DEVIANCE THEORIES
which the poor may attain wealth, albeit illegiti- Subcultural explanations have their origin in
mately. Thus, ‘‘strain’’ theories of deviance inter- two distinct sociological traditions. The first is
pret behaviors such as illegal drug selling, prostitu- writing on the properties of delinquent gangs that
tion, and armed robbery as innovative adaptations identifies a distinct lower-class culture of gang
to blocked opportunities for legitimate economic members that encourages aggression, thrill seek-
or occupational success. Similarly, the theories ing, and antisocial behavior (e.g., Miller 1958).
interpret violent crimes in terms of the frustra- The second is writing on cultural conflict that
tions of poverty, as acts of aggression triggered by recognizes that within complex societies there will
those frustrations (Blau and Blau 1982). Much of occur contradictions between the conduct norms
the current research in this tradition is examining of different groups. Thorsten Sellin (1938) sug-
the exact mechanisms by which poverty and eco- gests that in heterogeneous societies several differ-
nomic inequality influence rates of deviant behavior. ent subcultures may emerge, each with its own set
of conduct norms. According to Sellin, the laws
Although once considered a leading theory of and norms applied to the entire society do not
deviance, strain theory has come under criticism necessarily reflect cultural consensus but rather
for its narrow focus on poverty as the primary the values and beliefs of the dominant social groups.
cause of deviant behavior. Recent efforts have
sought to revise and extend the basic principles of Subcultural theories emerging from these two
the theory by expanding and reformulating ide- traditions argue that deviance is the product of a
as about strain. Robert Agnew (1992) has made cohesive set of values and norms that favors devi-
the most notable revisions to the theory. His ant behavior and is endorsed by a segment of the
reformulation emphasizes social psychological, rath- general population. Perhaps most prominent
er than structural, sources of strain. Agnew also among the theories is Marvin Wolfgang and Fran-
broadens the concept of strain, arguing that pover- co Ferracuti’s (1967) writing on subcultures of
ty may be a source of strain, but it is not the only criminal violence. Wolfgang and Ferracuti reason
source. Three sources of strain are important: that there may exist a distinct set of beliefs and
failure to achieve positively valued goals, removal expectations within the society; a subculture that
of positively valued stimuli, and confrontation promotes and encourages violent interactions. Ac-
with negative stimuli. The first type of strain, cording to Wolfgang and Ferracuti, this violent
failure to achieve positively valued goals, may be subculture is pervasive among blacks in the United
the result of a failure to live up to one’s expecta- States and may explain extremely high rates of
tions or aspirations. Strain may also result if an criminal homicide among young black males.
individual feels that he or she is not being treated
in a fair or just manner. The removal of a positively Although Wolfgang and Ferracuti offer lit-
valued stimulus, such as the death of a family tle material specifying the subculture’s precise
member or the loss of a boyfriend or girlfriend, causes, or empirical evidence demonstrating the
can also result in strain. Finally, strain can also be pervasiveness of subcultural beliefs, other writers
produced by the presentation of negative stimuli, have extended the theory by exploring the rela-
such as unpleasant school experiences. Thus, al- tionship between beliefs favoring violence and
though this reformulation of strain theory retains such factors as the structure of poverty in the
the notion that deviance is often the result of United States (Curtis 1975; Messner 1983), the
strain, the concept of strain is broadened to in- history of racial oppression of blacks (Silberman
clude multiple sources of strain. 1980), and ties to the rural South and a southern
subculture of violence (Gastil 1971; Erlanger 1974).
The second set of macro-level origin theories Even these writers, however, offer little empirical
examine the role of culture in deviant behavior. evidence of violent subcultures within U.S. society.
Although not ignoring structural forces such as
poverty in shaping deviance, this class of theories A third class of theories about the macro-level
reasons that there may exist cultures within the origins of deviance began with the work of sociolo-
larger culture that endorse or reinforce deviant gists at the University of Chicago in the 1920s.
values; deviant subcultures that produce higher Unlike strain and subcultural theories, these stress
rates of deviance among those segments of the the importance of the social integration of neigh-
population sharing subcultural values. borhoods and communities—the degree to which
664
DEVIANCE THEORIES
neighborhoods are stable and are characterized by controls in the form of parental supervision are
a homogenous set of beliefs and values—as a force either weak or entirely absent.
influencing rates of deviant behavior. As levels of
integration increase, rates of deviance decrease. Similarly, Robert Crutchfield (1989) argues
Based in the early work of sociologists such as that the structure of work opportunities in areas
Clifford Shaw and Henry McKay, the theories may have the same effect. Areas characterized
point to the structure of social controls in neigh- primarily by secondary-sector work opportunities—
borhoods, arguing that neighborhoods lacking in low pay, few career opportunities, and high em-
social controls are ‘‘disorganized,’’ that is, areas in ployee turnover—may tend to attract and retain
which there is a virtual vacuum of social norms. It persons with few stakes in conventional behav-
is in this normative vacuum that deviance flour- ior—a ‘‘situation of company’’ in which deviance
ishes. Therefore, these theories view deviance as a is likely to flourish.
property of areas or locations rather than specific
groups of people. Recent writing from the disorganization per-
spective has also taken the form of ethnographies;
Early writers in the ‘‘disorganization’’ tradi- qualitative studies of urban areas and the deviance
tion identified industrialization and urbanization producing dynamics of communities. As Sullivan
as the causes of disorganized communities and (1989, p. 9) states, ethnographies describe the
neighborhoods. Witnessing immense growth in community ‘‘as a locus of interaction, intermedi-
eastern cities such as Chicago, these writers ar- ate between the individual and the larger society,
gued that industrial and urban expansion create where the many constraints and opportunities of
zones of disorganization within cities. Property the total society are narrowed to a subset within
owners move from the residential pockets on the which local individuals choose.’’ At the heart of
edge of business and industrial areas and allow Sullivan’s argument is the idea that social networks
buildings to deteriorate in anticipation of the ex- in neighborhoods are important in understanding
pansion of business and industry. This process of whether individuals are capable of finding mean-
natural succession and change in cities disrupts ingful opportunities for work. For example, youth
traditional mechanisms of social control in neigh- were less likely to turn to crime in those neighbor-
borhoods. As property owners leave transitional hoods where they could take advantage of family
areas, more mobile and diverse groups enter. But and neighborhood connections to blue collar jobs.
the added mobility and diversity of these groups Because of the greater employment opportunities
translate into fewer primary relationships—fami- in these neighborhoods, even youth who become
lies and extended kinship and friendship networks. involved in crime were less likely to persist in high-
And as the number of primary relationships de- risk criminal behaviors.
cline, so will informal social controls in neighbor-
hoods. Hence, rates of deviance will rise. Similarly, Jay MacLeod (1995) attempts to ex-
plain how the aspirations of youth living in urban
Recent writing from this perspective focuses areas have been ‘‘leveled,’’ or reduced to the point
on the mechanisms by which specific places in where the youths have little hope for a better
urban areas become the spawning grounds for future. In an analysis of two urban gangs, MacLeod
deviant acts (Bursik and Webb 1982; Bursik 1984; argues that the youths’ family and work experi-
and others). For example, Rodney Stark (1987) ences, along with their relationships with their
argues that high levels of population density are peers, help explain why a predominantly white
associated with particularly low levels of supervi- gang had lower aspirations and engaged in more
sion of children. With little supervision, children delinquent and antisocial behavior than the other
perform poorly in school and are much less likely gang, predominantly comprised of African Ameri-
to develop ‘‘stakes in conformity’’—that is, emo- cans. According to MacLeod, the parents of white
tional and psychological investments in academic youth were much less likely to discipline their
achievement and other conforming behaviors. children or to encourage them to achieve and do
Without such stakes, children and adolescents are well in school. Also, white youth had more experi-
much more likely to turn to deviant alternatives. ence on the job market than the African American
Thus, according to Stark, rates of deviance will be youth. This contributed to a more pessimistic
high in densely populated areas because social outlook and a lowering of their future aspirations.
665
DEVIANCE THEORIES
Finally, MacLeod argues that the white youths’ individuals may become deviant. These theories
immersion in a subculture, which emphasized re- assume the existence of a homogeneous, pervasive
jecting the authority of the school, reinforced their set of norms in society and proceed to explain why
negative attitudes to a much greater extent than persons or entire groups of persons violate the
the African American peer group. norms. There exist two important traditions with-
in this category of theories. The first tradition
In sum, theories of the macro-level origins of involves ‘‘social learning theories’’—explanations
deviance argue that many of the causes of deviance that focus on the mechanisms through which peo-
may be found in the characteristics of groups ple learn the techniques and attitudes favorable to
within society, or in the characteristics of geo- committing deviant acts. The second tradition
graphic areas and communities. They offer expla- involves ‘‘social control theories’’—explanations
nations of group and areal differences in devi- that emphasize factors in the social environment
ance—for example, why some cities have relatively that regulate the behavior of individuals, thereby
higher rates of crime than other cities or why preventing the occurrence of deviant acts.
blacks have higher rates of serious interpersonal
violence than other ethnic groups. These theories Edwin Sutherland’s (1947) theory of differen-
make no attempt to explain the behavior of indi- tial association laid the foundation for learning
viduals or the occurrence of individual deviant theories. At the heart of this theory is the assump-
acts. Indeed, they reason that deviance is best tion that deviant behavior, like all other behaviors,
understood as a property of an area, community, is learned. Further, this learning occurs within
or group, regardless of the individuals living in the intimate personal groups—networks of family mem-
area or community, or the individuals comprising bers and close friends. Thus, according to these
the group. theories individuals learn deviance from persons
closest to them. Sutherland specified a process of
The theories’ implications for public policy differential association, reasoning that persons
focus on the characteristics of geographic areas become deviant in association with deviant others.
and communities that lead to deviance. The im- Persons learn from others the techniques of com-
pact of change on neighborhoods, for example, mitting deviant acts and attitudes favorable to the
can be reduced if the boundaries of residential commission of those acts. Further, Sutherland
areas are preserved. By preserving such bounda- reasoned that persons vary in their degree of
ries, communities are less likely to become transi- association with deviant others; persons regularly
tional neighborhoods that foster deviance and exposed to close friends and family members who
crime. Also, by maintaining residential properties held beliefs favoring deviance and who committed
people become invested in their own community, deviant acts would be much more likely than oth-
which helps foster the mechanisms of informal ers to develop those same beliefs and commit
social control that make deviance less likely. deviant acts.
Strengthening schools and other stabilizing insti-
tutions in neighborhoods, such as churches and Sutherland’s ideas about learning processes
community centers, can also contribute to a reduc- have played a lasting role in micro-level deviance
tion in deviance. Finally, establishing networks for theories. Central to his perspective is the view that
jobs and job placement in disadvantaged areas beliefs and values favoring deviance are a primary
may increase the opportunities of employment cause of deviant behavior. Robert Burgess and
among youth. If they succeed in increasing em- Ronald Akers (1966) and subsequently Akers (1985)
ployment, the networks should decrease the chances extended Sutherland’s ideas, integrating them with
that youth will turn to careers in crime. principles of operant conditioning. Reasoning that
learning processes may best be understood in
MICRO-LEVEL ORIGINS OF DEVIANCE terms of the concrete rewards and punishments
given for behavior, Burgess and Akers argue that
Many explanations of deviance argue that its causes deviance is learned through associations with oth-
are rooted in the background or personal circum- ers and through a system of rewards and punish-
stances of the individual. Micro-level origin theo- ments, imposed by close friends and relatives, for
ries have developed over the past fifty years, identi- participation in deviant acts. Subsequent empiri-
fying mechanisms by which ordinarily conforming cal studies offer compelling support for elements
666
DEVIANCE THEORIES
of learning theory (Matsueda 1982; Akers et al. to conformity—psychological investments or stakes
1979; Matsueda and Heimer 1987). people hold in a conforming lifestyle. Those hav-
ing weak attachments—that is, people who are
Some examples may be useful at this point. insensitive to the opinions of conforming others—
According to the theory of differential association, and who have few stakes in conformity, in the form
juveniles develop beliefs favorable to the commis- of commitments to occupation or career and edu-
sion of delinquent acts and knowledge about the cation, are more likely than others to deviate (see,
techniques of committing deviant acts from their e.g., Paternoster et al. 1983; Thornberry and
closest friends, typically their peers. Thus, suffi- Christenson 1984; Liska and Reed 1985). In effect,
cient exposure to peers endorsing beliefs favoring these individuals are ‘‘free’’ from the constraints
deviance who also have knowledge about the com- that ordinarily bond people to normative con-
mission of deviant acts will cause the otherwise formity. Conversely, individuals concerned about
conforming juvenile to commit deviant acts. Thus, the opinions of conforming others and who have
if adolescent peer influences encourage smoking, heavy psychological investments in work or school
drinking alcohol, and other forms of drug abuse— will see the potential consequences of deviant
and exposure to these influences occurs frequent- acts—rejection by friends or loss of a job—as
ly, over a long period of time, and involves rela- threatening or costly, and consequently will re-
tionships that are important to the conforming frain from those acts.
adolescent—then he or she is likely to develop
beliefs and values favorable to committing these A related concern is the role of sanctions in
acts. Once those beliefs and values develop, he or preventing deviant acts. Control theorists like
she is likely to commit the acts. Hirschi reason that most people are utilitarian in
their judgments about deviant acts, and thus evalu-
The second class of micro-level origin theo- ate carefully the risks associated with each act.
ries, control theories, explores the causes of devi- Control theories typically maintain that the threat
ance from an altogether different perspective. of sanctions actually prevents deviant acts when
Control theories take for granted the existence of the risks outweigh the gains. Much of the most
a cohesive set of norms shared by most persons in recent writing on sanctions and their effects has
the society and reason that most persons want to stressed the importance of perceptual processes in
and will typically conform to these prevailing so- decisions to commit deviant acts (Gibbs 1975,
cial norms. The emphasis in these theories, unlike 1977; Tittle 1980; Paternoster et al. 1982, 1987;
learning theories, is on the factors that bond indi- Piliavin et al. 1986; Matsueda, Piliavin, and Gartner
viduals to conforming lifestyles. The bonds act as 1988). At the heart of this perspective is the rea-
social and psychological constraints on the indi- soning that individuals perceiving the threat of
vidual, binding persons to normative conformity sanctions as high are much more likely to refrain
(Toby 1957; Hirschi 1969). People deviate from from deviance than those perceiving the threat as
norms when these bonds to conventional lifestyles low, regardless of the actual level of sanction threat.
are weak, and hence, when they have little restrain-
ing influence over the individual. Among control Writing from the social control perspective
theorists, Travis Hirschi (1969) has made the great- attempts to build on and extend the basic assump-
est contributions to our knowledge about bonding tions and propositions of control theory. Michael
processes and deviant behavior. Writing on the Gottfredson, in conjunction with Hirschi, has de-
causes of delinquency, he argued that four aspects veloped a general theory of crime that identifies
of bonding are especially relevant to control theo- ‘‘low self-control,’’ as opposed to diminished so-
ry: emotional attachments to conforming others, cial control, as the primary cause of deviant behav-
psychological commitments to conformity, involve- ior (Hirschi and Gottfredson 1987; Gottfredson
ments in conventional activities, and beliefs consis- and Hirschi 1990). Arguing that all people are
tent with conformity to prevailing norms. inherently self-interested, pursuing enhancement
of personal pleasure and avoiding pain, Gottfredson
Among the most important of the bonding and Hirschi suggest that most crimes, and for that
elements are emotional attachments individuals matter most deviant acts, are the result of choices
may have to conforming others and commitments to maximize pleasure, minimize pain, or both.
667
DEVIANCE THEORIES
Crimes occur when opportunities to maximize In sum, micro-level origin theories look to
personal pleasure are high and when the certainty those aspects of the individual’s social environ-
of painful consequences is low. Further, people ment that influence her or his likelihood of devi-
who pursue short-term gratification with little con- ance. Learning theories stress the importance of
sideration for the long-term consequences of their deviant peers and other significant individuals,
actions are most prone to criminal behavior. In and their impact on attitudes and behaviors favor-
terms of classical control theory, these are indi- able to the commission of deviant acts. These
viduals who have weak bonds to conformity or theories assume that the social environment acts
who disregard or ignore the potentially painful as an agent of change, transforming otherwise
consequences of their actions. They are ‘‘relatively conforming individuals into deviants through peer
unable or unwilling to delay gratification; they are influences. People exposed to deviant others fre-
indifferent to punishment and the interests of quently and sufficiently, like persons exposed to a
others’’ (Hirschi and Gottfredson 1987, pp. 959–960). contagious disease who become ill, will become
deviant themselves. Control theories avoid this
Building on traditional control theory, Charles ‘‘contagion’’ model, viewing the social environ-
Tittle (1995) reasons that it helps explain why ment as a composite of controls and restraints
individuals conform, but it also helps to explain cementing the individual to a conforming lifestyle.
why they engage in deviant behavior. Tittle (1995, Deviance occurs when elements of the bond—
p. 135) argues that ‘‘the amount of control to aspects of social control—are weak or broken,
which an individual is subject, relative to the amount thereby freeing the individual to violate social
of control he or she can exercise, determines the norms. Sanctions and the threat of sanctions are
probability of deviance occurring as well as the particularly important to control theories, a cen-
type of deviance likely to occur.’’ Conformity re- tral part of the calculus that rational actors use in
sults when individuals are subjected to and exert choosing to commit or refrain from committing
roughly equal amounts of control—there is ‘‘con- deviant acts.
trol balance.’’ According to Tittle, however, indi-
viduals who are subjected to more control than The policy implications of micro-level origin
they exert will be motivated to engage in deviance theories are obvious. If, as learning theories argue,
in order to escape being controlled by others. deviance is learned through association with devi-
ant peers, then the way to eliminate deviance is to
Robert Sampson and John Laub (1993) have assist youths in resisting deviant peer influences
also expanded on the basic propositions of control and helping them to develop attitudes that disap-
theory. In their research, Sampson and Laub focus prove of deviant behavior. Control theories, on
on stability and change in the antisocial behavior the other hand, suggest that deviance can be re-
of individuals as they grow from juveniles to adults. duced with programs that help families develop
Sampson and Laub argue that family, school, and stronger bonds between parents and children.
peer relationships influence the likelihood of devi- Control theory also implies that programs that
ant behavior among juveniles. In particular, Sampson help youths develop stronger commitments to
and Laub argue that the structure of the family conventional lines of activity and to evaluate the
(e.g., residential mobility, family size) affects fami- costs and benefits of deviant acts will also result in
ly context or process (e.g., parental supervision, a reduction of problematic behavior.
discipline), which, in turn, makes deviance among
children more or less likely. Many adolescent de- MICRO-LEVEL REACTIONS TO DEVIANCE
linquents grow up to become adult criminals be-
cause their juvenile delinquency makes the forma- Unlike micro-level origin theories, micro-level re-
tion of adult social bonds to work and family less action theories make no assumptions about the
likely. Despite this continuity in antisocial behav- existence of a homogeneous, pervasive set of norms
ior from adolescence to adulthood, however, in society. These theories take an altogether differ-
Sampson and Laub argue that many juvenile delin- ent approach to explaining deviant behavior, view-
quents do not commit deviant acts as adults be- ing deviance as a matter of definition; a social
cause they develop adult social bonds, such as status imposed by individuals or groups on others.
attachment to a spouse or commitment to a job. Most argue that there exists no single pervasive set
668
DEVIANCE THEORIES
of norms in society and that deviant behavior may initial deviant acts. According to Lemert (1951, p.
best be understood in terms of norms and their 287), deviance may often involve instances where
enforcement. These theories typically stress the ‘‘a person begins to employ his deviant behav-
importance of labeling processes—the mechanisms ior. . . as a means of defense, attack or adjustment
by which acts become defined or labeled as ‘‘devi- to the. . . problems created by the consequent
ant—and the consequences of labeling for the social reactions to him.’’ Therefore, a cause of
person so labeled. Many of these theories are deviant careers is negative social labeling; instanc-
concerned with the development of deviant life- es where reactions to initial deviant acts are harsh
styles or careers; long-term commitments to devi- and reinforce a ‘‘deviant’’ self-definition. Such
ant action. labeling forces the individual into a deviant social
role, organizing his or her identity around a pat-
One of the most important writers in this tern of deviance that structures a way of life and
tradition is Howard Becker (1963). Becker argues perpetuates deviant behavior (Becker 1963; Schur
that deviance is not a property inherent in any 1971, 1985).
particular form of behavior but rather a property
conferred on those behaviors by audiences wit- Perhaps the most significant developments in
nessing them. Becker (1963, p. 9) notes that ‘‘. . . this tradition have contributed to knowledge about
deviance is not a quality of the act the person the causes of mental illness. Proponents of micro-
commits, but rather a consequence of the applica- level reaction theories argue that the label ‘‘mental
tion by others of rules and sanctions to an ‘offend- illness’’ can be so stigmatizing to those labeled,
er.’ The deviant is one to whom that label has been especially when mental-health professionals im-
successfully applied; deviant behavior is behavior pose the label, that they experience difficulty re-
that people so label.’’ Thus, Becker and others in turning to nondeviant social roles. As a result, the
this tradition orient the study of deviance on rules labeling process may actually exacerbate mental
and sanctions, and the application of labels. Their disorders. Former mental patients may find them-
primary concern is the social construction of devi- selves victims of discrimination at work, in person-
ance—that is, how some behaviors and classes of al relationships, or in other social spheres (Scheff
people come to be defined as ‘‘deviant’’ by others 1966). This discrimination, and the widespread
observing and judging the behavior. belief that others devalue and discriminate against
mental patients, may lead to self-devaluation and
Building on the idea that deviance is a proper- fear of social rejection by others (Link 1982, 1987).
ty conferred on behavior that is witnessed by a In some instances, this devaluation and fear may
social audience, Becker (1963) also developed a be associated with demoralization of the patient,
simple typology of deviant behavior. The dimen- loss of employment and personal income, and the
sions upon which the typology is based are wheth- persistence of mental disorders following treat-
er or not the individual is perceived as deviant and ment (Link 1987).
whether or not the behavior violates any rule.
Conforming behavior is behavior that does not Hence, micro-level reaction theories reason
violate any rules and is not perceived as deviant. that deviant behavior is rooted in the process by
Individuals in the opposite scenario, in which the which persons define and label the behavior of
person both violates rules and is perceived by others as deviant. The theories offer explanations
others as deviant, Becker labeled pure deviants. of individual differences in deviance, stressing the
Some individuals, according to Becker, may be importance of audience reactions to initial deviant
perceived as deviant, even though they have not acts. However, these theories make no attempt to
violated any rules. Becker identified these indi- explain the origins of the initial acts (Scheff 1966).
viduals as the falsely accused. Finally, the secret devi- Rather, they are concerned primarily with the
ant is one who has violated the rules, but, nonethe- development and persistence of deviant careers.
less, is not perceived by others as being deviant.
Micro-level reaction theories have very differ-
Equally important is the work of Edwin Lemert ent implications for public policy than macro- and
(1951). Stressing the importance of labeling to micro-level origins theories. Micro-level reaction
subsequent deviant behavior, he argues that re- theories argue that unwarranted labeling can lead
petitive deviance may arise from social reactions to
669
DEVIANCE THEORIES
to deviant careers. In effect, the reaction to devi- management and control. In societies or communities
ance can cause deviant behavior to escalate. Thus, characterized by rigid economic stratification, elites
in order to reduce deviance, agencies of social are likely to impose formal social control in order
control must adopt policies of nonintervention. to preserve the prevailing economic order.
Rather than being formally sanctioned and la-
beled as deviant, nonintervention policies must Conflict theories stress the importance of the
encourage diversion and deinstitutionalization. For- political structure of society and focus on the
mal sanctioning must be highly selective, focusing degree of threat to the hegemony of political
only on the most serious and threatening devi- elites, arguing that elites employ formal social
ant acts. controls to regulate threats to political and social
order (Turk 1976; Chambliss 1978; Chambliss and
MACRO-LEVEL REACTIONS TO DEVIANCE Mankoff 1976). According to these theories, threat
varies in relation to the size of the problem popula-
The final class of theories looks to the structure of tion, with large problem populations substantially
economic and political power in society as a cause more threatening to political elites than small
of deviant behavior. Macro-level reaction theo- populations. Thus, elites in societies and commu-
ries—either Marxist or other conflict theories— nities in which those problem populations are
view deviance as a status imposed by dominant large and perceived as especially threatening are
social classes to control and regulate populations more likely to process members of the problem
that threaten political and economic hegemony. populations as deviants than in areas where such
Like micro-level reaction theories, these theories problems are small.
view deviance as a social construction and accord
greatest importance to the mechanisms by which Much of the writing in this tradition has ad-
society defines and controls entire classes of be- dressed the differential processing of people de-
havior and people as deviant in order to mediate fined as deviant. Typically, this writing has taken
the threat. However, these theories reason that the two forms. The first involves revisionist histories
institutional control of deviants has integral ties to linking the development of prisons, mental asylums,
economic and political order in society. and other institutions of social control to structur-
al changes in U.S. and European societies. These
Marxist theories stress the importance of the histories demonstrate that those institutions often
economic structure of society and begin with the target the poor and chronically unemployed inde-
assumption that the dominant norms in capitalist pendent of their involvement in crime and other
societies reflect the interests of the powerful eco- deviant acts, and thereby protect and serve the
nomic class; the owners of business. But contem- interests of dominant economic and political groups
porary Marxist writers (Quinney 1970, 1974, 1980; (Scull 1978; Rafter 1985).
Spitzer 1975; Young 1983) also argue that modern
capitalist societies are characterized by large ‘‘prob- A second and more extensive literature in-
lem populations’’—people who have become dis- cludes empirical studies of racial and ethnic dis-
placed from the workforce and alienated from the parities in criminal punishments. Among the most
society. Generally, the problem populations in- important of these studies is Martha Myers and
clude racial and ethnic minorities, the chronically Suzette Talarico’s (1987) analysis of the social and
unemployed, and the extremely impoverished. structural contexts that foster racial and ethnic
They are a burden to the society and particularly to disparities in the sentencing of criminal offenders.
the capitalist class because they create a form of Myers and Talarico’s research, and other studies
social expense that must be carefully controlled if examining the linkages between community social
the economic order is to be preserved. structure and differential processing (Myers 1987,
1990; Peterson and Hagan 1984; Bridges, Crutchfield,
Marxist theories reason that economic elites and Simpson 1987; Bridges and Crutchfield 1988),
use institutions such as the legal, mental-health, demonstrate the vulnerability of minorities to dif-
and welfare systems to control and manage socie- ferential processing during historical periods and
ty’s problem populations. In effect, these institu- in areas in which they are perceived by whites as
tions define and process society’s problem popula- serious threats to political and social order. In
tions as deviant in order to ensure effective effect, minorities accused of crimes during these
670
DEVIANCE THEORIES
periods and in these geographic areas are per- justice system, can also help to reduce the dispro-
ceived as threats to white hegemony, and there- portionate processing of less powerful groups,
fore become legitimate targets for social control. such as racial minorities, as deviant.
In addition to studying the connections be- NEW THEORETICAL DIRECTIONS
tween community social structure and the differ-
ential processing of racial and ethnic minorities, A recurring issue in the study of deviance is the
researchers have also begun to examine how court contradictory nature of many deviance theories.
officials’ perceptions of offenders can influence The theories often begin with significantly differ-
disparities in punishments. Bridges and Steen ent assumptions about the nature of human be-
(1998), for example, show how court officials’ havior and end with significantly different conclu-
perceptions of white and minority youths differ, sions about the causes of deviant acts. Some scholars
and how these different perceptions contribute to maintain that the oppositional nature of these
different recommendations for sentencing. Pro- theories—the theories are developed and based
bation officers often attribute the offenses of mi- on systematic rejection of other theories (Hirschi
nority youths to internal characteristics of the 1989)—tends toward clarity and internal consis-
youths (i.e., aspects of their personality), while tency in reasoning about the causes of deviance.
attributing the offenses of white youths to external However, other scholars argue that this oppositional
characteristics (i.e., aspects of their environments). nature is intellectually divisive—acceptance of one
As a result of these differential attributions, mi- theory precludes acceptance of another—and ‘‘has
nority youths are perceived as more threatening, made the field seem fragmented, if not in disar-
more at risk for re-offending than whites and more ray’’ (Liska, Krohn, and Messner 1989, p. 1).
likely to receive severe recommendations for
sentences. A related and equally troublesome problem is
the contradictory nature of much of the scientific
Thus, macro-level reaction theories view devi- evidence supporting deviance theories. For each
ance as a by-product of inequality in modern theory, there exists a literature of studies that
society, a social status imposed by powerful groups supports and a literature that refutes major argu-
on those who are less powerful. Unlike micro-level ments of the theory. And although nearly every
reaction theories, these theories focus on the forms theory of deviance may receive empirical confir-
of inequality in society and how entire groups mation at some level, virtually no theory of devi-
within the society are managed and controlled as ance is sufficiently comprehensive to withstand
deviants by apparatuses of the state. Like those empirical falsification at some other level. The
theories, however, macro-level reaction theories difficult task for sociologists is discerning whether
make little or no attempt to explain the origins of and under what circumstances negative findings
deviant acts, claiming instead that the status of should be treated as negating a particular theory
‘‘deviant’’ is, in large part, a social construction (Walker and Cohen 1985).
designed primarily to protect the interests of the
most powerful social groups. The primary concern In recent years, these two problems have re-
of these theories is explicating the linkages be- newed sociologists’ interest in deviance theory
tween inequality in society and inequality in the and, at the same time, suggested new directions
labeling and processing of deviants. for the development of theory. The oppositonal
nature of theories has spawned interest in theo-
Since macro-level reaction theories view devi- retical integration. Many scholars are dissatisfied
ance as a status imposed by powerful groups on with classical theories, arguing that their predic-
those with less power, the most immediate policy tive power is exceedingly low (see Elliott 1985;
implication of these theories is that imbalances in Liska, Krohn, and Messner 1989). Limited to a few
power and inequality must be reduced in order to key explanatory variables, any one theory can
reduce levels of deviance and levels of inequality in explain only a limited range and amount of devi-
the sanctioning of deviance. More effective moni- ant behavior. And because most scholars reason
toring of government agencies that are used to that the causes of deviance are multiple and quite
control problem populations, such as the criminal complex, most also contend that it may be ‘‘neces-
sary to combine different theories to capture the
671
DEVIANCE THEORIES
entire range of causal variables’’ (Liska, Krohn, to theory development encourages scholars to
and Messner 1989, p. 4). explore more fully the strengths and limitations of
their own theories. This approach will require
Because it combines the elements of different more complete elaboration of extant theory, ex-
theories, the new theory will have greater explana- plicitly specifying those circumstances under which
tory power than theories from which it was de- each theory may be meaningfully tested and thus
rived. However, meaningful integration of devi- falsified. The result will be a greater specification
ance theories will require much more than the of each theory’s contribution to explanations of
simple combination of variables. Scholars must deviant behavior.
first reconcile the oppositional aspects of theories,
including many of their underlying assumptions These two directions have clear and very dif-
about society, the motivations of human behavior, ferent implications for the development of de-
and the causes of deviant acts. For example, learn- viance theory. Theoretical integration offers
ing theories focus heavily on the motivations for overarching models of deviant behavior that cut
deviance, stressing the importance of beliefs and across classical theories, combining different lev-
values that ‘‘turn’’ the individual to deviant acts. In els of explanation and causal focuses. If funda-
contrast, control theories accord little importance mental differences between theories can be recon-
to such motivations, examining instead those as- ciled, integration is promising. The specification
pects of the social environment that constrain of scope conditions offers greater clarification of
people from committing deviant acts. Reconciling existing theories, identifying those conditions un-
such differences is never an easy task, and in some der which each theory most effectively applies.
instances may be impossible (Hirschi 1979). Although this direction promises no general theo-
ries of deviance, it offers the hope of more mean-
The problem of contradictory evidence sug- ingful and useful explanations of deviant behavior.
gests a related but different direction for deviance
theory. Theories may vary significantly in the con- REFERENCES
ditions—termed scope conditions—under which they
apply (Walker and Cohen 1985; Tittle 1975; Tittle Agnew, Robert 1992 ‘‘Foundation for a General Strain
and Curran 1988). Under some scope conditions, Theory of Crime and Delinquency.’’ Criminology
theories may find extensive empirical support, 30:47–88.
and under others virtually none. For instance,
macro-level origin theories concerned with the Akers, Ronald L. 1985 Deviant Behavior: A Social Learn-
frustrating effects of poverty on deviance may ing Approach. Belmont, Calif.: Wadsworth.
have greater applicability to people living in dense-
ly populated urban areas than those living in rural ———, Marvin D. Krohn, Lonn Lanza-Kaduce, and
areas. The frustration of urban poverty may be Marcia Radosevich 1979 ‘‘Social Learning and Devi-
much more extreme than in rural areas, even ant Behavior: A Specific Test of a General Theory.’’
though the actual levels of poverty may be the American Sociological Review 44:636–655.
same. As a result, the frustrations of urban poverty
may be more likely to cause deviant adaptations in Becker, Howard 1963 The Outsiders. Chicago: University
the form of violent crime, drug abuse, and vice of Chicago Press.
than those of rural poverty. In this instance,
‘‘urbanness’’ may constitute a condition that acti- Blau, Judith, and Peter Blau 1982 ‘‘Metropolitan Struc-
vates strain theories linking poverty to deviance. ture and Violent Crime.’’ American Sociological Review
Obviously, the same theories simply may not apply 47:114–128.
in rural areas or under other conditions.
Bridges, George S., and Robert D. Crutchfield 1988
Effective development of deviance theory will ‘‘Law, Social Standing and Racial Disparities in Im-
require much greater attention to the specifica- prisonment.’’ Social Forces 66:699–724.
tion of such scope conditions. Rather than com-
bining causal variables from different theories as ———, and Edith Simpson 1987 ‘‘Crime, Social Struc-
integrationists would recommend, this approach ture and Criminal Punishment.’’ Social Problems
34:344–361.
———, and Sara Steen 1998 ‘‘Racial Disparities in Offi-
cial Assessments of Juvenile Offenders: Attributional
Stereotypes as Mediating Mechanisms.’’ American
Sociological Review 63:554–570.
672
DEVIANCE THEORIES
Burgess, Robert L., and Ronald Akers 1966 ‘‘A Differen- Lemert, Edwin 1951 Social Pathology. New York: Mc-
tial Association-Reinforcement Theory of Criminal Graw-Hill.
Behavior.’’ Social Problems 14:128–147.
Link, Bruce 1982 ‘‘Mental Patient Status, Work and
Bursik, Robert J. 1984 ‘‘Urban Dynamics and Ecological Income: An Examination of the Effects of a Psychiat-
Studies of Delinquency.’’ Social Forces 63:393–413. ric Label.’’ American Sociological Review 47:202–215.
———, and J. Webb 1982 ‘‘Community Change and ——— 1987 ‘‘Understanding Labeling Effects in the
Ecological Studies of Delinquency.’’ American Journal Area of Mental Disorders: An Assessment of the
of Sociology 88:24–42. Effects of Expectations of Rejection.’’ American Socio-
logical Review 52:96–112.
Chambliss, William 1978 On the Take. Bloomington:
Indiana University Press. Liska, Allen E., Marvin D. Kroh, and Steven S. Messner
1989 ‘‘Strategies and Requisites for Theoretical Inte-
———, and Milton Mankoff 1976 Whose Law, What gration in the Study of Deviance and Crime.’’ In
Order? New York: Wiley. Steven S. Messner, Marvin D. Krohn, and Allen E.
Liska, eds., Theoretical Integration in the Study of Devi-
Cloward, Richard 1959 ‘‘Illegitimate Means, Anomie ance and Crime: Problems and Prospects. Albany, N.Y.:
and Deviant Behavior.’’ American Sociological Review SUNY Press.
24:164–176.
MacLeod, Jay 1995 Ain’t No Makin’ It. Boulder, Colo.:
———, and Lloyd E. Ohlin 1960 Delinquency and Oppor- Westview.
tunity. New York: Free Press.
Matsueda, Ross L. 1982 ‘‘Testing Control Theory and
Crutchfield, Robert D. 1989 ‘‘Labor Stratification and Differential Association: A Causal Modeling Ap-
Violent Crime.’’ Social Forces 68:489–513. proach.’’ American Sociological Review 47:36–58.
Curtis, Lynn 1975 Violence, Race and Culture. Lexington, ———, and Karen Heimer 1987 ‘‘Race, Family Struc-
Mass.: Heath. ture and Delinquency: A Test of Differential Associa-
tion and Social Control Theories.’’ American Socio-
Elliott, Delbert 1985 ‘‘The Assumption That Theories logical Review 52:826–840.
can be Combined with Increased Explanatory Power:
Theoretical Integrations.’’ In Robert F. Meier, ed., ———, Irving Piliavin, and Rosemary Gartner 1988
Theoretical Methods in Criminology. Beverly Hills, ‘‘Ethical Assumptions Versus Empirical Research.’’
Calif.: Sage. American Sociological Review 53:305–307.
Erlanger, Howard 1974 ‘‘The Empirical Status of the Merton, Robert K. 1938 ‘‘Social Structure and Ano-
Subculture of Violence Thesis.’’ Social Problems mie.’’ American Sociological Review 3:672–682.
22:280–292.
——— 1964 Social Theory and Social Structure. New York:
Gastil, Raymond 1971 ‘‘Homicide and a Regional Cul- Free Press.
ture of Violence.’’ American Sociological Review
36:412–427. Messner, Steven F. 1983 ‘‘Regional and Racial Effects on
the Urban Homicide Rate: The Subculture of Vio-
Gibbs, Jack P. 1975 Crime, Punishment and Deterrence. lence Revisited.’’ American Journal of Sociology
New York: Elsevier. 88:997–1,007.
——— 1977 ‘‘Social Control, Deterrence, and Perspec- Miller, Walter B. 1958 ‘‘Lower Class Culture as a Gener-
tives on Social Order.’’ Social Forces 62:359–374. ating Milieu of Gang Delinquency.’’ Journal of Social
Issues 14:5–19.
Gottfredson, Michael R., and Travis Hirschi 1990 A
General Theory of Crime. Palo Alto, Calif.: Stanford Myers, Martha 1987 ‘‘Economic Inequality and Dis-
University Press. crimination in Sentencing.’’ Social Forces 65:746–766.
Hirschi, Travis 1969 Causes of Delinquency. Berkeley: ——— 1990 ‘‘Economic Threat and Racial Disparities in
University of California Press. Incarceration: The Case of Postbellum Georgia.’’
Criminology 28:627–656.
——— 1979 ‘‘Separate and Unequal is Better.’’ Journal
of Research in Crime and Delinquency 16:34–37. ———, and Suzette Talarico 1987 Social Contexts of
Criminal Sentencing. New York: Springer-Verlag.
——— 1989 ‘‘Exploring Alternatives to Integrated Theo-
ry.’’ In Steven F. Messner, Marvin D. Krohn, and Paternoster, Raymond L., Linda Saltzman, Gordon P.
Allen E. Liska, eds., Theoretical Integration in the Study Waldo, and Theodore Chiricos 1982 ‘‘Perceived Risk
of Deviance and Crime: Problems and Prospects. Albany, and Deterrence: Methodological Artifacts in Deter-
N.Y.: SUNY Press. rence Research.’’ Journal of Criminal Law and Crimi-
nology 73:1,243–1,255.
———, and Michael R. Gottfredson 1987 ‘‘Causes of
White Collar Crime.’’ Criminology 25:949–974.
673
DIFFUSION THEORIES
——— 1983 ‘‘Perceived Risk and Social Control.’’ Jour- Tittle, Charles R. 1975 ‘‘Deterrents or Labeling?’’ Social
nal of Criminal Law and Criminology 74:457–480. Forces 53:399–410.
Peterson, Ruth D., and John Hagan 1984 ‘‘Changing ——— 1980 Sanctions and Social Deviance. New York:
Conceptions of Race: Toward an Account of Anoma- Praeger.
lous Findings of Sentencing Research.’’ American
Sociological Review 49:56–70. ——— 1995 Control Balance: Toward a General Theory of
Deviance. Boulder, Colo.: Westview Press.
Piliavin, Irving, Rosemary Gartner, Craig Thornton,
and Ross L. Matsueda 1986 ‘‘Crime, Deterrence and ———, and Barbara A. Curran 1988 ‘‘Contingencies for
Rational Choice.’’ American Sociological Review Dispositional Disparities in Juvenile Justice.’’ Social
51:101–120. Forces 67:23–58.
Quinney, Richard 1970 The Social Reality of Crime. Bos- Toby, Jackson 1957 ‘‘Social Disorganization and Stakes
ton: Little, Brown. in Conformity: Complimentary Factors in the Preda-
tory Behavior of Hoodlums.’’ Journal of Criminal Law
——— 1974 Critique of Legal Order. Boston: Little, Brown. and Criminology and Police Science 48:12–17.
——— 1980 Class, State and Crime. New York: Longman. Turk, Austin T. 1976 ‘‘Law as a Weapon in Social
Conflict.’’ Social Problems 23:276–291.
Rafter, Nicole Hahn 1985 Partial Justice: Women in State
Prisons, 1800–1935. Boston: Northeastern Universi- Walker, Howard, and Bernard P. Cohen 1985 ‘‘Scope
ty Press. Statements.’’ American Sociological Review 50:288–301.
Sampson, Robert J., and John H. Laub 1993 Crime in the Wolfgang, Marvin E., and Franco Ferracuti 1967 The
Making: Pathways and Turning Points Through Life. Subculture of Violence: Towards an Integrated Theory in
Cambridge, Mass.: Harvard University Press. Criminology. London: Tavistock.
Scheff, Thomas 1966 Being Mentally Ill. Chicago: Aldine. Young, Peter 1983 ‘‘Sociology, the State and Penal
Relations.’’ In David Garland and Peter Young, eds.,
Schur, Edwin 1971 Labeling Deviant Behavior. New York: The Power to Punish. London: Heinemann.
Harper and Row.
GEORGE S. BRIDGES
——— 1985 Labeling Women Deviant: Gender, Stigma and SCOTT A. DESMOND
Social Control. New York: Harper and Row.
DIFFERENTIAL ASSOCIATION
Scull, Andrew T. 1978 Museums of Madness. London:
Heinemann. See Crime, Theories of; Deviance Theories.
Sellin, Thorsten 1938 Culture, Conflict and Crime. New DIFFUSION THEORIES
York: Social Science Research Council.
The concept of diffusion inherently focuses upon
Shaw, Clifford 1930 The Jack-Roller. Chicago: University process. Diffusion refers to the dissemination of
of Chicago Press. any physical element, idea, value, social practice,
or attitude through and between populations. Dif-
———, and Henry McKay 1942 Juvenile Delinquency and fusion is among the rare concepts used across the
Urban Areas. Chicago: University of Chicago Press. physical, natural, and social sciences, as well as in
the arts. Diffusion is most closely associated with
Silberman, Charles 1980 Criminal Justice, Criminal Vio- the social sciences, particularly rural sociology,
lence. New York: Vintage. anthropology, and communication. Diffusion think-
ing offers a logic through which to describe and
Spitzer, Steven 1975 ‘‘Toward a Marxian Theory of perhaps explain myriad types of change that in-
Deviance.’’ Social Problems 22:638–651. volve equally diverse foci, ranging from the adop-
tion of internet technology (Adams 1997), to the
Stark, Rodney 1987 ‘‘Deviant Places: A Theory of the spread of belief systems (Dean 1997).
Ecology of Crime.’’ Criminology 25:893–910.
Work connected to the concept of diffusion is
Sullivan, Mercer L. 1989 Getting Paid: Youth Crime and arguably structured as theory. Certainly there is no
Work in the Inner City. Ithaca, N.Y.: Cornell Universi-
ty Press.
Sutherland, Edwin H. 1947 Principles of Criminology, 4th
ed. Philadelphia: Lippincott.
Thornberry, Terence P, and R. L. Christenson 1984
‘‘Unemployment and Criminal Involvement: An In-
vestigation of Reciprocal Causal Structures.’’ Ameri-
can Sociological Review 49:398–411.
674
DIFFUSION THEORIES
single, unified, deductively structured collection proposed the notion of diffusion as a means of
of propositions, widely regarded by social scien- explaining the appearance of similar culture ele-
tists as identifying the principal mechanisms of ments in different groups and of understanding
diffusion that could be employed across all sub- the progressive alteration of elements within the
stantive areas. There are however, many distinct same group. As the twentieth century began, diffu-
collections of propositions, a few well-tested over sion arose as an alternative to evolution as a basis
decades, that describe different diffusion phenome- for understanding cultural differences and change.
na in different content areas. Indeed, in the area of Evolutionists argued that cultural similarities prob-
innovation diffusion, Everett Rogers (1983) pro- ably arose through independent invention. Those
duced a formal theory that is broadly recognized, who embraced diffusion presented it as a more
often tested, and that has been adapted to other parsimonious explanation, emphasizing that traits
content areas including disaster research and tech- and institutions could pass between groups by
nology transfer. Part of the problem in formalizing means of contact and interaction.
diffusion theories is that the concept does not
inherently specify content (rather a framework or Historical Development. The English anthro-
process to structure thinking). However, it cannot pologists W. J. Perry and Elliot Smith devised the
be examined empirically without tying it to some most extreme position on cultural diffusion. These
substance. That is, studies of diffusion focus on scholars held that human culture originated in
something (a technology, idea, practice, attitude, Egypt and progressively diffused from that center
etc.) that is being diffused. Consequently, the re- over the remainder of the earth. In Germany, Fritz
search on diffusion that would drive theory build- Graebner (1911) argued that critical aspects of
ing has remained scattered in the literatures of cultures—toolmaking, for example—originated in a
different sciences, and as such it is not readily small number of geographically isolated societies.
pulled together. These conditions do not facilitate This hypothesis formed the basis for culture circles
ready assembly of information that would encour- (‘‘kulturkreise’’), collections of societies sharing
age creation of a general theory of diffusion. Thus, similar cultures. Unlike British diffusionists who
one must look to the growth of formal theory in emphasized tracking the movement of single cul-
sub-areas, although Rogers (in press) has begun to ture elements, Graebner and others in his tradi-
cross content spheres. tion focused on the dissemination of collections of
elements or cultural complexes.
There are at least three traditions or theory
families that can be historically discerned in the American anthropologists are credited with
study of diffusion. Rogers (1983, p. 39) has point- developing a social scientifically workable concept
ed out that for many years these traditions re- of diffusion. Franz Boas (1896) conceived of diffu-
mained largely distinct, with little overlap and sion as a viable mechanism for culture exchanges
cross-fertilization. Since the late l970s the level of among geographically adjacent areas. His view
research and theoretical isolation has decreased, figured prominently in the intellectual move away
leading to an enhanced awareness among the from the deterministic view of diffusion proposed
perspectives and some integration of empirical by early British anthropologists. Alfred Kroeber
findings into more general theoretical statements. (1923, p. 126) and Robert Lowie (1937, p. 58)—
The three theory families are: (1) cultural diffu- students of Boas—subsequently developed a posi-
sion; (2) diffusion of innovations; and (3) collec- tion called moderate diffusionism, which is currently
tive behavior. widely accepted in anthropology. This position
allowed for the coexistence of a variety of mecha-
CULTURAL DIFFUSION nisms of change and transfer—independent in-
vention, acculturation, etc.—in addition to diffusion
The earliest social scientific use of the term diffu- in accounting for culture change and differentia-
sion is found in Edward Tylor’s (1865) treatment of tion. Clark Wissler (1929), a Kroeber contempo-
culture change. Anthropologists have long attempt- rary, established an empirical basis for culture
ed to explain similarities and differences among diffusion by identifying ten culture areas (regions
cultures, especially those that were geographically with similar cultural inventories) in North and
adjacent. Tylor’s work on culture change first South America and the Caribbean.
675
DIFFUSION THEORIES
Theory in Cultural Diffusion. In terms of over time the sociological focus became identify-
theory development, cultural diffusion is the actual ing social psychological motivations and mecha-
movement of a given social institution or physical nisms supporting the diffusion process (Park and
implement, while stimulus diffusion is the exchange Burgess 1921, p. 20). Recently, sociological work
or movement of the principle upon which an insti- directly on culture is bifurcated, with one group of
tution or implement is based. In the cultural diffu- scientists still emphasizing the social psychological
sion literature, scholars have enumerated assump- issues in culture meaning (Wuthnow and Witten
tions, stated principles, and reviewed empirical 1988), and another more concerned with structur-
work with the objective of identifying propositions al (mathematical or statistical) models of culture
tested repeatedly and not found to be false. In- processes themselves (Griswold 1987). Neither
deed, beginning with the work of early twentieth group has especially focused on diffusion theory
century anthropologists, one can identify at least as a mechanism to track or identify the content-
five broadly accepted and empirically supported outcomes of culture change.
claims that form the core of what is called cultural
diffusion theory. First, borrowed elements usually DIFFUSION OF INNOVATIONS
undergo some type of alteration or adaptation in
the new host culture. Second, the act of borrowing The diffusion of innovations has historically fo-
depends on the extent to which the element can be cused on the spread of an idea, procedure, or
integrated into the belief system of the new cul- implement within a single social group or between
ture. Third, elements that are incompatible with multiple groups. For the most part, scholars of this
the new culture’s prevailing normative structure tradition define diffusion as the process through
or religious belief system are likely to be rejected. which some innovation is communicated within a
Fourth, acceptance of an element depends upon social system. Also important is the notion of a
its utility for the borrower. Finally, cultures with a time dimension reflecting the rate of diffusion,
history of past borrowing are more likely to bor- and the importance of the individual adopter (or
row in the future. These claims constitute the non-adopter) reflecting the role of social influence.
‘‘core propositions’’ of culture diffusion theory;
over the years, each has been qualified and elabo- The study of innovation diffusion began rath-
rated upon, and corollaries have been created er narrowly, grew to dominate the field of rural
(Stahl 1994). sociology for a time, contracted in popularity for
many years, and then spawned wide interest across
Currently, diffusion is seen as a mechanism several disciplines. Innovation diffusion study con-
for culture change that typically accounts for a tains several groups: those who focus on content
large proportion of any particular culture invento- or the specific innovation being diffused; those
ry. The deterministic, linear view of diffusion has who emphasize theoretical elaborations of generic
been discredited by the empirical record. The principles of innovation diffusion; and those con-
concept of culture diffusion as a means of under- cerned with creating structural models to track
standing cultural inventories is entrenched in the diffusion. Particularly in the past decade, the lit-
field of cultural anthropology. As the twenty-first erature has seen much cross-fertilization, although
century dawns, a principal controversy among mathematical modelers tend to appear less often
cultural anthropologists centers on the definition in the work of other diffusion scholars. Although
of culture, rather than upon the acceptability of the roots of innovation diffusion theory are seen
culture diffusion. Diffusion theory remains promi- to be largely in rural sociology, more recently the
nent in the archaeology literature, particularly as a field has become distinctly interdisciplinary with
means of tracing culture inventories for groups major advancements made especially in the disci-
over time (Posnansky and DeCorse 1986). pline of communication.
Sociologists were initially involved in the use Historical Development. The definitive histo-
of cultural diffusion theory as a means of looking ry of the diffusion of innovations as a paradigm
at cultural change (largely in terms of nonmaterial was published by Thomas Valente and Everett
culture) in the United States. Initially theory drawn Rogers (1995). The roots of innovation diffusion
from anthropology was used (Chapin 1928), but
676
DIFFUSION THEORIES
are usually traced to Gabriel Tarde (1890) who Indeed, the replications that these studies repre-
didn’t use the term diffusion, but was the first to sent substantially facilitated the later sophisticated
address the notions of adopters and the role of theoretical work initiated in the early 1960s (Rog-
social influence in adoption, as well as to identify ers 1962), and continued in the 1980s (Rogers
the S-shaped curve associated with the rate of an 1983, 1988).
innovation’s adoption. The formative empirical
work on innovation diffusion can be traced to The 1960s marked the beginning of the de-
Bryce Ryan’s Iowa State University-based study of cline of the central role of rural sociologists in
hybrid corn seeds (published with Gross in 1943), innovation diffusion research. In large part this
and Raymond Bowers’s (1937) study of the accept- was due to changes in the field of rural sociology,
ance and use of ham radio sets. For more than two but it also reflected the increasing involvement of
decades following this pioneering work, the study researchers from other disciplines, changing the
of innovation diffusion and particularly theory sheer proportion of rural sociologists working on
development took place within the context of innovation diffusion. After more than two decades
rural sociology. This circumstance was a function of extensive research on the diffusion of agricul-
of a variety of forces, principal among which were tural innovations, rural sociologists—like other
the location of rural sociologists in land grant social scientists of the time—began to devote more
institutions charged with the dissemination of ag- time to the study of social problems and the conse-
ricultural innovations to farmers (Hightower 1972) quences of technology. Indeed, Crane (1972) ar-
and the communication and stimulation accorded gued that around 1960 rural sociologists began to
by the North Central Rural Sociology Committee’s believe that the critical questions about innovation
(a regional professional society) formation of a diffusion had already been answered. Although
special subcommittee to deal with the issue of the late 1960s saw rural sociologists launch a series
diffusion of agricultural innovations (Valente and of diffusion studies on agricultural change in the
Rogers 1995, p. 254). international arena (particularly Latin America,
Asia and Africa), by 1965 research on diffusion of
Most scholars agree that contemporary views innovations was no longer dominated by members
of innovation diffusion grew from hybrid corn of that field. Of course, innovation diffusion re-
seeds; specifically the research on adoption done search by rural sociologists has continued, includ-
by Bryce Ryan and Neal Gross (1943). These stud- ing studies of the impacts of technological innova-
ies ultimately defined most of the issues that occu- tion diffusion, and diffusion of conservation prac-
pied diffusion researchers and builders of innova- tices and other ecologically-based innovations
tion diffusion theory for decades to come: the role (Fliegel 1993).
of social influence, the timing of adoptions, the
adoption process itself, and interactions among The infusion of researchers from many disci-
adopter characteristics and perceived characteris- plines studying a variety of specific innovations
tics of the innovation. From the middle 1940s initiated the process of expanding the empirical
through the 1950s, rural sociologists vigorously testing of innovation diffusion tenets. This began
developed a body of empirical information on the with studies in education addressing the diffusion
diffusion of innovations. Most of these studies of kindergartens and driver education classes in
remained tied to agriculture and farming, and the 1950s, as well as Richard Carlson’s (1965)
focused on the diffusion of new crop management study of the diffusion of modern math. Another
systems, hybridizations, weed sprays, insect man- major contribution came from the area of public
agement strategies, chemical fertilizers, and ma- health. Elihu Katz, Herbert Menzel, and James
chinery. A common criticism of the studies of this Coleman launched extensive studies of the diffu-
era is that many of the studies seem to be almost sion of a new drug (the antibiotic tetracycline);
replications of the Ryan and Gross work, the main first in a pilot study (Menzel and Katz 1955) and
difference among them being the specific innova- then in studies of four Illinois cities (Coleman,
tion studied. While it is true that these studies tend Menzel, and Katz 1957; Coleman, Katz, and Menzel
to share a common methodology and linear con- 1966). This research greatly expanded knowledge
ception of diffusion, it is also true that they pro- of interpersonal diffusion networks, and in par-
vide a strong foundation of empirical case studies. ticular its influence in adoption. Interestingly, as
Elihu Katz, M. L. Levine, and Harry Hamilton
677
DIFFUSION THEORIES
(1963) indicated, the drug studies truly represent- have studied the diffusion of many target materi-
ed an independent replication of the principles of al elements, phenomena, and other intangibles,
innovation diffusion developed by rural sociolo- but they have continued to produce theoretical
gists because the public health researchers were statements dealing with communication channels,
unaware of the agricultural diffusion research. diffusion networks, interpersonal influence and
Other studies in the public health arena focused the innovation-decision process. Finally, the tech-
on dissemination of new vaccines, family plan- nology dissemination and transfer issues have in-
ning, and new medical technology. volved work by geographers, engineers, and oth-
ers beginning in the 1970s. The primary focus of
Beginning in the late 1960s, there was a sub- such studies has been the spread or dissemina-
stantial increase in the amount of diffusion re- tion of technology (Sahal 1981) and the develop-
search in three disciplines: business marketing, ment of network models of innovation diffusion
communication, and transportation-technology (Valente 1995).
transfer. Marketing research principally address-
ed the characteristics of adopters of new products Theory in Innovation Diffusion. The theo-
and the role of opinion leaders in the adoption retical work of Everett Rogers initially resulted in
process (Howes 1996). This literature is based the collection of knowledge gained from the rural
almost exclusively on commercial products, rang- sociology tradition, then facilitated the transition
ing from coffee brands and soap to touch-tone to communication perspectives, and now has served
telephones, the personal computer, and internet as the mainstay of what is developing as a more
services. The studies tend to be largely atheoretical, cross-disciplinary focus on innovation diffusion.
methodologically similar, and aimed simply at us- His contribution is twofold. First, he created in-
ing knowledge of diffusion either to improve mar- ventories of findings from many disciplines and
keting and sales of the product or to describe from many types of innovation. These inventories
product dissemination. provided impetus for the development of a defini-
tion of innovation diffusion that was not bound by
In sharp contrast, work done on innovation discipline. Second, Rogers assembled and refined
diffusion by scholars trained in communication theoretical structures aimed at explaining the princi-
has been considerably more theoretically orient- pal features of innovation diffusion. The theoreti-
ed. Throughout the 1960s, universities in America cal work has cemented a core of knowledge and
began to establish separate departments of com- principles that are widely identified (and used
munication (Rogers 1994). Since diffusion of inno- empirically) as the bases of the diffusion of innova-
vations was widely seen as one type of communica- tions. Rogers’s (1983) theory includes eighty-one
tion process, scholars in these new departments generalizations (propositions) that have under-
adopted this type of research as one staple of their gone empirical testing.
work. Beginning with studies of the diffusion of
news events (Deutschmann and Danielson 1960), The theory of innovation diffusion may be
this research tradition has branched out to study understood as capturing the innovation-decision
the dissemination of a wide variety of specific process, innovation characteristics, adopter char-
innovations (McQuail 1983, p. 194). Scholars work- acteristics, and opinion leadership. The innova-
ing in this tradition have been principally responsi- tion-decision process represents the framework
ble for the progressive refinements of formalized on which diffusion research is built. It delineates
theory of innovation diffusion. Everett Rogers has the process through which a decision maker (rep-
consistently remained the leader in theory devel- resenting any unit of analysis) chooses to adopt,
opment in communication, revising and extend- reinvent (modify), or reject an innovation. This
ing his 1962 book Diffusion of Innovations with process consists of five stages. Knowledge is the
help from co-author Floyd Shoemaker to pro- initial stage when the decision maker detects the
duce Communication of Innovations in 1971. Subse- existence of the innovation and learns of its func-
quently, Rogers (1983, in press) restored the origi- tion. In the persuasion stage, the decision maker
nal title Diffusion of Innovations, broadened the forms a positive or negative attitude toward the
theoretical base and incorporated diffusion stud- innovation. The third stage, decision, deals with the
ies and thinking from other disciplines. Gener- decision-maker’s choice to accept or reject the
ally, Rogers and other communication scholars
678
DIFFUSION THEORIES
innovation. Implementation, the fourth stage, fol- Identifying the characteristics of people who
lows a decision to accept and involves putting the adopt innovations raises the question of interper-
innovation into some use (in either its accepted sonal influence. Three issues are addressed in the
form or some modified form). During the final development of propositions about the role of
stage of confirmation, decision makers assess an interpersonal influence in the innovation decision
adopted innovation, gather information from sig- process: information flow, opinion leadership, and
nificant others, and choose to continue to use the diffusion networks. Over time, information flow
innovation as is, modify it (reinvention), or reject has been seen as a ‘‘hypodermic needle’’ model, a
it. While some have criticized the stage model as two-step flow (to opinion leaders, then other
too linear, Rogers (1983) has convincingly argued adopters), and a multi-step flow. Currently, infor-
that existing formulations afford a degree of inter- mation flows are seen as multi-step in nature and
pretative and predictive flexibility that averts his- are described in terms of homophily and heterophily—
torical problems with stage models in social science. the degree to which pairs of interacting potential
adopters are similar or dissimilar. Opinion leader-
Different innovations have different proba- ship denotes the degree to which one member of a
bilities of adoption and hence, different adoption social system can influence the attitude and behav-
rates. That is, they travel through the innovation- ior of others. This concept is presently discussed
decision process at varying speeds. The literature relative to spheres of influence, wherein a given
demonstrates that five characteristics of innova- person may be a leader or follower depending
tions influence the adoption decision. Compatibili- upon the part of the diffusion network being
ty refers to the congruence between an innovation referenced. The diffusion or communication net-
and the prevailing norms, values, and perceived work is the structural stage upon which social
needs of the potential adopter. Higher levels of influence takes place. Considerable attention has
compatibility are associated with greater likeli- been devoted to developing analysis strategies and
hood of adoption. Innovation complexity, on the tactics for such networks (Wigand 1988).
other hand, is negatively associated with adoption.
The extent to which use of an innovation is visible COLLECTIVE BEHAVIOR
to the social group—called observability—is posi-
tively related to adoption. Relative advantage refers While diffusion is not a commonly used term
to the extent to which an innovation is perceived in collective behavior, processes of diffusion are
to be ‘‘better’’ than the idea, practice, or element important in connection with understanding crowds,
that it replaces. Higher relative advantage increas- fashion, and some aspects of disaster behavior.
es the probability of adoption. Finally, trialability— In all cases, analytic concern centers on the dis-
the extent to which an innovation may be experi- semination of emotions, social practices, or physi-
mented with—also increases the probability of cal elements through a collectivity. The study of
adoption. human behavior in disasters is recent and
multidisciplinary. In this field there has been a
The third component of diffusion of innova- concern with diffusion in the classic sense of track-
tion theory addresses adopter characteristics. ing ideas and practices through networks. The
Adopter categories are classifications of individu- principle foci of research have been the adoption
als by how readily they adopt an innovation. Rog- of protective measures and the dissemination of
ers (1983, p. 260) identifies nine socioeconomic warning messages (Lindell and Perry 1992), with
variables, twelve personality variables, and ten per- the aim of research being both the development of
sonal communication characteristics that have been general theories of protective behaviors and more
demonstrated to bear upon adoption choices. In effective protection of endangered populations.
general, the literature holds that early adopters are
more likely to be characterized by high socioeco- All three diffusion theory traditions converge
nomic status, high tolerance of uncertainty and in the study of crowd behavior. In proposing
change, low levels of fatalism and dogmatism, high imitation as an explanatory mechanism for crowd
integration into the social system, high exposure actions, Gabriel Tarde (1890, p. 45) drew upon
to mass media and interpersonal communication Edward Tylor’s concept of cultural diffusion. Sub-
channels, and frequent engagement in informa- sequently, Gustave LeBon (1895) and Gabriel Tarde
tion seeking.
679
DIFFUSION THEORIES
(1901) approached crowd behavior in terms of Dean, John 1997 ‘‘The Diffusion of American Culture in
social contagion: rapid dissemination of emotions Western Europe since World War Two.’’ Journal of
among interacting people. Although Floyd Allport American Culture 20:11–32.
(1924) and Herbert Blumer (1939) extended and
formalized the concept of contagion, it has been Deutschmann, P. J. , and W. A. Danielson 1960 ‘‘Diffu-
largely displaced as a theory of crowd behavior by sion of Knowledge of the Major News Story.’’ Journal-
convergence theory (Turner and Killian 1987, p.19). ism Quarterly 37:345–355.
Changes in dress have been conceptualized as Fliegel, Frederick 1993 Diffusion Research in Rural Soci-
diffusion processes. Alfred Kroeber (1919) stud- ology. Westport, Conn.: Greenwood.
ied fashion cycles which he believed diffused system-
atically through civilizations. Katz and Lazarsfeld Graebner, Fritz 1911 Methode der Ethnologie. Heidelberg,
(1955, p. 241) moved away from the initial concern Germany: Carl Winter.
with movement of fashion through networks to
focus more on social influence. Herbert Blumer Griswold, William 1987 ‘‘A Methodological Framework
(1969) firmly established social psychological mo- for the Sociology of Culture.’’ Sociological Methodolo-
tivations as the basis for fashion behavior. Current gy 14:1–35.
theoretical work on fashion continues to empha-
size social psychological approaches wherein fash- Hightower, James 1972 Hard Tomatoes, Hard Times.
ion diffusion issues are peripheral (Davis 1985; Cambridge, Mass.: Shenkman.
Nagasawa, Hutton, and Kaiser 1991).
Howes, David 1996 Cross-Cultural Consumption. New
REFERENCES York: Routledge.
Adams, Tyronne 1997 ‘‘Using Diffusion of Innovations Katz, Elihu, and Paul Lazarsfeld 1955 Personal Influence.
Theory to Enrich Virtual Organizations in Cyberspace.’’ New York: Free Press.
Southern Communication Journal 62:133–141.
Katz, Elihu, M. L. Levine, and H. Hamilton 1963 ‘‘Tradi-
Allport, Floyd 1924 Social Psychology. Boston: Hough- tions of Research on Diffusion of Innovation.’’ Ameri-
ton-Mifflin. can Sociological Review 28:237–253.
Blumer, Herbert 1939 ‘‘Collective Behavior.’’ In Robert Kroeber, Alfred 1919 ‘‘On the Principle of Order in
Park, ed., Outline of the Principles of Sociology. New Civilization as Exemplified by Changes of Fashion.’’
York: Barnes and Noble. American Anthropologist 21:235–263.
——— 1969 ‘‘Fashion.’’ Sociological Quarterly 10:275–291. ——— 1923 Anthropology. New York: Harcourt.
Boas, Franz 1896 The Limitations of the Comparative LeBon, Gustave 1895 Psychologie des foules. Paris: Alcan.
Method of Anthropology. New York: Alfred Knopf.
Lindell, Michael, and Ronald Perry 1992 Behavioral
Bowers, Raymond 1937 ‘‘The Direction of Intra-Societal Foundations of Community Emergency Planning. Wash-
Diffusion.’’ American Sociological Review 2:826–836. ington, D.C.: Hemisphere Publishers.
Carlson, Richard 1965 Adoption of Educational Innova- Lowie, Robert 1937 The History of Ethnological Theory.
tions. Eugene: University of Oregon Center for the New York: Rinehart and Company.
Advanced Study of Educational Administration.
McQuail, Denis 1983 Mass Communication Theory. Bever-
Chapin, F. Stuart 1928 Cultural Change. New York: ly Hills, Calif.: Sage.
Century.
Menzel, Herbert, and Elihu Katz 1955 ‘‘Social Relations
Coleman, James, Herbert Menzel, and Elihu Katz 1957 and Innovation in the Medical Profession.’’ Public
‘‘The Diffusion of an Innovation among Physicians.’’ Opinion Quarterly 19:337–352.
Sociometry 20:253–270.
Nagasawa, Richard, Sandra Hutton, and Susan Kaiser
——— 1966 Medical Innovation: A Diffusion Study. New 1991 ‘‘A Paradigm for the Study of the Social Mean-
York: Bobbs-Merrill. ing of Clothes.’’ Clothing and Textiles Research Journal
10:53–62.
Crane, Diana 1972 Invisible Colleges. Chicago: University
of Chicago Press. Park, Robert, and Ernest Burgess 1921 Introduction to the
Science of Sociology. Chicago: University of Chica-
Davis, Fred 1985 ‘‘Clothing and Fashion as Communica- go Press.
tion.’’ Symbolic Interaction 5:111–126.
Posnansky, M., and C. R. DeCorse 1986 ‘‘Historical
Archaeology in Sub-Saharan Africa.’’ Historical Ar-
chaeology 20:1–14.
Rogers, Everett 1962, 1995 Diffusion of Innovations. New
York: Free Press.
680
DISASTER RESEARCH
——— 1988 ‘‘The Intellectual Foundation and History change consequences of a munitions ship explo-
of the Agricultural Extension Model.’’ Knowledge sion in the harbor of Halifax, Canada. Pitirim
9:492–510. Sorokin (1942) two decades later wrote Man and
Society in Calamity that mostly speculated on how
——— 1994 A History of Communication Study. New war, revolution, famine, and pestilence might af-
York: Free Press. fect the mental processes and behaviors, as well as
the social organizations and the cultural aspects of
———, and Floyd Shoemaker 1971 Communication of impacted populations. However, there was no build-
Innovations. New York: Free Press. ing on these pioneering efforts.
Ryan, Bryce, and Neal Gross 1943 ‘‘The Diffusion of It was not until the early 1950s that disaster
Hybrid Seed Corn in Two Iowa Communities.’’ Ru- studies started to show any continuity and the
ral Sociology 8:15–24. accumulation of a knowledge base. Military inter-
est in possible American civilian reactions to post-
Sahal, Devendra 1981 Patterns of Technological Innova- World War II threats from nuclear and biological
tion. Reading, Mass.: Addison-Wesley. warfare led to support of academic research on
peacetime disasters. The National Opinion Re-
Stahl, Ann 1994 ‘‘Innovation, Diffusion and Culture search Center (NORC) undertook the key project
Contact.’’ Journal of World Prehistory 8:51–112. at the University of Chicago between 1950 and
1954. This work, intended to be multidisciplinary,
Tarde, Gabriel 1890 Les Lois de l’imitation. Paris: Alcan. became dominated by sociologists as were other
concurrent studies at the University of Oklahoma,
——— 1901 L’opinion et la foule. Paris: Alcan. Michigan State, and the University of Texas
(Quarantelli 1987, 1994). The NORC study not
Turner, Ralph, and Lewis Killian 1987 Collective Behav- only promoted field research as the major way for
ior. Englewood Cliffs, N.J.: Prentice-Hall. gathering data, but also brought sociological ideas
from the literature on collective behavior and
Tylor, Edward 1865 Early History of Mankind and the notions of organizational structure and functions
Development of Civilisation. London: John Murray. into the thinking of disaster researchers (Dynes
1988; Dynes and Tierney 1994).
Valente, Thomas 1995 Network Models of the Diffusion of
Innovations. Cresskill, N.J.: Hampton. While the military interest quickly waned, re-
search in the area obtained a strategic point of
———, and Everett Rogers 1995 ‘‘The Origins and salience and support when the U.S. National Acade-
Development of the Diffusion of Innovations Para- my of Sciences in the late 1950s created the Disas-
digm.’’ Science Communication 16:242–273. ter Research Group (DRG). Operationally run by
sociologists using the NORC work as a prototype,
Wigand, Rolf 1988 ‘‘Communication Network Analy- the DRG supported field research of others as well
sis.’’ In G. Goldhaber and G. Barnett, eds., Handbook as conducted its own studies (Fritz 1961). When
of Organizational Communication. Norwood, N.J.: Ablex the DRG was phased out in 1963, the Disaster
Publishing. Research Center (DRC) was established at the
Ohio State University. DRC helped the field of
Wissler, Clark 1929 An Introduction to Social Anthropolo- study to become institutionalized by its continu-
gy. New York: Henry Holt. ous existence to the present day (having moved to
the University of Delaware in 1985). In its thirty-six
Wuthnow, Robert, and Marsha Witten 1988 ‘‘New Di- years of existence DRC has trained dozens of
rections in the Study of Culture.’’ Annual Review of graduate students, built the largest specialized
Sociology 14:49–67. library in the world on social aspects of disasters,
produced over six hundred publications and about
RONALD W. PERRY three dozen Ph. D. dissertations (see http://
www.udel.edu/DRC/homepage.htm), continual-
DISASTER RESEARCH ly and consciously applied a sociological perspec-
tive to new disaster research topics, initiated an
SOCIOHISTORY OF THE FIELD
Descriptions of calamities exist as far back as the
earliest human writings, but systematic empirical
studies and theoretical treatises on social features
of disasters appeared only in the twentieth centu-
ry. The first major publications in both instances
were produced by sociologists. Samuel Prince
(1920) wrote a doctoral dissertation in sociology at
Columbia University that examined the social
681
DISASTER RESEARCH
interactive computer net of researchers in the view was followed by notions of disasters as phe-
area, and intentionally helped to create interna- nomena that resulted in significant disruptions of
tional networks and critical masses of disaster social life, which, however, might not involve a
researchers. physical agent of any kind (e.g., a false rumor
might evoke the same kind of evacuation behavior
DRC was joined in time in the United States by that an actual threat would). Later, disasters came
two other major social science research centers to be seen as crises resulting either from certain
(both currently headed by sociologists). The Natu- social constructions of reality, or from the applica-
ral Hazards Center at the University of Colorado tion of politically driven definitions, rather than
has as part of its prime mission the linking of necessarily from one initial and actual social dis-
disaster researchers and research-users in policy ruption of a social system. Other researchers equat-
and operational areas. The Hazards Reduction ed disasters with occasions where the demand for
and Recovery Center at Texas A & M University emergency actions by community organizations
has a strong multidisciplinary orientation. The exceeds their capabilities for response. By the late
organization of these groups and others studying 1980s, disasters were being seen as overt manifes-
disasters partly reflects the fact that the sociologi- tations of latent societal vulnerabilities, basically of
cal work in the area was joined in the late 1960s by weaknesses in social structures or systems (Schorr
geographers with interest in natural hazards (Cut- 1987; Kreps 1989).
ter 1994), in the 1980s by risk analysts (including
sociologists such as Perrow 1984; Short 1984) Given these variants about the concept, it is
especially concerned with technological threats, not surprising that currently no one formulation is
and later by political scientists who initially were totally accepted within the disaster research com-
interested in political crises (Rosenthal and Kouzmin munity (see Quarantelli 1998 where it is noted that
1993). More important, in the 1980s disaster re- postmodernistic ideas are now also being applied).
search spread around the world, which led to the However, there would be considerable agreement
development of a critical mass of researchers. This that the following is what is involved in using the
culminated in 1986 in the establishment within the term ‘‘disaster’’ as a sensitizing concept: Disasters
International Sociological Association of the Re- are relatively sudden occasions when, because of
search Committee on Disasters (# 39) (http:// perceived threats, the routines of collective social
sociweb.tamu.edu/ircd/), with membership in over units are seriously disrupted and when unplanned
thirty countries; its own professional journal, The courses of action have to be undertaken to cope
International Journal of Mass Emergencies and Disas- with the crisis.
ters (www.usc.edu/dept/sppd/ijmed); and a news-
letter, Unscheduled Events. At the 1998 World Con- The notion of ‘‘relatively sudden occasions’’
gress of Sociology, this committee organized indicates that disasters have unexpected life histo-
fourteen separate sessions with more than seventy- ries that can be designated in social space and
five papers from several dozen countries. The time. Disasters involve the perceptions of dangers
range of papers reflected that the initial focus on and risks to valued social objects, especially people
emergency time behavior has broadened to in- and property. The idea of disruption of routines
clude studies on mitigation and prevention, as well indicates that everyday adjustive social mechanisms
as recovery and reconstruction. cannot cope with the perceived new threats. Disas-
ters necessitate the emergence of new behaviors
CONCEPTUALIZATION OF ‘‘DISASTER’’ not available in the standard repertoire of the
endangered collectivity, a community, which is
Conceptualizations of ‘‘disaster’’ have slowly evolved usually the lowest social-level entity accepted by
from employing everyday usages of the term, researchers as able to have a disaster. In the proc-
through a focus on social aspects, to attempts to ess of the refinement of the concept, sociologists
set forth more sociological characterizations. The have almost totally abandoned the distinction be-
earliest definitions equated disasters with features tween ‘‘natural’’ and ‘‘technological’’ disasters, de-
of physical agents and made distinctions between rived from earlier notions of ‘‘acts of God’’ and
‘‘acts of God’’ and ‘‘technological’’ agents. This ‘‘man-made’’ happenings. Any disaster is seen as
inherently social in nature in origin, manifesta-
tion, or consequences. However, there is lack of
682
DISASTER RESEARCH
consensus on whether social happenings involving there usually is little organizational planning for
intentional, deliberate human actions to produce disasters. Even agencies that plan tend to think of
social disruptions such as occur in riots, civil dis- disasters as extensions of everyday emergencies
turbances, terrorist attacks, product tampering or and fail, according to researchers, to recognize the
sabotage, or wars, should be conceptualized as qualitative as well as quantitative differences be-
disasters. The majority who oppose their inclusion tween routine crises and disaster occasions. In
argue that conflict situations are inherently differ- disasters the responding organizations have to
ent in their origins, careers, manifestations, and quickly relate to more and different groups than
consequences. They note that in disaster occasions normal, adjust to losing part of their autonomy to
there are no conscious attempts to bring about overall coordinating groups, apply different per-
negative effects as there are in conflict situations formance standards and criteria, operate within a
(Quarantelli 1993). Nevertheless, there is general closer-than-usual public and private interface, and
agreement that both conflict- and consensus-type cannot function well when their own facilities and
crises are part of a more general category of operations may be directly impacted by the disas-
collective stress situations, as first suggested by ter agent.
Allan Barton (1969).
Communities. Low priority usually is given to
MAJOR RESEARCH FINDINGS preparing localities for disasters, and when there
is some effort it is usually independent of general
While the research efforts have been uneven, much community development and planning. This re-
has been learned about the behavior of individuals flects the reactive rather than proactive orienta-
and households, organizations, communities, and tion of most politicians and bureaucrats and the
societies in the pre-, trans-, and postimpact time fact that the issue of planning very seldom be-
periods (Quarantelli and Dynes 1977; Kreps 1984; comes a matter of broad community interest as
Drabek 1986). A separation of the disaster-plan- would be indicated by mass media focus, discus-
ning cycle into mitigation, preparedness, response, sions in the political arena, or the existence of
and recovery phases has won partial acceptance at advocacy groups. Efforts to initiate overall disaster
some policy and operational levels in the United preparedness often are hindered by prior organi-
States. However, international usage of the terms zational and community conflicts, cleavages, and
is far from total and there also is disagreement disputes. Starting in the 1990s, major systematic
regarding what should be considered under miti- efforts from the top down have been made in a few
gation. Therefore we will continue to discuss find- countries to push for the implementation of local
ings under the older ‘‘time’’ breakdown. mitigation measures. For programs to be imple-
mented, however, people must accept the realistic
Preimpact behavior. Individuals and households. criticism that while a disaster may be a high-im-
Most residents show little concern about disasters pact, it is a very low-probability event.
before they happen, even in risk-prone areas and
where threats are recognized. Citizens tend to see Societies. Generally, disaster planning does not
disaster planning as primarily a moral even more rank very high on the agenda of most societies.
than a legal responsibility of the government. Very However, increasingly there are exceptions in de-
few households ever plan in any concrete way for veloping countries when major recurrent disasters
possible disasters. Exceptions to these passive atti- have had major impact on the gross national prod-
tudes are where there are many recurrent experi- uct and developmental programs. Also, in devel-
ences of disasters as occur in some localities, where oped societies certain even distant catastrophes
disaster subcultures (institutionalized expectations) such as a Bhopal or Chernobyl can become sym-
have developed, and where potential disaster set- bolic occasions that lend impetus to instituting
tings (such as at chemical complexes or nuclear preparedness measures for specific disaster agents.
plants) are the focus of activist citizen groups. Increasingly too, attention to national-level disas-
ter planning has increased as citizens have come to
Organizations. Except for some disaster-orient- expect their governments to provide more securi-
ed groups such as police and fire departments, ty in general for the population. Also, mitigation
or prevention of disasters is being given higher
priority than in the past.
683
DISASTER RESEARCH
Transemergency Period Behavior. Individu- problems associated with strained organizational
als and households. When disasters occur, individu- relationships created by new disaster tasks and by
als generally react very well. They are not para- the magnitude of a disaster impact.
lyzed by a threat but actively seek relevant information
and attempt to do what they can in the emergency. Communities. Since disasters almost always cut
Victims while usually very frightened, not only act across formal governmental boundaries, problems
positively but also show little deviant behavior; of coordination among different impacted politi-
they extremely seldom break in panic flight; they cal entities are all but inevitable. The greater the
do not act irrationally especially from their per- impact of a disaster, the more there will be the
spective; and they very rarely engage in antisocial emergence of new and adaptive community struc-
activities, although stories of such contrary behav- tures and functions, especially emergent groups
ior as looting may circulate very widely. Prosocial (that is, those without any preimpact existence).
behavior especially comes to the fore, with the The greater the disaster also, the more organized
initial search and rescue being undertaken quickly improvisations of all kinds appear accompanied
and mostly by survivors in the immediate area. by pluralistic decision making. In addition, the
Most sudden needs, such as emergency housing, mass convergence of outside but nonimpacted
are met by kin and friends rather than official personnel and resources on impacted communi-
relief agencies. Family and household relation- ties, while functional in some ways, creates major
ships are very important in affecting immediate coordination problems.
reactions to disasters such as whether evacuation
will occur or if warnings will be taken seriously, Societies. Few societies ignore major disasters,
because mass media reports are filtered through but this sometimes occurs especially in the case of
primary ties. slow and diffuse occasions such as droughts and
famines, especially if they primarily affect sub-
Organizations. As a whole, organizations do groups not in the mainstream of a developing
not react as well to disasters as do individuals and country. In responding to domestic disasters, typi-
households. But while there are many organiza- cally massive help is provided to impacted areas.
tional problems in coping with the emergency Increasingly, most societies, including governmen-
time demands of a disaster, these difficulties are tal officials at all levels, obtain their view of what is
often not the expected ones. Often it is assumed happening in their disasters from mass media
that if there has been organizational disaster plan- accounts (what has been called the ‘‘CNN syn-
ning, there will be successful crisis or emergency drome’’); this also affects what is often remem-
management. But apart from the possibility of bered about the occasions. There is also a spread-
planning being poor in the first place, planning is ing belief, so far unsupported by research, that
not management and the former does not always new technologies—especially computer-related
translate well into the latter in community disas- ones—will allow major improvements in disaster
ters. There typically are problems in intra- and planning and management.
interorganizational information flow, and in com-
munication between and to organizations and the Postimpact Behavior. Individuals and house-
general public. Groups initially often have to strug- holds. Overall, there is little personal learning as a
gle with major gaps in knowledge about the im- result of undergoing a single disaster. While the
pacts of a disaster. There can be organizational experience of a major disaster is a memorable one
problems in the exercise of authority and decision from a social-psychological point of view, there are
making. These can stem from losses of higher- seldom lasting and widespread negative behavior-
echelon personnel because of overwork, conflict al consequences. Disasters very seldom produce
regarding authority over new disaster tasks, and new psychoses or severe mental illnesses. They
clashes over organizational jurisdictional differ- often, but not always, generate subclinical, short-
ences. Generally, there is much decentralization of lived and self-remitting surface reactions, such as
organizational response which in most cases is loss of appetite, sleeplessness, and anxiety. More
highly functional. Organizations operating with a common are many problems in living that stem
command and control model of response do not more from inefficient and ineffective relief and
do well at emergency times. There often too are recovery efforts of helping organizations rather
684
DISASTER RESEARCH
than from the direct physical impacts of disasters. societies and very small countries, this is not neces-
However, not all postimpact effects are negative; sarily true; a catastrophic disaster may reduce the
sometimes, the experience of undergoing a disas- gross national product five to ten percent as well as
ter results in positive self-images and closer social producing tens of thousands of casualties. Never-
ties among victims. theless, changes or improvement in national disas-
ter planning often does not occur except in certain
Organizations. Organizational changes, wheth- cases such as after the Mexico City earthquake
er for planning for disasters or for other purposes where an unusual set of circumstances existed,
in the postimpact period, is not common and including a ‘‘political will’’ to do something. But
selective at best. Most modifications are simply increasingly, in the aftermath of major disasters, to
accelerations of noncrisis-related changes already the extent that planning is instituted or improved,
planned or underway. Postimpact discussion of it is being linked to developmental planning, a
how to improve disaster planning seldom gets move strongly supported by international agen-
translated into concrete actions (unlike civil distur- cies such as the World Bank.
bances which in American society in the 1960s led
to many changes in organizations). However, over- UNIVERSALITY OF GENERALIZATIONS
all, both in the United States and elsewhere, there FROM AN UNEVEN RESEARCH BASE
have been in recent decades the growth of small,
locally based, formal social groups primarily con- Cross-societal and comparative research increased
cerned with emergency time disaster planning and markedly in the 1990s. Studies have ranged from
management. Partly because of seemingly con- cooperative work on local mass-media reporting
stantly escalating economic losses, certain busi- of community disasters in Japan and the United
nesses in such sectors as banking and insurance States (Mikami, Hiroi, Quarantelli, and Wenger
have increasingly become interested in disaster 1992) and flood responses and crisis management
preparedness and recovery. in four Western European countries (Rosenthal
and Hart 1998), to comparisons of perceptions of
Communities. There are selective longer-run recurrent floods in Bangladesh by European engi-
outcomes and changes in communities that have neers and local residents (Schmuck-Widmann
been impacted by disasters. There can be accelera- 1996), and cross-national analyses of post-disaster
tion of some ongoing and functional community political unrest in a dozen countries (Olson and
trends (e.g., in local governmental arrangements Drury 1997), as well as methodological issues in-
and power structures), and generation of some volved in cross-societal research in Italy, Mexico,
limited new patterns (e.g., in providing local men- Turkey, Peru, the United States, and Yugoslavia
tal-health services or some mitigation measures (Bates and Peacock 1993). However, this kind of
such as floodproofing regulations). On the other comparative empirical research so far has been
hand, particularly as the result of rehousing and limited. Furthermore, although the bulk of disas-
rebuilding, there can be magnifications of preimpact ters occur in developing countries, the majority of
community conflicts as well as the generation of studies from which the generalizations advanced
new ones; some of the latter is manifested in blame have been derived, have been done in developed
assignation, which, however, tends to deflect at- societies. Thus, the question of the universality of
tention away from social structural flaws to mass- disaster behaviors in different social systems has
media–influenced search for individual scapegoats. increasingly been raised. Some universals appear
It is also being recognized after disasters that to have been found: Prosocial rather than antiso-
changes in technology that create diffuse networks cial behavior clearly predominates in responses
and systems, such as among lifeline organizations, everywhere; household members and significant
are increasingly creating the need for regional others are crucial in validating warning messages,
rather than just community-based disaster planning. and the larger kin system is vital in providing
emergency assistance; emergent groups always ap-
Societies. In developed societies, there are few pear at the height of the crisis period; organiza-
long-run negative consequences of disaster losses tions have relatively more difficulty in adjusting to
whether of people or property, since such effects and coping with disasters than do individuals and
are absorbed by the larger system. In developing
685
DISASTER RESEARCH
small groups; the disaster recovery period is fraught world increasingly becomes dependent on com-
with problems at the household, organization, and puters that are bound to fail somewhere at some
community levels; mitigation measures are given key point, with drastic consequences for social
little priority even in disaster-prone localities; and systems. In addition, the newer threats are fre-
social change is seldom an outcome of most quently dangerous at places and times distant
disasters. from their initial source or origin as dramatized by
the Chernobyl nuclear radiation fallout in Europe-
Generalizations of a more limited nature also an countries and smog pollution episodes such as
seem to exist. There are social-system–structure- forest fires in Indonesia which had negative effects
specific behaviors. For example, there often is a in many Southeast Asian countries.
major delay in the response to catastrophic disas-
ters from centralized, compared to decentralized, On the other hand, there is increasing con-
governmental systems. There also may be cultural- cern and attention being paid to disaster planning
ly specific differences. For example, reflecting cul- of all kind. The future augers well for more and
tural values, individual volunteers in disasters very better planning. Citizens almost everywhere are
rarely appear in some societies such as Japan coming to expect that their governments will take
whereas they are typical in almost all American steps to protect them against disasters; this is often
disasters. actualized in planning for emergency prepared-
ness and response. Whereas two decades ago a
THE FUTURE number of societies had no preimpact disaster
planning of any kind, this is no longer the case. A
There is a dialectical process at work: There will be symbolic manifestation of this trend was the proc-
more and worse disasters at the same time that lamation by the United Nations of the 1990s as
there will be more and better planning. Why more The Decade for Natural-Disaster Reduction. This
and worse disasters? Risks and threats to human international attention accelerated efforts to pre-
beings and their societies are increasing. Tradi- vent, prepare for, respond to, and recover from
tional natural-disaster agents, such as earthquakes disasters. The effect was especially notable in the
and floods, will simply have more to impact as the increased disaster planning in developed countries.
result of normal population growth and higher,
denser concentration of inhabitants in risk-prone In developed societies, a focus on disaster
localities, such as floodplains or hurricane-vulner- planning and crisis management had started earli-
able shorelines that otherwise are attractive for er, partly as a result of sociological and related
human occupancy. There is an escalating increase research. By the 1980s, social scientists were in-
in certain kinds of technological accidents and creasingly influencing policies, political agenda
mishaps in the chemical, nuclear, and hazardous- settings, and operational matters regarding disas-
waste areas that were almost unknown before ters. This can be seen in a variety of ways. Social
World War II. There are technological advances scientists were represented on almost all national
that create risks and complexities to old threats committees set up for the U. N. Decade, and
such as when fires are prevented in high-rise build- contributed significantly to the reports prepared
ings by constructing them with highly toxic materi- to mark the midpoint of the decade. The Board on
als, or when the removal of hazardous substances Natural Disasters, established in 1992 in the U.S.
from solid sewage waste generates products that National Academy of Sciences, always has had
contain dangerous viruses and gases. New versions members from sociology and related disciplines.
of old threats are also appearing, such as the Sociologists have had major roles in national-disas-
increasing probability of urban rather than rural ter legislation in Greece and Italy. American social
droughts, or the potential large-scale collapse of science disaster researchers typically testify before
the infrastructure of older metropolitan area life- state and congressional committees considering
line systems. Finally, there is the continual devel- disaster-related laws and policies. Many sociologi-
opment of newer kinds of risks ranging from the cal disaster researchers provide both paid and
biological threats that are inherent in genetic engi- unpaid consultant services to international, na-
neering, to the crises that will be generated as the tional, and local public and private groups in-
volved in disaster-related activities. In places such
686
DISASTER RESEARCH
as the United States, Australia, New Zealand, and ‘‘grounded theory’’ approach and the employ-
some European countries, the emergency-man- ment of inductive analytical models), the theoreti-
agement and disaster-planning community has be- cal ideas used (e.g., the notion of emergence from
come more open to recognizing the practical im- collective-behavior thinking and the idea of infor-
plications of social science research. mal and formal structures of organizations), and
the general perspectives employed (e.g., that there
However, a caveat is in order for the generally can be latent as well as dysfunctional aspects of any
correct view that studies by sociologists and others behavior and that societies and communities have
have increasingly influenced policy decisions and a social history that is not easily set aside). In the
operational activities in the disaster area. Research volume entitled Sociology of Disasters: Contributions
results and questions are sometimes more than of Sociology to Disaster Research (Dynes, De Marchi,
counterbalanced by other factors: These include and Pelanda 1987), these and other contributions
the vested interests of powerful professional and to disaster theory, disaster research methods, dis-
bureaucratic elites to maintain traditional stances, aster models, and disaster concepts are set forth in
resistance to seriously questioning the viabilities considerable detail.
and competencies of specific organizations, and
an unwillingness to face up to false assumptions of The relationship has not been one-sided, since
some cultural beliefs and values. Research to the disaster research has also contributed to sociology.
contrary, for example, has had little effect on the The field of collective behavior has been most
fad-like spread of the ‘‘Incident Command Sys- influenced and this has been explicitly noted
tem’’ as a model for the emergency time opera- (Wenger 1987). Other significant contributions
tions of organizations, or an ever-spreading ac- include the study of formal organizations, social
ceptance that victims are likely to suffer posttraumatic roles, social problems, organizational and social
stress disorders, or the current common belief change, mass communications, medical sociology,
that mitigation measures are necessarily a better and the urban community (see Drabek 1986; Dynes,
strategy for disaster planning than giving priority De Marchi, and Pelanda 1987; Dynes and Tierney
to improving resilience and response to crises. 1994). A symposium on social structure and disas-
ter, coattended by disaster researchers and promi-
RELATIONSHIP TO SOCIOLOGY nent sociological theorists, examined how disaster
studies not only are informed by but could also
Although not true everywhere, sociologists have inform sociological theory; the proceedings were
been increasingly accepted as having an important published in Social Structure and Disaster (Kreps
contribution to make to disaster planning and 1989). It is also perhaps of interest that for several
management. In part this stems from the fact that decades now, many introductory sociology text-
in many countries such as Germany, Italy, Russia, books have a section on disaster behavior, usually
and the United States, they have played the lead in the collective behavior chapter.
role among social scientists in undertaking disas-
ter studies. While many reasons account for this, REFERENCES
probably the crucial factor has been that much in
general sociology can be used in doing research in Barton, Allan 1969 Communities in Disaster: A Sociological
the area. Analysis. Garden City, N.Y.: Anchor.
There has been a close relationship between Bates, F., and W. Peacock 1993 Living Conditions, Disas-
disaster studies and sociology from the earliest ters and Development: An Approach to Cross-Cultural
days of work in the area. In part this is because Comparisons. Athens, Ga.: University of Georgia Press.
sociologists, being among the leading pioneers
and researchers in the area, have tended to use Cutter, Susan,ed. 1994 Environmental Risks and Hazards.
what they could from their discipline. Thus, soci- Englewood Cliffs, N.J.: Prentice-Hall.
ology has contributed to the research techniques
used (e.g., field studies and open-ended interview- Drabek, Thomas 1986 Human System Responses to Disas-
ing), the research methodology utilized (e.g., the ters: An Inventory of Sociological Findings. New York:
Springer Verlag.
Dynes, R. R.,ed. 1988 ‘‘Disaster Classics Special Issue.’’
International Journal of Mass Emergencies and Disasters
6:209–395.
687
DISCRIMINATION
———, and K. Tierney, eds. 1994 Disasters, Collective Schorr, J. 1987 ‘‘Some Contributions of German
Behavior, and Social Organization. Newark, Del.: Uni- Katastrophensoziologie to the Sociology of Disas-
versity of Delaware Press. ter.’’ International Journal of Mass Emergencies and
Disasters 5:115–135.
Dynes, R. R., B. De Marchi, and C. Pelanda (eds.) 1987
Sociology of Disasters: Contributions of Sociology to Disas- Short, James F. 1984 ‘‘The Social Fabric at Risk: Toward
ter Research. Milan: Franco Angeli. the Social Transformation of Risk Analysis.’’ Ameri-
can Sociological Review 49:711–725.
Fritz, C. 1961 ‘‘Disasters.’’ In R. K. Merton and R. A.
Nisbet, eds., Contemporary Social Problems. New York: Sorokin, Pitirim 1942 Man and Society in Calamity. New
Harcourt. York: Dutton.
Kreps, G. 1984 ‘‘Sociological Inquiry and Disaster Re- Wenger, Dennis 1987 ‘‘Collective Behavior and Disaster
search.’’ Annual Review of Sociology 10:309–330. Research.’’ In R. R. Dynes, B. De Marchi, and C.
Pelanda, eds., Sociology of Disasters: Contribution of
——— 1989 Social Structure and Disaster. Newark, Del.: Sociology to Disaster Research, 213–238. Milan: Fran-
University of Delaware Press. co Angeli.
Mikami, S., O. Hiroi, E. L. Quarantelli, and D. Wenger E. L. QUARANTELLI
1992 ‘‘A Cross-Cultural Comparative Study of Mass
Communication in Disasters.’’ Bulletin of Faculty of DISCRIMINATION
Sociology. Toyo University 29:59–202.
Discrimination, in its sociological meaning, in-
Olson, Richard, and A. Cooper Druy 1997 ‘‘Un-Thera- volves highly complex social processes. The term
peutic Communities: A Cross-National Analysis of derives from the Latin discriminatio, which means
Post-Disaster Political Unrest.’’ International Journal to perceive distinctions among phenomena or to
of Mass Emergencies and Disasters 15:221–238. be selective in one’s judgment. Cognitive psychology
retains the first of these meanings, popular usage
Perrow, C. 1984 Normal Accidents: Living With High-Risk the second. Individual behavior that limits the
Technologies. New York: Basic Books. opportunities of a particular group is encompassed
in many sociological considerations of discrimina-
Prince, S. 1920 Catastrophe and Social Change. New York: tion. But exclusively individualistic approaches are
Columbia University Press. too narrow for robust sociological treatment. In-
stead, sociologists understand discrimination not
Quarantelli, E. L. 1987 ‘‘Disaster Studies: An Analysis of as isolated individual acts, but as a complex system
the Social Historical Factors Affecting the Develop- of social relations that produces intergroup ineq-
ment of Research in the Area.’’ International Journal uities in social outcomes.
of Mass Emergencies and Disasters 5:285–310.
This definitional expansion transforms ‘‘dis-
——— 1993 ‘‘Community Crises: An Exploratory Com- crimination’’ into a truly sociological concept. But
parison of the Characteristics and Consequences of in its breadth, the sociological definition leaves
Disasters and Riots.’’ Journal of Contingencies and room for ambiguity and controversy. Obstacles to
Crisis Management 1:67–78. consensus on what constitutes discrimination stem
from two sources—one empirical, the other ideo-
——— 1994 ‘‘Disaster Studies: The Consequences of logical and political. First, deficiencies in analysis
the Historical Use of a Sociological Approach in the and evidence limit our ability to trace thoroughly
Development of Research.’’ International Journal of the dynamic web of effects produced by discrimi-
Mass Emergencies and Disasters 12:25–49. nation. Second, because social discrimination is
contrary to professed national values and law, a
———,ed. 1998 What is a Disaster? Perspectives on the judgment that unequal outcomes reflect discrimi-
Question. London: Routledge. nation is a call for costly remedies. Variable will-
ingness to bear those social costs contributes to
———, and R. R. Dynes 1977 ‘‘Response to Social Crisis dissension about the extent of discrimination.
and Disaster.’’ Annual Review of Sociology 3:23–49.
Rosenthal, U., and P. Hart, eds. 1998 Flood Response and
Crisis Management in Western Europe: A Comparative
Analysis. New York: Springer Verlag.
Rosenthal, U., and A. Kouzmin 1993 ‘‘Globalizing an
Agenda for Contingencies and Crisis Management.’’
Journal of Contingencies and Crisis Management 1:1–12.
Schmuck-Widmann, Hanna 1996 Living with the Floods:
Survival Strategies of Char-Dwellers in Bangladesh. Ber-
lin: FDCL.
688
DISCRIMINATION
The broadest sociological definitions of dis- causal chain, the original injury is often perpetu-
crimination assume that racial minorities, women, ated and magnified by unwitting accomplices.
and other historical target groups have no inher- Intentionality criteria deny that the continuing
ent characteristics that warrant inferior social out- disadvantage is a legacy of discrimination.
comes. Thus, all inequality is seen as a legacy of
discrimination and a social injustice to be reme- Years ago, Williams outlined the concept dif-
died. By contrast, political conservatives favor a far ferently: ‘‘Discrimination may be said to exist to
narrower definition that limits the concept’s scope the degree that individuals of a given group who
by including only actions intended to restrict a are otherwise formally qualified are not treated in
group’s chances. For solid conceptual reasons, conformity with these nominally universal institu-
sociologists seldom follow suit (but see Burkey tionalized codes.’’ (Williams 1947, p. 39, italics add-
1978, p. 79). First, an intentionality criterion re- ed). For Antonovsky (1960, p. 81), discrimination
turns the concept to the realm of psychology and involves ‘‘. . . injurious treatment of persons on
deflects attention from restraining social struc- grounds rationally irrelevant to the situation’’ (ital-
ture. Second, the invisibility of intentions creates ics added). Economists use starker terms. Becker
insuperable obstacles to documenting discrimination. (1968, p. 81) held that economic discrimination
occurs ‘‘. . . against members of a group whenever
DIRECT AND INDIRECT DISCRIMINATION their earnings fall short of the amount ’warranted’
by their abilities’’ (italics added).
Sociologists’ understanding of intricate societal
patterns sensitizes them to the fact that disadvan- Two problems arise with these definitions.
tage accruing from intentional discrimination typi- First, the assessment of ‘‘abilities’’ and of what
cally cumulates, extends far beyond the original treatment is ‘‘rationally’’ relevant or ‘‘warranted’’
injury, and long outlives the deliberate perpetra- is no easy task. Critical examination of common
tion. Many sociologists distinguish between direct practice uncovers many instances where formal
and indirect discrimination (Pettigrew 1985). Di- qualifications and ‘‘nominally universal institutional-
rect discrimination occurs at points where ine- ized codes’’ prove not to provide a logical basis for
quality is generated, often intentionally. When distinctions. Employment testing litigation dem-
decisions are based explicitly on race, discrimina- onstrates that when hiring criteria once legiti-
tion is direct. Indirect discrimination is the per- mized by tradition or ‘‘logic’’ are put to scientific
petuation or magnification of the original injury. test, they often fail to predict job performance in
It occurs when the inequitable results of direct the assumed fashion. Analogous fallacies have been
discrimination are used as a basis for later decisions identified in the conventional wisdom guiding
(‘‘past-in-present discrimination’’), or decisions in admission to advanced education. Hence, nomi-
linked institutions (‘‘side-effect discrimination’’) nally universalistic standards may provide an alto-
(Feagin and Feagin 1986). Hence, discrimination gether illogical basis for decision making. If such
is indirect when an ostensibly nonracial criterion misguided selection procedures also work to the
serves as a proxy for race in determining social disadvantage of historical victims of discrimina-
outcomes. tion, these practices are not exempted from the
charge of discrimination by their universalistic facade.
To illustrate with wages, direct discrimination
exists when equally qualified blacks and whites or The second problem with these definitions is
men and women are paid at different rates for the that they ignore another, prevalent form of in-
same work. Indirect discrimination exists when direct discrimination. Even where nominally
the two groups are paid unequally because prior universalistic standards do serve some legitimate
discrimination in employment, education, or hous- social function, such as selecting competent work-
ing created apparent differences in qualifications ers, adverse impact of these standards on those
or channeled the groups into better- and worse- who bear the cumulated disadvantage of historical
paying jobs. This direct/indirect distinction re- discrimination cannot be disregarded.
sembles the legal distinction between disparate
treatment and disparate impact. While intention- The complexity of discrimination and unre-
al direct discrimination may have triggered the solved issues about its definition impede easy ap-
plication of social science methods to inform insti-
tutional policy. Apparently rigorous quantitative
689
DISCRIMINATION
analyses often only camouflage the crucial issues, of inequity that the institution is obliged to reme-
as critical examination of wage differential decom- dy, the answers are neither unequivocal nor simple.
positions reveals.
If the university’s promotion system has oper-
THE DECOMPOSITION APPROACH ated fairly, a gender gap reflecting differences in
rank may be warranted. If gender bias has existed
Assessments of discrimination produced by de- in the university’s promotion system, depressing
composing gross race or gender differences in the average academic rank of women faculty, the
wages or other social outcomes are common in resulting deficit in women’s salaries reflects indi-
sociology (e.g., Corcoran and Duncan 1978; Farley rect discrimination. Attention should then be di-
1984, Rosenfeld and Kalleberg 1990) as well as rected to the offending promotion processes. How-
in economics (e.g., Gill 1989). One segment of ever, direct salary adjustments may also be in
the gross intergroup differential is defined by its order, because gender bias in promotions weak-
empirical linkage to ‘‘qualifications’’ and other ens the link between rank and merit. Inequitable
factors deemed legitimate determinants of social depression of women’s ranks would not necessari-
rewards. The second, residual segment not de- ly lessen their actual contributions to the faculty,
monstrably linked to ‘‘legitimate’’ determinants of just their status. Using rank as a universalistic
the outcomes often is presented as the estimate of determinate of salary would then undermine the
discrimination. However, in the absence of better goal this practice is claimed to promote—the match-
information than usually available and greater ing of rewards to contributions.
agreement on what constitutes discrimination, no
unique estimate is possible. Through their choice Salary differences tied to the age differential
of control variables to index ‘‘legitimate’’ determi- of female and male faculty also raise troublesome
nants of social outcomes and their interpretation questions. If the relative youth of women faculty
of findings, researchers wittingly or unwittingly reflects lower retention and higher turnover as a
shape their answers. Any appearance of scientific result of discriminatory review processes or gener-
certitude is an illusion. For example, estimated ally inhospitable conditions, salary differentials
proportions of the gender earnings gap caused by tied to age differences are again examples of indi-
discrimination in the United States range from rect discrimination. The evidence would signal a
Sanborn’s (1969) 10 percent or less to Blinder’s need for institutional efforts to improve the reten-
(1973) 100 percent. Predictably, each has been tion of women faculty. But here it is not clear that
challenged (Bergman and Adelman 1973; Rosensweig salary adjustments are warranted. Because faculty
and Morgan 1976). contributions may be a function of experience,
application of the universalistic age criterion is
An Illustrative Decomposition Study. An analy- arguably reasonable. Any gender gap in salary tied
sis of gender differentials in faculty salaries at a to age differentials could, then, be both a legacy of
large university illustrates the difficulty of separat- discrimination and a reasonable conditioning of
ing ‘‘legitimate’’ wage differentials from inequity rewards on contributions. Complicating the issue
(Taylor 1988). About 90 percent of a $10,000 further is the fact that affirmative action efforts
gender difference in faculty salaries was empirical- often meet with greatest success in recruiting jun-
ly linked to three factors widely considered legiti- ior candidates. Thus, without supplementary data,
mate determinants of faculty pay: academic rank, it is not even clear whether an age-linked gender
age, and discipline. Women tend to hold lower gap in salary reflects continuing institutional dis-
academic rank, to be younger, and more often to crimination or affirmative hiring.
be affiliated with poorly-paid disciplines than men.
Insofar as women’s lower salaries are linked to Gender differences in salary associated with
rank, age, and discipline, is the salary differential discipline present even more complicated
untainted by gender discrimination? Convention- interpretational problems. Women and men are
al wage differentials imply an unequivocal yes. If a distributed across academic disciplines in a fash-
simple answer is given, it should be no. But in ion that mirrors the gender distribution across
truth, when policy makers ask for dollar estimates occupations. Disciplinary differences in average
salary likewise mirror wage differentials across
occupations. But are the differing occupational
690
DISCRIMINATION
distributions simply a matter of gender differ- complex nature of discrimination and unresolved
ences in preferences or abilities with no implica- issues of remedy.
tion of discrimination? Or are women steered
away from lucrative fields, so that gender differen- 1. In American society today, the injuries of
tials in salary linked to disciplinary affiliation rep- indirect discrimination are often far more
resent indirect discrimination in training, recruit- extensive than those of direct discrimina-
ment, and hiring? Or does the pattern of occupational tion. This conclusion does not imply that
wage differentials mirrored in disciplinary differ- direct discrimination no longer exists
ences represent direct discrimination, an influ- (Reskin 1998). The continued operation of
ence of gender composition per se on occupation- direct forms of discrimination is indicated
al wage structures (England et al. 1988)? In the by employment complaint records. Ameri-
latter case, assignment of responsibility for reme- can women and minorities have filed
dy presents particular problems. The university, almost 1.5 million job discrimination com-
like any other single employer, is simultaneously plaints since 1965 (Blumrosen 1996, p. 4).
vulnerable to competitive forces of the wider labor In 1994 alone, over 150,000 such com-
market and a constituent element of that market. plaints were filed; 91,000 to local and
Defiance of the market by a single organization is state agencies and 64,000 to the U.S.
costly to that organization; adherence by all or- Equal Employment Opportunity Commis-
ganizations to the broader occupational wage struc- sion (Leonard 1994, p. 24). Of course, not
ture perpetuates gender inequity and carries broad- all complaints reflect genuine discrimina-
er costs. tion; on the other hand, not all discrimina-
tion prompts formal complaints. Many
This faculty salary study did not examine the major corporations were found guilty of
role of scientific ‘‘productivity.’’ But the inclusion direct race or gender discrimination in the
of productivity measures among the control vari- 1990s. And employment audits using pair-
ables raises further difficulties. On standard ‘‘pro- ed, equally qualified applicants reveal
ductivity’’ measures, women faculty average lower widespread direct discrimination (Reskin
scores than men (Fox 1990). Thus, an institutional 1998, pp. 25–29).
study of salary differentials might find some seg-
ment of the male/female salary gap linked to 2. Apparently reasonable universalistic princi-
productivity differences. Fox’s research demon- ples may on closer examination be un-
strates, however, that gender differences in scien- necessary or even disfunctional. Scru-
tific productivity reflect contrasting levels of re- tiny of employment criteria prompted
sources that institutions provide to male and female by the Supreme Court’s 1971 Griggs v.
faculty. Like age and rank, the gender difference Duke Power Co. decision has provided
in productivity may itself be a product of institu- useful models for challenging nominally
tional discrimination. Thus, salary differentials universalistic standards. Where it is possi-
based on male/female productivity differences ble to substitute standards that do as well
also may represent indirect discrimination. The or better at screening or evaluation with-
new ingredient here is that institutional shaping of out adversely affecting historical targets of
productivity is subtle. Scientific productivity is discrimination, there are gains for all
ordinarily seen as an outgrowth of talent and involved.
effort, not potentially gender-biased institutional
resource allocation. Solid documentation of this 3. When gaps in actual qualifications are a
indirect discrimination process offers another chal- legacy of discrimination, more extensive
lenge for researchers. remedies are needed. Where training
deficits impair employability, or inade-
COMPLEXITIES OF DISCRIMINATION quate preparation impedes admission to
AND REMEDY higher education, attention should be
given to the earlier schooling processes
Critical reflection on this decomposition study that generated these deficiencies. This
highlights a set of interrelated points about the form of remedy aids future generations. In
the meantime, compensatory training can
691
DISCRIMINATION
reduce the liabilities of those who have markets could not explain the sudden improve-
already fallen victim to inferior schools. ments. It was federal anti-discrimination programs,
they concluded, that made a decisive contribution
4. Microcosms cannot escape the discrimina- to the gains.
tory impact of the societal macrocosm.
Just as salary differences across academic More general assessments also show that anti-
disciplines reflect general occupational discrimination legislation did reduce direct job
wage structures, institutions are often both discrimination nationally (Burstein 1985). It did
prey to and participant in broader social not, however, eliminate the problem. Nor did such
forces. Narrow, legalistic approaches to laws effectively attack indirect discrimination. For
remedy are inadequate for addressing this this more difficult problem, affirmative action pro-
dynamic of discrimination. grams were needed and have had some success
(Reskin 1998). The resistance to such programs,
5. Empirical research on group discrimina- however, underscores the difficulty of establishing
tion must mirror the phenomenon in effective remedies for the more subtle forms of
its variety and complexity. The regres- discrimination.
sion decomposition approach has proven
useful but has its limitations (see also DISCRIMINATION IN WESTERN EUROPE
Dempster 1988, and the ensuing commen-
tary). Regression analyses could provide Beyond racial and gender discrimination in the
more pertinent information if based on United States, the same basic concerns and princi-
more homogeneous job groups (Conway ples arise for other nations and targets. Discrimi-
and Roberts 1994) and on structural nation against Western Europe’s new immigrant
equation models that test reciprocal causa- minorities is pervasive (Castles 1984; MacEwen
tion. Most important, if the aim is to 1995; Pettigrew 1998). Both direct and indirect
guide policy, a framework far more com- discrimination are involved, though the indirect
plex than the dichotomous discrimination- forms are largely unrecognized in Europe.
or-not approach is required. The sociologi-
cal arsenal of methods offers other prom- Investigators have repeatedly uncovered di-
ising approaches. Research that traces the rect discrimination in England (Amin et al. 1988;
actual processes of institutional discrimina- Daniel 1968; Gordon and Klug 1984; Smith 1976).
tion is essential (e.g., Braddock and Controlled tests reveal the full litany of discrimina-
McPartland 1987, 1989). Also needed is tory forms involving employment, public accom-
attention to victims’ perceptions of dis- modations, housing, the courts, insurance, banks,
crimination (e.g., Feagin and Sikes 1994) even car rentals. Employment discrimination pos-
and investigation of the changes generated es the most serious problem. In every European
by anti-discrimination efforts. Another ap- Union nation, minorities have far higher unem-
proach involves cross-national comparative ployment rates than the majority group. In 1990 in
research, which we consider below. the Netherlands, Moroccans and Turks had unem-
ployment rates above 40 percent compared with
EFFECTIVE REMEDIAL INTERVENTIONS the native Dutch rate of 13 percent (Pettigrew and
Meertens 1996). During the 1974–1977 recession,
Direct racial and gender discrimination in the West German manufacturing reduced its labor
United States has declined in recent decades— force by 765,000—42 percent of whom were for-
more slowly in the 1980s and 1990s than in the eign workers (Castles 1984, p. 148).
1960s and 1970s. But what caused this decline?
Many factors were involved, but governmental As in the United States, there are many rea-
intervention was an important impetus. For exam- sons for minority unemployment disparities. The
ple, blacks in South Carolina made dramatic eco- ‘‘last-in, first-out’’ principle selectively affects the
nomic gains in manufacturing during the late younger minority workers. Typically less skilled,
1960s. Heckman and Payner (1989) demonstrated they are more affected by job upgrading. Minori-
that human capital, supply shifts, and tight labor ties also are more likely to be in older, declining
692
DISCRIMINATION
industries. But these patterns are not accidental. 1991). Without supporting evidence, this ration-
Planners put minorities into these industries for alization supports quotas and dispersal policies
cheaper labor precisely because of their decline. In that limit minority access to suitable housing.
addition, these multiple factors offer insufficient
explanations for the greater unemployment of The Netherlands, United Kingdom, and Swe-
minorities. Veenman and Roelandt (1990) found den have enacted anti-discrimination legislation
that education, age, sex, region, and employment that specifically applies to the new immigrant mi-
level explained only a small portion of the differ- norities. And the Dutch have instituted modest
ential unemployment rates in the Netherlands. affirmative action programs for women and mi-
norities (De Vries and Pettigrew 1994). Not coinci-
Indirect discrimination arises when the ina- dentally, these countries make citizenship easier to
bility to obtain citizenship restricts the opportuni- obtain than Germany. Yet this legislation has been
ties of non-European Union minorities. It limits largely ineffective for two interrelated reasons.
their ability to get suitable housing, employment, First, European legal systems do not allow class
and schooling. A visa is necessary in order to travel action suits—a forceful North American weapon
to other European Union countries. In short, the to combat discrimination. Second, European ef-
lives of Europe’s non-citizens are severely circum- forts rely heavily on individual complaints rather
scribed (Wilpert 1993). Castles (1984) contends than systemic remediation. Britain’s 1976 Act gave
that the guest-worker system that brought many of the Commission for Racial Equality the power to
the immigrants to Europe was itself a state-con- cast a broad net, but individual complaints remain
trolled system of institutional discrimination. It the chief tool (MacEwen 1995).
established the newcomers as a stigmatized ‘‘out-
group’’ suitable for low-status jobs but not citizen- It is a sociological truism that individual ef-
ship. Widespread indirect discrimination was in- forts are unlikely to alter such systemic phenome-
evitable for these victims of direct discrimination. na as discrimination. Mayhew (1968) showed how
individual suits and complaints are largely non-
Anti-discrimination remediation has been large- strategic. Minorities bring few charges against the
ly ineffective in Europe. Basic rights in Germany worst discriminators, because they avoid applying
are guaranteed only to citizens. So, the disadvan- for jobs with them in the first place. Complaints
tages of non-citizenship include limited means to about job promotion are common, but they are
combat discrimination. There is extensive Ger- made against employers who hire minorities. Thus,
man legislation to combat anti-Semitism and Nazi effective anti-discrimination laws must provide
ideology, but these laws have proved difficult to broad powers to an enforcement agency to initiate
apply to non-citizens. The German constitution strategic, institutionwide actions that uproot the
explicitly forbids discrimination on the basis of structural foundations of discrimination. Mayhew’s
origin, race, language, beliefs, or religion—but American analysis proves to be just as accurate in
not citizenship. Indeed, the Federal Constitutional Western Europe.
Court has ruled that differential treatment based
on citizenship is constitutional if there is a reason- CONCLUSIONS
able basis for it and if it is not wholly arbitrary.
Hence, a German court upheld higher taxes for A comprehensive understanding of societal dis-
foreign bar owners than German bar owners. And crimination in both North America and Western
restaurants can refuse service to Turks and oth- Europe must encompass two propositions.
ers on the grounds that their entry might lead
to intergroup disturbances (Layton-Henry and 1. The long-lasting character of discrimina-
Wilpert 1994). tion means that the effects typically outlive
the initiators of discriminatory practices.
Few means of combatting discrimination are Apart from its importance to the law, this
available in France either. Commentators often feature of modern discrimination has
view discrimination as ‘‘natural,’’ as something critical implications for sociological theory.
universally triggered when a ‘‘threshold of toler- Discrimination is fundamentally norma-
ance’’ (seuil de tolerance) is surpassed (MacMaster tive; its structural web operates in large
part independent of the dominant group’s
693
DISCRIMINATION
present ‘‘tastes,’’ attitudes, or awareness. Employment Opportunities: Research on Labor Mar-
Hence, models based primarily on indi- ket and Institutional Barriers.’’ Journal of Social Is-
vidual prejudice or ‘‘rationality,’’ wheth- sues 43:5–39.
er psychological or economic, will uni-
formly understate and oversimplify the ——— 1989 ‘‘Social Psychological Processes that Per-
phenomenon. petuate Racial Segregation: The Relationship Be-
tween School and Employment Desegregation.’’ Jour-
2. Discrimination is typically cumulative and nal of Black Studies 19:267–289.
self-perpetuating. For example, an array of
research on black Americans has demon- Burkey, R. 1978 Ethnic and Racial Groups: The Dynamics
strated that neighborhood racial segrega- of Dominance. Menlo Park, Calif.: Cummings.
tion leads to educational disadvantages,
then to occupational disadvantage, and Burstein, P. 1985 Discrimination, Jobs, and Politics. Chica-
thus to income deficits (Pettigrew 1979, go: University of Chicago Press.
1985). To be effective, structural reme-
dies must reverse this ‘‘vicious circle’’ Castles, S. 1984 Here for Good: Western Europe’s New
of discrimination. Affirmative action pro- Ethnic Minorities. London: Pluto Press.
grams are one such remedy.
Conway, D. A., H.V. and Roberts 1994 ‘‘Analysis of
Seen in sociological perspective, then, dis- Employment Discrimination through Homogeneous
crimination is considerably more intricate and Job Groups.’’ Journal of Econometrics 61:103–131.
entrenched than commonly thought. The com-
plexity of discrimination presents major challenges Corcoran M., and G. J. Duncan 1978 ‘‘Work History,
to social-scientific attempts to trace its impact. Labor Force Attachment, and Earning Differences
This complexity also precludes any one-to-one Between Races and Sexes.’’ Journal of Human Resourc-
correspondence between perpetration and respon- es 14:3–20.
sibility for remedy. Broad social programs will be
necessary if the full legacy of direct and indirect Daniel, W. W. 1968 Racial Discrimination In England.
discrimination is finally to be erased. Harmondsworth, UK: Penguin.
REFERENCES Dempster, A. P. 1988 ‘‘Employment Discrimination and
Statistical Science.’’ Statistical Science 3:149–195.
Amin, K., M. Fernandes, and P. Gordon 1988 Racism
and Discrimination in Britain: A Select Bibliography, De Vries, S., and T. F. Pettigrew 1994 ‘‘A Comparative
1984–87. London: Runnymede Trust. Perspective on Affirmative Action: Positieve aktie in
the Netherlands.’’ Basic and Applied Social Psychology
Antonovsky, A. 1960 ‘‘The Social Meaning of Discrimi- 15:179–199.
nation.’’ Phylon 11:81–95.
England, P., G. Farkas, B. Kilbourne, and T. Dou 1988
Becker, G. 1968 ‘‘Economic Discrimination.’’ In D. L. ‘‘Explaining Occupational Sex Segregation and Wages:
Sills, ed., International Encyclopedia of the Social Sci- Findings from a Model with Fixed Effects.’’ American
ences, vol. IV. New York: Macmillan. Sociological Review 53:544–558.
Bergman, B. R., and I. Adelman 1973 ‘‘The 1973 Report Farley, R. 1984 Blacks and Whites: Narrowing the Gap?
of the President’s Council of Economic Advisors: Cambridge, Mass.: Harvard University Press.
The Economic Role of Women.’’ American Economic
Review 63:509–514. Feagin, J. R., and C. B. Feagin 1986 Discrimination
American Style: Institutional Racism and Sexism, 2d ed.
Blinder, A. S. 1973 ‘‘Wage Discrimination: Reduced Malabar, Fla.: Krieger.
Form and Structural Estimates.’’Journal of Human
Resources 8:436–455. Feagin, J. R., and M. P. Sikes 1994 Living with Racism:
The Black Middle-Class Experience. Boston, Mass.: Bea-
Blumrosen, A. W. 1996 Declaration. Statement submit- con Press.
ted to the Supreme Court of California in response
to Proposition 209 (Sept. 26, 1996). Fox, M. F. 1991 ‘‘Gender, Environmental Milieux, and
Productivity in Science.’’ In J.Cole, H. Zuckerman,
Braddock, J. H. II, and J. M. McPartland 1987 ‘‘How and J. Bruer, eds., The Outer Circle: Women in the
Minorities Continue to be Excluded from Equal Scientific Community. New York: Norton.
Gill, A. M. 1989 ‘‘The Role of Discrimination in Deter-
mining Occupational Structure.’’ Industrial and La-
bor Relations Review 42:610–623.
Gordon, P., and F. Klug 1984 Racism and Discrimination
in Britain: A Select Bibliography, 1970–83. London:
Runnymede Trust.
694
DIVISION OF LABOR
Heckman, J. J., and B. S. Payner 1989 ‘‘Determining the Conference of the Association for Public Policy Analy-
Impact of Federal Antidiscrimination Policy on the sis and Management. Seattle, Wash.
Economic Status of Blacks: A Study of South Caroli- Veenman, J., and T. Roelandt 1990 ‘‘Allochtonen:
na.’’ American Economic Review 79:138–177. Achterstand en Achterstelling.’’ In J. J. Schippers,
ed., Arbeidsmarkt en Maatschappelijke Ongelijkheid.
Layton-Henry, Z., and C. Wilpert 1994 Discrimination, Groningen, The Netherlands: Wolters-Noordhoff.
Racism and Citizenship: Inclusion and Exclusion in Brit- Williams, R. M., Jr. 1947 The Reduction of Intergroup
ain and Germany. London: Anglo-German Founda- Tensions. New York: Social Science Research Council.
tion for the Study of Industrial Society. Wilpert, C. 1993 ‘‘Ideological and Institutional Founda-
tions of Racism in the Federal Republic of Germa-
Leonard, J. S. 1994 ‘‘Use of Enforcement Techniques in ny.’’ In J. Solomos and J. Wrench, eds., Racism and
Eliminating Glass Ceiling Barriers.’’ Report to the Migration in Western Europe. Oxford, UK: Berg.
Glass Ceiling Commission. U. S. Dept. of Labor,
Washington, D.C. THOMAS F. PETTIGREW
MARYLEE C. TAYLOR
MacEwen M. 1995 Tackling Racism in Europe: An Exami-
nation of Anti-Discrimination Law in Practice. Washing- DISENGAGEMENT THEORY
ton, D.C.: Berg.
See Aging and the Life Course; Retirement.
MacMaster N. 1991 ‘‘The ‘Seuil de Tolerance’: The Uses
of a ‘Scientific’ Racist Concept.’’ In M. Silverman, DISTRIBUTION-FREE
ed., Race, Discourse and Power in France. Aldershot, STATISTICS
UK: Avebury.
See Nonparametric Statistics.
Mayhew, L. H. 1968 Law and Equal Opportunity: A Study
of the Massachusetts Commission Against Discrimination. DISTRIBUTIVE JUSTICE
Cambridge, Mass.: Harvard University Press.
See Human Rights/Children’s Rights; Social
Pettigrew, T. F. 1979 ‘‘Racial Change and Social Policy.’’ Justice.
Annals of the American Association of Political and Social
Science 441:114–131. DIVISION OF LABOR
——— 1985 ‘‘New Black-White Patterns: How Best to The division of labor is a multifaceted concept that
Conceptualize Them?’’ Annual Review of Sociology applies to several levels of analysis: small groups,
11:329–346. families, households, formal organizations, socie-
ties, and even the entire ‘‘world system’’ (Wallerstein
——— 1998 ‘‘Responses to the New Minorities of West- 1979). Each level of analysis requires a slightly
ern Europe.’’ Annual Review of Sociology 24:77–103. different focus; the division of labor may refer to
the emergence of certain roles in groups, to the
———, and R. W. Meertens 1996 ‘‘The Verzuiling relative preponderance of industrial sectors (pri-
Puzzle: Understanding Dutch Intergroup Relations.’’ mary, secondary, tertiary) in an economy, to the
Current Psychology 15:3–13. distribution of societal roles by sex and age, or to
variability among occupational groupings. Soci-
Reskin, B. F. 1998 The Realities of Affirmative Action in ologists at the micro-level are concerned with ‘‘who
Employment. Washington, D.C.: American Sociologi- does what,’’ while macrosociologists focus on the
cal Association. larger structural issues of societal functions.
Rosenfeld, R. A., and A. L. Kalleberg 1990 ‘‘A Cross-
national Comparison of the Gender Gap in Income.’’
American Journal of Sociology 96:69–105.
Rosensweig, M. R., and J. Morgan 1976 ‘‘Wage Discrimi-
nation: A Comment.’’ Journal of Human Resourc-
es 11:3–7.
Sanborn, H. 1969 ‘‘Pay Differences Between Men and
Women.’’ Industrial and Labor Relations Review
17:534–550.
Smith, D. J. 1976 The Facts of Racial Disadvantage: A
National Survey. London: PEP.
Taylor, M. C. 1988 ‘‘Estimating Race and Sex Inequity in
Wages: Substantive Implications of Methodological
Choices.’’ Paper presented at the 1989 Research
695
DIVISION OF LABOR
All of the major sociological theorists consid- outside the home. This reflects the ‘‘provider’’ vs.
ered the division of labor to be a fundamental ‘‘homemaker’’ role distinction that formerly char-
concept in understanding the development of mod- acterized most nuclear families in industrialized
ern society. The division of labor in society has societies. Another major family role, occupied
been a focus of theoretical debate for more than a almost exclusively by the wife, is that of ‘‘kinkeeper.’’
century, with some writers most concerned about The maintenance of family traditions, the record-
hierarchical divisions, or the vertical dimension, ing of important family anniversaries, and the
and others emphasizing the heterogeneity or the coordination of visits between households are all
horizontal dimension, of a social system. The analy- vital elements of this role. The classic works by
sis here is primarily theoretical, and so does not Bott (1957) and Blood and Wolfe (1960) docu-
deal with debates about the technical problems of mented the spousal division of labor as it was
measurement (but see Baker 1981; Clemente 1972; influenced by wider social networks and the rela-
Gibbs and Poston 1975; Land 1970; Rushing and tive power of the husband and wife. Kamo (1988)
Davies 1970). has noted that cultural ideology is a strong factor
in the determination of spousal roles, although it
THE DIVISION OF LABOR IN SMALL is not entirely independent of the resources avail-
GROUPS able to husband and wife.
Most small groups exhibit ‘‘role differentiation,’’ Within a household, both age and sex struc-
as first described by Simmel (1890), Whyte (1943), ture the division of labor. Young children have few
and Bales and Slater (1955). The research on responsibilities, while parents have many, and males
informal work groups stresses a hierarchical form and females tend to do very different tasks. For
of the division of labor, or the emergence of example, White and Brinkerhoff (1981) studied
leadership; in particular, the emergence of both a the reported allocation of a variety of household
‘‘task leader’’ and a ‘‘social leader.’’ It appears that tasks among a sample of Nebraska households,
defining group goals and enforcing norms is a type and found that the youngest cohort of children
of activity incompatible with maintaining group (aged two to five) performed relatively few house-
cohesion, and therefore most groups have a shared hold tasks and showed little differentiation by sex.
leadership structure. Even in same-sex groups, this Older children, however, diverged considerably in
structure of coleadership is apparent; also, the their roles (males doing more outdoor work, fe-
degree of role differentiation appears be greater males more responsible for cooking or childcare).
in larger groups. The functional need for a social At the other end of the lifespan, the roles of the
leader depends in part on the degree to which grandparent generation were strongly influenced
group members are task oriented, and on the by geographical proximity, the ages of the grand-
degree to which the task leader is perceived as a parents, and the ages of the children. Divorce
legitimate authority (with the power to reward and among the parental generation can also substan-
punish other group members). Burke (1969) has tially impact the roles adopted by grandparents
also noted that a ‘‘scapegoat’’ role may emerge (see Bengston and Robertson 1985).
within task groups, reducing the need for a social
leader to maintain harmony among the rest of the FORMAL ORGANIZATIONS AND
group members. OCCUPATIONAL SPECIALIZATION
HOUSEHOLD AND FAMILY DIVISION Formal organizations are always structured by an
OF LABOR explicit division of labor (an organization chart
contains the names of functions, positions, or
When the family or household is the unit of analy- subunits). The horizontal differentiation in an
sis, issues concerning the division of tasks between organization, or task specialization, is normally
spouses are of primary interest to sociologists. based on functional units that are of roughly equiva-
One major finding has been the persistence of lent levels in the hierarchy of authority. The divi-
disproportionately high levels of traditional house- sion of labor may also be constructed on a geo-
work by the wife, even when she is employed graphical basis, with the extreme examples being
696
DIVISION OF LABOR
transnational corporations that draw raw resourc- THEORETICAL APPROACHES
es and labor from the developing nations, and
management from the developed nations. Adam Smith, in The Wealth of Nations (1776),
produced the classic statement of the economic
Labor is also organized hierarchically, into efficiencies of a complex division of labor. He
levels of authority. Max Weber was one of the first observed that the manufacture of steel pins could
sociologists to analyze the emergence of modern be more than two hundred times more productive
bureaucracies, where the division of labor is fun- if each separate operation (and there were more
damental. Weber’s approach suggests a maximum than a dozen) were performed by a separate work-
possible level of specialization, so that each posi- er. The emergence of the systematic and intention-
tion can be filled by individuals who are experts in al division of labor probably goes back to prehis-
a narrow area of activity. An extreme form of the toric societies, and is certainly in evidence in the
bureaucratic division of labor was advocated by ancient civilizations of China, Egypt, Greece, and
Frederick Taylor’s theory of ‘‘Scientific Manage- Rome. The accumulation of an agricultural sur-
ment’’ (1911). By studying in minute detail the plus and the establishment of markets both creat-
physical motions required to most efficiently oper- ed and stimulated the differentiation of producers
ate any given piece of machinery, Taylor pio- and consumers. Another phase in the transforma-
neered ‘‘time and motion studies.’’ Along with tion in Europe was the decline of the ‘‘craft guilds’’
Henry Ford, he also made the assembly line a that dominated from the twelfth to the sixteenth
standard industrial mode of production in mod- centuries. When labor became free from the
ern society. However, this form of the division of constraints of the journeyman/apprentice system
labor can lead to isolation and alienation among and when factories began to attract large numbers
workers, and so more recent organizational strate- of laborers, the mechanization of modern work
gies, in some industrial sectors, emphasize the flourished. (For economic and political perspec-
‘‘craft’’ approach, in which a team of workers tives on the division of labor see Krause 1982;
participate more or less equally in many aspects Putterman 1990).
of the production process (see Blauner 1964;
Hedley 1992). Early sociologists such as Herbert Spencer
considered the growth of societies as the primary
SOCIETAL DIVISION OF LABOR determinant of the increased specialization and
routinization of work; they also emphasized the
Macrosociologists measure the societal division of positive impacts of this process. Spencer, like oth-
labor in a number of ways, most commonly by er functionalists, viewed human society as an or-
considering the number of different occupational ganic system that became increasingly differentiat-
categories that appear in census statistics or other ed as it grew in size, much as a fertilized egg
official documents (see Moore 1968). These lists develops complex structures as it develops into a
are often implicitly or explicitly ranked, and so full-fledged embryo. In his Principles of Sociology
both horizontal and vertical division of labor can (1884), Spencer considered the evolution of hu-
be analyzed. This occupational heterogeneity is man society as a process of increasing differentia-
dependent in part on the official definitions, but tion of structure and function.
researchers on occupational prestige (the vertical
dimension) have shown comparable levels in in- Karl Marx(1867) argued that the increasing
dustrialized societies such as the United States, division of labor in capitalist societies is a primary
Canada, England, Japan, Sweden, Germany, and cause of alienation and class conflict, and there-
France (see Treiman 1977). The ‘‘world system’’ as fore is a force in the eventual transformation to a
described by Wallerstein (1979) is the upper limit socialist/communist society. In fact, a specific ques-
of the analysis of the division of labor. Entire tion asking for details of the division of labor
societies are characterized as ‘‘core’’ or ‘‘periph- appeared on one of the earliest questionnaire
ery’’ in Wallerstein’s analysis of the global implica- surveys in sociology done by Marx in 1880. Marx
tions of postindustrialism. and his followers called for a new form of the
division of labor, supported by an equalitarian
ethos, in which individuals would be free to choose
697
DIVISION OF LABOR
their productive roles; labor would not be alienat- ‘‘organic solidarity.’’ The former is found in small-
ing because of the common ideology and sense of er, less-advanced societies where families and vil-
community. lages are mostly self-sufficient, independent, and
united by similarities. The latter is found in larger,
Weber (1947) painted a darker picture when urbanized societies where specialization creates
he documented the increasing ‘‘rationalization’’ interdependence among social units.
of society, especially the ascendence of the bureau-
cratic division of labor with its coordinated system Following Spencer’s lead, Durkheim noted
of roles, each highly specialized, with duties speci- that the specialization of functions always accom-
fied in writing and incumbents hired on the basis panies the growth of a society; he also observed
of their documented competence at specific tasks. that increasing population density—the urbaniza-
The ideal-type bureaucracy was in actuality subject tion of society that accompanies modernization—
to the negative consequences of excessive speciali- greatly increases the opportunities for further in-
zation, however. Weber pointed out that the ‘‘iron creases in the division of labor.
cage’’ placed stifling limits on human freedom
within the organization, and that decisions by It should be noted that the shift to a modern
bureaucrats often became so rule bound and in- division of labor could not have occurred without
flexible that the clients were ill served. a preexisting solidarity; in his chapter on ‘‘organic
and contractual solidarity’’ he departed from
Georg Simmel’s ‘‘differentiation and the prin- Spencer’s utilitarian explanation of social cohe-
ciple of saving energy’’ (1976 [1890]) is a little- sion, and noted that the advanced division of labor
known essay that similarly describes the inevitable can occur only among members of an existing
problems that offset the efficiencies gained by the society, where individuals and groups are united
division of labor; he called these ‘‘friction, indi- by preexisting similarities (of language, relig-
rectness, and superfluous coordination.’’ He also ion, etc.).
echoed Marx and Engels when he described the
effects of high levels of differentiation upon the A sense of trust, obligation, and interdepend-
individual: ence is essential for any large group in which there
are many diverse roles; indirect exchanges occur;
. . . differentiation of the social group is and individuals form smaller subgroupings based
evidently directly opposed to that of the on occupational specialization. All of these changes
individual. The former requires that the create high levels of interdependence, but with
individual must be as specialized as possible, increasing specialization, and different world views
that some single task must absorb all his develop, along with different interests, values, and
energies and that all his impulses, abilities and belief systems. This is the problem Durkheim saw
interests must be made compatible with this one in the shift from mechanical to organic solidarity;
task, because this specialization of the indi- he feared the ‘‘anomie’’ or lack of cohesion that
vidual makes it both possible and necessary to might result from a multiplicity of views, languag-
the highest degree for him to be different from es, and religions within a society (as in the France
all other specialized individuals. Thus the of his times, and even more so today). The prob-
economic setup of society forces the individual lems of inequality in modern industrial society
for life into the most monotonous work, the were not lost on Durkheim, either; he noted how
most extreme specialization, because in this the ‘‘pathological form of the division of labor’’
way he will acquire the skill which makes posed a threat to the full development of social
possible the desired quality and cheapness of solidarity (see Giddens 1971). Although many sim-
the product. (Simmel 1976, p. 130) plistic analyses of Durkheim’s approach suggest
otherwise, he dealt at length with the problems of
While the foregoing theorists contributed sub- ‘‘the class war’’ and the need for justice and
stantially to the understanding of the division of fraternity.
labor, Emile Durkheim’s The Division of Labor in
Society (1893) stands as the classic sociological The division of labor is treated as a key ele-
statement of the causes and consequences of the ment in Peter Blau’s book, Inequality and Heteroge-
historical shift from ‘‘mechanical solidarity’’ to neity (1977). This important work emphasizes the
primacy of differentiation (division of labor) as an
698
DIVISION OF LABOR
influence on mobility, prejudice, conflict, affilia- • Rising levels of education and qualifi-
tion, intermarriage, and inequality. In Blau’s cations promote an advanced division
scheme, a society may be undifferentiated (all of labor.
persons or positions are independent, self-suffi-
cient, etc.) or strongly differentiated (a high de- • Large work organizations promote the
gree of specialization and interdependence). This division of labor in society.
differentiation may be a matter of degree, on
dimensions such as authority, power, prestige, • Linguistic heterogeneity impedes the divi-
etc.; this constitutes inequality. Or the differentia- sion of labor.
tion may be a matter of kind, such as occupational
differentiation—the division of labor. • The more the division of labor is in
the form of specialization rather than
Blau, like Durkheim, distinguished two major routinization, the higher are rates of
types of division of labor: routinization and expert associations among different occupations,
specialization. which produces higher integration.
The two major forms of division of labor are • The more the division of labor intersects
the subdivision of work into repetitive routines with other nominal parameters (includ-
and its subdivision into expert specialities. . . ing kinship, language, religion, ethnicity,
when jobs are divided into repetitive routines, etc.) the greater is the probability that
the training and skills needed to perform them intergroup relations strengthen society’s
are reduced, whereas when they are divided integration.
into fields of specialists, the narrower range of
tasks permits greater expertness to be acquired • The smaller an organization, the more
and applied to the work, increasing the its internal division of labor increas-
training and skills required to perform it. es the probabilities of intergroup and
(Blau 1977, p. 188) interstratum associations, and therefore
the higher the degree of integration.
Blau has shown that the division of labor (Summarized from Blau 1977, pp.
always increases inequality in the organization. 214–215)
The managerial and technical experts coordinate
the increasing number and diversity of routinized Blau concludes by noting that ‘‘Advances in
positions. As organizations and societies increase the division of labor tend to be accompanied by
in size and population density, the division of decreases in various forms of inequality but by
labor increases. At the same time, the forces of increases in inequality in power. Although the
industrialization and urbanization require and en- advancing division of labor does not generate the
courage further specialization, and in most cases, growing concentration of power, the two are likely
increase inequality, and indirectly, social integra- to occur together, because the expansion of work
tion. Some of the confounding boundaries in- organizations promotes both.’’ (p. 214).
clude the degree of linguistic, ethnic, or cultural
heterogeneity (all of which can inhibit integra- CONCLUSION
tion), and social or geographic mobility (which can
increase integration). Ford’s moving assembly lines began to produce
the frames for Model T automobiles in 1913, at a
In a rare display of explicitly stated definitions rate of about one every two working days, but
and propositions, Blau created a landmark theory within months, refinements on the assembly proc-
of social organization. Here are the most impor- ess reduced this to four units per day. This eightfold
tant of Blau’s assumptions and theorems relating increase in efficiency was accompanied by a de-
to the division of labor: crease in the price of the cars and indirectly stimu-
lated a very large industry. Now, automated facto-
• The division of labor depends on opportu- ries, using robotics and highly specialized computer
nities for communication. systems, have dramatically increased the efficiency
of the automobile industry. The effects on morale
• Population density and urbanization in- and the environment, however, appear to be less
crease the division of labor. salutary.
699
DIVORCE
(SEE ALSO: Bureaucracy, Complex Organizations, Conver- Marx, K. (1867) 1977 Capital: A Critique of Political
gence Theories, Family and Household Structure, Family Economy. New York: Vintage Books.
Roles, Industrial Sociology, Industrialization, Parental Roles,
Social Change, Social Structure, Technology and Society, and Moore, W.E. 1968 Economy and Society. New York: Ran-
Work and Occupations.) dom House.
REFERENCES Putterman, L. 1990 Division of Labor and Welfare: An
Introduction to Economic Systems. New York: Oxford
Baker, P.M. 1981 ‘‘The Division of Labor: Interdepend- University Press.
ence, Isolation, and Cohesion in Small Groups.’’
Small Group Behavior 12(1):93–106. Rushing, W.A., and V. Davies 1970 ‘‘Note on the Mathe-
matical Formalization of a Measure.’’ Social Forces
Bales, R.F,. and P.E. Slater 1955 ‘‘Role Differentiation in 48(March):394–396.
Small Decision-Making Groups.’’ In T. Parsons et al.,
eds., Family, Socialization, and Interaction Process. Simmel, G. (1890) 1976 ‘‘On Social Differentiation.’’ In
Glencoe, Ill.: Free Press. P. Lawrence, ed., Georg Simmel, Sociologist and Europe-
an. New York: Nelson.
Bengston, V.L., and J.F. Robertson 1985 Grandparenthood.
Beverly Hills, Calif.: Sage. Smith, Adam (1776) 1985 An Inquiry into the Nature and
Causes of the Wealth of Nations. New York: Modern
Blau, P.M. 1977 Inequality and Heterogeneity. New York: Library.
Free Press.
Smith, D., and R. Snow 1976 ‘‘The Division of Labor:
Blauner, R. 1964 Alienation and Freedom. Chicago: Uni- Conceptual and Methodological Issues.’’ Social Forces
versity of Chicago Press. 55:520–528.
Blood, R., and D. Wolfe 1960 Husbands and Wives: The Spencer, H. 1884 Principles of Sociology. New York: Ap-
Dynamics of Married Living. New York: Free Press. pleton-Century-Crofts.
Bott, E. 1957 Family and Social Network. London: Tavistock Taylor, F. 1911 Scientific Management. New York: Harper.
Publications.
Treiman, D. 1977 Occupational Prestige in Comparative
Burke, P.J. 1969 ‘‘Scapegoating: An Alternative to Role Perspective. New York: Academic Press.
Differentiation.’’ Sociometry 32(June):159–168.
Wallerstein, I. 1979 The Capitalist World-Economy. New
Clemente, R. 1972 ‘‘The Measurement Problem in the York: Cambridge University Press.
Analysis of an Ecological Concept: The Division of
Labor.’’ Pacific Sociological Review 15:30–40. Weber, M. (1908) 1947 The Theory of Social and Economic
Organization. New York: Free Press.
Durkheim, E. (1893) 1933 The Division of Labor in Society.
New York: Free Press. Whyte, W.F. 1943 Street Corner Society. Chicago: Chicago
University Press.
Gibbs, J.P., and D.L. Poston 1975 ‘‘The Division of
Labor: Conceptualization and Related Issues.’’ Social PAUL MORGAN BAKER
Forces 53 (March) 468–476.
DIVORCE
Giddens, A. 1971 Capitalism and Modern Social Theory:
Analysis of the Writings of Marx, Durkheim, and Max Divorce is of sociological significance for several
Weber. Cambridge: Cambridge University Press. reasons. To begin, divorce rates are often seen as
indicators of the health of the institution of mar-
Hedley, R.A. 1992 Making a Living: Technology and riage. When divorce rates rise or fall, many soci-
Change. New York: Harper Collins. ologists view these changes as indicating some-
thing about the overall quality of marriages or,
Kamo, Y. 1988 ‘‘Determinants of Household Labor: alternatively, the stability of social arrangements
Resources, Power, and Ideology.’’ Journal of Family more generally. Viewed from another perspective,
Issues 9:177–200. divorce interests sociologists as one of several
important transitions in the life course of individu-
Krause, E.A. 1982 The Division of Labor: A Political als. The adults and children who experience di-
Perspective. Westport, Conn.: Greenwood Press. vorce have been studied to understand both the
causes and consequences. From this perspective, a
Land, K.C. 1970 ‘‘Mathematical Formalization of divorce is as much an event in the biography of
Durkheim’s Theory of Division of Labor.’’ E.F.
Borgatta, ed., Sociological Methodology 1970, 257–282.
San Francisco: Jossey-Bass.
700
DIVORCE
family members, as other life-course transitions and were effected by bills to dissolve a particular
(remarriage, childbirth, and retirement). The so- marriage. Still, though legal, divorce was very rare.
ciological interest in divorce also focuses on the During the seventeenth century, there were fifty-
social trends it is part of, figuring prominently in four petitions for divorce in Massachusetts, of
any sociological analysis of industrialization, pov- which forty-four were successful (Phillips 1988, p.
erty rates, educational attainment, strategies of 138). The middle colonies provided annulments
conflict resolution, or law. or divorces for serious matrimonial offenses such
as prolonged absence or bigamy. The southern
For sociologists, divorce may characterize an colonies afforded no provisions for divorce
individual, a family, a region, a subgroup, a histori- whatsoever.
cal period, or an entire society. It may be studied as
either the cause or consequence of other phe- Post-Revolutionary War. Immediately after
nomena. Still, the overriding concern of almost all the Revolutionary War, without British legal im-
research on this topic has been the increase in pediments to divorce, the states began discussion
divorce over time. Divorce is now almost as com- of laws to govern divorce. In New England and the
mon as its absence in the lives of recently married middle states, divorce became the province of
couples. The National Center for Health Statistics state courts while in the more restrictive southern
estimates that 43 percent of marriages begun in states it was more often a legislative matter. By the
the early 1990s will end in divorce (NCHS 1998), a turn of the nineteenth century, almost all states
significant decline from the estimates of 50 per- had enacted some form of divorce law. And by the
cent to 65 percent in the late 1980s (Martin and middle of the century, even southern states were
Bumpass 1989). The decline in divorce rates in the operating within a judicial divorce system.
recent past is probably a result of the aging of the
post-World War II Baby Boom generation who are The shift to judicial divorce is significant. By
no longer at high risk of divorce because of their removing divorce deliberations from legislatures,
age. It is also possible that American marriages are states were forced to establish grounds that justi-
becoming somewhat more stable than they were a fied a divorce. Such clauses reflected the prevail-
decade ago. Still, the fluctuations in divorce rates ing sentiments governing normative marriage—
one decade to the next do not mask the more they indicated what was expected of marriage at
general trend for the past two centuries. Under- the time. And by investing judges with the authori-
standing the increase in divorce has been the ty to interpret and adjudicate, such changes signifi-
larger sociological endeavor regardless of the par- cantly liberalized the availability of divorce. North-
ticular perspective employed. A historical account ern and southern states permitted divorces for
of trends is necessary before considering contem- specific offenses such as adultery, desertion, biga-
porary issues associated with divorce. my, and increasingly with time, cruelty. In the
newer frontier western states, grounds resembled
A BRIEF HISTORICAL RECORD OF those of the East plus ‘‘any other cause for which
DIVORCE IN AMERICA the court shall deem it proper that the divorce
shall be granted’’ (Phillips 1988, p. 453).
The Colonial Period. Divorce was not legal in any
but the New England settlements. The Church of Throughout the nineteenth century, there was a
England allowed for legal separations (a mensa et gradual liberalization of divorce laws in the United
thoro), but not for divorce. The New England States and a corresponding increase in divorce as
Puritans who first landed at Plymouth in 1621, well. Where divorces totaled a few hundred at the
however, were disenchanted with this, as well as beginning of the nineteenth century, the numbers
many other Anglican doctrines. Divorce was per- grew exponentially as the century wore on; 7,380
mitted on the grounds of adultery or seven-year divorces in 1860, 10,962 in 1870, 19,663 in 1880,
desertion as early as 1639 in Plymouth. Other New 33,461 in 1890, and 55,751 in 1900 (U.S. Bureau of
England colonies followed similar guidelines. Di- the Census 1975). These figures assume greater
vorce governed by rudimentary codified law was significance when growth in population is removed
granted by legislative decree. Individual petitions from them. Whereas the divorce rate (number of
for divorce were debated in colonial legislatures divorces per 1,000 marriages) was but 1.2 in 1869,
it had climbed to 4.0 by 1900. In short, the increase
701
DIVORCE
in divorce outstripped the increase in population established in the latter nineteenth century. Two
several times. world wars and the Great Depression interrupted
gradually increasing divorce rates, however. Dur-
A number of factors have been identified as ing each war and during the Depression, divorce
causes of such dramatic increases. In part, these rates dropped. After each, rates soared before
can be described as social changes, which made falling to levels somewhat higher than that which
marriage less essential. The growth of wage labor preceded these events. Sociological explanations
in the nineteenth century afforded women an for these trends focus on women’s employment
alternative to economic dependence on a hus- opportunities. Women’s labor force participation
band. In an economy dominated by individuals permits the termination of intolerable unions. The
rather than families, marriage was simply less es- separations, hastily timed marriages, and sexual
sential. Life as a single individual gradually lost its misalliances characteristic of wartime were also
legal or social stigma (New England settlements undoubtedly factors in the post-war divorces rates.
had forbidden solitary dwelling while southern Further, the increases in divorce following these
communities had taxed it heavily). difficult times may be seen, in part, as a delayed
reaction. Once the Depression or war was over,
More important, however, were fundamental the reservoir of impending divorces broke. And
shifts in the meaning of marriage. Divorce codes finally, postwar optimism and affluence may have
reflected the growing belief that marriages should contributed to an unwillingness to sustain an un-
be imbued with heavy doses of affection and equali- happy marriage.
ty. Divorce grounds of cruelty or lack of support
indicate that marriage was increasingly viewed as a The second half of the century witnessed even
partnership. Where a century earlier men had more dramatic increases in divorce. With the ex-
been granted greater discretion in their personal ception of the peculiar 1950s (for an explanation
lives, latter nineteenth-century morality attacked of this anomaly, see Cherlin 1992), the trend for
such double standards. Men were not necessarily the second half of the 1900s was a regular and
less culpable than women for their vices. Victorian exponential growth in divorce until around 1980,
morality stressed the highest standards of sexual at which point the increase stopped.
behavior for both husbands and wives. Changing
divorce codes coincided with the passage of laws Though specific explanations for the increase
restricting husbands’ unilateral control over their in divorces during the twentieth century vary,
wifes’ property. The passage of married women’s several themes may be noted. First, marriage has
property acts throughout the nation in the latter lost much of its central economic and social signifi-
nineteenth century acknowledged married wom- cance—especially for women. For example, di-
en’s claims to property brought to or acquired in vorce was undoubtedly inhibited by the fact that
marriage. By 1887, thirty-three states and the Dis- prior to the twentieth century, custody of children
trict of Columbia gave married women control was uniformly awarded to fathers (since they were
over their property and earnings (Degler 1980, p. 332). legally responsible for financial support). With the
acceptance of Freudian ideas of psychosexual de-
Divorce codes including omnibus grounds such velopment and similar ideas about intellectual and
as ‘‘cruelty’’ (which could justify a divorce from a cognitive growth, the so-called Tender Years Doc-
drunkard husband, for example) may be viewed as trine became accepted practice in courts during
reflecting a Victorian American belief that women the early 1900s which then awarded custody to
were morally sensitive and fragile, and in need of mothers as regularly as they had once done to
protection (Phillips 1988, p. 500). More particular- fathers. And as it became more commonplace,
ly, the growing use of offenses against the intimate remarriage began to lose some of its stigma. All
and emotional aspects of marriage reflected a these changes made it possible for women to
growing belief that such things constituted matri- divorce their husbands if they wished. But why did
monial essentials. If a failure of intimacy could so many wish to obtain divorces?
justify the dissolution of a marriage, then intimacy
may be viewed as a core expectation of marriage. The simplest explanation is that more divorce
is a consequence of higher expectations of mar-
The Twentieth Century. The first half of the riage. More and more grounds for divorce are
twentieth century was a continuation of trends
702
DIVORCE
developed as there are higher and higher expecta- twentieth century, in short, is when most signifi-
tions for what a marriage should be. In the nine- cant changes in divorce rates occurred. And the
teenth century, drunkenness, cruelty, and failure changes noted in America were seen in most other
to provide were added to more traditional grounds Western nations.
of adultery and desertion. In the early twentieth
century, cruelty was continually redefined to in- Between World Wars I and II, there were
clude not only physical, but mental cruelty as well. widespread changes in divorce laws that reflected
changing beliefs about matrimony and its essen-
The post-war surges in divorce created suffi- tials. The strains of war and the associated prob-
cient numbers of divorced persons so that the lems that produced more divorces made the prac-
practice lost much of its stigma. The increase in tice more conspicuous and consequently more
divorce becomes more understandable when the acceptable. There is no doubt one cause of divorce
loss of stigma is considered alongside the increase is divorce. When obscure, the practice was stigma-
in women’s employment since the mid 1960s. tized and there was little to counter stereotypes
When women are employed, there is less con- associated with its practice. When divorce became
straint on them to remain in a marriage. But there more commonplace, it lost some of its stigma.
is also less constraint on their husbands who will
not be required to support their employed ex- Social changes pertaining to women’s roles
wives after a divorce. are a large part of the story of divorce during the
postwar era. One sign of these changes was the
Since 1970, divorce has been fundamentally growth, throughout the West, of women’s labor
redefined. No-fault divorce laws passed since the force participation. But the most conspicuous sym-
early 1970s have defined as unacceptable those bol of the changing role of women was the passage
marriages in which couples are ‘‘incompatible,’’ of suffrage legislation throughout the Western
have ‘‘irreconcilable differences’’ or in which the world. Before 1914, women were permitted to
marriage is ‘‘irretrievably broken.’’ Prior to the no- vote only in New Zealand, Australia, Finland, Nor-
fault regime, divorces required proof of a fault way, and eleven western U.S. states. In the United
(crime) on the part of one spouse. The court States, women were enfranchised in 1920. In Brit-
decided whether to grant the divorce. Divorce ain, Sweden, Germany, and many other European
proceedings were intentionally adversarial. To- countries, suffrage passed soon after World War I.
day, the non-adversarial grounds for divorce are
almost entirely based on the failures of emotional Divorce laws, similarly, were altered between
essentials. Emotional marital breakdown may have the wars in accordance with changing views of
been a feature of large numbers of marriages in marriage and the role of women. The British
earlier historical periods. Only now, however, is Parliament enacted divorce reform in 1937 by
such a situation viewed as solely sufficient grounds significantly extending the grounds for divorce
for terminating the marriage. (including cruelty) and granting women new op-
tions for filing for divorce. Scotland reformed its
DIVORCE IN THE WEST divorce laws in 1938 by extending grounds for
divorce to include failures of emotional essen-
Any theory of divorce must be able to account for tials—cruelty and habitual drunkenness, for exam-
the broad similarities in historical (twentieth cen- ple. In 1930, the Canadian Parliament for the first
tury) trends throughout the entire Western world. time empowered judicial magistrates to grant di-
These similarities exist despite notable differences vorce rather than requiring legislative decrees.
in national economies, forms of government, and And the Spanish divorce law of 1932 was the most
the role of the church. The trends are well known. liberal in contemporary Europe—providing di-
There was very little divorce until the end of the vorce by mutual consent (Phillips 1988, p. 539).
nineteenth century, a slow but constant growth in Even Nazi Germany permitted no-fault divorce by
divorce rates through the first half of the twentieth 1938 (though divorce law was aimed at increasing
century (interrupted by two world wars and an the number of Aryan children born).
international economic depression), and signifi-
cant increases in divorce rates since the 1960s. The Following World War II, divorce rates through-
out the Western world stabilized after an initial
increase. The low divorce rates, high fertility, and
703
DIVORCE
Divorce Rates in the United States
1940–1995
Divorces per 1000 married women 24.0
22.0
20.0 1945 1950 1955 1960 1965 1970 1975 1980 1985 1995
18.0
16.0
14.0
12.0
10.0
8.0
6.0
1940
Figure 1
SOURCE: Vital Statistics of the United States
lower age at marriage that characterized all West- of no-fault divorce laws signaled a profound shift
ern nations after World War II are trends that have in the way divorce was to be handled.
not been adequately explained. Whether these
trends reflected the consequences of war, the Most significantly, divorce became the pre-
effects of having grown up during the worldwide rogative of the married couple with little involve-
depression, or a short-term rise in social conserva- ment of the state. No-fault divorce laws do not
tism is now debated. Regardless of the cause, the require either spouse to be guilty of an offense.
decade of the 1950s is universally regarded as a Instead, they focus on the breakdown of the emo-
temporary aberration in otherwise long-term and tional relationship between the spouses. These
continuous twentieth century trends. Not until the statutes typically require a period of time during
1960s were there additional significant changes in which the spouses do not live together. Beyond
divorce laws or divorce rates. that, evidence must be adduced to substantiate
one or both spouses’ claim that the marriage is
The 1960s were years of significant social irretrievably broken. The significance of no-fault
change in almost all Western nations. The demo- divorce lies entirely in the fact that decisions about
graphic consequences of high fertility during the divorce are no longer the prerogative of the state
1950s became most apparent in the large and or church but rather of the married couple.
vocal youth movements challenging conventional
sexual and marital norms, censorship, the war in The passage of no-fault divorce laws in the
Vietnam, and educational policies. Challenges to West is properly viewed as a response to changing
institutional authority were commonplace. Divorce behaviors and attitudes. Indeed, social science
laws were not immune to the general liberalization. research has shown that divorce rates began to
‘‘Between 1960 and 1986 divorce policy in almost increase significantly prior to passage of such laws
all the countries of the West was either completely and did not change any more dramatically after-
revised or substantially reformed’’ (Phillips 1988, ward (Stetson and Wright 1975).
p. 562). Most such reforms occurred in the late
1960s to the late 1970s. Unlike earlier divorce law The changes in divorce law and actual divorce
reforms, those during the post-World War II era behaviors in the West are a reflection of the redefi-
did not extend the grounds for divorce so much as nition of marriage. More vulnerable and fragile
they redefined the jurisdiction over it. The passage emotional bonds have replaced the economic
constraints that once held spouses together. The
704
DIVORCE
availability of gainful employment for women makes have higher divorce rates than those who never
marriage less essential and divorce more possible. cohabited (Nock 1995). The reason is still unclear.
Indeed, the significant changes in women’s social Research shows that cohabiting individuals are
positions and the corresponding changes in nor- less committed to the idea of marriage or marital
mative expectations (i.e., gender) have been the permanence. They are also less religious and tend
subject of significant sociological research. These to be drawn from lower social classes (both of
changes are recognized as fundamentally altering which are associated with higher divorce rates)
almost all social institutions. Marriage is no exception. (Nock 1995). Cohabitation appears to foster (or
reflect) a belief that problems in intimate relation-
The redefinition of marriage in the latter twenti- ships are solvable by ending the relationship. When
eth century throughout the West reflects the pro- such beliefs are carried into marriage, the result is
found changes in relationships between men and higher divorce rates.
women that have occurred. No longer an econom-
ic institution, marriage is now defined by its emo- Race correlates with divorce—even after con-
tional significance. Love and companionship are trols are imposed for socioeconomic correlates of
not incidents of the institution. Rather, they are race—with black individuals having divorce rates
essentials. Meeting these high expectations may be approximately twice those of whites. However,
difficult, but sustaining them is certainly more so. such large differences associated with race are
recent in origin. Not until the late 1950s did
Taken together, the changes in the second significant differences in divorce, separation, and
half of the twentieth century may be summarized other marital statuses emerge between blacks and
as redefining the meaning of marriage. Children whites, even though a pattern of marginally higher
are not economic assets. Spouses are not econom- marital disruption has been found among blacks
ic necessities. Marriage is a conjugal arrangement for at least a century. Such findings suggest that
where the primary emphasis is on the relation- the differences stem more from contemporary
ship between husband and wife. The reasons for than historical circumstances. As Cherlin suggests,
divorce are direct consequences of the reasons the recent changes in black Americans’ family
for marriage. As one changes so does the other. situations resemble those of other racial and eth-
Since it is more difficult to accomplish and sustain nic groups, though they are more pronounced.
matrimonial essentials, it is easier to terminate The restructuring of the American economy, the
the legal framework surrounding them. Divorce decline in semi-skilled jobs, and the rise in service
has become less costly (financially, legally, and occupations has resulted in higher rates of black
reputationally) as marriage has become more so male unemployment or low wages, and better
(in terms of the investments required to accom- opportunities for black women. ‘‘Faced with diffi-
plish what is expected of it). cult times economically, many blacks responded
by drawing upon a model of social support that
CORRELATES OF DIVORCE was in their cultural repertoire, a way of making it
from day to day passed down by African Ameri-
Sociologists have documented a number of demo- cans who came before them. This response relied
graphic and personal characteristics that correlate heavily on extended kinship networks and de-
with the probability of divorce. These include emphasized marriage’’ (Cherlin 1992, p. 113).
early age at marriage, premarital births, premarital
cohabitation, divorce from a previous marriage, CONSEQUENCES OF DIVORCE
and low educational attainments. Social class is
inversely related to divorce, yet wives’ employ- For Children. A central concern of much of the
ment significantly increases divorce probabilities recent research on divorce is how children fare.
(see Huber and Spitze 1988 for a review). Developmental psychologists describe five ways in
which marital disruption may affect children’s
Half of all recent marriages began with cohabi- adjustment. First, some adults and some children
tation (unmarried couples living together) (Bumpass are more vulnerable to the stress and strain of
and Sweet 1989). Repeated national studies have divorce. Personality characteristics, ethnicity, or
found that married couples who cohabited (either
with each other, or with others) before marrying
705
DIVORCE
Divorce in Selected Countries, 1995
5
4.5
4
Divorces per 1000 people 3.5
3
2.5
2
1.5
1
0.5
0 Japan Czech Germany Hungary Nether- Spain Sweden Switzer-
Mexico Canada United Republic lands land
States
Figure 2
SOURCE: United Nations 1996 Demographic Yearbook
age for example, may make some individuals more the chances of premarital cohabitation, and in-
susceptible to negative outcomes. Second, the ab- creases the odds of premarital pregnancies or
sence of one parent, per se, may affect children’s fatherhood (Cherlin, Kiernan, and Chase-Lansdale
adjustment to divorce. Boys, in particular, appear 1995). The effects of divorce in young adulthood
to benefit from the presence of a male adult. include higher rates of unemployment and lower
Third, the loss of income creates many indirect educational attainments. Divorce weakens young
problems for children, including changes in resi- people’s connections to their friends and neigh-
dence, school, neighborhood, and peer networks. bors due to higher rates of residential mobility
Fourth, divorce often diminishes the custodial (McLanahan and Sandefur 1994). Following di-
parent’s ability to provide supportive and appro- vorce, many children are subjected to changes in
priate parenting, especially if depression follows residence, often to disadvantaged neighborhoods
marital disruption. And finally, negative, conflictual, where peers have lower educational prospects.
and dysfunctional family relationships between The lack of connections to others affects parents’
parents, parents and children, and siblings are ability to monitor their children. It also limits
probably the most damaging consequence of di- young people’s knowledge about local employ-
vorce for children. (Hetherington, Bridges, and ment opportunities.
Insabella 1998).
The changed economic circumstances caused
Longitudinal research has shown that child- by divorce affects children in many indirect ways.
ren who experience divorce differ from others The loss of available income may affect the quality
before the disruption occurs. Cherlin showed that of schools children attend if custodial parents
children whose parents were still married, but who move to poorer neighborhoods. The lack of in-
would later divorce, showed more behavior prob- come may limit children’s opportunities for extra-
lems and did less well in school than children curricular activities (e.g., travel, or music lessons).
whose parents would remain married (Cherlin et The need for income often compels custodial
al. 1991). parents to work more hours, reducing their ability
to monitor children’s after-school activities.
Even after such predisruption differences are
considered, divorce takes a toll in the lives of In their socioeconomic attainments, children
children who experience it. Divorce significantly who experienced their parents’ divorce average
increases the chances that young people will leave one to two fewer years of educational attainment
their homes due to friction with a parent, increases than children from intact homes (Krein and Beller
706
DIVORCE
1988; Hetherington, Camara, and Featherman and Featherman 1983, p. 218). In matters of disci-
1983). Such effects are found even after rigorous pline, single mothers have been found to rely on
controls are imposed for such things as race, sex, restrictive (authoritarian as opposed to authorita-
years since the divorces, age at time of divorce, tive) disciplinary methods—restricting children’s
parental income, parental education, number of freedom and relying on negative sanctions—a pat-
siblings, region of residence, educational materi- tern psychologists believe reflects a lack of authori-
als in the home, or the number of years spent in ty on the part of the parent (Hetherington 1972).
the single-parent family. There are comparable Whatever else it implies, the lack of generational
effects of divorce on occupational prestige, in- boundaries means a less hierarchical family and
come and earnings, and unemployment (Nock 1988). less authoritative generational distinctions.
White women who spent some childhood time The institutional contexts within which achieve-
in a single-parent family as a result of divorce are ment occurs, however, are decidedly hierarchical
53 percent more likely to have teenage marriages, in nature. Education, the economy, and occupa-
111 percent more likely to have teenage births, tions are typically bureaucratic structures in which
164 percent more likely to have premarital births, an individual is categorically subordinate to a su-
and 92 percent more likely to experience marital perior—an arrangement Goffman described as an
disruptions than are daughters who grew up in ‘‘eschelon authority structure’’ (1961, p. 42). The
two-parent families. The effects for black women nuclear family has been described as producing in
are similar, though smaller. Controls for a wide children the skills and attitudes necessary for com-
range of background factors have little effect on petition within such eschelon authority relation-
the negative consequences of divorce. Further, ships in capitalist production and family childrearing.
remarriage does not remove these effects of di- ‘‘The hierarchical division of labor (in the econo-
vorce. And there is no difference between those my) is merely reflected in family life’’ (Bowles and
who lived with their fathers and those who lived Gintis 1976, p. 144–147). The relative absence of
with their mothers after divorce. Experiencing clear subordinate-superordinate relationships in
parents’ divorce has the same (statistical) conse- single-parent families has been argued to inade-
quences as being born to a never-married mother quately socialize children, or place them in a disad-
(McLanahan and Bumpass 1988; McLanahan and vantageous position when and if they find them-
Sandefur 1994). selves in hierarchical organizations (Nock 1988).
Such large and consistent negative effects have For Adults. A wide range of psychological
eluded simple explanation. Undoubtedly much of problems has been noted among divorcing and
the divorce experience is associated with the alter- recently divorced adults. A divorce occasions
ed family structure produced—in almost 90 per- changes in most every aspect of adult life; resi-
cent of all cases a single-mother family—and the dence, friendship networks, economic situation,
corresponding changes in family functioning. Such and parental roles. Marriage in America makes
a structure is lacking in adult role models, in significant contributions to individual well-being.
parental supervision, and in hierarchy. On this last Thus, regardless of the quality of the marriage that
dimension, research has shown that divorced wom- ends, emotional distress is a near-universal experi-
en and their children are closer (less distinguished ence for those who divorce (Weiss 1979). Anxiety,
by generational distinctions) to one another than anger, and fear are dominant psychological themes
is true in intact families. Parent and child are immediately before and after divorce. At least for a
drawn together more as peers, both struggling to year or two after divorce, men and women report
keep the family going. The excessive demands on psychosomatic symptoms of headaches, loss of
single parents force them to depend on their appetite, overeating, drinking too much, trem-
children in ways that parents in intact families do bling, smoking more, sleeping problems, and nerv-
not, leading to a more reciprocal dependency ousness (Group for the Advancement of Psychia-
relationship (Weiss 1975, 1976). Single mothers try 1980).
are ‘‘. . . likely to rely on their children for emo-
tional support and assistance with the practical The emotional problems occasioned by di-
problems of daily life’’ (Hetherington, Camara, vorce are accompanied by major changes in eco-
nomic situations, as well—especially for women.
707
DIVORCE
The vast majority of those involved in divorce family status—especially divorce and remarriage—
experience a significant decline in their immediate are the most important cause of change in family
standard of living. This problem is especially acute economic well-being and poverty among women
for women who—in almost 90 percent of cases— and children (1984).
assume custody of children. Immediately after a
divorce, women suffer an average 30 percent to 40 Single-parent families in America have grown
percent decline in their overall standards of living dramatically as a result of increasing divorce rates.
(Hoffman and Duncan 1988; Peterson 1989). Ei- And even though most divorced persons remarry,
ther in anticipation of or as a consequence of Bumpass has shown that the average duration of
divorce, there is typically an increase in divorced marital separation experienced by children under
women’s labor force participation. Analyzing na- age 18 was 6.3 years and 7.5 years for whites and
tional longitudinal data, Peterson estimates that blacks respectively. In fact 38 percent of white and
one year before the divorce decree (when most 73 percent of black children are still in a single-
divorcing individuals are separated), women’s av- parent family 10 years after the marital disrup-
erage standard of living (total family income divid- tion—a reflection of blacks’ lower propensity to
ed by the poverty threshold for a family of a remarry and their longer intervals between di-
particular size) is 70 percent of its level in the vorce and remarriage (1984). The role of divorce
previous year. As a consequence of increased hours in the formation of single-parent families differs
worked, the standard of living increases one year by race. Among all single-parent white families, 25
after divorce and by five or six years after divorce, percent are maintained by never-married moth-
‘‘the standard of living of divorced women is about ers, 47 percent by divorced (or separated) moth-
85 percent of what it had been before separation’’ ers, 7 percent by never-married men, and 13 per-
(1989, p. 48). Women who have not been em- cent by divorced or separated men. Among black
ployed during their marriages, however, are par- single-parent families, 59 percent are maintained
ticularly hard-hit; the majority ending up in poverty. by never-married women, 28 percent by divorced
or separated women, 4 percent by never-married
Child support payments are not a solution to men, and only 3 percent by divorced or separated
the economic problems created by divorce for two men. Divorce is the primary route to single-parent-
reasons. First, about one-quarter (24 percent) of hood for white mothers, whereas out-of-wedlock
women due child support receive none (39 per- childbearing is for black mothers (U.S. Bureau of
cent of men awarded child support receive none). the Census 1998c, Table 11; 1998d).
Another one-quarter receive less than the court-
ordered amount. In 1991, the average amount of Families headed by single women with child-
child support received by divorced mothers was ren are the poorest of all major demographic
$3,011 per year ($2,292 for men) (U.S. Bureau of groups regardless of how poverty is measured.
the Census 1995). About 16.7 million, or 85 per- Combined with frequent changes in residence and
cent of the 19.8 million children in single-parent in employment following divorce, children and
families in 1997 were living with the mother; 60 mothers in such households experience signifi-
percent of whom were divorced (U.S. Bureau of cant instabilities—a fact reflected in the higher
the Census 1998a). Their median family income rates of mental health problems among such wom-
was $22,999 compared with $34,802 for those in en (Garfinkel and McLanahan 1986, pp. 11–17).
single-father situations, and $51,681 for children
in households where both parents were present CONCLUSION
(U.S. Bureau of the Census 1998a). Families head-
ed by single mothers are the most likely to be in High rates of remarriage following divorce clearly
poverty, and represent 55 percent of all poor indicate that marital disruption does not signify a
families. In 1997, a third (31.6 percent) of all rejection of marriage. There is no evidence of
single-mother families were in poverty compared widespread abandonment of conjugal life by Ameri-
to 5.2 percent of two-parent families (U.S. Bureau cans. Admittedly, marriage rates have dropped in
of the Census 1998b). Analyzing national longitu- recent years. However, such changes are best seen
dinal data, Duncan concluded that changes in to be the result of higher educational attainments,
708
DIVORCE
occupational commitments, and lower fertility ex- are not necessarily co-residents with their child-
pectations; not a rejection of marriage per se. ren. And new categories of ‘‘quasi’’ kin are invent-
Rather, increasing divorce rates reflect the fact ed to accommodate the complex connections
that marriage is increasingly evaluated as an en- among previously married spouses and their new
tirely emotional relationship between two per- spouses and children. In many ways, divorce itself
sons. Marital breakdown, or the failure of mar- has become a dominant institution in American
riage to fulfill emotional expectations, has come society. It is, however, significantly less structured
increasingly to be a cause for divorce. Since the by consensual normative beliefs than the family
1970s, our laws have explicitly recognized this as institutions to which it is allied.
justification for terminating a marriage—the best
evidence we have that love and emotional close- REFERENCES
ness are the sine qua non of modern American
marriage. Contemporary divorce rates thus signal Bennett, Neil G., David E. Bloom, and Patricia H. Craig
a growing unwillingness to tolerate an unsatisfying 1989 ‘‘The Divergence of Black and White Marriage
emotional conjugal relationship. Patterns.’’ American Journal of Sociology 95 (3):692–772.
The consequences of divorce for children are Bowles, Samuel, and Herbert Gintis 1976 Schooling in
difficult to disentangle from the predictable changes Capitalist America. New York: Basic Books.
in household structure. Whether the long-term
consequences are produced by the single-parent Bumpass, Larry L. 1984 ‘‘Children and Marital Disrup-
situation typically experienced for five to ten years, tion: A Replication and Update.’’ Demography 21:71–82.
or from the other circumstances surrounding di-
vorce is not clear. It is quite apparent, however, ———, and James A. Sweet 1989 ‘‘National Estimates of
that divorce occasions significant instabilities in Cohabitation.’’ Demography 26:615–625.
children’s and mothers’ lives.
Cherlin, Andrew J. 1981 Marriage, Divorce, and Remarriage.
Our knowledge about the consequences of Cambridge, Mass.: Harvard University Press.
divorce for individuals is limited at this time by the
absence of controlled studies that compare the ———, F.F. Furstenberg, Jr., P.L. Chase-Lansdale, K.E.
divorced to the nondivorced. Virtually all research Kiernam, P.K. Robins, D.R. Morrison, and J.O. Teitler
done to date follows the lives of divorced individu- 1991 ‘‘Longitudinal Studies of Effects of Divorce on
als without comparing them to a comparable group Children in Great Britain and the United States.’’
of individuals who have not divorced. A related Science 252 (June 7):1386–1389.
concern is whether the consequences of divorce
reflect the experience itself, or whether they re- Cherlin, Andrew J., Kathleen E. Kiernam, and P. Lind-
flect various selection effects. That is, are people say Chase-Lansdale 1995 ‘‘Parental Divorce in Child-
who divorce different from others to begin with? hood and Demographic Outcomes in Young Adult-
Are their experiences the results of their divorce, hood.’’ Demography 32 (No. 3):299–318.
or of antecedent factors?
Degler, Carl N. 1980 At Odds: Women and the Family in
When almost half of all marriages are predict- America from the Revolution to the Present. New York:
ed to end in divorce, it is clear that marital disrup- Oxford University Press.
tion is a conspicuous feature of our family and
kinship system. Divorce creates new varieties of Duncan, Greg J. 1984 Years of Poverty, Years of Plenty. Ann
kin not traditionally incorporated in our domi- Arbor, Mich.: University of Michigan, Institute for
nant institutions. The rights and obligations at- Social Research.
tached to such kinship positions as spouse of the
noncustodial father are ambiguous—itself a source Garfinkel, Irwin, and Sara S. McLanahan 1986 Single
of problems. The social institution of the family is Mothers and their Children. Washington, D.C.: The
redefined as a consequence of divorce. Entering Urban Institute Press.
marriage, for example, is less commonly the begin-
ning of adult responsibilities. Ending marriage is Goffman, Irving 1961 Asylums. New York: Anchor.
less commonly the consequence of death. Parents
Group for the Advancement of Psychiatry 1980 Divorce,
Child Custody, and the Family. San Francisco: Jossey-Bass.
Hetherington, E. Mavis 1972 ‘‘Effects of Paternal Ab-
sence on Personality Development in Adolescent
Daughters.’’ Developmental Psychology 7:313–326.
———, Margaret Bridges, and Glendessa M. Insabella
1998 ‘‘What Matters? What Does Not? Five Perspec-
tives on the Association Between Marital Transitions
709
DRUG ABUSE
and Children’s Adjustment.’’ American Psychologist 53 ——— 1998c ‘‘Household and Family Characteristics:
(No.2):167–184. 1997.’’ Current Population Reports. P-20, No. 515.
Hetherington, E.M., K.A. Camara, and D.L. Featherman ——— 1998d ‘‘Marital Status and Living Arrangements:
1983 ‘‘Achievement and Intellectual Functioning of 1997.’’ Current Population Reports. P-20, No. 506.
Children in One Parent Households.’’ In J. T. Spence,
ed., Achievement and Achievement Motives, 205–284 Weiss, Robert 1975 Marital Separation. New York: Ba-
San Francisco: W. H. Freeman. sic Books.
Hoffman, Saul D., and Greg J. Duncan 1988 ‘‘What ——— 1976 ‘‘The Emotional Impact of Marital Separa-
are the Economic Costs of Divorce?’’ Demography tion.’’ Journal of Social Issues 32:135–145.
25:641–645.
——— 1979 Going it Alone: The Family Life and Social
Huber, Joan, and Glenna Spitze 1988 ‘‘Trends in Family Situation of the Single Parent. New York: Basic Books.
Sociology.’’ In N. J. Smelser, ed., Handbook of Sociolo-
gy. Beverly Hills, Calif.: Sage. STEVEN L. NOCK
ALISON BURKE
Krein, Sheila F., and Andrea H. Beller 1988 ‘‘Education-
al Attainment of Children from Single-Parent Fami- DRAMATURGY
lies: Differences by Exposure, Gender, and Race.’’
Demography 25:221–234. See Symbolic Interaction Theory.
Martin, Teresa C., and Larry L. Bumpass 1989 ‘‘Recent DRUG ABUSE
Trends in Marital Disruption.’’ Demography 26 (1):37–51.
Drug abuse has been a major social problem in the
McLanahan, Sara S., and Larry Bumpass 1988 ‘‘Inter- United States for almost a century and we are now
generational Consequences of Family Disruption.’’ in the second decade of a continuing war on drugs.
American Journal of Sociology 94 (1):130–152. Drug abuse is a health and criminal justice prob-
lem that also has implications for nearly every
National Center for Health Statistics 1998 FASTATS: facet of social life. It is a major element in the high
‘‘Divorce.’’ www.cdc.gov/nchswww/fastats/divorce.htm cost of health care, a central reason for the United
States’s extraordinarily high rate of incarceration,
Nock, Steven L. 1988 ‘‘The Family and Hierarchy.’’ and a focus of intensive education and treatment
Journal of Marriage and the Family 50 (Nov):957–966. efforts. Substance abuse is an equal-opportunity
problem that affects both high- and low-income
——— 1995 ‘‘A Comparison of Marriages and Cohabiting persons, although its consequences are most often
Relationships.’’ Journal of Family Issues 16:53–76. felt by those persons and communities that have
the lowest social capital.
Peterson, Richard R. 1989 Women, Work, and Divorce.
Albany: State University of New York Press. Substance abuse, with its connotations of disap-
proval or wrong or harmful or dysfunctional usage
Phillips, Roderick 1988 Putting Asunder: A History of of mood-modifying substances, is a term that was
Divorce in Western Society. New York: Cambridge developed in the United States. The more neutral
University Press. term, dependence, is often used in other countries.
Addiction, which formerly communicated the de-
Stetson, Dorothy M., and Gerald C. Wright, Jr. 1975 velopment of tolerance after use and a physical
‘‘The Effects of Law on Divorce in American States.’’ withdrawal reaction after a drug became unavail-
Journal of Marriage and the Family (August):537–547. able, has assumed less explicit meanings. Whatev-
er terminology is employed, there is intense socie-
U.S. Bureau of the Census 1975 Historical Statistics of the tal concern about the use of psychoactive mood-
United States: Colonial Times to 1970. Part I. Washing- altering substances that involve loss of control.
ton, D.C.: U.S. Government Printing Office. This concern is manifest particularly for young
people in the age group most likely to use such
——— 1989 ‘‘Child Support and Alimony: 1985.’’ Cur- substances. Society is concerned that adolescents
rent Population Reports. P-23, No. 154.
——— 1989 ‘‘Studies in Marriage and the Family.’’
Current Population Reports. P-23, No. 162.
——— 1995 ‘‘Who Receives Child Support?’’ Statistical
Brief. www.census.gov/socdemo/www/chldsupp.html
——— 1998a ‘‘Money Income in the United States:
1997.’’ Current Population Reports. P-60, No. 200.
——— 1998b ‘‘Poverty in the United States: 1997.’’
Current Population Reports. P-60, No. 201.
710
DRUG ABUSE
and young adults, who should be preparing them- individuals used heroin in 1997, with the estimat-
selves for crucial educational, vocational, and oth- ed number of new users at the highest level in
er significant life choices, are instead diverted by thirty years.
the use of controlled substances.
Data on incidence and prevalence of use must
The United States has the highest rate of drug be interpreted in terms of social structure. Thus,
abuse of any industrialized country and, not sur- one out of five of the American troops in Vietnam
prisingly, spends more public money than any were addicted to heroin, but follow-up studies one
other country to enforce laws that regulate the use year after veterans had returned to the United
of psychoactive drugs. Its efforts to control drug States found that only 1 percent were addicted
abuse reach out across its borders. The United (Robins, Helzer, Hesselbrook et al. 1980). In Viet-
States also plays a critical role in developing knowl- nam, heroin use was typically found among enlist-
edge about substance abuse; more than 85 percent ed men and not among officers. Knowing such
of the world’s drug abuse research is supported by aspects of social setting and role can help in under-
the National Institute on Drug Abuse. standing the trends and can contribute to under-
standing the use of other substances in other
EPIDEMIOLOGY situations. In any setting, the frequency of sub-
stance use, the length of time over which it was
Information on incidence and prevalence of drug taken, the manner of ingestion, whether it was
use and abuse derives from a range of sourc- used by itself or with other substances, its relation-
es: surveys of samples of households and schools; ship to criminal activity and other user characteris-
hospital emergency room and coroners’ reports; tics (e.g., mental illness), the degree to which its
urine testing of samples of arrestees; treatment use was out of control, the setting, and whether it
programs; and ethnographic studies. Such was part of a group activity are also important.
epidemiological information enables us to assess
drug abuse programs and decide on allocation of Rates of use by subgroup can vary greatly.
resources (Winick 1997). Thus, for example, prevalence rates of drug use
are higher among males than females and highest
Since World War II, the peak years for illicit among males in their late teens through their
drug use were in the late 1970s, when approxi- twenties. Over half the users of illicit drugs work
mately 25 million persons used a proscribed sub- full time. About one-third of homeless persons
stance in any thirty-day period. Overall illicit drug and more than one-fourth of the mentally ill are
use has been declining since 1985. The yearly physically or psychologically dependent on illicit
National Household Survey on Drug Abuse, which drugs. The first survey of mothers delivering
is the most influential source of epidemiology liveborns, in 1993, found that 5.5 percent had
data, reported that in 1997 marijuana was used by used illicit drugs at some time during their preg-
11.1 million persons or 80 percent of illicit drug nancy. A survey of college students reported that
users (Office of Applied Studies 1999). Sixty per- in the previous year, 26.4 percent had used mari-
cent only used marijuana but 20 percent used it juana and 5.2 percent had used cocaine. National
along with another illicit substance. During the Household Survey data indicate that use of illicit
1990s, the rate of marijuana initiation among youths drugs by persons over thirty-five, which was 10.3
aged twelve to seventeen reached a new high, of percent in 1979, jumped to 29.4 percent by 1991
approximately 2.5 million per year. The level of and was 33.5 percent in 1997.
current use of this age group (9.4 percent) is
substantially less than the rate in 1979 (14.2 percent). Rates of cigarette smoking are of interest be-
cause of their possible relationship to the use of
Twenty percent of illicit drug users in 1997, other psychoactive substances. Approximately one-
ingested a substance other than marijuana in the eighth of cigarette smokers also use illicit drugs. In
month preceding the interviews. Some 1.5 million a typical month in 1997, 30 percent of Americans,
Americans, down from 5.7 million in 1985, used or 64 million, had smoked cigarettes and one-fifth
cocaine in the same period; the number of crack of youths between the ages of twelve and seven-
users, approximately 600,000, has remained near- teen, were current smokers. Almost half of all
ly constant for the last ten years. At least 408,000 American adults who ever smoked have stopped
711
DRUG ABUSE
smoking. Drug abusers may also be involved with is the establishment of needle exchanges, for in-
alcohol. jecting users of heroin and other drugs, in order to
minimize the possibility of HIV transmission re-
POLICY sulting from the sharing of infected needles. The
use of needles to inject illegal substances has been
A central contributor to current American policy linked to one-third of the cumulative number of
toward mood-modifying drugs was the Harrison AIDS cases in the United States. In the United
Act of 1914, which prevented physicians from States, the use of federal government money for
dispensing narcotics to addicts (Musto 1987). The needle exchanges is prohibited, although there
Marijuana Tax Act of 1937 and strict penalties for are approximately 1.3 million injecting drug users.
sale and possession of narcotics that were imposed Critics of these programs believe that such ex-
by federal legislation in 1951 and 1966 expanded changes increase heroin use and send a latent
punitive strategies. An important change took place message that it is acceptable to use drugs like
in 1971, when President Nixon—who had cam- heroin. Harm reductionists disagree and argue
paigned vigorously against drug use—established that needle exchanges lead to a decline in rates of
a national treatment network. Nixon was the only HIV infection without encouraging use.
president to devote most of the federal drug budg-
et to treatment; his successors have spent most of Another policy disagreement between Ameri-
the budget on law enforcement. ca and other countries involves marijuana. In the
United States many federal benefits, including
In 1972, the Commission on Marijuana and student loans, are not available to those convicted
Drug Abuse recommended a dual-focused policy of marijuana crimes. In contrast, marijuana has
that is both liberal and hard-line. The policy, which been decriminalized in a number of Western Eu-
continues to the present, is liberal in that users ropean countries, including Italy, Spain, and Hol-
who need help are encouraged to obtain treat- land. It is openly available in coffee houses in
ment. But it is hard-line because it includes harsh Holland, where officials believe that its use is
criminal penalties for drug possession and sales. relatively harmless and can deter young people
As a result, nearly two-thirds of the federal re- from using heroin or cocaine. In America, mari-
sources devoted to drug use are now spent by the juana is viewed by federal authorities as possibly
criminal justice system to deter drug use and hazardous and a potential ‘‘stepping stone’’ to
implement a zero-tolerance philosophy. heroin or cocaine use, and approximately 695,000
persons were arrested for its possession in 1997.
President Carter’s 1977 unsuccessful attempt
to decriminalize marijuana was the only effort by a Other countries have experimented with ways
national political leader to lessen harsh penalties to make drugs such as heroin legally available,
for drug possession. Between 1981 and 1986, Presi- albeit under control. Thus, in Switzerland, heroin
dent Reagan doubled enforcement budgets to addicts have been legally maintained. In England,
fight the ‘‘war on drugs.’’ Politicians generally methadone (a heroin substitute) can be obtained
have felt that the traditional hard-line policy served by prescription from a physician. In the United
their own and the country’s best interests and States, by contrast, an addict must enroll in a
there has been limited national support for legali- program to be able to receive methadone.
zation or decriminalization (Evans and Berent 1992).
In the United States prevention of drug abuse
Originating in several European countries, has never been as important a policy dimension as
the policy of harm reduction has, during the last treatment or law enforcement, in part because it
decade, generated growing interest in the United requires legislators to commit resources in the
States as a politically viable alternative to legaliza- present to solve a future problem. Prevention has,
tion (Heather, Wodak, Nadelmann et al. 1998). It thus, accounted for less than one-seventh of the
attempts to understand drug use nonevaluatively drug abuse budget. Because of the variety of pre-
in the context of people’s lives and to urge that the vention approaches and because of the American
policies that regulate drug use should not lead to local approach to education, there are many view-
more harm than the use of the substance itself points on how to conduct programs that will pre-
causes. A representative harm-reduction initiative vent young people from becoming drug users and
712
DRUG ABUSE
abusers. An information-didactic approach, often Preventing illicit drugs from entering the Unit-
with the assistance of law enforcement personnel, ed States is difficult because of heavily trafficked,
has been traditional. A role-training, peer-orient- long, porous borders. Large tax-free profits pro-
ed, values-clarification, alternatives, affective-edu- vide incentives for drug entrepreneurs to develop
cation approach emerged in the 1970s, along with new ways to evade customs barriers, process the
psychological inoculation. Addressing the social drug for the market, and sell it (Johnson, Goldstein,
structure and family in which young people live, Preble et al. 1985). For example, approximately
and targeted community action, attracted substan- seven-eighths of the retail price represents profit
tial support in the 1980s and 1990s. after all costs of growing, smuggling, and process-
ing cocaine for illegal sale in the United States. The
National policy toward drug use is systemati- increasing globalization of the world economy
cally promulgated by the Office of National Drug further facilitates the international trade in illicit
Control Policy (1999). The office has established substances.
the goal of reducing drug use and availability by 50
percent and reducing the rate of related crime and A key component in efforts to reduce the
violence by 30 percent by 2007. It is proposed that supply of stimulants, depressants, and hallucino-
these goals will be achieved by expanding current gens is the Comprehensive Drug Abuse and Con-
approaches. trol Act of 1970, which established a national
system of schedules that differentiated the public
Although drug abuse has been called ‘‘the health threat of various drugs of abuse. This law,
American disease,’’ physicians have had little im- which has been modified over the years, classifies
pact on policy. Between 1912 and 1925, clinics in controlled substances into five categories, based
various states dispensed opiates to users. More on their potential for abuse and dependency and
recently, however, the federal government has their accepted medical use. Schedule I products,
opposed making marijuana available for medici- such as peyote, have no acceptable safe level of
nal purposes, even to treat persons with terminal medical use. Schedule II products, such as mor-
or debilitating illnesses. Nevertheless, eleven states phine, have both medicinal value and high abuse
decriminalized marijuana possession in the 1970s potential. Schedule III substances, such as am-
and others, by referendum vote in the 1990s, have phetamines, have medical uses but less abuse po-
permitted physicians to recommend and patients tential than categories I or II. Also acceptable
to use marijuana medically. medicinally, Schedule IV substances, such as phe-
nobarbital have low abuse potential, although the
CONTROL potential is higher than Schedule V products, such
as narcotics that are combined with non-narcotic
In the United States, programs to control the active ingredients. Conviction for violation of fed-
supply of mood-modifying substances are intend- eral law against possession or distribution of sched-
ed to interdict the importation of illicit materials, uled products can lead to imprisonment, fines,
enforce the laws, and cooperate with other coun- and asset forfeiture.
tries that are interested in minimizing the availa-
bility of controlled substances. In addition to illicit Ever since it assumed a major role in promot-
substances (such as heroin, that has no established ing the Hague Opium Convention of 1912, the
medical use), prescription products can be abused. United States has been a leader in the internation-
These include substances such as barbiturates, al regulation of drugs of abuse. The United States
that are used without medical supervision in an convened the 1961 Single Convention on Narcotic
inappropriate manner. The problem also includes Drugs and the 1971 Convention on Psychotropic
over-the-counter drug products that are not used Substances. Some countries, like England and Hol-
for the purpose for which they were manufac- land, subscribe to the treaties but interpret them
tured. Some nondrug substances like airplane mod- more liberally than does the United States. The
el glue and other inhalants that can provide a United States has also provided technical assist-
‘‘high’’ and are difficult to regulate, are also con- ance, financing, and encouragement to other coun-
sidered part of the country’s substance abuse burden. tries to minimize the growth of drugs such as
cocaine and marijuana. Programs have been con-
ducted in Mexico and Turkey to eradicate these
713
The words you are searching are inside this book. To get more targeted content, please make full-text search by clicking here.
Encyclopedia of sociology by Edgar F. Borgatta, Rhonda J. V. Montgomery (z-lib.org)
Discover the best professional documents and content resources in AnyFlip Document Base.
Search
Encyclopedia of sociology by Edgar F. Borgatta, Rhonda J. V. Montgomery (z-lib.org)
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 650
- 651 - 700
- 701 - 750
- 751 - 756
Pages: