. In the second section, I
bility (Shamir 2008)one response to the dan- describe the more than fourfold increase in ref-
gers of technological systems. Invoking culture erences to safety culture that appeared in pop-
as both the explanation and remedy for techno- ular and academic literature between 2000 and
logical disasters obscures the different interests 2007. Organizing the work in terms of three
and power relations enacted in complex organi- commonly deployed conceptions, I then de-
zations. Although it need not, talk about culture scribe culture as causal attitude, as engineered
often focuses attention primarily on the low- organization, and as emergent. Relying on a
level workers who become responsible, in the conception of culture as an indissoluble dialec-
last instance, for organizational consequences, tic of system and practice, both a product and
including safety. Rather than forgoing partic- context of social action, I argue that the first
ularly dangerous technologies or doing less in two perspectives not only fail to provide per-
order to reduce vulnerabilities to natural, indus- suasive accounts, but reproduce individualist
trial, or terrorist catastrophes, talk about safety and reductionist epistemologies that are un-
culture reinforces investments in complex, hard able to reliably explain social or system perfor-
to control systems as necessary and manageable, mance. Although invocation of safety culture
as well as highly profitable (for a few), although seems to recognize and acknowledge systemic
unavoidably and unfortunately dangerous (for processes and effects, it is often conceptualized
many) (Perrow 2007). At the same time, talk to be measurable and malleable in terms of the
of safety culture suggests that the risks asso- attitudes and behaviors of individual actors, of-
ciated with increased efficiency and profitabil- ten the lowest-level actors, with least authority,
ity can be responsibly managed and contained. in the organizational hierarchy. The third cate-
The literature on safety culture traces its prove- gory of culture as emergent and indeterminate
nance to the copious work on risk assessment critiques claims that safety culture can be confi-
and systems analysis, system dynamics, and sys- dently instrumentalized to prevent catastrophic
tems engineering that became so prevalent over outcomes from complex technologies. This sec-
the past 30 years.2 At the outset, paying attention suggests that future research on safety in
tion to culture seems an important and valu- complex systems should explore just those fea-
able modification to what can be overly abstract tures of complex systems that are elided in the
and asocial theories of work and organization. talk of safety culture: normative heterogeneity
Despite this important correction, research on and cultural conflict, competing sets of inter-
safety culture usually ignores the historical- ests within organizations, and inequalities in
political context, the structural relationships, power and authority. Rather than imagine com-
plex yet homogeneous local cultures, research
should explore how struggles among competing
2
interests are part of the processes of cultural
Risk and systems analysis pervades contemporary organiza-
tions from manufacturing, transportation, and communica- production and how normative heterogene-
tions to finance, health, and education. ity, structured competition, and countervailing
www.annualreviews.org Safety and Culture 343

ANRV381-SO35-17 ARI 5 June 2009 9:28
centers of power can contribute to, rather than However, accidents alone cannot be driving
undermine, safer technologies. the recent attention to safety culture. Techno-
logical accidents are not new phenomena, and
safety has been a lively concern since the middle
HISTORICAL SHIFTS: of the nineteenth century, if not earlier. Indeed,
CONSTRUCTING AND in some accounts, much of the regulatory appa-
DECONSTRUCTING ratus of the modern state was institutionalized
SAFETY NETS to protect against the injurious consequences of
Why has attention to safety culture arisen at industrial production by setting minimally safe
this historical moment? conditions of work, establishing private actions
Any answer must begin by acknowledg- at law, and spreading the risks (of what could
ing the technological catastrophes of the past not be prevented) through fair trade practices,
40 years: Three Mile Island, Bhopal, Cher- workmens compensation, and pension systems,
nobyl, the Challenger and Columbia accidents as well as labor unions, private mutual help,
at NASA, the Exxon Valdez oil spill, oil rig and insurance. Safety was one of several objec-
accidents, Buffalo Creek, contaminated blood tives promoted by the system of instruments
transfusions, and a host of less spectacular dis- regulating relations between capital and labor
asters (Ballard 1988; Davidson 1990; Erikson (cf. Baker 2002, Ewald 2002, Friedman 1967,
1978; Fortun 2001; Jasanoff 1994; Keeble 1991; Orren 1991, Welke 2001, Witt 2004).
Kurzman 1987; Medvedev 1992; Petryna 2002; In a sense, the invention of risk,3 and with it
Rees 1994; Setbon 1993; Stephens 1980; Stern widespread insurance and regulation of work-
1976/2008; Vaughan 1996, 2003, 2006; Walker places, products, and markets, created the ba-
2004). sis of a new social contract. Responsibility was
In each instance, the accident was usually transferred from the person to the situation
explained as just that, an accidentnot a sys- the job, the firm, the union, or the collective
tem or design failure, but the result of some nationforgoing reliance on any individuals
extraneous mistake or mismanagement of a ba- behavior, whether worker or boss. Eschewing
sically well-conceived technology. Because the interest in specific causality, and thus individ-
systems in which the accidents occurred are om- ual liability, this collectivized regime acknowl-
nipresent, the recurring accidents undermine edged a general source of insecurity in tech-
confidence that catastrophes can be avoided. nology and responded with a set of generalized
Alongside concerns about genetically modified responses, albeit after extended and sometimes
foods, the toxicity of commonly used house- tragic struggle. Where responsibility had pre-
hold products, the migration of various syn- viously rested on the idea of proximate cause
thetic compounds from plants through animals and a selective distribution of costs based on
into the human body, the rapid spread of dis- liability as a consequence of imprudence, the
ease and contamination through porous and late nineteenth and early twentieth century in-
swift global transportation routes, and human- dustrial and business regulation redistributed
produced environmental degradation, techno- costs to collectivities, offering compensation
logical accidents feed a deepening mistrust of and reparation, if not safety and security. Re-
science ( Jasanoff 2005). If, as Hughes (1998) sponsibility was no longer the attribute of a
suggests, the invention of postmodern systems subject, but rather a consequence of a social
rescued Prometheus from the technological
disillusionment of the 1960s and 1970s, per-
haps the promotion of safety culture responds 3
Accounts vary as to the moment when probabilistic calcu-
to a renewed technological skepticism in the lation about hazardous events became a recognized practice
twenty-first century. (see Hacking 1990, 2003).
344 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
fact (Ewald 2002, p. 279). One was no longer & Lester 2004; Lash & Wynne 1992, p. 4).
responsible because one is free by nature and Heimer (1985) identified the illusory nature
could therefore have acted differently, but be- of this supposed realism in her prescient anal-
cause society judges it fair to place responsi- ysis of the reactive nature of risk, demon-
bility in a particular social location, that is, to strating how risk (probabilities of threats to
cause a particular person or collectivity to bear safety and security) would necessarily elude our
the financial costs of the injury. In short, the grasp because each effort to control risk trans-
costs of technological consequences were dis- formed its probabilities in an ever-escalating
persed, the source and foundation of respon- spiral.
sibility . . . displaced from the individual onto Embracing risk also refers to the specific
society (p. 279). policies and techniques instituted over the past
Talk about safety culture offers a new twist, several decades to undo the system of collective
or possible reversion, in the allocation of re- security. Across a wide range of institutions, of-
sponsibility for technological failures, a return ficials are now as concerned about the perverse
to the nineteenth century and earlier regimes effects of . . . risk shifting [i.e., risk sharing], as
of individual responsibility, but in a context of they are about the risks [probabilities of hazard]
more hazardous and global technologies. Af- being shifted (Baker & Simon 2002, p. 4). In
ter several decades of sustained attack by ad- place of the regime of risk containment, propo-
vocates seeking supposedly more efficient and nents of flexibility argue that safety and security
just allocations of goods through unregulated can be achieved more effectively by embracing
markets, the regime of collective responsibil- and privatizing risk.
ity has been dismantled, replaced by one of Although pro-privatization market policies
institutional flexibility. Rather than attempt- that attempt to make people more individu-
ing to mitigate and distribute risk, contempo- ally accountable for risk (Baker & Simon 2002,
rary policies and practices embrace risk (Baker p. 1) are often justified as natural and efficient,
& Simon 2002, p. 1). Embracing risk means there is nothing natural about them (Klein
to conceive and address social problems in 2007, Mackenzie 2006). Just as risk-spreading
terms of riskcalculated probability of haz- was achieved through the efforts of financial and
ard (Heimer 1988, Simon 1988). Human life, moral entrepreneurs to transform common, of-
including the prospects of human autonomy ten religious, conceptions of morality, respon-
and agency, is now conceived in very much the sibility, and money (Becker 1963; Zelizer 1979,
same way and analyzed with the same tools we 1997), contemporary risk-embracing policies
employ to understand and manipulate physical are also the outcome of ideological struggles. If
matter: ordered in all its important aspects by in the nineteenth century marketing life insur-
instrumental and probabilistic calculation and ance required a modification in what it meant
mechanical regulation (Bittner 1983). to protect ones family by providing materially
Unfortunately, risk analysis and discourse for them after death rather than seeming to
narrow consideration of legitimate alterna- earn a profit from death, so too risk-embracing
tives while nonetheless sustaining the appear- policies in the twentieth and twenty-first cen-
ance of broad pluralism (cf. Habermas 1975). turies require a similar redefinition in what it
Because of the assumption that realism resides means to be responsible, productive citizens.
exclusively in science, reflexive observation Contemporary moral entrepreneurs energeti-
and critique as well as unmeasured variables cally promote risk taking rather than risk shar-
are excluded from official risk discourses. As a ing as morally desirable; the individual more
consequence, allegedly empirical analyses be- effectively provides for family security, it is
come solipsistic, focusing exclusively on the claimed, by participating in a competitive, ex-
methods and epistemologies that are inter- panding, market economy than by relying on
nal to technological instrumentalism (Deutch government-constructed safety nets.

ANRV381-SO35-17 ARI 5 June 2009 9:28
This moral entrepreneurship directs our of the regulatory regime. For whatever reasons,
attention to safety culture because the concept ideological or coincidental, the focus on mar-
arises as a means of managing technological ket competition as the central guarantor of pro-
risk, just as the previous security regime has ductivity and efficiency overlooked constituent
been successfully dismantled. This is not to structural features of the regime of government
say that the nineteenth to twentieth century regulation, insurance, and liability that miti-
regulatory system was perfect, nor as good as gated risk by promoting countervailing inter-
it might have been, nor that it prevented or ests in safety and responsibility.
repaired all or most technological damage. It Notably, the nineteenth to twentieth cen-
was, however, a means of distributing, if not tury solidarity regime was not only a paradigm
preventing, the costs of injuries. Yet, for most of of compensation but also one of prevention
the twentieth century, risk analysts themselves (Ewald 2002, p. 281). Bottom-line profit taking
expended a good part of their energy attacking required diligent efforts not simply to estimate
this system, legitimating the risks undertaken, costs and prices but also to prevent losses, that
reassuring the public that they were nonethe- is, accidents and disasters. A host of institutional
less being protected, and second-guessing the practices and organizations promoted respon-
regulatory agencies attempts to do a very sibility by enacting prevention, in this way re-
difficult job (Perrow 1999 [1984], p. 307). ducing costs and increasing profit. For example,
Paradoxically, many risk analysts regularly
the great life insurance companies were pi-
assessed the risks of regulation more negatively
oneers in epidemiology and public health.
than the risks of the hazards themselves (e.g.,
The fire insurance industry formed Under-
Deutch & Lester 2004).
writers Laboratories, which tests and certi-
With a commitment to the idea of effi-
fies the safety of household appliances and
cient markets, critics of regulation produced
other electrical equipment. Insurance compa-
accounts of government regulation as publicly
nies seeking to cut their fire losses formed the
sanctioned coercion sought by private firms to
first fire departments. More recently, health
consolidate market power, inhibit price com-
insurance companies have been behind many
petition, and limit entry. As a result, critics ar-
efforts to compare, test, and measure the ef-
gued, the system produced inefficiencies, a lack
fectiveness of medical procedures (Baker &
of price competition, higher costs, and over-
Simon 2002, p. 8; cf. Knowles 2007a,b).
capitalization ( Joskow & Noll 1977, Joskow
& Rose 1989; cf. Schneiberg & Bartley 2008). Under the solidarity regime, industries, indi-
Interestingly, these challenges to government vidual firms, and labor unions collectively pro-
regulation rarely valued as highly consumer ser- moted forms of social control, workplace disci-
vice, product quality, and environmental pro- pline, and self-governance that were expected
tection that were also promoted by regulation. to reduce injuries and thus costs for the various
The accounts of corporate capture undermin- organizations (Ericson et al. 2003). Minimally,
ing regulatory effectiveness (Bernstein 1955; they identified the worst offenders.
Derthick & Quirk 1985; Peltzman et al. 1989; Insurance companies have traditionally also
Vogel 1981, 1986) also ignored the new so- taken precautions to mitigate financial losses
cial regulation in safety, consumer protection, not only through safer practices but through
and civil rights. Perhaps the focus on market investment of premiums and reinsurance. The
control, and a latent hostility to the struggles post-1929 American banking and financial
between labor and capital and between man- industry regulations purposively segregated dif-
ufacturers and consumers that became ideo- ferent financial functions and markets to pre-
logically entwined with the struggles against vent excessive losses in one activity from
regulation, blinded scholars to non-economic contaminating related industries and parallel
variables such as safety that had also been part silos in the financial markets. However, since
346 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
the systematic deconstruction of this regulatory Although strict liability is not the generously
regime began in the 1980s, insurance firms, like absurd protector of irresponsibility that critics
many corporations, have become ever more fi- claim it to be (Burke 2004, Holtom & McCann
nancialized, earning profit more directly from 2004), there is no doubt that the twentieth cen-
investments in global financial markets than tury produced, by any measure, a great deal
from selling insurance. With the invention of more law (Galanter 2006, p. 5). The legal pro-
derivatives and similar instruments, a wider ar- fession exploded from 1 lawyer for every 627
ray of firms have been transformed into finan- Americans in 1960 to 1 lawyer for every 264
cial rather than productive entities. Financial- in 2006. Spending on law increased, as did cel-
ization means that capital and business risks are ebration of lawyers and legal work in popular
disaggregated, recombined in heterogeneous media and film ( J. Silbey 2001, 2004, 2005,
assets that are bought and sold globally, and 2007a,b).
distributed among myriad other firms, share- In canonical Newtonian fashion, the expan-
holders, and markets. Losses in these assets sion of law caused an energetic backlash. The
are supposedly protected through insurance early and mid-twentieth century cries that the
swaps. There is an indirect but substantial con- legal system failed to provide justice to the
sequence for safety in this financialized system weakgave way to a responsive critique that
because there is less interest in the reliability of the nation was afflicted by too much law
the specific products manufactured or services (Galanter 2006, p. 5, citing Galanter 1994). One
offered. Less financial risk means reduced at- alleged legal crisis followed another, from prod-
tention to the associated practices that encour- uct liability to overcrowded courts to medical
age risk prevention and enhance safety.4 malpractice (Baker 2005). Calls for tort reform
Finally, we cannot ignore the role of civil lit- and informal dispute resolution as alternatives
igation as part of the twentieth century solidar- to litigation became common, the centerpiece
ity regime and its twenty-first century demise. of organized professional and political cam-
The expansion of rights and remedies that be- paigns (Burke 2004, Silbey & Sarat 1988). With
gan slowly with the New Deal but grew rapidly Ronald Reagans election to the U.S. presidency
postWorld War II came with a great a burst of in 1980 and subsequent Republican presidents,
legalization. While regulation proliferated, nominees to the federal courts were systemati-
extending to aspects of life previously unsu- cally screened for their ideological conformity
pervised by the state (Galanter 2006, p. 4), with a less law, less rights agenda. By Septem-
civil litigation independently generated rights. ber 2008, 60% of active federal judges with
Although some commentators describe this as this agenda had been appointed, and, as a con-
a litigation explosion (Friedman 1985, Kagan sequence, the federal courts have joined the
2003, Lieberman 1981), it is actually a shift: movement to embrace risk, becoming another
from contract litigation dominating in the nine- voice promoting individual, rather than shared,
teenth century to tort litigation predominat- assumption of risk (Scherer 2005).
ing in the twentieth century (Galanter 1983). Thus, from the middle nineteenth through
the late twentieth centuries, industrial and in-
surance firms, individual families, the civil liti-
4
The financial downturn that escalated to a worldwide cri- gation system, governmental regulatory agen-
sis in 2008 can be attributed in part to just these practices. cies, and labor unions built and sustained a
In the financial markets, not only was the safety of the pro- safety net of collective responsibility; they re-
duced material goods less salient, but the safety or security
of the financial assets was of less concern because of default inforced each other within a tapestry of or-
swaps, hedging, and insurance on bets that finally unraveled. ganizations and institutions whose interests
Rather than encouraging responsibility, the layered system competed, yet coalesced to support relatively
of disaggregation and recombination buttressed by hedges
and insurance undermined critical or responsible decision safer practices. The demise of those structural
making. components is precisely what underwrites the

ANRV381-SO35-17 ARI 5 June 2009 9:28
contemporary focus on safety culture as a means expression of responsibilization, this neo-

of managing technological hazards. If we do not liberal technique of governance. Without nec-
have empowered regulatory agencies, judicial essarily intending to promote policies of dereg-
support for tort litigation, organized labor, and ulation and privatization, the celebration of
insurance companies with a financial interest in safety culture as a means of managing the
the safety and longevity of their customers, we hazardous consequences of complex systems
have lost a good part of what made the previous expresses what Weber described as an elec-
paradigm work to the extent it did for as long tive affinity, phenomena that do not necessar-
as it did. ily cause one another but nonetheless vary to-
Talk of safety culture flourishes at the very gether. In the next section, I explore calls for
moment when advocates extend the logic of in- and accounts of safety culture to extract from
dividual choice, self-governance, and rational this diverse literature the purported meanings
action from the market to all social domains. and relationships of safety and responsibility.
Just as historic liberalism was concerned with

setting limits on the exercise of political or
public authority, viewing unwarranted inter- TALK ABOUT SAFETY CULTURE
ventions in the market as harmful, contem- Between 2000 and 2007, academic literature
porary neoliberalism5 promotes markets as a and popular media exploded with references
principle not only for limiting government but to safety culture. Over 2250 articles in news-
also for rationalizing authority and social rela- papers, magazines, scholarly journals, and law
tions in general (Shamir 2006, p. 1). Through reviews in an eight-year period included ref-
a process of so-called responsibilization, pre- erences to safety culture, whereas only 570
disposing social actors to assume responsibility references were found in the prior decade.
for their actions (Shamir 2008, p. 10), these Before 1980, I could find no references in pop-
policies simultaneously empower individuals to ular or academic literature.6 Although the un-
discipline themselves while distributing, as in precedented appearance and the rapidly esca-
the nineteenth century prudential regime, to lating use of the concept seem to support my
each the costs of that discipline and the conse- hypothesis of ideological affinities between talk
quences for the lack thereof (Rose 1989, 1999). about safety culture and the dismantling of the
As a concept, responsibilization names efforts regulatory state, we should look more closely
to both cultivate and trust the moral agency of at what people say to interpret what they mean
rational actors as the foundation of individual when they speak about safety culture.
and collective well-being (Shamir 2008, p. 11). The earliest uses of safety culture in newspa-
Because the propagation and inculcation of pers and popular media invoke the term primar-
safety culture is only one approach to enhancing ily in discussions of nuclear power, energy gen-
the reliability and safety of complex technolo- eration, and weapons production to describe
gies, it is not unreasonable to wonder whether within organizations an ingrained philosophy
safety culture, focused on individual partici- that safety comes first (Diamond 1986). One
pants self-determined contributions to the sys- non-nuclear reference to a British railroad ac-
tem as a whole, might not be described as an cident is illustrative because, even in this less
common venue, a deteriorating safety culture
was offered as the explanation for what went
5
The term neoliberalism is conventionally used to refer to the
policies advocating deregulation, privatization, and reliance
on markets for both distribution and coordination, but also
6
includes a set of fiscal, tax, and trade liberalization policies I searched LexisNexis, JSTOR, and the Engineering Village
that is sometimes referred to as the Washington Consensus databases for the years between 1945 and 2008, using the
because of support by the International Monetary Fund and phrases safety culture, safety (and) culture, and culture of
the World Bank. safety within two words of each other.
348 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
wrong and should be improved to prevent fu- elsewhere. Its difficult to enforce our culture
ture accidents.7 Mechanical error compounded on another country, Holtzman said, espe-
by lax management processes was named as the cially when the other country seems willing to
cause of the accident. Nonetheless, the judge take risks in exchange for speedy technological
heading the accident inquiry focused his rec- advance (Kiefer 1984).
ommendations for improving the safety culture This use of safety culture to name variations
not on the management of the system or the in national cultures, reminiscent of historic jus-
communications processes within the railroad tifications for colonial rule, did not stick. Very
hierarchy, but on the laborers, calling for radi- quickly, it became apparent that the preexist-
cal improvements in recruiting and training and ing safety problems in the Bhopal plant were
an end to excessive overtime (Diamond 1986). not peculiar to Bhopal, or to India. Although
Although talk about safety culture emerged Union Carbide had insisted that the conditions
during the 1980s when major accidents at in Bhopal were unique, one of its sister plants
Three Mile Island, Bhopal, and Chernobyl in Institute, West Virginia, produced a sim-
weakened public confidence in complex tech- ilar accident just eight months later (Perrow
nologies, only well into the 1990s did talk about 1999 [1984], p. 358). Although an Occupational
safety culture become a common phenomenon. Safety and Health Administration (OSHA) in-
Although thousands of newspaper articles were spection had previously declared the West
written about the March 28, 1979, partial melt- Virginia plant in good working order, the
down of Unit 2 at the Three Mile Island nuclear OSHA inspection following the explosion
power plant in Dauphin County, Pennsylvania, declared that this was an accident waiting
none spoke about the plants safety culture. We to happen, citing hundreds of longstand-
first see accounts of lax safety culture following ing, constant, willful, violations (quoted in
the December 3, 1984, explosion of a Union Perrow 1999 [1984], p. 359). Clearly, the dif-
Carbide plant synthesizing and packaging the ferent national cultures of India and the United
pesticide methyl isocyanate in Bhopal in the States could not explain these accidents, which
Indian state of Madhya Pradesh. seemed to have had some other source. No one
In these early references, the phrase is in- mentioned the role of lax inspections as part of
voked primarily to denote culture in its more the safety culture. With the exception of one
colloquially circulating meaning: to suggest story about how E.I. DuPont de Nemours &
that nations vary in their respect for safety. Co is recognized within industry for its [exem-
Because the Indian partners in the Union plary] safety practices (Brooks et al. 1986), the
Carbide plant did not share the American cul- early references in popular media to safety cul-
ture (which implicitly valued safety), they were, ture do little more than invoke the term. They
by inference, responsible for the accident. John provide little specification of what activities, re-
Holtzman, spokesman for the Chemical Man- sponsibilities, or symbolic representations con-
ufacturers Association in Washington, DC, tribute to a safety culture.
pointed to the differences in safety culture be- In professional and scholarly literature, the
tween the US and other countries. . . . We have phrase safety culture first appears in a 1986
a certain sense of safety. You see it in campaigns report of the International Atomic Energy
like buckle up. Its not necessarily the same Agency (IAEA) on the Chernobyl accident.
Three years later, a second reference by the
U.S. Nuclear Regulatory Commission (1989)
7
During the morning rush hours of December 12, 1988, states that plant management has a duty and
35 people were killed and another 100 injured when one obligation to foster the development of a safety
commuter train rammed the rear of a stopped commuter culture at each facility and throughout the fa-
train, outside busy Clapham Junction in south London. The
wreckage was then struck by a freight train (Associated Press cility, that assures safe operations. After five
1989). years in common usage, an IAEA report defined

ANRV381-SO35-17 ARI 5 June 2009 9:28
safety culture as that assembly of character- human interactions share symbolic and cogni-
istics and attitudes in organizations and indi- tive resources, many cultural resources are dis-
viduals which establishes that, as an overriding crete, local, and intended for specific purposes.
priority, nuclear power safety issues receive at- Nonetheless, it is possible (c) to observe general
tention warranted by their significance (IAEA patterns so that we are able to speak of a culture,
1991, p. 8; 1992). As Perin (2005) comments in or cultural system, at specified scales and lev-
her detailed study of four nuclear power plants, els of social organization. System and practice
Determining that significance in particular are complementary concepts: each presupposes
contexts is . . . the crux of the quandary (p. 14). the other (Sewell 2005, p. 164),8 although
For the past two decades, researchers have the constituent practices are neither uniform,
been actively engaged in analyses of safety cul- logical, static, nor autonomous. As a collec-
ture, with the vast majority of work produced in tion of semiotic resources deployed in interac-
engineering, management, and psychology, and tions (Swidler 1986), culture is not a power,
a smattering of mostly critical work produced something to which social events, behaviors,
in sociology and political science. If we look institutions, or processes can be causally at-
across these fields, we find variation in the ways tributed; it is a context, something within which
in which safety culture is invoked, although [events, behaviors, institutions, and processes]
there is a great deal of conceptual importation can be intelligiblythat is, thicklydescribed
from the social sciences to what we may think (Geertz 1973, p. 14). (d ) Variation and conflict
of as applied social science in engineering and concerning the meaning and use of these sym-
management. bols and resources is likely and expected be-
cause at its core, culture is an intricate system
of claims about how to understand the world
The General Concept of Culture and act in it (Perin 2005, p. xii; cf. Helmreich
Culture is an actively contested concept; its 2001).
importation into organizational and engineer-
ing analyses is equally contentious. Confusion
derives in part from intermingling two mean- Culture as Causal Attitude
ings of culture: a concrete world of beliefs and For some authors, safety culture is understood
practices associated with a particular group and as a measurable, instrumental source composed
an analytic tool of social analysis referring to of individual attitudes and organizational be-
a system of symbols and meanings and their havior, or conversely as a measurable product
associated social practices, both the product of values, attitudes, competencies, and behav-
and context of social action. The analytic con- iors that are themselves the cause of other ac-
cept is invoked (a) to recognize signs, perfor- tions (Cox & Cox 1991, Geller 1994, Glennon
mances, actions, transactions, and meanings as 1982, Lee 1996, Ostrom et al. 1993). In
inseparable, yet (b) to disentangle, for the pur- both uses, culture determine[s] the commit-
pose of analysis [only], the semiotic influences ment to, and the style and proficiency of,
on action from the other sorts of influences an organizations health and safety programs
demographic, geographical, biological, techno-
logical, economic, and so onthat they are nec-
essarily mixed with in any concrete sequence of 8
The employment of a symbol, Sewell (2005, p. 164) writes,
behavior (Sewell 2005, p. 160). Thus, orga- can be expected to accomplish a particular goal only because
symbols have more or less determinate meaningsmeanings
nizational culture and safety culture are terms specified by their systematically structured relations to other
used to emphasize that organizational and sys- symbols. But it is equally true that the system has no exis-
tem performances are not confined to formally tence apart from the succession of practices that instantiate,
reproduce, ormost interestinglytransform it. Hence, a
specified components, nor to language alone. system implies practice. System and practice constitute an
Although formal organizational attributes and indissoluble duality or dialectic.
350 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
(Reason 1997, p. 194, citing Booth, UK Health (Beamish 2002; Bourrier 1996; Carroll 1998a,b;
and Safety Commission 1993). Whether the Cooper 2000; Schein 1992), and both organi-
first mover or an intermediate mechanism, an zational and safety culture developed alongside
ideal safety culture is the engine that continues concepts of organizational climate and safety
to propel the system toward the goal of max- climate, generating a bewildering mix of con-
imum safety health, regardless of the leader- cepts and measures. Numerous efforts have at-
ships personality or current commercial con- tempted to parse these terms, with negligible
cerns (Reason 1997, p. 195). Culture as the theoretical advance (Denison 1996, Zhang et al.
ultimate, intermediate, or proximate cause of- 2002).
ten leaves unspecified the particular mechanism To some extent, the conceptual puzzle is en-
that shapes the safe or unsafe outcomes of the ergized by occupational and professional com-
organization or technology (but see Glennon petitions, different disciplinary communities
1982, Zohar 1980), with much of the man- pushing in one direction or another, using pre-
agement and engineering literatures debating ferred concepts and tools to authorize expert
exactly this: how to operationalize and mea- advice about how to design systems, assess per-
sure both the mechanism and the outcome. formance, and manage them on the basis of this
Clearly, this conception of safety culture be- information (Abbott 1988). Although organiza-
lies exactly that thick description of practice tional and safety culture can and should be nor-
and system that cultural analysis entails (Fischer matively neutral, the terms have usually been
2006, Geertz 1973, Silbey 2005b). deployed to emphasize a positive aspect of or-
A persistent muddle in this usage derives, in ganizations, one that leads to increased safety
part, from the aggregation over time and across by fostering, with minimal surveillance, an effi-
professional communities of concepts devel- cient and reliable workforce sensitized to safety
oped to name the emergent properties of social issues. The framing generates ellipses that in-
interactions not captured by the specification vite further conceptual elaboration to account
of components, stakeholders, objectives, func- for what has been excluded in the particular
tions, and resources of formal organizations. normative tilt of the concept. Although some
There seems to be a recurring cycle in which authors view culture as something that can be
heretofore unnamed or unperceived phenom- changedmanaged to improve organizational
ena are recognized as playing a role in organized performanceand seek to develop models to
action. A construct is created to name what ap- generate more effective safety culture (Carroll
pear to be stable, multidimensional, shared fea- 1998b, Cooper 2000), others adopt more disin-
tures of organized practices that had not yet terested formulations (Beamish 2002, OReilly
been captured by existing categories and mea- & Chatman 1996).
sures. All this is fine and congruent with the Guldenmunds systematic review of the lit-
best sociology. However, once the phenomena erature through 2000 describes organizational
are named, some researchers attempt to spec- and safety culture as general frames determin-
ify and measure them more concretely; dis- ing organizational and safety climates.
parate results generate continuing debate about
different conceptualizations and measurement The term organizational climate was coined to
tools (Cooper 2000, Guldenmund 2000). As refer to a global, integrating concept underly-
empirical results outpace the purportedly de- ing most organizational events and processes.
scriptive models, new constructs are offered to Nowadays, this concept is referred to by the
name the persistent, yet elusive effluent of un- term organizational culture whereas organi-
predicted events, now hypothesized as intan- zational climate has come to mean more and
gible cultural causes, fueling additional debate. more the overt manifestation of culture within
Thus, talk of safety culture emerged as a sub- an organization. Therefore, climate follows
set from prior talk about organizational culture naturally from culture, or, put another way,

ANRV381-SO35-17 ARI 5 June 2009 9:28
organizational culture expresses itself through and management toolmakers (Humphrey et al.
organizational climate (Guldenmund 2000, 2007, Morgeson & Humphrey 2006). The
p. 221). resulting literature is littered with competing
models, instruments, types of analysis, and
Summarizing across dozens of uses, Zhang measures, providing much occupation but
et al. (2002, p. 8) suggest that safety culture be unreliable instruction or guidance (Cooper
understood as 2000, Guldenmund 2000), offering more heat
than light (OReilly & Chatman 1996, p. 159).
the enduring value and priority placed on Nonetheless, the repeated efforts to specify and
worker and public safety by everyone in ev- measure safety culture in terms of individual
ery group at every level of an organization. It attitudes and behaviors to foster a reliable
refers to the extent to which individuals and workforce who will commit to personal re-
groups will commit to personal responsibil- sponsibility (Zhang et al. 2002) illustrates well
ity for safety, act to preserve, enhance and affinities between policies of responsibilization
communicate safety concerns, strive to ac- and advocacy of safety culture.
tively learn, adapt and modify (both individual The Report of the BP U.S. Refineries Inde-
and organizational) behavior based on lessons pendent Safety Review Panel (Baker Panel 2007,
learned from mistakes, and be rewarded in a hereafter BP Report 2007) marks, perhaps, the
manner consistent with these values. quintessence of talk about safety culture. Fol-
lowing a catastrophic accident at a BP refin-
Although these uses of safety culture refer ery in Texas City, Texas, on March 23, 2005,
to the shared values, beliefs, assumptions, resulting in 15 deaths and more than 170 in-
and norms which may govern organizational jured persons, as well as significant economic
decisionmaking . . . about safety (Ciaverelli & loss, the U.S. Chemical Safety and Hazard In-
Figlock 1996), much of the research observes, vestigation Board recommended, with explicit
measures, and assesses safety culture through urgency, that BP initiate its own parallel inves-
survey instruments collecting individual ex- tigation into its safety management practices.9
pressions, attitudes, and beliefs, or what others The BP Report, issued 21 months later, de-
define as safety climate. In effect, the terms scribed what it repeatedly called a damaged
are collapsed, so that safety climate becomes safety culture. The phrase safety culture ap-
the temporal state measure of safety culture, pears 3 times on the opening page and more
assessed and evaluated in terms of the degree than 390 times in the approximately 150-page
of coherence and commonality among indi- document. Clearly, safety culture has become
vidual perceptions of the organization (Zhang the mantra for technologically complex and
et al. 2002, p. 10). Some studies recognize the hazardous organizations.
inadequacy of assessing a diffuse, emergent The report claims early and often that BP
phenomenon such as culture through individ- has come to appreciate the importance of cul-
ual measures, and as a consequence add a group tural factors in promoting good process safety
or aggregate measure to designate that which performance (BP Report 2007, p. 59) but
is shared (Cox & Cox 1991) or applies to the nonetheless adopted a rather shallow notion
group (Lee 1996), the set (Pidgeon 1997, 1998), of safety culture that focused on individual ac-
or the assembly. Because a good part of the lit- tions rather than on systemic processes. Thus,
erature on safety culture seeks to develop tools alongside safety culture, the first of its three
to improve organizational performance, small
linguistic variations in conceptualization of the
9
often intangible system of signs and practices, BP had experienced two other fatal safety incidents in 2004,
a major process-related hydrogen fire on July 28, 2005, and
even if not named as such, become critically de- another serious incident on August 10, 2005 (BP Report
terminant variations for empirical researchers 2007).
352 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
high-level findings, the report identified the get for investment. Because BP relied heavily
need for process safety management systems on individual injury rates to assess safety perfor-
and performance evaluation, corrective action, mance and because these personal safety indica-
and corporate oversight. The report repeatedly tors showed improvement, BP was mistakenly
states that, in contrast to the individualized no- confident that it was addressing process (design
tion of safety promoted by BP, safety is rather and management) risks. The report nicely high-
the responsibility of the corporate board, which lights a feature of the culture that mistakenly
must exercise leadership by establishing safety defined safety as individual responsibility.
as a core value across all its refineriesexactly
what it had failed to do. Absent a healthy safety
culture, even the best safety management sys- Culture as Engineered Organization
tems will be largely ineffective (BP Report Other scholars speak less about organiza-
2007, p. 59). tional or safety culture in general than specif-

Thus, the report specifically distinguishes ically about an organizations learning culture
between personal safety (slips and falls) and pro- (Carroll 1998a,b), especially in high-reliability
cess safety, that is, safety built into the design organizations (HROs) (Eisenhardt 1993, Klein
and engineering of the facilities, management, et al. 1995, La Porte & Consolini 1991, La
and maintenance, exactly what BP had failed to Porte & Rochlin 1994, Roberts & Rousseau
do. In its own inquiry, BP had used interview 1989, Roberts et al. 1994, Rochlin et al. 1987,
and survey data that revealed significant vari- Schulman 1993, Weick 1987, Weick et al.
ation in attitudes and perceptions about safety 1999). Like the previous category, however, the
within and across five plant sites. On the ba- main focus of these authors has been to under-
sis of these attitudinal variations, BP declared stand how culture leads to particular outcomes,
its process safety damaged. By relying on at- specifically, reliability and efficiency. Again, cul-
titudes as indicators, BP clearly identified in- ture is instrumentalized in order to manipulate
dividuals rather than the system as the central and manage its consequences. This work dif-
safety focus. In contrast, the panel report de- fers from the previous category, however, by its
scribed systemic problems such as underfunded explicit articulation of the organizational con-
management of the U.S. plants and underre- figuration and practices that should make orga-
sourced safety programs alongside an abun- nizations more reliably safe. Nonetheless, the
dance of discrete safety initiatives that over- HRO literature also seems to invoke a notion
loaded and underresourced management and of culture as homogeneous and instrumentally
workers. Similarly, in keeping with the respon- malleable.
sibilization of lower-level workers, BP had cited HRO analysts suggest that good organiza-
worker fatigue as one of the root causes of the tional design with built-in redundancies, de-
Texas City accident, despite the fact that fa- centralized decision making for prompt in situ
tigue was common across sites, including those responses, and extensive training alongside
that did not experience a major accident. The trial-and-error learning can create high re-
panel report argued that BPs focus on individ- liability, that is, safety, even in organiza-
ual behaviors and errors ignored and failed to tions with particularly hazardous technologies
address the root causes of accidents: fatigue and (e.g., Marone & Woodhouse 1986, Weick &
sensory overload due to management policies, Sutcliffe 2001). Continuous operations and
in this instance, specifically policies that relied learning that allow for backup to compen-
on routine overtime to meet production needs sate for failures will lead, HRO theorists ar-
rather than on hiring additional employees. By gue, to reduced error rates and safer outcomes.
emphasizing personal safety, the report claims Organizational learning takes place through
that BP leadership failed to establish process trial and error, supplemented by anticipa-
safety or system safety as a core value and tar- tory simulations. Although HRO scholars most

ANRV381-SO35-17 ARI 5 June 2009 9:28
often describe the prevalence of these condi- (but see Klein et al. 1995). Nonetheless, slip-
tions in military-style organizations, each of pages toward instrumental conceptions of cul-
which holds itself to a failure-free standard of ture and aspirations for homogeneously dis-
performance, (e.g., nuclear submarines, nuclear tributed cognitive capacities also appear in the
power plants, aircraft carriers, space shuttles, corpus of HRO scholarship. Because the HRO
and air traffic control), they argue that such is offered as a model for reliably safe perfor-
organizational practices and what are called mance, it becomes essential to operationalize
processes of collective mindfulness are appro- with increasing specificity the particular mech-
priate in nonmilitary organizations as well. anisms that will ensure that performance, push-
HROs provide, authors claim, a unique win- ing a less reductionist model of culture toward
dow into organizational effectiveness under try- the same limitations as the culture-as-cause
ing, dangerous (Weick et al. 1999, p. 81), and literature. Although many organizations share
high-velocity environments (Eisenhardt 1993). the named characteristics of high reliability,

HRO scholars focus on what they claim is these are apparently insufficient for prevent-
a distinctive though not unique set of cogni- ing accidents because not all such organiza-
tive processes that prevail in the better plants tions are reliably safe. When organizations
and systems. These five orientations (preoccu- that ought to be highly reliablebecause they
pation with and proxies for failure, reluctance exhibit the specified characteristicsare not,
to simplify interpretations, sensitivity to op- authors either claim poor management or al-
erations, commitment to and capabilities for ter the criteria of success. What is now called
resilience, and resistance to over-structure or high-reliability theory (HRT) is a response that
preference to under-specify the system) lead transforms a prescription into a hypothesis, in
to mindfulnessa feature of safe organizations some cases by transforming the independent
(Weick et al. 1999, pp. 83, 88). Mindfulness can variables naming organizational processes, and
be understood in terms of the quality and al- at other times reframing the dependent variable
location of scarce attention, the repertoire of or the definition of reliability. Early formula-
action capabilities, or active information tions emphasized the total elimination of error,
searching (Westrum 1997), but is perhaps most absence of trial-and-error learning, a closed sys-
concisely described by Vaughan (1996, chap- tem buffered from environmental stresses, and
ter 4) as interpretive work directed at weak a singular focus on safety (Weick 1987, Weick &
signals. In contrast, Roberts 1993). Later versions basically inverted
the criteria to value the role of trial-and-error
when fewer cognitive processes are activated learning, learning from failures, the importance
less often, the resulting state is one of mind- of exogenous influences such as regulations and
lessness characterized by reliance on past cate- public perception, and the importance of mul-
gories, acting on autopilot, and fixation on tiple objectives alongside safety (e.g., safety and
a single perspective without awareness that service) (LaPorte & Consolini 1991, LaPorte
things could be otherwise. . . . [T]o say that an & Rochlin 1994).
organization is drifting toward mindlessness Empirical tests of HRT have challenged its
is simply another way of saying that the or- own reliability as well as validity. For exam-
ganization is drifting toward inertia without ple, Klein et al. (1995) compared the organiza-
consideration that things could be different tional cultures in a range of HROs to other or-
(Weick et al. 1999, p. 91). ganizations. Unlike most organizations, HROs
showed few hierarchical differences in cultural
If many culture-as-cause analyses describe norms, although there were differences across
safety culture shaping members safety atti- HROs. In a study restricted to two nuclear
tudes and behavior, HRO analyses adopt a plants, Bourrier (1996) found that the orga-
less reductionist or determinist epistemology nizations use quite different strategies in their
354 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
search for reliability and effectiveness, includ- increased the number of nuclear weapons rou-
ing coordination of workers and structuring of tinely deployed and circulating the globe.
tasks, variables specifically excluded in the basic Sagan (1993) also identifies conventional
HRO model. Although the lack of hierarchical features of organizations that impede learn-
variation emphasized the cultural homogene- ing, for example, the persistence and limitations
ity within individual organizations, research of bounded rationality that pervades garbage
failed to demonstrate a shared culture across can processes (Cohen et al. 1972). These adap-
organizations. tive, yet unscriptable decision-making practices
In perhaps the most important empirical prevail in many complex, porous organizations
tests, Sagan (1993) failed to substantiate HRTs where unstable environments, unclear goals,
fundamental premises. Using Freedom of In- misunderstanding, mis-learning, and happen-
formation Act petitions, Sagan scoured previ- stance prevail.
ously classified archives to discover why there Sagan shows that in each of the high-
have been no unintended explosions of nu- alert moments of crisis, protocols for ensur-
clear weapons. Is this a uniquely safe systema ing against accidental detonation were violated.
model of high reliability? Sagans analysis of two Because the supposed protections were not op-
moments of near nuclear war (Cuban Missile erating, only chance prevented nuclear detona-
Crisis, 1973 October defcon alert) and the loss tion. He demonstrates that there was no learn-
of a nuclear armed aircraft (1968 bomber crash ing among analysts or high-level officers from
near Thule Air Base, Greenland) reveals that one incident to the next. More importantly,
the system is anything but reliably safe. Even from my perspective, and a point I return to
the necessary, if not sufficient, conditions for below, Sagans work stresses the importance of
HRO failed; there were no accidents, how- competing group interests that undermined not
ever, although there were repeated near misses. only commitments to safety, but also HRTs
Sagans data discredit the fundamental features notions of homogeneous organizational culture
of the high-reliability model. He argues that and self-reflexive learning. Group rivalries led
redundancy, promoted by HRT to prevent ac- to limited communication, burying informa-
cidents, is often the cause of problems, espe- tion about what actually happened and imped-
cially when redundancy is added on rather than ing development of shared interpretations that
designed into systems. The bomber crash was can promote improvement over time. Concerns
the result of planned redundancy that was itself about organizational surveillance, fear of pub-
the source of near disaster. Planes loaded with licity about near misses, and inter- and intraser-
nuclear weapons routinely fly as a form of a re- vice rivalries produced a culture of informa-
dundant triad of U.S. strategic bombers, sub- tional secrecy among the military that leads to
marine launch ballistic missiles (SLBMs) and even more near misses10 (cf. Galison 2004).
intercontinental ballistic missiles (ICBMs), to
ensure that retaliation would be possible un-
der any conceivable circumstances in which the Culture as Emergent
U.S. might be attacked (Sagan 1993, p. 157). and Indeterminate
When a Strategic Air Command bomber with If optimism characterizes HRT, in effect sug-
nuclear weapons crashed on January 21, 1968, gesting that if we only try harder we will have
the conventional high explosive in all four of
the nuclear bombs went off. No nuclear det-
onation occurred but radioactive debris was 10
It may be worth remembering that just this kind of orga-
dispersed over a wide expanse (Sagan 1993, nizational competition and secrecy among law enforcement
agencies contributed to the failure to respond effectively to
p. 156; 1996). This near miss would not have intelligence information prior to the 9/11 World Trade Cen-
occurred without the built-in redundancy that ter disaster.

ANRV381-SO35-17 ARI 5 June 2009 9:28
virtually accident-free systems, more skeptical coupled and complex systems have limited tem-
scholars believe that no matter how hard we poral slack, substitutability, and response op-
try we will still have accidents because of intrin- tions. This conception of normal accidents
sic characteristics of complex/coupled systems directly challenges the high-reliability model
(Perrow 1999 [1984], p. 369). For those of intense discipline, rigid socialization, and
who eschew reductionist and instrumental isolation. However, even in analyzing military
conceptions, culture is understood to be organizations, which emphasize discipline, so-
emergent and indeterminate, an indissoluble cialization, and relative isolation from environ-
dialectic of system and practice. As such, the mental contamination, Sagan successfully chal-
consequences of safety culture cannot be engi- lenged HRT and extended the normal accidents
neered and only probabilistically predicted with model by emphasizing issues of bounded ra-
high variation from certainty. For scholars who tionality and interest competition within and
adopt this constitutive perspective, safety as a between organizations. Finally, in a very dif-
form of organizational expertise is . . . situated ferent setting (the 1980s savings and loan cri-
in the system of ongoing practices. . . . [S]afety- sis), Mezias (1994) also showed how tight cou-
related knowledge is constituted, institu- pling and complexity increased simultaneously
tionalized, and continually redefined and to produce catastrophic results in a nonmechan-
renegotiated within the organizing process ical but nonetheless complex technology.11
through the interplay between action and In a series of thickly described accounts,
reflexivity. Here, safety practices have both Vaughan (1996, 1999, 2003, 2004, 2005a,b,
explicit and tacit dimensions, [are] relational 2006) has provided close, carefully nuanced
and mediated by artifacts, . . . material as well analyses of how the routine features of bu-
as mental and representational (Gherardi & reaucratic organizations that make for effec-
Nicolini 2000, p. 329). Rather than a specific tive coordination across persons, times, and
organization of roles and learning processes tasks nonetheless lead to mistakes, misconduct,
or a measurable set of attitudes and beliefs, and disaster. Bridging micro and macro per-
safety is understood as an elusive, inspirational spectives, Vaughans work proposes a series
asymptote, and more often only one of a num- of mechanisms that prevent well-intentioned
ber of competing organizational objectives. actors and well-designed organizations from
In his original formulation of the theory of achieving desired objectives. Despite signifi-
normal accidents, Perrow (1999 [1984], p. 94) cant differences between loosely coupled net-
identified those intrinsic characteristics of so- works of heterogeneously distributed, and of-
ciotechnical systems that challenge aspirations ten collegially connected, communities of di-
to total safety, breeding failure and catastro- verse participants (Hughes 1998) and tightly
phe. Where system components are complexly coupled complex systems (Perrow 1999 [1984]),
organized (i.e., with many interacting param- both display consistent cognitive patterns that
eters and subsystems, indirect and inferential undermine safety and render accidents normal.
sources of information, feedback loops, and per- Vaughan (1999) describes these practices as the
sonnel isolation), tight coupling among the sub- dark side of organizations. However, rather
units undermines the ability to recover from than focusing on hidden information, Vaughan
inevitable malfunctions. Tightly coupled sys- emphasizes the interpretive flexibility in all
tems have little slack and more invariant se- processes; she demonstrates how the cognitive
quences in time-dependent processes, usually
permitting only one way to reach the pro-
duction goal. Thus, when things go wrong, 11
I write this essay as the financial crisis of 2008 is unfold-
and they always do, if only because of the vari- ing. One cannot help but notice the unfortunate parallels
with previous moments when supposedly expert technolo-
ability in component life spans or unobserved gies failed to perform as their promoters and beneficiaries
faults in minor or major parts, these tightly insisted they would.
356 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
construction of situations is at the heart of the Eden 2006). In these studies, researchers de-
safety problem and why organizations do not scribe how:
learn from their mistakes. 1. Linguistic schema, formal categories,
Unusual events and accidents are generated embedded norms, and familiar artifacts
by the same cognitive processes that enable provide both fixed and flexible frames of
the ordinary, routine interactions of daily life. reference with which people apprehend
Vaughan documents the ways that participants and interpret information system perfor-
interpret uncertain and anomalous events mances, risks, and safety (Clarke 1993,
as routine, thus failing to identify emerging Heimer 1988, Kahneman et al. 1982,
disasters. This interpretive construction is an Pfohl 1978, Starbuck & Milliken 1988).
irreducible feature of organizational processes 2. Information that might shape more cau-
because it is also a necessary feature of all social tious and responsive interpretations is of-
action. Cognitive processes homogenize, or ten missing, actively buried (Sagan 1993),
normalize, across differences, so that each event or discredited (Vaughan 1996, 2003).
is not perceived as unique but is categorized, Some knowledge is removed or seg-
and responded to, as an example of something mented by the distributed work pro-
known and familiar for which interpretations cesses and organizational norms of se-
and established responses exist. Thus, action crecy that impede the communication or
can proceed across time and space rather than understanding that is vital for our safety
as if each moment, phenomenon, or interaction (Galison 2004).
was being experienced for the first time. (This 3. Dangers that are neither spectacular,
is exactly the import of defining culture as both sudden, nor disastrous, or that do not
system and practice. Each act is interpretable resonate with symbolic fears, can remain
only as part of a system; the system is produced ignored and unattended, and as a conse-
through myriad individual actions.) However, quence are not interpreted or responded
routinized habits and tacit knowledge that are to as safety hazards (Alvarez & Arends
fundamental constituents of sociality 2000, Brown & Mikkelsen 1990, Glassner
mechanisms for assembling the social 1999). For example, Beamish (2002)
(Latour 2005)also efface particular dif- describes oil spilling continuously for
ferences that can, and sometimes do, have 38 years in the Guadalupe Dunes between
catastrophic consequences. Los Angeles and San Francisco and how
Within rich ethnographies of sociotechni- agencies geared to answer dramatic and
cal systems, scholars display the local enact- sudden pollution events lacked the frame-
ment of more general representational prac- works to recognize or tools to respond to
tices that constitute, reconstitute, and reform ongoing, routine environmental degra-
the cultural system in which safety is valued, if dation by continuously leaking pipes.
not consistently achieved (Clarke 1989, 1999; From the 1950s until the 1990s, the spill
Gusterson 1998; Perin 2005; Vaughan 1996). was ignored, existing physically but not
This research varies by the specific foci of inter- in any organizationally cognizable form.
pretation: on artifacts and objects (Hilgartner Because it was continuous, ongoing for
1992, Schein 1992) including, for example, ra- 20 years or more, it was in effect routine
diation (Hacker 1987), asbestos (Maines 2005), and interpreted as such. However, when
o-rings (Gieryn & Figert 1990), oil (Beamish the spill was no longer just a set of
2002), system conditions or signals (e.g., Perin distributed puddles but began to appear
2005, Walker 2004), the significance of a par- in the ocean nearby, impressions of what
ticular event (Galison 1997, pp. 35262; Gieryn was normal quickly changed (Beamish
& Figert 1990), repeated events (Sagan 1993; 2000, p. 481). Yet, because there was
Vaughan 2003, 2005a), or imagined events (e.g., no category for perceiving, naming, and

ANRV381-SO35-17 ARI 5 June 2009 9:28
responding to slow, long-term environ- superbugs, killer kids, or teenage

mental degradation, the spill was again moms rather than more immediate, em-
normalized through that part of the oil pirically demonstrable threats to well-
field subculture that expected rapidly being and safety such as poverty or guns.
unfolding disasters. It was reinterpreted He suggests that the media, ever in
as an emergency, an available category search of salacious stories that will in-
for those with expertise in short-lived but crease market share, are simultaneously
possibly big oil spills. Interpreted as an the promulgators and debunkers of fear-
emergency, the experts responded with mongering. He suggests that misplaced
standardized responses that turned out fears are propagated by those who seek
to be inappropriate for the long-term to profit by selling protections against
system failure and environmental degra- that which is feared, generating both de-
dation that had actually taken place. mand (creating fears) and supply (safety).
Vaughans (2003, 2005b) analyses of In contrast, Clarke (2006) suggests that

the Challenger and Columbia disasters although the harbingers of impending
provide parallel examples of situations catastrophe are more reasonable and pre-
in which patterned, systemic condi- scient than many imagine, the ubiquity
tions are repeatedly misperceived and of worst cases . . . renders them ordinary
misinterpreted and, when disaster strikes, and mundaneno longer able to shock.
are reinterpreted once again, in these Cumulatively, there is a loss of public trust
cases oppositely, not as emergencies and an increase in the likelihood of insti-
but as random, incidental, contingent tutional failure (Freudenberg 1993); this
occurrences rather than the product of generalized loss of trust in institutions ex-
long-term systemic processes. plains three times as much of the variation
4. Organizational structures, roles, and rou- in public fears as do sociodemographic
tines shape interpretations so that differ- and ideological variables.
ent organizational routines produce very
different understandings of risk and er-
ror. In a comparison of NASAs organi-
zational structure with the FAA (Federal CONCEPTUAL CONUNDRUMS,
Aviation Administration, National Air IDEOLOGICAL ELISIONS,
Traffic System), Vaughan (2005a) shows AND STRUCTURAL SUPPORTS
how invariant and open discussion of FOR SAFETY
the most minor variations or mishaps in Talk about safety culture presents a series of
the Air Traffic Control system and not conundrums. First, safety is defined and mea-
in NASA facilitates effective self-scrutiny sured more by its absence than by its presence
and sensitivity to mishap. (Reason 1999, p. 4); we are safe because there
5. The larger macrosocietal and popular are no accidents. As non-events, we pay little at-
culture embeds particular interpretations tention to near misses. Belief in the attainabil-
of risk and safety (Douglas 1985, Douglas ity of absolute safety . . . impede[s] the achieve-
& Wildavsky 1982, Giddens 1999), and ment of realizable safety goals (Perrow 2007;
repeated organizational and institutional Reason 1999, p. 11). By attempting to institu-
failures breed generalized and dispropor- tionalize an absence (no accidents), safety cul-
tionate fear and uncertainty. For exam- ture chases an ever-receding chimera, observ-
ple, Glassner (1999) argues that fears able only when it ceases to exist. If absolute
are generally focused on the wrong safety is chimerical, and if systems are never
things: on chimerical dangers such as perfect, some suggest that research should
358 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
focus instead on adaptability and resilience. For come to constitute the system in practice.
example, Dekker (2006, pp. 83, 86) calls for re- Despite this well-documented understanding
search on the drift into failure, that is, the of organizational behavior, many engineering
ways in which a systems protective mechanisms models fail to describe the way work is actually
slowly push it toward the boundary between done, offering instead what turn out to be
resilient adaptability and failure. Rather than largely imaginary accounts of work and system
wait for the safety silence to be broken and performance (Sosa et al. 2003). [Pilot training
rather than model safety failures, Dekker sug- is a notable exception, routinely instructing
gests that engineers should model the ordinary pilots to differentiate when to follow or break
routines and micro decisions that often do not protocol (Galison 2004, Gladwell 2008).]
lead to failure but may nonetheless be linked Even some who promote resilience in place
to macrolevel drift. Thus, some resilience en- of safety culture engineering and recognize
gineers call for close, detailed observation of so- safety to be an emergent system property
ciotechnical systems, for studies of the underly- context rather than causeoffer what turn out
ing dynamic relationships within organizations, to be merely more complex models of dynami-
especially on decisions that would relax produc- cally intercollated feedback loops. Like Dekker
tion pressures and consequent risks. To become (2006), Leveson et al. (2006) also suggest that a
wiser in the ways of the world (Perin 2005, resilient system should include systematic self-
p. xix), engineers should learn something reflection. They differ, however, by proposing
meaningful about insider interpretations, about not deep ethnography, nor skepticism concern-
peoples changing (or fixed) beliefs and how ing information and communication, but me-
they do or do not act on them (Dekker 2006, chanical observation, modeled in terms of a par-
p. 86), and in this way achieve an observable allel control system that adjusts for variation in
(resilience), rather than spectral (safety), system the development and behavior of a system from
condition. Eschewing the reductionist concep- the engineered design. By constant comparison
tions of safety culture, resilience engineering of behavior to design, Leveson et al. suggest
joins the interpretive turn and ends up call- that we can build resilience into safety-critical
ing for thick description; engineering becomes systems. They adapt standard cybernetic mod-
ethnography. els with system-dynamic models that continu-
Second, measures designed to enhance ously and automatically modify system speci-
a systems safetydefenses, barriers, and fications. The research displays pervasive and
safeguardscan also bring about its destruc- persistent refusal to accept the basic feature of
tion (Reason 1999, p. 6). Although most complex organizations and sociotechnical sys-
engineering-based organizations believe that tems: They are continually in the making, con-
safety is best achieved through a prede- structed and reconstructed in every moment
termined consistency of their processes and with every act. Each new safety process or pro-
behaviors, . . . it is the uniquely human ability to cedure, each specification of the system, rein-
vary and adapt actions to suit local conditions stantiates the system that was into something
that preserves system safety in a dynamic and that issomething new, if not different (Ewick
uncertain world (Reason 1999, p. 9; Hollnagel & Silbey 1998, pp. 4344). Because adjustment
et al. 2006). Thus, organizations routinely to the new model always includes some adap-
succeed, and recover from near disaster, tation (with implied variation and innovation),
because workers do not follow predetermined specification of the system is always pushing
protocols or designs; instead, they interpret against an asymptotic aspiration of full informa-
rules and recipes, adapt resources to innovative tion. Thus, there remains an unwarranted con-
uses, develop work-arounds, and invent in fidence in the ability to marshal information, as
situ many of the routines that ultimately well as its credibility (Perin 2005), to control

ANRV381-SO35-17 ARI 5 June 2009 9:28
system behavior that is belied by the history of that later generate catastrophic hazards. For ex-
system failure, inviting, Perrow (2007) claims, ample, in his study of nuclear weapon deploy-
the next catastrophe.12 ment, Sagan (1993) noted that (a) American
Quite noticeably, the discussions of safety radar installations had been installed across
culture ignore those features of complex or- the northern hemisphere to observe possible
ganizations and technological systems from launches from the Soviet Union and could not
which cultural schemas and interpretations of- observe or monitor missile launches from the
ten emerge: normative heterogeneity, compet- southern hemisphere, including launches from
itive and conflicting interests, and inequalities Cuba during the missile crisis of October 1962;
in power and authority. Thus, what is specifi- (b) strategic planning and nuclear safety re-
cally missing from accounts of safety culture is lied on a design that limited detonation in
attention to the mechanisms and processes that the absence of at least two independent deci-
produce systemic meanings, including under- sion makers, but during the crisis of October
standings of risk, safety, authority, and control. 1962, planes were launched with armed nuclear
A reflexive, historically grounded, empirical re- weapons, with only a single pilot having the
search agenda should address these issues. I of- ability to detonate; and (c) planners generated
fer two suggestions about what such approaches at least 10 scenarios of common-mode failures13
might explore. that might provoke an unintended detonation
of a nuclear weapon; in none of these imaginar-
ies did the analysts conceive the configuration
Challenging Hegemonic of events that actually occurred in 1968 when
Normalization a B-52 bomber crashed near Thule Air Base,
Research on accidents and disasters has re- Greenland, with four thermonuclear weapons
peatedly demonstrated what sociologists have aboard. [The conventional high-explosive ma-
known for close to a century: All purposive so- terials detonated upon impact, but the bomb
cial action has unintended consequences, and, had been designed to withstand the heat and
although social action is inherently variable, so- pressure of a crash. This important safety fea-
cial solidarity and coordination are sustained ture worked (Sagan 1993, p. 180).] Confident
by perceptually, conceptually, and morally nor- that the enemy was on the other side of the
malizing the variation. Thus, we fail to dis- globe rather than 90 miles away, that redundant
tinguish novel or threatening from familiar security systems reinforced rather than under-
and manageable events, productively innovative mined each other, and that the collective imag-
from functionally destructive deviance. This is ination of the defense planners could antici-
true in simple as well as complex relationships pate any confluence of events, the United States
and situations. This is true in planning and in managed only accidentally to avoid unintended
implementation. Thus, in studies of hazardous nuclear detonation.
technologies, researchers have documented the In similar lines of analysis, Perin (2005,
ways in which the most rational and rigorous p. 5) describes how nuclear power planners
analysts regularly fail to imagine contingencies had imagined myriad possible problems but not
what actually occurred in 2002. Although leaks
had been a generic problem known to the indus-
12
Clarke & Short (1993, p. 375) suggest an additional co- try since 1990, at the Davis-Besse Station on
nundrum deriving from the fact that we must respond to Lake Erie near Toledo, Ohio, in 2002, leaks had
accidents and disasters through organizations that may be
precisely the wrong social instruments for such response and
may themselves be an independent source of risk. Organi-
zations are built on predictability, but accidents by definition
13
involve unpredictability. . . . Organizations are organized to Common-mode refers to a system component that serves
be inflexible, when flexibility is exactly what unpredictabil- multiple other components such that, if it fails, the other
ity requires (p. 392). modes or components also fail.
360 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
eaten completely through the 6.63 inch carbon opportunities for sharing concerns about dis-
steel [vessel head] down to a thin [3/16 inch] in- tractions. In her reimagined culture of con-
ternal liner of stainless steel. No engineer had trol, information channels would be laid across
ever considered that nozzle lead deposits could functional boundaries, and observational and
eat into the carbon steel of the reactor vessel. interpretive competencies would become high
Finally, Eden (2006) shows how, for more priority for staff.
than half a century, government analysts failed
to predict and plan for the consequences of nu-
clear fire during a nuclear war. Because organi- Power Differentials and
zations focus on and try to institutionalize what Structured Inequality
they do well, they often fail to value what lies One is hard-pressed to find a reference to
outside their normal view and capacities. Hav- power, group interests, conflict, or inequal-
ing developed expertise in precision bombing, ity in the literature promoting safety cul-
the Air Force also developed parallel skill in ture. This may be the most striking feature
predicting accuracy and damage, but failed to of this research field. This is not to say that
imagine the fire that would follow, although there is no recognition of hierarchy. Indeed,
much of the World War II bombing damage the proponents of safety culture recognize the
was due to fire. Working only with what they greater authority and resources of top-level
knew best and for which they had secure bud- management and recommend using it to insti-
gets, they produced a rather poor representa- tute organizational change from the top down,
tion of the world as it had been (e.g., in Dresden, mandated by organizational leaders, even if de-
Hiroshima, and Nagasaki) or might be. signed by hired consultants. Indeed, the con-
To challenge the processes of normaliza- sistent valorization of clear lines of hierarchy
tion that impede recognition of hazardously accompanies a surprising failure to see how
deviant events, future research might attempt this same hierarchy undermines communica-
to map more systematically not only the ubiq- tion and self-reflection about hazards. Recog-
uity of and variations within such processes, nizing the greater power of management, safety
but most importantly the conditions and re- culture advocates nonetheless fail to adequately
sources that challenge hegemonic normaliza- recognize the diminished power of those in
tion (Ewick & Silbey 1995, 2003). If hegemony subordinate positions (Edwards 1979, Hodson
refers to that which is unthinkable, and safety 2001). As a consequence, organizations often
demands seeing what is not therean accident attempt to institute safety culture by addressing
in the makingthen research needs to iden- only one facet of the organization at a time, for
tify the processes that successfully unsettle or- example, peoples attitudes, behaviors, coordi-
ganizational routines to make the unthinkable nating structures (Cooper 2000), management
cognizable and the invisible apparent. Recalling messages, or organizational symbols, without
that hegemony is what goes without saying, by considering dependencies and interdependen-
articulating what is taken for granted and con- cies. Vaughan (1996, 2003) and Perin (2005) re-
ventionally unspoken, closely observed ethnog- fer at length to the dysfunctional safety con-
raphy can identify the moments in which critical sequences of the hierarchical credibility gap
self-reflection emerges to unsettle convention that derives from the embedded, but unac-
and make space for innovative practices (Kelty knowledged, stratification. Lower-level actors
2008, Suchman 2006). From her deep ethnog- are often repositories of critical information and
raphy of nuclear power plants, Perin (2005) counterhegemonic views, yet are often unable
identified moments when unexpected knowl- to persuade higher-ups in the organization of
edge flowed from one group to another, when either the credibility of their knowledge or rel-
outside observers brought new perspectives on evance of their perspectives. To the extent that
routines, and when in-house meetings provided the consequences of hazardous technologies are

ANRV381-SO35-17 ARI 5 June 2009 9:28
promoted and managed by advocating safety the institutionalized mechanisms, such as

culture that elides issues of power and inequal- worker expertise, oversight, and solidarity, that
ity, it becomes, intended or not, an ideological help mitigate risk. One cannot help notice, for
project (Silbey 1998). example, the rapidity with which particularly
Proponents of safety culture ignore the fact hazardous technologies, such as oil refineries,
that although safety has mutual benefits, all are are bought and sold. In the financialized world,
not made equally better off, even from a mu- a refinery is an asset, not an organization, nei-
tually beneficial objective. If management does ther a community nor a complex system. With
not regard workers as substitutable costs, whose each shift in ownership comes a new manage-
functions are more economically purchased ment regime with a new set of procedures,
through outsourcing, an opposite management policies, and practices; new IT systems; and dif-
theory imagines that all members of the orga- ferent safety regimes.
nization are similarly situated, with commen- Attention to power and inequality suggests
surate interests, trust, and loyalty. From this several lines of research. Studies might pro-
latter, enlightened leadership perspective, cor- ductively bring literature on financial risk and
porate managers are safety stewards promoting material hazards into conversation with each
a generally shared and valued objective and can other. Surely they have been interacting with
thus expect organizational members to follow each other during the past decade, most ob-
enthusiastically; after all, we all benefit from viously when changes in firm ownership bring
increased safety. Yet the differential interests changes in personnel and policies. Future re-
of upper-level managers and lower-level work- search might also explore how safety culture
ers are systematically elided through popular discourse operates in different stratification sys-
ideologies and representational practices that tems. For example, where there is deep poverty
insist on our mutual self-interest. By assum- and low education, obvious human need may
ing a similarity of interest, managers fail to outweigh concerns about safety, and rational
address the differential positions and, more im- discourse may function merely as a concession
portantly, fail to recognize the differential re- to external or symbolic constituencies. Where
sources workers bring to the organization that there is more education and material abun-
can be mobilized in the service of greater safety. dance, talk of safety culture may obscure the
Moreover, with the decline in unioniza- inherent risks of complex systems, disguising
tion, lower-level workers lack the institution- them behind a facade of personal risk and indi-
alized base from which to make their voices vidual deviance.
heard in their own interest or in the inter- Most importantly, however, research should
ests of the firm, including their knowledge of explore ways in which differentially situated in-
safe or unsafe operating conditions. Research terests might be mobilized to produce counter-
has shown, for example, that safety violations, vailing power. Where the relation between the
accidents, and product defects increase with source and victims of hazardous risks are bound
outsourcing (Kochan et al. 1992). Because of by neither space (across geographic bound-
a lack of oversight, integration across func- aries), nor time (across generations), security
tions, and an intimate knowledge of the pro- and safety certainly seem elusive (Beck 1992).
duction process, contract suppliers are unable Some recent litigation campaigns suggest tac-
to provide the mitigation of hazard that had tics that, if not directly controlling hazardous
been supplied in the past through shop-floor, systems, might nonetheless highlight the links
rather than upper-level-management, steward- among dispersed organizations, technologies,
ship. Furthermore, as firms spread more of their and collectively as well as locally experienced
capital risks through innovative financial instru- harms. Deterritorialized risks can be brought
ments, they have inadvertently broken the or- to earth, so to speak, apprehended and localized
ganizational field, one might say, by eschewing by mass or class action litigations, such as in the
362 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
asbestos and tobacco litigations or in lawsuits for the consequences of complex technolo-
against oil companies for global warming. gies resides in a cultural ether, everywhere or
nowhere. If the ether proves elusive, the ex-
planation of operator error is always available.
CONCLUSION In seventeenth century England, when experi-
Safety culture is a particularly narrow at- ments went wrong in performances before the
tempt to tame Prometheus, where the central Royal Society, the air-pump exploding for ex-
problematicassembling the social (Latour ample, assistants and craftspeople were blamed
2005)is assumed rather than explored. In its for the failure rather than the gentlemen sci-
most common invocations, safety culture be- entists or even the artifacts themselves (Shapin
comes either a thing or an ether. Culture names & Schaffer 1985). Because technologies con-
what is left over after you forgot what it was cretize the scientific theories and social rela-
you were originally trying to learn (OReilly tions, hierarchy, and authority of the organi-
& Chatman 1996, p. 159). Rather than address zations assembled around them, it was difficult
the structural and historical conditions that ei- to point toward systemic failures in the appara-
ther sustain or impede safe organizational per- tus without undermining the scientist design-
formance, culture becomes a supplement, the ers. Since the seventeenth century, Prometheus
detritus of social transactions. As the phenom- has become omnipresent, if not omnipotent,
ena continually recede before efforts to con- scientific authority even more secure; yet, four
trol them, research advocating safety culture centuries on, we still focus on the assistant, fail-
seems, in the end, to suggest that responsibility ing once again to tame Prometheus.
DISCLOSURE STATEMENT
The author is not aware of any biases that might be perceived as affecting the objectivity of this
review.
ACKNOWLEDGMENTS
I am particularly grateful to Joelle Evans and Ayn Cavicchi for their excellent research assistance
and to Sandy Brown for initiating my work on this topic. I also appreciate the comments and
suggestions I received at various stages of writing from Bernard Harcourt, Hugh Gusterson,
Tom Baker, Ronen Shamir, Tanu Agrawal, Tanina Rostain, Douglas Goodman, Chris Kelty,
Diane Vaughan, Don Lessard, Nelson Repenning, Nancy Leveson, Matt Richards, Xaq Frolich,
Ruthanne Huising, John Paul Ferguson Keith Bybee, Carroll Seron, Michael Fischer, Connie
Perin, Jean Jackson, Chihyung Jeon, Susan Sterrett, Jason Jay, Kieran Downes, Sophia Roosth,
David Kaiser, and Stefan Helmreich.
LITERATURE CITED
Abbott AD. 1988. The System of Professions: An Essay on the Division of Expert Labor. Chicago: Univ. Chicago
Press
Alvarez R, Arends J. 2000. Fire, earth and water: an assessment of the environmental, safety and health im-
pacts of the Cerro Grande fire on Los Alamos National Laboratory, a Department of Energy facility. Con-
cerned Citizens for Nuclear Safety and Nuclear Policy Project. http://www.nuclearactive.org/docs/
CerroGrandeindex.html
Associated Press. 1989. Inquiry finds laxity at root of 35 rail deaths last year. Assoc. Press, Nov. 7

ANRV381-SO35-17 ARI 5 June 2009 9:28
Baker Panel. 2007. The Report of the BP U.S. Refineries Independent Safety Review Panel. http://www.bp.com/
bakerpanelreport
Baker T. 2002. Risk, insurance and the social construction of responsibility. See Baker & Simon 2002,
pp. 3351
Baker T. 2005. The Medical Malpractice Myth. Chicago: Univ. Chicago Press
Baker T, Simon J, eds. 2002. Embracing Risk: The Changing Culture of Insurance and Responsibility. Chicago:
Univ. Chicago Press
Ballard GM. 1988. Nuclear Safety After Three Mile Island and Chernobyl. London: Elsevier Appl. Sci.
Beamish TD. 2000. Accumulating trouble: complex organization, a culture-of-silence, and a secret spill. Soc.
Probl. 47(4):47398
Beamish TD. 2002. Silent Spill: The Organization of an Industrial Crisis. Cambridge, MA: MIT Press
Beck U. 1992. Risk Society: Towards a New Modernity, transl. M Ritter. London: Sage
Becker H. 1963. The Outsiders: Studies in the Sociology of Deviance. New York: Free Press
Bernstein M. 1955. Regulating Business by Independent Commission. Princeton, NJ: Princeton Univ. Press
Bittner E. 1983. Technique and the conduct of life. Soc. Probl. 30:24961
Bourrier M. 1996. Organizing maintenance work at two American nuclear power plants. J. Conting. Crisis
Manag. 4:10412
Brooks K, Lazorko L, Zygmunt J, Hill J, Savage P, et al. 1986. Tighter controls on chemical hazards. Chem.
Week, Jan. 1/Jan. 8, p. 21
Brown P, Mikkelsen EJ. 1990. No Safe Place: Toxic Waste, Leukemia and Community Action. Berkeley: Univ.
Calif. Press
Burke TF. 2004. Lawyers, Lawsuits, and Legal Rights: The Battle over Litigation in American Society. Berkeley:
Univ. Calif. Press
Carroll J. 1998a. Organizational learning activities in high-hazard industries: the logics underlying self-analysis.
J. Manag. Stud. 35:699717
Carroll J. 1998b. Safety culture as an ongoing process: culture surveys as opportunities for enquiry and change.
Work Stress 12:27284
Chabon M. 2008. In priceland. New York Rev. Books, May 1
Ciaverelli A, Figlock R. 1996. Organizational factors in aviation accidents. In Proceedings of the Ninth Annual
Symposium on Aviation Psychology, pp. 103335. Columbus, OH: Dep. Aviat.
Clarke L. 1989. Acceptable Risk?: Making Decisions in a Toxic Environment. Berkeley: Univ. Calif. Press
Clarke L. 1993. The disqualification heuristic: when do organizations misperceive risk? In Research in Social
Problems and Public Policy, ed. RT Youn, WF Freudenberg, 5:289312. Greenwich, CT: JAI
Clarke L. 1999. Mission Improbable: Using Fantasy Documents to Tame Disaster. Chicago: Univ. Chicago Press
Clarke L. 2006. Worst Cases: Terror and Catastrophe in the Popular Imagination. Chicago: Univ. Chicago Press
Clarke L, Short JF. 1993. Social organization and risk: some current controversies. Annu. Rev. Sociol. 19:37599
Cohen MD, March JG, Olsen JP. 1972. A garbage can model of organizational choice. Adm. Sci. Q. 17:125
Cooper MD. 2000. Towards a model of safety culture. Saf. Sci. 36:11136
Cox S, Cox T. 1991. The structure of employee attitudes to safety: a European example. Work Stress 5:93106
Davidson A. 1990. In the Wake of the Exxon Valdez: The Devasting Impact of the Alaska Oil Spill. San Francisco:
Sierra Club Books
Dekker S. 2006. Resilience engineering: chronicling the emergence of confused consensus. See Hollnagel
et al. 2006, pp. 7792
Denison DR. 1996. What is the difference between organizational culture and organizational climate? A
natives point of view on a decade of paradigm wars. Acad. Manag. Rev. 21:61954
Derthick M, Quirk PJ. 1985. The Politics of Deregulation. Washington, DC: Brookings Inst.
Deutch JM, Lester RK. 2004. Making Technology Work: Applications in Energy and the Environment. Cambridge,
UK: Cambridge Univ. Press
Diamond S. 1986. Chernobyl causing big revisions in global nuclear power policies. New York Times, Oct. 27,
p. 1
Douglas M. 1985. Risk as a forensic resource. Daedalus 119:116
Douglas M, Wildavsky A. 1982. Risk and Culture. Berkeley: Univ. Calif. Press
364 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
Eden L. 2006. Whole World on Fire: Organizations, Knowledge and Nuclear Weapons Devastation. Ithaca, NY:
Cornell Univ. Press
Edwards R. 1979. Contested Terrain: The Transformation of the Workplace in the Twentieth Century. New York:
Basic Books
Eisenhardt K. 1993. High reliability organizations meet high velocity environments: common dilemmas in
nuclear power plants, aircraft carriers, and microcomputer firms. In New Challenges Understanding Orga-
nizations, ed. KH Roberts, pp. 11735. New York: Macmillan
Ericson R, Doyle A, Barry D. 2003. Insurance as Governance. Toronto: Univ. Toronto Press
Erikson K. 1978. Everything in Its Path: Destruction of Community in the Buffalo Creek Flood. New York: Simon
& Schuster
Ewald F. 2002. The return of Descartess malicious demon: an outline of a philosophy of precaution. See
Baker & Simon 2002, pp. 273301
Ewick P, Silbey SS. 1995. Subversive stories and hegemonic tales: toward a sociology of narrative. Law Soc.
Rev. 29(2):197226
Ewick P, Silbey SS. 1998. The Common Place of Law: Stories from Everyday Life. Chicago: Univ. Chicago Press
Ewick P, Silbey SS. 2003. Narrating social structure: stories of resistance to legal authority. Am. J. Sociol.
109:132872
Fischer MMJ. 2006. Culture and cultural analysis. Theory Cult. Soc. 23:36064
Fortun K. 2001. Advocacy After Bhopal. Chicago: Univ. Chicago Press
Freudenburg WR. 1993. Risk and recreancy: Weber, the division of labor, and the rationality of risk perception.
Soc. Forces 71:90932
Friedman L. 1967. Legal rules and the process of social change. Stanford Law Rev. 19:786
Friedman L. 1985. Total Justice: What Americans Want from the Legal System and Why. Boston: Beacon
Galanter M. 1983. Reading the landscape of disputes: what we know and dont know (and think we know)
about our allegedly contentious and litigious society. UCLA Law Rev. 31:471
Galanter M. 1994. Predators and parasites: lawyer-bashing and civil justice. Ga. Law Rev. 28:63381
Galanter M. 2006. Law, anti-law and social science. Annu. Rev. Law Soc. Sci. 2:116
Galison P. 1997. Image and Logic: A Material Culture of Microphysics. Chicago: Univ. Chicago Press
Galison P. 2004. Removing knowledge. Crit. Inq. 31:22943
Geertz C. 1973. The Interpretation of Cultures: Selected Essays. New York: Basic Books
Geller ES. 1994. Ten principles for achieving a total safety culture. Prof. Saf. 39:1824
Gherardi S, Nicolini D. 2000. To transfer is to transform: the circulation of safety knowledge. Organization
7:32948
Giddens A. 1999. Risk and responsibility. Mod. Law Rev. 62(1):110
Gieryn TF, Figert SE. 1990. Ingredients for a theory of science in society: O-rings, ice water, C-clamp,
Richard Feynman and the press. In Theories of Science in Society, ed. SE Cozzens, TF Gieryn, pp. 6797.
Bloomington: Ind. Univ. Press
Gladwell M. 2008. Outliers. Boston: Little Brown
Glassner D. 1999. The Culture of Fear: Why Americans Are Afraid of the Wrong Things. New York: Basic Books
Glennon DP. 1982. Measuring organizational safety climate. Aust. Saf. News. Jan./Feb., pp. 2328
Guldenmund FW. 2000. The nature of safety culture: a review of theory and research. Saf. Sci. 34:21557
Gusterson H. 1998. Nuclear Rites. Berkeley: Univ. Calif. Press
Gusterson H. 2007. Anthropology and militarism. Annu. Rev. Anthropol. 36:15575
Habermas J. 1975. The Legitimation Crisis. Boston: Beacon
Hacker B. 1987. The Dragons Tail: Radiation Safety in the Manhattan Project, 19421946. Berkeley: Univ. Calif.
Press
Hacking I. 1990. The Taming of Chance. Cambridge, UK: Cambridge Univ. Press
Hacking I. 2003. Risk and dirt. In Risk and Morality, ed. R Ericson, A Doyle, pp. 2247. Toronto: Univ. Toronto
Press
Heimer C. 1985. Reactive Risk and Rational Action: Managing Moral Hazard in Insurance Contracts. Berkeley:
Univ. Calif. Press
Heimer C. 1988. Social structure, psychology and the estimation of risk. Annu. Rev. Sociol. 14:491519

ANRV381-SO35-17 ARI 5 June 2009 9:28
Helmreich S. 2001. After culture: reflections on the apparition of anthropology in artificial life, a science of
simulation. Cult. Anthropol. 16:61328
Hilgartner S. 1992. The social construction of risk objects: or, how to pry open networks of risk. Organizations,
Uncertainties and Risk, ed. J Short, L Clarke, pp. 3953. Boulder, CO: Westview
Hodson R. 2001. Dignity at Work. Cambridge, UK: Cambridge Univ. Press
Hollnagel E, Woods DD, Leveson N, eds. 2006. Resilience Engineering: Concepts and Precepts. Aldershot, UK:
Ashgate
Holtom W, McCann M. 2004. Distorting the Law: Politics, Media and the Litigation Crisis. Chicago: Univ.
Chicago Press
Hughes TP. 1998. Rescuing Prometheus. New York: Vintage Books
Humphrey SE, Nahrgan JD, Morgeson FP. 2007. Integrating motivational, social and contextual work design
features: a meta-analytic summary and theoretical extension of the work design literature. J. Appl. Psychol.
92:133256
Int. Atomic Energy Agency (IAEA). 1991. Safety Culture. A Report by the International Nuclear Safety Advisory
Group. Safety Ser. 75-INSAG-4. Vienna, Austria: IAEA
Int. Atomic Energy Agency (IAEA). 1992. Information sheet. Tech. Comm. Meet. Impact Manag. Organ. Safe
Ooperation of Nuclear Power Plants. Ref. J4-TC-804. Vienna, Austria. Aug.Sept.
Jacobs MD, Hanrahan NW, eds. 2005. The Blackwell Companion to the Sociology of Culture. Malden, MA:
Blackwell
Jasanoff S, ed. 1994. Learning from Disaster. Philadelphia: Univ. Pa. Press
Jasanoff S. 2005. Designs on Nature: Science and Democracy in Europe and the United States. Princeton, NJ:
Princeton Univ. Press
Joskow P, Noll R. 1977. Regulation in theory and practice. In Studies in Public Regulation, ed. G Fromm,
pp. 165. Cambridge, MA: MIT Press
Joskow P, Rose N. 1989. The effects of economic regulation. In Handbook of Industrial Organization,
ed. R Schmalensee, R Willing, pp. 1450506. Amsterdam: North-Holland, Elsevier
Kagan RA. 2003. Adversarial Legalism: The American Way of Law. Cambridge, MA: Harvard Univ. Press
Kahneman D, Slovic P, Tversky A, eds. 1982. Judgement Under Uncertainty: Heuristics and Biases. Cambridge,
UK: Cambridge Univ. Press
Keeble J. 1991. Out of the Channel: The Exxon Valdez Oil Spill in Prince William Sound. New York: HarperCollins
Kelty C. 2008. Allotropes of fieldwork in nanotechnology. In Emerging Conceptual, Ethical and Policy Issues in
Bionanotechnology, Vol. 101, ed. F Jotterand, pp. 15780. Dordrecht/London: Springer
Kiefer F. 1984. Bhopal tragedy resounds for other multinationals. Christ. Sci. Monit., Dec. 12, p. 1
Klein N. 2007. The Shock Doctrine: The Rise of Disaster Capitalism. New York: Metropolitan Books
Klein RL, Bigley GA, Roberts KH. 1995. Organizational culture in high reliability organizations: an extension.
Hum. Relat. 48:77193
Knowles S. 2007a. Engineers in disaster: underwriters laboratories and the invention of risk engineering in the United
States. Unpubl. manuscr.
Knowles S. 2007b. What is a disaster? Social science disaster research in the postwar United States. Unpubl. manuscr.
Kochan T, Smith M, Wells JC, Rebitzer JB. 1992. Managing the safety of contingent workers: a study in contract
workers in the petrochemical industry. Sloan Sch. Work. Pap. 3442-92-BPS, Cambridge, MA
Kurzman D. 1987. A Killing Wind: Inside Union Carbide and the Bhopal Catastrophe. New York: McGraw-Hill
LaPorte T, Consolini P. 1991. Working in practice but not in theory: theoretical challenges of high-reliability
organizations. J. Public Adm. Res. Theory 1:1948
LaPorte TR, Rochlin G. 1994. A rejoinder to Perrow. J. Conting. Crisis Manag. 2:22127
Lash SB, Wynne B. 1992. Introduction. See Beck 1992, pp. 18
Latour B. 2005. Reassembling the Social. Oxford: Oxford Univ. Press
Lee TR. 1996. Peceptions, attitudes and behaviour: the vital elements of a safety culture. Health Saf. Oct:115
Leveson N, Dulac N, Zipkin D, Cutcher-Gershenfeld J, Carroll J, Barrett B. 2006. Engineering resilience
into safety-critical systems. See Hollnagel et al. 2006, pp. 95123
Lieberman JK. 1981. The Litigious Society. New York: Basic Books
MacKenzie D. 2006. An Engine Not a Camera: How Financial Models Shape Markets. Cambridge, MA: MIT
Press
366 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
Maines R. 2005. Asbestos and Fire: Technological Trade-Offs and the Body at Risk. New Brunswick, NJ: Rutgers
Univ. Press
Marone J, Woodhouse E. 1986. Averting Catastrophe: Strategies for Regulating Risky Technologies. Berkeley: Univ.
Calif. Press
Medvedev ZA. 1992. The Legacy of Chernobyl. New York: W.W. Norton
Mezias S. 1994. Financial meltdown as normal accident: the case of the American savings and loan industry.
Acc. Organ. Soc. 19:18192
Morgeson FP, Humphrey SE. 2006. The work design questionaire (WDQ): developing and validating a
comprehensive measure for assessing job design and the nature of work. J. Appl. Psychol. 91:132139
OReilly C, Chatman J. 1996. Culture as social control: corporations, cults, and commitment. Res. Organ.
Behav. 18:157200
Orren K. 1991. Belated Feudalism: Labor, Law and Liberal Development in the United States. Cambridge, UK:
Cambridge Univ. Press
Ostrom L, Wilhelmsen C, Kaplan B. 1993. Assessing safety culture. Nucl. Saf. 34:16372
Peltzman S, Levine ME, Noll RG. 1989. The economic theory of regulation after a decade of deregulation.
Brookings Pap. Econ. Act. Microecon. 1989:159

Perin C. 2005. Shouldering Risks: The Culture of Control in the Nuclear Power Industry. Princeton, NJ: Princeton
Univ. Press
Perrow C. 1999 (1984). Normal Accidents: Living with High-Risk Technologies. Princeton, NJ: Princeton Univ.
Press. 2nd ed.
Perrow C. 2007. The Next Catastrophe: Reducing Our Vulnerabilities to Natural, Industrial, and Terrorist Disasters.
Princeton, NJ: Princeton Univ. Press
Petryna A. 2002. Life Exposed: Biological Citizens After Chernobyl. Princeton, NJ: Princeton Univ. Press
Pfohl S. 1978. Predicting Dangerousness. Lexington, MA: Lexington Books
Pidgeon NF. 1991. Safety culture and risk management in organizations. J. Cross Cult. Psychol. 22:12940
Pidgeon N. 1997. The limits to safety culture: politics, learning and man-made disasters. J. Conting. Crisis
Manag. 5:114
Pidgeon N. 1998. Safety culture: key theoretical issues. Work Stress 13:20216
Reason JT. 1997. Managing the Risks of Organizational Accidents. Aldershot, UK: Ashgate
Reason JT. 1999. Safety paradoxes and safety culture. Int. J. Inj. Control Saf. Promot. 7:314
Rees J. 1994. Hostages of Each Other: The Transformation of Nuclear Safety Since Three Mile Island. Chicago: Univ.
Chicago Press
Roberts KH, Rousseau DM. 1989. Research in nearly failure-free, high-reliability organizations. IEEE Trans.
Eng. Manag. 36:13239
Roberts KH, Rousseau DM, LaPorte TR. 1994. The culture of high reliability: quantitative and qualitative
assessment aboard nuclear powered aircraft carriers. J. High Technol. Manag. Res. 5:14161
Rochlin GI, LaPorte TR, Roberts KH. 1987. The self-designing high-reliability organization: aircraft carrier
flight operations at sea. Nav. War Coll. Rev. 49:7690
Rose N. 1989. Governing the Soul: The Shaping of the Private Self. London: Routledge
Rose N. 1999. The Powers of Freedom. Cambridge/New York: Cambridge Univ. Press
Sagan S. 1993. The Limits of Safety: Organizations, Accidents and Nuclear Weapons. Princeton, NJ: Princeton
Univ. Press
Sagan SD. 1996. When redundancy backfires: why organizations try harder and fail more often. Presented at Annu.
Meet. Am. Polit. Sci. Assoc., San Francisco, Sept.
Schein EH. 1992. Organizational Culture and Leadership. San Francisco: Jossey-Bass
Scherer N. 2005. Scoring Points: Politicians, Activists and the Lower Federal Court Appointment Process. Stanford,
CA: Stanford Univ. Press
Schneiberg M, Bartley T. 2008. Organizations, regulation, and economic behavior: regulatory dynamics and
forms from the 19th to 21st century. Annu. Rev. Law Soc. Sci. 4:3161
Schulman P. 1993. The analysis of high reliability organizations: a comparative framework. In New Challenges
Understanding Organizations, ed. KH Roberts, pp. 3354. New York: Macmillan
Setbon M. 1993. Pouvoirs Contre Sida: De la Transfusion Sanguine au Depistage: Decisions et Pratique en France,
Grande-Bretagne et Suede. Paris: Ed. Seuil

ANRV381-SO35-17 ARI 5 June 2009 9:28
Sewell WH. 2005. Logics of History. Chicago: Univ. Chicago Press

Shamir R. 2006. Neo-liberal moment. Presented at Annu. Meet. Law Soc. Assoc., Las Vegas, May
Shamir R. 2008. The age of responsibilization: on market-embedded morality. Econ. Soc. 37:119
Shapin S, Schaffer S. 1985. The Leviathan and the Air-Pump. Princeton, NJ: Princteon Univ. Press
Silbey J. 2001. Patterns of courtroom justice. J. Law Soc. 28:97
Silbey J. 2004. Judges as film critics: new approaches to filmic evidence. Univ. Mich. J. Law Reform 37:493
Silbey J. 2005. Filmmaking in the precinct house and the genre of documentary film. Columbia J. Law Arts
29:107
Silbey J. 2007a. Criminal performances: film, autobiography and confession. New Mexico Law Rev. 37:189
Silbey J. 2007b. Truth tales and trial films. Loyola Law Rev. 40:551
Silbey SS. 1998. Ideology, power and justice. In Justice and Power in Socio-Legal Studies, ed. AD Sarat, B Garth,
pp. 272308. Evanston, IL: Northwest. Univ. Press
Silbey SS. 2001. Legal culture and consciousness. In International Encyclopedia of the Social and Behavioral Sciences,
pp. 862329. Amsterdam: Elsevier Sci.

Silbey SS. 2005a. After legal consciousness. Annu. Rev. Law Soc. Sci. 1:32368
Silbey SS. 2005b. Everyday life and the constitution of legality. In The Blackwell Companion to the Sociology of
Culture, ed. M Jacobs, N Hanrahan, pp. 33245. Oxford: Blackwell
Silbey SS, Sarat A. 1988. Dispute processing in law and legal scholarship: from institutional critique to the
reconstitution of the juridical subject. Univ. Denver Law Rev. 66:43798
Simon J. 1988. The ideological effects of actuarial practices. Law Soc. Rev. 22:771800
Sosa M, Eppinger SD, Rowles C. 2003. The misalignment of product architecture and organizational structure
in complex product development. Manag. Sci. 50:167489
Starbuck W, Milliken F. 1988. Challenger: fine-tuning the odds until something breaks. J. Manag. Stud.
25:31940
Stephens M. 1980. Three Mile Island. New York: Random House
Stern GM. 1976/2008. The Buffalo Creek Disaster. New York: Random House
Suchman L. 2006. Human-Machine Configurations: Plans and Situated Actions. Cambridge, UK: Cambridge
Univ. Press
Swidler A. 1986. Culture in action: symbols and strategies. Am. Sociol. Rev. 51:27386
U.S. Nucl. Regul. Comm. 1989. Policy statement on the conduct of nuclear power plant operations. 54 Fed.
Reg. 3424, Jan. 24
Vaughan D. 1996. The Challenger Launch Decision: Risky Technology, Culture, and Deviance at NASA. Chicago:
Chicago Univ. Press
Vaughan D. 1999. The dark side of organizations: mistake, misconduct and disaster. Annu. Rev. Sociol. 25:271
305
Vaughan D. 2003. History as cause: Columbia and Challenger. Chapter 8. Rep. Columbia Accid. Investig. Board
1:195204
Vaughan D. 2004. Theorizing disaster: analogy, historical ethnography, and the Challenger accident.
Ethnography 5:31345
Vaughan D. 2005a. Organizational rituals of risk and error. In Organizational Encounters with Risk, ed. B Hutter,
MK Power, pp. 3366. Cambridge, UK: Cambridge Univ. Press
Vaughan D. 2005b. System effects: on slippery slopes, negative patterns, and learning by mistake. In Organi-
zations at the Limit: NASA and the Columbia Accident, ed. WH Starbuck, F Farjoun, pp. 4159. Oxford:
Blackwell
Vaughan D. 2006. NASA revisited: theory, analogy, and public sociology. Am. J. Sociol. 112:35393
Vogel D. 1981. The new social regulation in comparative and historical perspective. In Regulation in Perspective,
ed. T McCraw, pp. 15585. Cambridge, MA: Harvard Univ. Press
Vogel D. 1986. National Styles of Regulation: Environmental Policy in Great Britain and the United States. Ithaca,
NY: Cornell Univ. Press
Walker JS. 2004. Three Mile Island: A Nuclear Disaster in Historical Perspective. Berkeley: Univ. Calif. Press
Weick K. 1987. Organizational culture as a source of high reliability. Calif. Manag. Rev. 29:11227
Weick K, Roberts KH. 1993. Collective mind in organizations: heedful interrelating on flight decks. Admin.
Sci. Q. 38(3):35781
368 Silbey
ANRV381-SO35-17 ARI 5 June 2009 9:28
Weick K, Sutcliffe K. 2001. Managing the Unexpected: Assuring High Performance in an Age of Complexity. San
Francisco: Wiley
Weick K, Sutcliffe KM, Obstfeld D. 1999. Organizing for high reliability: processes of collective mindfulness.
In Research in Organizational Behavior, ed. BM Staw, LL Cummings, 21:81123. Greenwich, CT: JAI
Welke BY. 2001. Recasting American Liberty: Gender, Race, Law and the Railroad Revolution, 18651920.
Cambridge, UK: Cambridge Univ. Press
Westrum R. 1997. Social factors in safety-critical systems. In Human Factors in Safety Critical Systems,
ed. F Redmill, J Rajan, pp. 23356. London: Butterworth-Heinemann
Witt JB. 2004. The Accidental Republic: Crippled Workingmen, Destitute Widows, and the Remaking of American
Law. Cambridge, MA: Harvard Univ. Press
Zelizer V. 1979. Morals and Markets. The Development of Life Insurance in the United States. New Brunswick,
NJ: Transaction Books
Zelizer V. 1997. The Social Meaning of Money: Pin Money, Paychecks, Poor Relief and Other Currencies. Princeton,
NJ: Princeton Univ. Press

Zhang H, Wiegmann DA, von Thaden TL, Sharma G, Mitchell AA. 2002. Safety Culture: A Concept in Chaos?
Santa Monica, CA: Hum. Factors Ergon. Soc.
Zohar D. 1980. Safety climate in industrial organizations: theoretical and applied implications. J. Appl. Psychol.
65:96102

AR348-FM ARI 2 June 2009 9:48
Annual Review
of Sociology
Contents Volume 35, 2009
Frontispiece
Herbert J. Gans ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! xiv

Prefatory Chapters
Working in Six Research Areas: A Multi-Field Sociological Career
Herbert J. Gans ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 1
Theory and Methods
Ethnicity, Race, and Nationalism
Rogers Brubaker ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !21
Interdisciplinarity: A Critical Assessment
Jerry A. Jacobs and Scott Frickel ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !43
Nonparametric Methods for Modeling Nonlinearity
in Regression Analysis
Robert Andersen ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !67
Gender Ideology: Components, Predictors, and Consequences
Shannon N. Davis and Theodore N. Greenstein ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !87
Genetics and Social Inquiry
Jeremy Freese and Sara Shostak ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 107
Social Processes
Race Mixture: Boundary Crossing in Comparative Perspective
Edward E. Telles and Christina A. Sue ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 129
The Sociology of Emotional Labor
Amy S. Wharton ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 147
Societal Responses toTerrorist Attacks
Seymour Spilerman and Guy Stecklov ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 167
Intergenerational Family Relations in Adulthood: Patterns, Variations,
and Implications in the Contemporary United States
Teresa Toguchi Swartz ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 191
v
AR348-FM ARI 2 June 2009 9:48
Institutions and Culture

Sociology of Sex Work
Ronald Weitzer ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 213
The Sociology of War and the Military
Meyer Kestnbaum ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 235
Socioeconomic Attainments of Asian Americans
Arthur Sakamoto, Kimberly A. Goyette, and ChangHwan Kim ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 255
Men, Masculinity, and Manhood Acts
Douglas Schrock and Michael Schwalbe ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 277
Formal Organizations
American Trade Unions and Data Limitations: A New Agenda

for Labor Studies
Caleb Southworth and Judith Stepan-Norris ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 297
Outsourcing and the Changing Nature of Work
Alison Davis-Blake and Joseph P. Broschak ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 321
Taming Prometheus: Talk About Safety and Culture
Susan S. Silbey ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 341
Political and Economic Sociology
Paradoxes of Chinas Economic Boom
Martin King Whyte ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 371
Political Sociology and Social Movements
Andrew G. Walder ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 393
Differentiation and Stratification
New Directions in Life Course Research
Karl Ulrich Mayer ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 413
Is America Fragmenting?
Claude S. Fischer and Greggor Mattson ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 435
Switching Social Contexts: The Effects of Housing Mobility and
School Choice Programs on Youth Outcomes
Stefanie DeLuca and Elizabeth Dayton ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 457
Income Inequality and Social Dysfunction
Richard G. Wilkinson and Kate E. Pickett ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 493
Educational Assortative Marriage in Comparative Perspective
Hans-Peter Blossfeld ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 513
vi Contents
AR348-FM ARI 2 June 2009 9:48
Individual and Society

Nonhumans in Social Interaction
Karen A. Cerulo ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 531
Demography
Social Class Differentials in Health and Mortality: Patterns and
Explanations in Comparative Perspective
Irma T. Elo ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 553
Policy
The Impacts of Wal-Mart: The Rise and Consequences of the Worlds

Dominant Retailer
Gary Gereffi and Michelle Christian ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 573
Indexes
Cumulative Index of Contributing Authors, Volumes 2635 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 593

Cumulative Index of Chapter Titles, Volumes 2635 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 597
Errata
An online log of corrections to Annual Review of Sociology articles may be found at

http://soc.annualreviews.org/errata.shtml
Contents vii

Taming Prometheus Talk About Safety Culture

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Taming Prometheus Talk About Safety Culture

Uploaded by

Copyright:

Available Formats

ANRV381-SO35-17 ARI 5 June 2009 9:28

Taming Prometheus: Talk

School of Humanities, Arts and Social Sciences, Massachusetts Institute of Technology,

Annu. Rev. Sociol. 2009. 35:34169 Key Words

The Annual Review of Sociology is online at Abstract

www.annualreviews.org Safety and Culture 343

www.annualreviews.org Safety and Culture 345

www.annualreviews.org Safety and Culture 347

contemporary focus on safety culture as a means expression of responsibilization, this neo-

Just as historic liberalism was concerned with

www.annualreviews.org Safety and Culture 349

www.annualreviews.org Safety and Culture 351

2007, p. 59). tional or safety culture in general than specif-

www.annualreviews.org Safety and Culture 353

high-velocity environments (Eisenhardt 1993). the named characteristics of high reliability,

www.annualreviews.org Safety and Culture 355

www.annualreviews.org Safety and Culture 357

responding to slow, long-term environ- superbugs, killer kids, or teenage

Vaughans (2003, 2005b) analyses of In contrast, Clarke (2006) suggests that

www.annualreviews.org Safety and Culture 359

www.annualreviews.org Safety and Culture 361

promoted and managed by advocating safety the institutionalized mechanisms, such as

www.annualreviews.org Safety and Culture 363

www.annualreviews.org Safety and Culture 365

Brookings Pap. Econ. Act. Microecon. 1989:159

www.annualreviews.org Safety and Culture 367

Sewell WH. 2005. Logics of History. Chicago: Univ. Chicago Press

pp. 862329. Amsterdam: Elsevier Sci.

NJ: Princeton Univ. Press

www.annualreviews.org Safety and Culture 369

Contents Volume 35, 2009

Herbert J. Gans ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! xiv

Institutions and Culture

American Trade Unions and Data Limitations: A New Agenda

Individual and Society

The Impacts of Wal-Mart: The Rise and Consequences of the Worlds

Gary Gereffi and Michelle Christian ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 573

Cumulative Index of Contributing Authors, Volumes 2635 ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! 593

An online log of corrections to Annual Review of Sociology articles may be found at

You might also like