MD1

Family Violence

Author(s): Richard J. Gelles

Source: Annual Review of Sociology , 1985, Vol. 11 (1985), pp. 347-367

Published by: Annual Reviews

Stable URL: https://www.jstor.org/stable/2083298

REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/2083298?seq=1&cid=pdf- reference#references_tab_contents You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of Sociology

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

Ann. Rev. Sociol. 1985. 11:347-67 Copyright ? 1985 by Annual Reviews Inc. All rights reserved

FAMILY VIOLENCE

Richard J. Gelles

College of Arts and Sciences, University of Rhode Island, Kingston, Rhode Island

02881

Abstract

This chapter reviews research on family violence. Once viewed as rare and

confined to a few mentally ill offenders, family violence has rapidly captured

public and social scientific attention. The review examines the "discovery" of family violence as a social and sociological problem. Among the major prob- lems that confront students of family violence are defining, both nominally and

operationally, child abuse, wife abuse, and violence. Access to cases, sam- pling, and measurement of violence are additional issues that are reviewed.

Research on family violence is described with a specific focus on the extent of

the various forms of family violence and the factors associated with violence in the home. Seven theoretical models that have been developed to analyze the

specific issue of family violence are briefly reviewed. The chapter concludes

by covering new issues in the field of family violence, including research on the

responses of victims of wife abuse, studies of the consequences of child abuse,

and evaluation studies of prevention and treatment programs.

THE DISCOVERY OF A LATENT FAMILY AND SOCIAL PROBLEM

Sociologists came late to the study of family violence. Social workers recog- nized but did not write about child abuse and child maltreatment in the

nineteenth century. Pediatric radiologists in 1946 identified the pattern of

broken bones and healed fractures that were thought to be caused by children's

caretakers (Caffey 1946, 1957; Silverman 1953). Physician C. Henry Kempe

and his colleagues published their seminal article, "The Battered-Child Syn-

drome," in the July, 1962, issue of the Journal of the American Medical

347

0360-0572/85/0815-0347$02.00

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

348 GELLES

Association. This article and an accompanying editorial served to focus medi-

cal and public attention on the problem of child abuse and shaped the way

scientists and the lay public thought about child abuse in the ensuing years

(Nelson 1984). By the mid-1960s there was a growing literature on child abuse;

yet, in the complete bibliography on child abuse published by the then US

Department of Health, Education, and Welfare in 1969, not a single entry was

authored by a sociologist.

Violence towards women, or wife abuse, received little public or scientific

attention until the 1970s. Yet here too the earliest writings were not by

sociologists. Wife abuse was initially approached from a psychiatric point of

view (Schultz 1960; Snell, Rosenwald & Robey 1964).

What sociological work there was on family violence prior to 1970 was

embedded in criminological studies of homicide (see for example Gillen 1946,

Palmer 1962, Pokorny 1965, Wolfgang 1958).

In spite of the historical and crosscultural evidence that the family has been

the scene of interpersonal violence for as long as we have had written records of

humankind and family life (Bakan 1971, DeMause 1974, 1975; Korbin 1981, Radbill 1980, Taylor & Newberger 1979, Shorter 1975), and despite the fact

that other disciplines had begun to study abuse and violence in families,

sociological research and recognition of these topics did not begin in earnest

until the 1970s. Why were sociologists late in identifying family violence as a

topic for investigation? What forces resulted in the discipline's change from

selective inattention and the assumption that family violence was a latent social

problem? How did family violence finally capture attention as a manifest social

problem?

Star (1980) claims that the silence that traditionally surrounded the issue of family violence was attributable to three factors: (a) lack of awareness, (b)

general acceptance, and (c) denial. Straus (1974) notes that three factors influenced the emergence of family violence as a social and sociological

problem: (a) the sensitization to violence that occurred in the late 1960s and

early 1970s due to the Vietnam war, political assassinations, violent social

protest, and the rise in the homicide and assault rates; (b) the reemergence of the

women's movement; and, (c) theoretical challenges to the consensus model of

society by proponents of the conflict and social action models of society. There are additional factors that may help explain why and how sociologists

first overlooked and then discovered family violence. First, violence in the

home is a family issue, a violence issue, and a criminological issue. While it

touches all these specialties in sociology, family violence finds no "home" in any one specialty. For the criminologist, family violence is but one form of criminal violent behavior. Students of family relations frequently group family violence into a residual "deviance in the family" category. A second problem is

that family violence is a private matter. Unlike assassination, riots, rebellion,

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 349

and street violence, most family violence is hidden behind the symbolic and

literal closed doors of the home (Gelles 1974). The private nature of family

violence not only hides the problem from public and scientific view, it also

makes the victims and offenders nearly inaccessible to many social researchers.

While social workers, psychiatrists, and physicians have access to participants

in family violence, sociologists' access to cases and subjects is frequently

limited to public instances of violence homicides, assaults, or officially

reported cases of child abuse. Social service personnel and hospital personnel

frequently have denied sociologists access to subjects on the grounds of

confidentiality.

The early 1970s saw a number of changes. First, all states instituted official

reporting of child abuse and neglect cases, and aggregate statistics on child

abuse became available for sociological analysis. Second, safe houses, or

battered wife shelters, were established to serve the needs of battered women.

Many of these shelters were established by feminist organizations, who granted female sociologists access to the shelters for purposes of sociological research.

The vast majority of published research on wife battering has been based on

samples drawn from such shelters (see for example Dobash & Dobash 1979,

Pagelow 1981). Third, exploratory studies of family violence (O'Brien 1971,

Levinger 1966, Steinmetz 1971, Straus 1971) demonstrated that research could

be conducted (using nonclinical samples) on family violence. These early studies paved the way by outlining appropriate methods and research strategies

for uncovering and examining instances of family violence.

A final constraint on sociological research was that the earliest writings on

child abuse and wife abuse shaped the scientific and public perception of the problem. Thus, because the medical/psychiatric model was the first applied to

studying abuse and violence in the family, there has been a tendency to see

abuse as a psychiatric problem. The first reports on child abuse not only profiled the suspected psychopathology that led to abuse, but also claimed that

social factors were essentially irrelevant as a generative cause (Galdston 1965,

Steele & Pollock 1974).

The Sociological Perspective

When sociologists did become involved in studying family violence, they

found that they had a twofold mandate. First, they were debunkers, critiquing

the early, medical-model dominated research. Sociologists critiqued research on child and wife abuse on methodological grounds small, nonrepresentative

samples, lack of comparison groups, and improper data analysis (see Gelles 1973 for a critique of the early child abuse research); and on theoretical

grounds-unicausal models and post-hoc theorizing (Dobash & Dobash 1979,

Gelles 1974). Sociologists attacked the myths that family violence was rare and

was confined to mentally disturbed people (Steinmetz & Straus 1974), that only

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

350 GELLES

poor people abused their wives and children; and that battered women liked

being abused. Sociologists also helped discover hidden forms of family vio-

lence, such as sibling violence (Steinmetz 1977), violence towards parents

(Steinmetz 1978c; Cornell & Gelles 1982a), and violence towards the elderly

(Cornell & Gelles 1982b; Steinmetz 1984).

A second mandate was the use of the sociological perspective and research

tools to help establish the nature, extent, and dynamics of family violence as a

social problem. Sociologists were consulted for research on incidence and

prevalence, for studies of the factors correlated with abuse and violence, and

for causal models and theories. In addition, social service and medical person-

nel, starving for information to help deal with the emotionally trying cases of

violence they encountered, sought advice about how best to treat and prevent

violence and abuse.

The first decade of research on family violence focused on four major issues:

(a) defining (nominally and operationally) abuse and violence; (b) attempting

to measure the incidence and prevalence of family violence; (c) assessing the

factors correlated with violence in the home; and (d) developing causal models

to explain family violence. The last four years have seen sociologists go beyond

these initial concerns and begin to assess the consequences of violence, the

dynamics of violence, and the impact of interventions on the occurrence of

violence in the home.

DEFINITIONAL ISSUES

One of the major problems confronting researchers who attempt to study

violence in the family has been the quagmire of definitional dilemmas encoun-

tered. The concepts "violence" and "abuse" have frequently been used inter-

changeably by those who study domestic violence. These concepts, however,

are not conceptually equivalent. Moreover, there are considerable variations in

how each concept is nominally defined.

Defining Abuse

The first form of family violence that was uncovered and recognized as a

problem was child abuse, or the battered child syndrome. The first widely

disseminated article defined the battered child syndrome as a clinical condition

(with diagnosable physical and medical symptoms) having to do with those who have been deliberately injured by physical assault by a parent or caretaker

(Kempe et al 1962). The term "battered child syndrome" quickly gave way to

the terms child abuse, child abuse and neglect, and child maltreatment. Abuse

was not only physical assault, but also malnutrition, failure to thrive, sexual

abuse, educational neglect, medical neglect, and mental abuse. The official

federal definition of child abuse, stated in the Federal Child Abuse Prevention and Treatment Act of 1974 (PL 93-237) was:

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 351

. .. the physical or mental injury, sexual abuse, negligent treatment, or maltreatment of a

child under the age of eighteen by a person who is responsible for the child's welfare under

circumstances which would indicate that the child's health or welfare is harmed or threatened

thereby."

This definition is consequential because it became the model for state

definitions, which are in turn the basis for state laws that require reporting of

suspected cases of child abuse and neglect. Also, the federally funded national

incidence survey of officially reported child abuse and neglect (Burgdorf 1980)

employed this nominal definition of child abuse. Thus, most official report data

on child abuse are influenced, to one degree or another, by the federal defini-

tion.

It would be a mistake, however, to assume that since there is a federal

definition of abuse, there is uniformity in how child abuse is nominally defined

by researchers. In point of fact, most studies of child abuse and violence

towards children cannot be compared to one another because of the wide

variation of nominal definitions employed by scholars. While some investiga-

tors study only violence towards children (see for example Gelles 1978), others

examine the full range of acts of commission and omission (see for example

Newberger et al 1977). Thus, reports of incidence, correlations, cause, and

effect vary from study to study for many reasons, one being that researchers

infrequently define child abuse the same way.

The new interest in cross-cultural research on child abuse further illustrates

the definitional problem (Gelles and Cornell 1983, Korbin 1981). Korbin

(1981) points out that since there is no universal standard for optimal child

rearing, there is no universal standard for what is child abuse and neglect. Thus,

those who seek to develop cross-cultural definitions of abuse face the dilemma of choosing between a culturally relative standard in which any behavior can be

abusive or nonabusive depending on the cultural context, or an idiosyncratic

standard whereby abusive acts are those behaviors at variance with the normal

cultural standards for raising children. Korbin and others have tended towards

using the latter standard as they develop and carry out cross-cultural research

(Korbin 1984).

To a lesser extent, the same definitional problems that have plagued the

study of violence towards children have been part of the development of research on violence towards women. Initial definitions of wife abuse focused

on acts of damaging physical violence directed toward women by their spouses or partners (Gelles 1974, Martin 1976). As wife abuse became recognized as a

social problem, the definition was sometimes broadened to include sexual abuse, marital rape, and even pornography (London 1978).

The major definitional controversy surrounding wife abuse emerged as a

result of published research on violence towards husbands (Steinmetz 1978a)

and the use of the terms domestic violence, family violence, and spouse abuse when referring to violence between adult partners or married couples.

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

352 GELLES

Sociologists and psychologists, including Dobash & Dobash (1979), Pagelow

(1979), and Wardell, Gillespie & Leffler (1983) have argued that to discuss

violence and abuse of adults is to miss the point that the preferential victims of

violence in the family are women. These authors argue that the true problem is

wife assault, wife abuse, or violence towards wives, and that evenhanded

attempts to discuss spousal violence misdirect scientific and public attention

and are an example of misogyny. This controversy will be reviewed in more

detail in the discussion of incidence and prevalence of violence.

Defining Violence

Violence has also proven to be a concept that is not easy to define. First,

violence has frequently been used interchangeably with the term "aggression."

While violence typically refers to a physical act, aggression frequently refers to

any malevolent act that is intended to hurt another person. The hurt may not be

only physical but may be emotional injury or material deprivation. Second,

because of the negative connotation of the term violence, some investigators

have tried to differentiate between hurtful violence and more legitimate acts.

Thus, Goode (1971) tried to distinguish between legitimate acts of force and

illegitimate acts of violence. Spanking a child who runs into a street might be

considered force, while beating the same child would be violence. Research on

family violence has demonstrated the difficulty in distinguishing between

legitimate and illegitimate acts, since offenders, victims, bystanders, and

agents of social control often accept and tolerate many acts that would be

considered illegitimate if committed by strangers (Gelles 1974, Steinmetz

1977, Straus et al 1980).

Additional theoretical and ideological concerns influence the ways in which

the concept of violence is defined. Violence is frequently a political concept

used to attract attention to undesirable behaviors or situations. Thus, some

members of the political left will define various federal programs, such as Aid

to Families with Dependent Children, as violent. Members of the political right will likewise claim that abortion is a violent act. The entire capitalist system has

been derided as violent.

One frequently used nominal definition of violence, proposed by Gelles & Straus (1979), defines violence as "an act carried out with the intention, or perceived intention of physically hurting another person." This definition includes spankings and shovings as well as other forms of behavior; injury and/or death are also included under this broad definition.

By and large, the remainder of this review and the majority of research on family violence focus on the most severe and abusive forms of violence. Injurious violence or violence that has the high potential for causing an injury

has captured the attention of scholars who measure the incidence of family violence, identify factors associated with violent behavior, and develop

theories to explain family violence.

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 353

METHODOLOGICAL APPROACHES: SOURCES OF DATA, OPERATIONAL DEFINITIONS, AND THE SOCIAL CONSTRUCTION OF FAMILY VIOLENCE

Sources of Data

Data on child abuse, wife abuse, and family violence have come from three

sources. Each source has certain advantages and specific weaknesses that

influence both the nature and generalizability of the findings derived from the

research.

CLINICAL SAMPLES The most frequent source of data on family violence are

clinical studies carried out by psychiatrists, psychologists, and counselors. The

clinical setting (including battered wife shelters) provides access to extensive,

in-depth information about particular cases of violence. The pioneer studies of

child abuse and wife abuse were based almost exclusively on such clinical

samples (Steele & Pollock 1974, Snell, Rosenwald & Robey 1964, Schultz

1960, Kempe et al 1962, Galdston 1965, Walker 1979). Such studies, while

important for breaking new ground and rich in qualitative data, cannot be used

to generalize information on the frequency of factors associated with violence or the representativeness of the findings or conclusions.

As stated earlier, studies of wife abuse and violence towards women have

relied heavily on samples of women who seek help at battered wife shelters (Dobash & Dobash 1979, Walker 1979, Giles-Sims 1983, Pagelow 1981).

Such samples are extremely important because they are the best and sometimes

the only way of obtaining detailed data on the most severely battered women.

Such data are also necessary to study the impact of intervention programs.

However, these data are not generalizable to all women who experience

violence; and the study designs (see for example Walker 1979) frequently fail

to employ comparison groups.

OFFICIAL STATISTICS The establishment of mandatory reporting laws for

suspected cases of child abuse and neglect made case-level and aggregate-level data on abuse available to researchers. The American Humane Association

collects data from each state on officially reported child abuse and neglect

(American Humane Association 1982, 1983), and the Federal government

sponsored its own national survey of officially reported child maltreatment

(Burgdorf, 1980). Official report data provide information on an extremely

large number of cases. But these cases are limited only to those known by

service providers. Incidence rates from these data are likely to be lower than the

true rates, and the data are biased in a number of ways (Finkelhor & Hotaling 1984). As with many other types of official records on deviant behavior, the

poor are typically overrepresented in official reports, as are ethnic and racial minorities (Gelles 1975, Newberger et al 1977).

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

354 GELLES

RANDOM SAMPLE SURVEYS The low base rate of most forms of abuse poses

a problem for those who desire to apply standard survey research methods to

studying family violence. The low base rate requires investigators to use

purposive or nonrepresentative sampling techniques to identify cases (such as drawing samples from social agencies or police reports) or to draw large

representative samples. The high costs associated with large samples may, in

turn, require that interviews be reduced in length. Some scholars are wary of

applying survey research to studying abuse and violence because they assume that subjects will not provide reliable or valid information to interviewers

(Pelton 1979). However, a number of random sample survey studies (Harris

1979, Straus et al 1980, US Department of Justice 1980, 1984) have been

conducted. While these studies derive data which are generalizable, the amount of information elicited is frequently limited. Moreover, operationalizing vio- lence and abuse in survey instruments frequently results in compromises.

Straus and his colleagues (1980), for example, introduced their violence

measure, the Conflict Tactics Scale, by setting the stage in the context of family

conflicts. Critics of this form of measurement note that such a procedure limits discussions of violence to conflict situations (Pagelow 1981, Dobash & Dobash

1979). Moreover, the Conflict Tactics Scales did not measure the context, consequences, or outcomes of violent acts or whether an injury resulted from the violence.

Operational Definitions

While there is considerable variation in nominal definitions of violence and

abuse, the reliance on official statistics and clinical samples has produced surprising similarity in the way that researchers in all disciplines operationally define family violence and especially child abuse. Child abuse, and other forms of family violence (especially abuse of the elderly) are typically defined operationally as those instances in which the victim becomes publically known and labeled by an official agency.

Researchers have criticized this uniformity of operationalization and claimed

that it is biased by the process by which cases come to public attention and are labeled (Gelles 1975). Support for such criticism comes from Turbett &

O'Toole's (1980) experiment which found that physicians are more likely to label minority children and children from lower-class families abused. Survey research conducted by Giovannoni & Becerra (1979) also found that attitudes

and definitions of child abuse vary by professional group and social status.

The Social Construction of Family Violence

The tendency to operationalize child abuse, wife abuse, and family violence as those cases that come to professional and official attention, or those individuals who choose to seek professional help or flee to a shelter, results in confounding

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 355

the factors leading people to come forward or be publically labeled with the

factors causing wife abuse, child abuse and family violence. Because research

results are used by clinicians, police, and other labelers to identify suspected

cases of family violence, the confounding thus becomes built into the system.

Research is based on official reports or cases that come to public attention.

Clinicians, agents of social control, and other "gatekeepers" read the literature

and use it to diagnose injuries and situations as either instances of abuse or

"accidents." Thus, the definition of the situation of violence and abuse is

caused and becomes causal. An emergency room physician may treat an

injured woman who is reluctant to discuss the nature of her injury. The

physician may draw on the literature on wife abuse to see if the woman fits the

"profile". If she does, she becomes a case. Later, a researcher may choose to

draw cases from the population of the emergency room, and include (or

exclude) the woman based on the diagnosis.

The social construction of family violence is consequential for estimating

extent, patterns, and causes, and also for studying the impact of intervention

and preventions. Researchers may study the impact of interventions and pre-

vention programs on officially known cases and then generalize to the wider

population of abuse and violence victims. Obviously, since the wider popula-

tion is quite distinct from those cases that are publically identified, the interven-

tions and prevention outcomes may not be generalizable to the undetected or hidden instances of domestic abuse.

THE EXTENT AND NATURE OF FAMILY VIOLENCE

Considerable effort has been expended on discussion, debate, and research

concerning the extent of family violence. Part of this effort is aimed at

exploding the myth that violence in the home is rare. Another goal has been to convince policymakers, opinion leaders, and the public that child abuse, wife

abuse, and other forms of domestic violence are extensive enough to be

considered legitimate social problems-especially since one part of the defini-

tion of a social problem is that a behavior is found harmful to a significant number of people (Merton & Nisbet 1976). Lastly, social scientists need data

on incidence when they plan social survey research. Numerous methods have been used to assess the extent of the various forms

of family violence. First, scientists employ educated guesses. The majority of estimates that were made about the extent of child abuse were little more than

guesstimates (see for example Fontana 1973, US Senate 1973, Pediatric News 1975). Others who have tried to base estimates on actual data have used survey data on physical punishment (Erlanger 1974, Stark & McEvoy 1970), homicide

data (Curtis 1974, Steinmetz & Straus 1974), assault (Pittman & Handy 1964, Boudouris 1971), and applicants for divorce (Levinger 1966, O'Brien 1971),

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

356 GELLES

or reports by subjects who know of a case of abuse in their community (Block &

Sinnott 1979, Gil 1970).

Three national surveys have been conducted with the purpose of assessing

the national incidence of domestic violence. Straus, Gelles & Steinmetz (1980)

surveyed a representative sample of 2143 couples. Data were collected on

violence towards children, violence between spouses, violence between sib-

lings, and violence towards parents. Data on the extent of domestic violence

have also been abstracted from the National Crime Survey (US Justice Depart-

ment 1980, 1984). These estimates are based on a sample of subjects 12 years

of age and older who report crimes against persons or households, whether

reported or unreported to the police. The sample includes some 60,000 housing

units, and data are available from 1973 to 1981. Lastly, the National Center on

Child Abuse and Neglect conducted a national incidence survey of child

maltreatment (Burgdorf 1980), and these data include acts of physical maltreat-

ment of children.

ESTIMATES OF INCIDENCE AND PREVALENCE

Estimates of incidence range from thousands to millions. Some authors claim

that family violence is an epidemic and that perhaps half of all wives are abused.

The survey by Straus and his colleagues found that the yearly incidence of

acts that they defined as abusive violence (all acts that had the high probability of causing an injury-whether or not an injury occurred) was about 3.8 to 4.0 instances per 100 individuals (Straus et al 1980). About 3.8% of children 3-17

years of age were victims of severe or abusive violence. The same percentage of wives were victims of abusive spousal violence. The percentage of husbands victimized was found to be higher than that of wives (Straus et al 1980,

Steinmetz 1978a), and this stirred up a storm of professional and public controversy. The data on husband abuse were attacked as invalid because the

survey did not measure the consequences of the violence (women are thought to

be the most likely to be injured) and because the survey did not assess the context of the violence (women are thought to use violence to protect them-

selves from violence initiated by their husbands; Pleck et al 1978, Jones 1980).' One of the main points of contention arising out of the incidence data on violence towards husbands is that the national incidence study indicating large

numbers of male victims did not square with findings from clinical research that

found few if any battered men. Obviously, the clinical studies employed differing definitions and sampling techniques than the national survey. Study- ing only women who seek aid from battered wife shelters produces different

'In addition to the methodological criticisms was the concern by those who feared that the issue

of husband abuse would take both attention and resources away from the problem of wife abuse.

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 357

samples and different results than national surveys. Few men are actually

interviewed in clinical studies, and data about male victimization is based on

their wives' or partners' reports.

The national incidence survey (Straus et al 1980) found that in one fourth of

the homes where there was couple violence, men were victims but not offend-

ers. In one fourth of the homes, women were victims but not offenders, and in

one half of the violent homes, both men and women were violent-although the

survey could not detect whether the wives' violence was retaliatory or in

self-defense.

Despite the clamor about husband abuse data, nominal and operational

definitions of violence, and plausible interpretations of the data, it does seem

clear that while women are the most likely victims of spousal violence, there are indeed men who are victimized. Little additional research has been con- ducted on violence towards husbands.

Sibling violence and abuse were found to be the most common forms of

domestic violence in Straus and his colleagues' survey (1980). Finally, the

investigators reported that child-to-parent violence was comparable to other

forms of family violence-3.5% of teenagers (1 5-17 years of age) used severe

violence towards a parent during the survey year.

Data from the National Crime Survey (US Department of Justice 1984)

indicate that the yearly incidence rate of domestic violence (among those 12

years of age and older) was 1.5 per 1000 people in the population.

The incidence survey by the National Center on Child Abuse and Neglect

yielded a figure of 10.5 children per 1000 who are known by agencies to be

maltreated (Burgdorf 1980). More than 200,000 children were victims of

physical assault.

There are few data on other forms of family violence. There have been some

preliminary attempts to assess the extent of the abuse of the elderly (see for

example Block & Sinnott 1979; and Cornell & Gelles 1982b for a review of

major studies). Additionally, a number of states have instituted mandatory

reporting laws for elder abuse; thus, some official report data on elder abuse are available.

One unintended outcome of the various attempts to measure the incidence of

family violence is that the effort has been partially counter-productive. While

the use of incidence estimates is partly designed to promote and maintain interest in the problem of family violence, the wide range of estimates has frequently led opinion and policymakers to conclude that the real magnitude of family violence remains unknown (Nelson 1984).

Factors Associated With Family Violence

Other reviews have carefully analyzed the factors associated with various forms

of family violence (for child abuse, see Maden & Wrench 1977, Parke &

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

358 GELLES

Collmer 1975; for spouse abuse, see Byrd 1979. For family violence in general

see Gelles 1980; Gelles 1982; Steinmetz 1978b).

The last two decades of research on the various aspects of family violence are

in agreement on one major point-there are a multitude of factors associated

with violence in the home. Nevertheless, the unusual, emotional, and bizarre

nature of some instances of family violence has led some researchers, popular

writers, and the public alike to grasp at single factor explanations. Mental

illness, character disorders, psychopathology, alcohol and drugs, stress, pover-

ty, growing up in a violent home, and diet have all been pointed to as the cause

of domestic strife. Social scientists are obviously aware of the drawbacks of

single factor explanations, but they too sometimes select explanations that are

single factor in nature. Dobash & Dobash argue that patriarchy and male

domination are the primary causes of violence towards wives (1979).

Among the factors that researchers have consistently found related to various

aspects of domestic violence are:

1. The cycle of violence-the intergenerational transmission of violence.

2. Low socioeconomic status.

3. Social and structural stress.

4. Social isolation and low community embeddedness.

5. Low self-concept.

6. Personality problems and psychopathology.

It is important to point out that the research that supports the claim for these

associations is far from definitive and is often based on empirical research

suffering from methodological problems. These include the fact that most

studies are based on "caught cases," that is, only cases that come to official

attention; studies frequently either do not include comparison groups or have comparison groups that are not equivalent; and studies frequently have small

samples and lack generalizability (Gelles 1982).

In addition to the methodological limitations, there are other problems. Associations found tend to be quite modest (perhaps as a result of the low base

rate of family violence and the multidimensional nature of the behavior).

Frequently, the publicly perceived strength of an association is based on how

often the finding is cited, not how strong the statistical association is or how

well the research meets the standards of scientific evidence. Houghton (1979)

calls this the "Woozle Effect" (based on the Winnie the Pooh story-Milne 1926). In the Woozle Effect, findings are initially stated with a qualification,

then repeated in the literature without the qualification. Reviews of the litera-

ture cite other reviews, and the strength of a finding grows without the original

qualifications or evidence being confirmed through replication. Fortunately, the majority of the factors cited in this section have been found to be associated

with family violence by investigators who have used various methods of

research and have drawn their data from all three sources of data.

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 359

CAUSAL MODELS

As noted previously, the field of family violence abounds with simplistic

theoretical models. In the earliest research reports the model advanced was

psychopathology-mental illness caused people to abuse their children, wives,

and parents (Steele & Pollock 1974). Other intraindividual models proposed

that family violence is caused by alcohol and drugs.

The major theoretical approaches to family violence have been reviewed

extensively elsewhere (Gelles 1980, Gelles 1982, Gelles 1983, Gelles & Straus

1979). Of interest here is the fact that students of family violence have chosen

to view family violence as a special case of violence that requires its own body

of theory to explain it (Gelles & Straus 1979). Thus, existing theories of

violence and aggression, such as frustration-aggression theory (Berkowitz

1962, Dollard et al 1939, Miller 1941), self-attitude theory (Kaplan 1962),

functional theory (Coser 1967), and culture of violence theory (Wolfgang &

Ferracuti 1967) have typically not been directly applied to the study of violence

in the home. The only existing theoretical model from the general study of

violence to be frequently applied to violence in the home has been social

learning theory (Bandura et al 1961, Singer 1971).

Students of family violence have justified the development of a special body

of theory to explain family violence on the basis of the high incidence of

violence in the home, because of the unique nature of the family as a small

group, and because of the distinctive nature of the family as a social institution

(Gelles & Straus 1979).

While special theories of family have been developed, the development has

gone in three different directions. Theories have been developed to explain the

abuse of children; there are theories that attempt to explain spouse abuse; and there are those theories that are designed to explain family violence in general.

Theories of Family Violence

Among the special approaches developed to explain family violence are:

resource theory (Goode 1971), general systems theory (Straus 1973), an

ecological model (Garbarino 1977), an exchange model (Gelles 1983), a

patriarchy explanation (Dobash & Dobash 1979), a sociobiological perspectiv-

e; an economic model; and a sociocultural explanation.

RESOURCE THEORY A resource theory, the first theoretical approach applied

to family violence, proposes that all social systems rest to some degree on force or its threat (Goode 1971). The use of violence depends on the resources a

participant in a system or family member can command. The more resources,

the more force can be used, but the less it actually is employed. Those with the

fewest resources tend to employ force and violence the most. For instance,

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

360 GELLES

husbands tend to resort to violence when they lack the traditional resources

associated with the culturally assumed dominant role of the male in the family

(Gelles 1974, O'Brien 1971).

GENERAL SYSTEMS THEORY Straus (1973) and Giles-Sims (1983) use a

social system approach to explain family violence. Here, violence is viewed as

a system product rather than the result of individual pathology. System opera-

tions can maintain, escalate, or reduce levels of violence.

AN ECOLOGICAL PERSPECTIVE Garbarino has proposed an ecological model

of child maltreatment. The model rests on three levels of analysis: the rela-

tionship between organism and environment; the interacting and overlapping

systems in which human development occurs; and environmental quality (Gar-

barino 1977). Garbarino proposes that maltreatment arises out of a mismatch of

parent, child, and family to neighborhood and community.

EXCHANGE THEORY Exchange theory (Gelles 1983) proposes that family

violence is governed by the principles of costs and rewards. Violence is used

when rewards are higher than costs. The private nature of the family, the

reluctance of social institutions and agencies to intervene in violence, and the

low risk of other interventions reduce the costs of violence. The cultural

approval of violence as both expressive and instrumental behavior raises the

potential rewards for violence.

PATRIARCHY Dobash & Dobash (1979) see abuse of women as a unique

phenomenon that is caused by the social and economic processes that directly

and indirectly support a patriarchal social order and family structure. Patriarchy

leads to the domination of women by men and explains the historical pattern of

systematic violence directed at women.

SOCIOBIOLOGY One of the newest theoretical models, a sociobiological

perspective (or evolutionary perspective) suggests that violence towards human or nonhuman primate offspring is the result of the reproductive success poten- tial of children and parental investment. The theory proposes that parents will not invest in children with low reproductive potential. Thus, children not

genetically related to the parent (e.g. stepchildren, adopted or foster children), or children with low reproductive potential (e.g. handicapped or retarded) are

at the highest risk for abuse and infanticide (Hrdy 1979, Burgess & Garbarino 1983). It logically follows the risk is greatest that adoptive parents, foster

parents, and caretakers not genetically related to the child (e.g. a boyfriend) will become abusers. Risk of abuse is high where there is lack of bonding

between parent and child and where paternity is highly uncertain (Burgess

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 361

1979). Large families can dilute parental energy and lower attachment to

children (Burgess 1979).

AN ECONOMIC MODEL The economic or social-structural model explains

that violence and abuse arise out of socially structured stress. Stress, such as

low income, unemployment, and illness, is unevenly distributed in the social

structure. When violence is the accepted response or adaptation to stress, stress

can lead to violence and abuse (Coser 1967, Gelles 1974).

A SOCIOCULTURAL EXPLANATION Finally, students of family violence have

explained the occurrence of violence by drawing on sociocultural attitudes and

norms concerning violent behavior. Societies, cultures, and subcultures that

approve of the use of violence are thought to have the highest rates of domestic

violence (Straus et al 1980).

SUMMARY The recency of family violence as an area of study for sociologists

is best exemplified by the limited level of theoretical development of the field.

The field has not developed to the point where theories have actually been

subjected to rigorous empirical testing. Individual theories have been tested in

limited ways (typically with small, nonrepresentative data sets), and there has

yet to be a critical test that pits one theory against another to see which best fits

the data. Various criticisms have been leveled against each of the theories. A

major limitation of ecological theory is the argument that the theory commits

the ecological fallacy by attributing aggregate level demographics to individual

level behavior. The patriarchy explanation suffers from being a "single factor"

explanation. Moreover, the variability of the independent variable (patriarchy)

has not been adequately specified by the theorists. In its present form, a

patriarchy theory is not amenable to an empirical test. This is also the case with

sociobiological explanations. Additionally, while sociobiological explanations

seem empirically to fit the data on infanticide, the theory does not logically

seem to explain nonlethal instances of child maltreatment; nonlethal violence

and neglect do not remove the child or increase the inclusive fitness of the other

children. Abuse and neglect may actually require a greater parental investment

of time and energy in the victimized child. The available data do fit the

propositions of resource, exchange, economic, and sociocultural explanations.

The next decade of research is likely to be devoted to testing theories and to

theoretical integration of the supported theories and propositions.

OTHER RESEARCH ISSUES

The most recent research on domestic violence has moved beyond the study of

incidence and correlations and the proposing of causal models. Researchers

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

362 GELLES

have begun to test theories, to study the consequences of family violence, and

to examine the impact of interventions on violence in the family. While a full

examination of all the developing research issues and questions is beyond the

scope of this paper, this review highlights some of the work that has addressed

key questions in the study of domestic violence.

The Responses of Victims of Wife Abuse

One question that is frequently raised concerning wife abuse is why the victims

choose to stay or to leave battering husbands. Common sense would argue that

a victim of a beating would flee to protect herself. Yet common sense overlooks

both the structural and interpersonal nature of family relations. One reason the

issue of staying or leaving has captured and continues to hold public and

scholarly attention is that some of the wives who have killed their husbands

have claimed self-defense and have drawn upon scholarly research on the

"battered wife syndrome" to justify their choice to kill rather than flee their

husbands. Walker's theory of "learned helplessness" (1979) is the frequent

explanation offered by those defending battered women. Walker claims that

battered women develop an inability to protect themselves from future assaults.

The beatings leave them with a diminished capacity to control events that go on

around them (Walker 1979).

Gelles (1976) found three factors influenced whether a woman stays or leaves a violent relationship: frequency and severity of violence is directly

related to leaving; limited education and poor occupational skills are indirectly

related to leaving; and women who experience violence in their childhoods are

more likely to remain in a violent relationship.

Pagelow's research (1981) did not support the hypothesis concerning fre-

quency and severity of violence or the hypothesis relating experience of violence to staying or leaving. Pagelow (1981) did find that diminished educa-

tional and occupational resources and skills inhibit women from leaving violent

relationships. Strube & Barbour's (1983) study of 98 women also found that

economically dependent women are more likely to remain with their abusive husbands. They also found evidence that commitment to the marital rela-

tionship was an important and independent factor related to the decision to

leave a violent relationship.

The results of research on women's decisions to stay or leave violent

relationships must be interpreted with caution, since these studies rely exclu-

sively on small, nonrepresentative samples. Differing results from study to study may be partially due to differing sampling procedures.

Bowker analyzed the situation of battered women from a different perspec- tive (1983). Bowker interviewed 136 women who reported that they had been

beaten and that their husbands had stopped being violent. Bowker attempted to identify what the battered women had done to get their husbands to cease the

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 363

violent behavior. Passive defense (defending oneself by covering the body with

one's hands, arms, and feet) was the most common personal strategy used by women; calling friends was the most commonly used informal strategy;

and social services were the most widely used formal source of help. Bow-

ker, however, found no single technique worked best. What mattered was

that women showed their determination that the violence must stop (Bowker 1983).

The Consequences of Child Maltreatment

Students of child abuse and child maltreatment have begun to turn their attention to the question of the consequences of being an abused child. The

clinical literature strongly suggests that abused children have higher rates of

developmental delays and difficulties and as adults they have higher rates of

drug abuse, alcohol abuse, criminal behavior, and psychiatric disturbances

(Smith, Hansen & Nobel 1973, Galdston 1975, Martin 1972).

Survey data also suggest that abused children have higher rates of juvenile delinquency (Alfaro 1977, Carr 1977).

Research on the consequences of abuse and violence typically suffers from

numerous methodological flaws (differing definitions, small clinical samples, no comparison groups, and retrospective data Garbarino & Plantz 1984). Preliminary data from a prospective study do point to deficits suffered by

abused children (Egeland & Jacobvitz 1984).

Preventing and Treating Family Violence

The study of family violence has always had an implicit applied mandate. The

search for knowledge has been carried out with the assumption that such knowledge could be useful in designing and implementing prevention and treatment programs. As yet, there are few empirical studies that assess the impact of existing prevention and treatment efforts in the fields of child abuse

and domestic violence. An evaluation of child abuse treatment programs was carried out in the late 1970s. The study found that while one-on-one therapy conducted by professional therapists was the most costly treatment, it was the least effective (in terms of recidivism data collected over a very limited time

period). The most effective intervention was group counseling provided by lay therapists. This program of intervention was also the least expensive interven- tion (Berkeley Planning Associates 1978).

Sherman & Berk (1984) conducted a natural experiment to assess the impact

of police interventions in cases of wife abuse. Of three randomly applied interventions arrest, removing the husband from the home, and trying to cool

down the situation arrest resulted in fewer calls for help and fewer instances of repeated violent behavior.

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

364 GELLES

CONCLUSION: FUTURE RESEARCH

The study of family violence is still relatively new. The psychologist Edward

Zigler once claimed that the current knowledge base on child abuse is about

equal to what we knew about mental illness in the late 1940s (1976). Zigler

could be either too pessimistic about family violence or too optimistic in his

assessment of our understanding of mental illness. Nevertheless, it is clear that

those who study domestic violence are still struggling with definitional issues,

methodological constraints and problems, and have yet to actually test the

major theoretical assumptions and models that have been developed.

The study of family violence continues to be an interdisciplinary effort.

Psychologists, sociologists, anthropologists, physicians, and social workers

frequently share interests, questions, models, and authorship of research re-

ports. In the future, we should expect to see continued efforts to study family

violence cross-culturally. Korbin (1981) has already contributed to that effort

and her work, and the work of others, is useful in challenging and informing

some of the sociocultural assumptions that developed out of research on

violence in the United States (Straus et al 1980). Research on family violence is

examining violence in alternative families (e.g. stepfamilies). Research con-

tinues to be conducted by those interested in how professionals label and

approach cases of domestic violence (e.g. Hampton & Newberger 1984). There

is considerable interest in the topic of sexual abuse, and marital rape, but space

limitations and definitional issues precluded reviewing that literature in this

article. Researchers have recently begun to try to collect data directly from men

who batter, rather than relying only on reports from the victims (e.g. Rouse

1984). There has also been an expansion of the study of violence between

partners by those interested in courtship violence (e.g. Makepiece 1981, 1983; Cate et al 1982). Sociologists are contributing as well to study of the conse-

quences of child maltreatment and are using survey data to complement clinical data on this topic.

The literature on family violence has expanded at a rapid pace over the last

few years. Sociologists can be expected to continue to be part of this interdisci-

plinary effort, as theoreticians, methodologists, and as debunkers of conven-

tional wisdom.

Literature Cited

Alfaro, J. 1977. Report on the relationship between child abuse and neglect and later socially deviant behavior. Presented at Ex- ploring the Relationship Between Child Abuse and Delinquency Symposium, Seat- tle, Univ. Wash.

American Humane Association. 1982. Nation- al Analysis of Official Child Neglect and Abuse Reporting. Englewood, Colo: Am. Humane Assoc.

American Humane Association 1983. High-

lights of Official Child Neglect and Abuse Reporting. Denver, Colo: Am. Humane Assoc.

Bakan, D. 1971. Slaughter of the Innocents. A Study of the Battered Child Phenomenon. Boston, Mass: Beacon

Bandura, A., Ross, D., Ross, S. 1961. Trans- mission of aggression through imitation of aggressive models. J. Abnorm. Soc. Psy- chol. 63:575-582

Berkeley Planning Associates. 1978. Executive

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 365

summary: Evaluation of the joint OCDISRS national demonstration program in child abuse and neglect. Mimeographed

Berkowitz, L. 1962. Aggression: A Social Psychological Analysis. New York: McGraw-Hill

Block, M., Sinnott, J. 1979. The Battered Elder Syndrome: An Exploratory Study. Un- published manuscript, Univ. Md.

Boudouris, J. 1971. Homicide and the family. J. Marriage Fam. 33:667-682

Bowker, L. H. 1983. Beating Wife-Beating. Lexington, Mass.: Lexington

Burgdorf, K. 1980. Recognition andReporting of Child Maltreatment. Rockville, Md: Westat

Burgess, R. L. 1979. Family violence: Some implications from evolutionary biology. Pre- sented at Ann. Meet. Amer. Soc. Criminolo- gy, Philadelphia, Penn.

Burgess, R. L., Garbarino, J. 1983. Doing what comes naturally? An evolutionary per- spective on child abuse. In The Dark Side of the Family: Current Family Violence Re- search, ed. D. Finkelhor, R. Gelles, M. Straus, G. Hotaling, pp. 88-101. Beverly Hills, Calif: Sage

Byrd, D. E. 1979. Intersexual assault: A re- view of empiricalfindings. Ann. Meet. East- ern Sociological Society, New York.

Caffey, J. 1946. Multiple fractures in the long bones of infants suffering from chron- ic subdural hematoma. Am. J. Roentge- nology, Radium Therapy, Nucl. Med. 58: 163-173

Caffey, J. 1957. Some traumatic lesions in growing bones other than fractures and dis- locations. Br. J. Radiol. 23:225-238

Carr, A. 1977. Some preliminary findings on the association between child maltreatment and juvenile misconduct in eight New York counties. Report to Admin. for Children, Youth and Families, National Center on Child Abuse and Neglect. Kingston, RI

Cate, R. M., Henton, J. M., Christopher, F. S., Lloyd, S. 1982. Premarital abuse: A so- cial psychological perspective. J. Fam. Iss. 3:79-90

Cornell, C. P., Gelles, R. J. 1982a. Adolescent to parent violence. Urban Soc. Change Rev. 15:8-14

Cornell, C. P., Gelles, R. J. 1982b. Elder abuse: The status of current knowledge. Fam. Rel. 31:457-465

Coser, L. A. 1967. Continuities in the Study of Social Conflict. New York: Free Press

Curtis, L. 1974. Criminal Violence: National Patterns and Behavior. Lexington, Mass: Lexington

De Mause, L. 1974. The History of Childhood. New York: Psychohistory Press

De Mause, L. 1975. Our forebearers made childhood a nightmare. Psychol. Today 8:85-87

Dobash, R. E., Dobash, R. 1979. Violence Against Wives. New York: Free

Dollard, J. C., Doob, L., Miller, N., Mowrer, O., Sears, R. 1939. Frustration andAggres- sion. New Haven, Conn: Yale Univ. Press

Egeland, B., Jacobvitz, L. 1984. Intergenera- tional continuity of parental abuse: Cause and consequence. Presented at Social Sci- ence Research Council Con. on Biosocial Perspectives on Child Abuse and Neglect, York, Me.

Erlanger, H. B. 1974. Social class and corporal punishment in childrearing: a reassessment. Am. Sociol. Rev. 39:68-85

Finkelhor, D., Hotaling, G. 1984. Sexual abuse in the national incidence study of child abuse and neglect: An appraisal. ChildAbuse Neglect: Int. J. 8:23-33

Fontana, V. 1973. Somewhere a Child Is Crying: Maltreatment-Causes and Preven- tion. New York: MacMillan

Galdston, R. 1965. Observations of children who have been physically abused by their parents. Am. J. Psychiatry 122:440-443

Galdston, R. 1975. Preventing abuse of little children: the parent's center project for the study and prevention of child abuse. Am. J. Orthopsychiatry 45:372-381

Garbarino, J. 1977. The human ecology of child maltreatment. J. Marriage Fam. 39:721-735

Garbarino, J., Plantz, M. C. 1984. An ecologi- cal perspective on the outcomes of child mal- treatment: What difference will the differ- ences make? Presented at Social Science Re- search Council Conference on Biosocial Perspectives on Child Abuse and Neglect. York, Me.

Gelles, R. J. 1973. Child abuse as psycho- pathology: A sociological critique and refor- mulation. Am. J. Orthopsychiatry. 43:611- 621

Gelles, R. J. 1974. The Violent Home: A Study of Physical Aggression Between Husbands and Wives. Beverly Hills, Calif: Sage

Gelles, R. J. 1975. The social construction of child abuse. Am. J. Orthopsychiatry 45:363- 371

Gelles, R. J. 1976. Abused wives: Why do they stay? J. Marriage Fam. 38:659-668

Gelles, R. J. 1978. Violence towards children in the United States. Am. J. Orthopsychiatry 48:580-592

Gelles, R. J. 1980. Violence in the family: A review of research in the seventies. J. Mar- riage Fam. 42:873-885

Gelles, R. J. 1982. Applying research on fami- ly violence to clinical practice. J. Marriage Fam. 44:9-20

Gelles, R. J. 1983. An exchange/social control theory. In The Dark Side of the Family: Cur- rent Family Violence Research. ed. D. Fink- elhor, R. Gelles, M. Straus, G. Hotaling, pp. 151-165. Beverly Hills, Calif: Sage

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

366 GELLES

Gelles, R. J., Cornell, C. P. 1983. Internation- al Perspectives on Family Violence. Lexing- ton, Mass: Lexington

Gelles, R. J., Straus, M. A. 1979. Determi- nants of violence in the family: Toward a theoretical integration. In Contemporary Theories About the Family (Vol. 1). ed. W. R. Burr, R. Hill, F. I. Nye, I. L. Reiss, pp. 549-581. New York: Free

Gil, D. 1970. Violence Against Children: Physical Child Abuse in the United States. Cambridge, Mass: Harvard Univ. Press

Giles-Sims, J. 1983. Wife-Beating: A Systems Theory Approach. New York: Guilford

Gillen, J. 1946. The Wisconsin Prisoner: Stud- ies in Crimogenesis. Madison, Wis: Univ. Wis. Press

Giovannoni, J. M., Becerra, R. M. 1979. De- fining Child Abuse. New York: Free

Goode, W. 1971. Force and violence in the family. J. Marriage Fam. 33:624-636

Hampton, R., Newberger, E. 1984. Child abuse incidence and reporting by hospitals: Significance of severity, class and race. Pre- sented at Second National Conference for Family Violence Researchers. Durham, NH

Harris, L., and associates. 1979. A Survey of Spousal Abuse Against Women in Kentucky. New York: Harris & Assoc.

Houghton, B. 1979. Research of research on women abuse. Presented at Ann. Meet. Am. Society of Criminology, Philadelphia, Penn.

Hrdy, S. B. 1979. Infanticide among animals: A review classification, and examination of the implications for reproductive strategies of females. Ethol. Sociobiol. 1: 13-40

Jones, A. 1980. Women Who Kill. New York: Holt, Rinehart, & Winston

Kaplan, H. 1982. Toward a general theory of psychosocial deviance: The case of aggres- sive behavior. Soc. Sci. Med. 6:593-617

Kempe, C. H., Silverman, F. N., Steele, B. F., Droegemueller, W., Silver, H. K. 1962. The battered child syndrome. J. Am. Med. Assoc. 181:107-112

Korbin, J. 1981. Child Abuse and Neglect: Cross-Cultural Perspectives. Berkeley, Calif: Univ. Calif. Press

Korbin, J. 1984. Child maltreatment and the cultural context: Current knowledge and fu- ture directions. Presented to Social Science Research Council Conference on Child Abuse and Neglect. York, Me.

Levinger, G. 1966. Sources of marital dissatis- faction among applicants for divorce. Am. J. Orthopsychiatry 26:803-897

London, J. 1978. Images of violence against women. Victimology 2:510-524

Maden, Marc F., Wrench, D. F. 1977. Signifi- cant findings in child abuse research. Victi- mology. 2:196-224

Makepiece, J. M. 1981. Courtship violence among college students. Fam. Rel. 30:97- 102

Makepiece, J. M. 1983. Life events stress and courtship violence. Fam. Rel. 32:101-109

Martin, D. 1976. Battered Wives. San Francis- co, Calif: Glide

Martin, H. P. 1972. The child and his develop- ment. In Helping the Battered Child and His Family, ed. C. H. Kempe, R. E. Helfer, pp. 93-114. Philadelphia: Lippincott

Merton, R. K., Nisbet, R. 1976. Contempor- ary Social Problems. New York: Harcourt Brace Jovanovich

Miller, N. 1941. The frustration-aggression hypothesis. Psychol. Rev. 48:337-342

Milne, A. A. 1926. Winnie-the-Pooh. New York: Dell

Nelson, B. J. 1984. Making an Issue of Child Abuse: Political Agenda Setting for Social Problems. Chicago: Univ. Chicago Press

Newberger, E. et. al. 1977. Pediatric social illness: Toward an etiologic classification. Pediatrics 60:178-185

O'Brien, J. 1971. Violence in divorce prone families. J. Marriage Fam. 33:692-698

Pagelow, M. 1979. Research on woman batter- ing. In Stopping Wife Abuse, ed. J. B. Flem- ing, pp. 334-349. New York: Anchor

Pagelow, M. 1981. Woman-Battering: Victims and Their Experiences. Beverly Hills, Calif: Sage

Palmer, S. 1962. The Psychology of Murder. New York: Crowell

Parke, R. D., Collmer, C. W. 1975. Child abuse: An interdisciplinary analysis. In Re- view of Child Development Research, Vol. 5, ed. M. Hetherington, pp. 1-102. Chicago: Univ. Chicago Press

Pediatric News. 1975. One child dies daily from abuse: Parent probably was abuser. Pediatric News, vol. 9:3

Pelton, L. G. 1979. Interpreting family vio- lence data. Am. J. Orthopsychiatry 49: 194

Pittman, D., Handy, W. 1964. Patterns in criminal aggravated assault. J. CriminalLaw Criminol. 55:462-470

Pleck, E. J., Pleck, M., Grossman, M., Bart, P. 1978. The battered data syndrome: A comment on Steinmetz's article. Victimolo- gy 2:680-683

Pokorny, A. D. 1965. Human violence: A com- parison of homicide, aggravated assault, suicide, and attempted suicide. J. Crim. Law, Criminol. 56:488-497

Radbill, S. 1980. A history of child abuse and infanticide. The Battered Child, ed. R. Hel- fer, C. Kempe, pp. 3-20. Chicago: Univ. Chicago Press

Rouse, L. 1984. Conflict tactics used by men in marital disputes. Presented at Second National Conference for Family Violence Researchers, Durham, NH

Schultz, L. G., 1960. The wife assaulter. J. Soc. Therapy. 6:103-111

Sherman, L. W., Berk, R. A. 1984. Deterrent

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms

FAMILY VIOLENCE 367

effects of arrest for domestic violence. Am. Sociol. Rev. 49:261-272

Shorter, E. 1975. The Making of the Modern Family. New York: Basic

Silverman, F. 1953. The roentgen manifesta- tions of unrecognized skeletal trauma in in- fants. Am. J Roentgenology, Radium Ther- apy, Nucl. Med. 69:413-427

Singer, J. 1971. The Control ofAggression and Violence. New York: Academic

Smith, S., Hanson, R., Noble, S. 1973. Par- ents of battered babies: A controlled study. Br. Med. J 4:388-391

Snell, J. E., Rosenwald, R. J., Robey, A. 1964. The wifebeater's wife: A study of family interaction. Archives Gen. Psychiatry 11:107-113

Star, B. 1980. Patterns of family violence. Soc. Casework 61:339-346

Stark, R., McEvoy, J. 1970. Middle class vio- lence. Psychol. Today 4:52-65

Steele, B. F., Pollock, C. 1974. A psychiatric study of parents who abuse infants and small children. In The Battered Child (2nd edi- tion). ed. R. Helfer, C. Kempe, pp. 89-134. Chicago: Univ. Chicago Press

Steinmetz, S. K. 1971. Occupation and physi- cal punishment: A response to Straus. J. Marriage Fami. 33:664-666

Steinmetz, S. K. 1977. The Cycle of Violence: Assertive, Aggressive, and Abusive Family Interaction. New York: Praeger

Steinmetz, S. K. 1978a. The battered husband syndrome. Victimology 2:499-509

Steinmetz, S. K. 1978b. Violence between family members. Marriage Fam. Rev. 1:1- 16

Steinmetz, S. K. 1978c. Battered parents. Soci- ety 15:54-55

Steinmetz, S. K. 1984. Family violence to- wards elders. In Violence Individuals and Families: A Handbook for Practitioners, ed. S. Saunders, A. Anderson, C. Hart, G. Rubenstein, pp. 137-163. Springfield, Ill: Thomas.

Steinmetz, S. K., Straus, M. A. 1974. Violence in the Family. New York: Harper, Row.

Straus, M. A. 1971. Some social antecedents of physical punishment: A linkage theory inter- pretation. J. Marriage Fam. 33:658-663

Straus, M. A. 1973. A general systems theory approach to a theory of violence between family members. Soc. Sci. Inf. 12:105-125

Straus, M. A. 1974. Forward. In The Violent Home: A Study of Physical Aggression Be- tween Husbands and Wives. R. J. Gelles, pp. 13-17. Beverly Hills, Calif: Sage

Straus, M. A., Gelles, R. J., Steinmetz, S. K. 1980. Behind Closed Doors: Violence in the American Family. Garden City, NY: Anchor/Doubleday

Strube, M. J., Barbour, L. S. 1983. The deci- sion to leave an abusive relationship: Eco- nomic dependence and psychological com- mitment. J. Marriage Fam. 45:785-793

Taylor, L., Newberger, E. H., 1979. Child abuse in the International Year of the Child. N. Engl. J. Med. 301:1205-1212

Turbett, J. P., O'Toole, R. 1980. Physician's recognition of child abuse. Presented at Ann. Meet. of American Sociological Associa- tion, New York

US Department of Justice 1980. Intimate Vic- tims: A Study of ViolenceAmong Friends and Relatives. Washington, DC: USGPO

US Department of Justice 1984. Family Vio- lence. Washington, DC: Bureau of Justice Statistics

United States Senate 1973. Hearing before the Subcommittee on Children and Youth of the Committee on Labor and Public Welfare. United States Senate, 93rd Congress First Session, on S .1191 Child Abuse Prevention Act. Washington, DC: USGPO

Walker, L. 1979. The Battered Woman. New York: Harper Row.

Wardell, L., Gillespie, D. L., Leffler, A. 1983. Science and violence against wives. In The Dark Side of the Family: Current Family Violence Research. ed. D. Finkelhor, R. Gelles, G. Hotaling, M. Straus, pp. 69-84. Beverly Hills, Calif: Sage

Wolfgang, M. 1958. Patterns in Criminal Homicide. New York: Wiley

Wolfgang, M., Ferracuti, F. 1967. The Subcul- ture of Violence. London: Tavistock

Zigler, E. 1976. Controlling child abuse in America: An effort doomed to failure. Pre- sented at First National Conference on Child Abuse and Neglect, Atlanta

This content downloaded from �������������75.89.149.244 on Sun, 21 Aug 2022 21:09:58 UTC�������������

All use subject to https://about.jstor.org/terms