ArticlePDF Available

Religion-related Child Physical Abuse: Characteristics and Psychological Outcomes



Content may be subject to copyright.
Religion and child abuse
Religion-related Child Physical Abuse:
Characteristics and Psychological Outcomes
Bette L. Bottoms
University of Illinois at Chicago
Michael Nielsen and Rebecca Murray
Georgia Southern University
Henrietta Filipas
University of Illinois at Chicago
In press, Journal of Aggression, Maltreatment, and Trauma and in Janet Mullins (Ed.),
Emerging trends and issues in the victimization of children. Binghamton, NY: Haworth.
Authors' note: We thank Leslie Hollins, Sara B. Kay, Kristina Almassy, Sharanya Gururajan,
Monica Antoun, Anisha Shetty, and Krissie Fernandez for dedicated research assistance.
Direct correspondence to Bette L. Bottoms at the University of Illinois at Chicago,
Department of Psychology (m/c 285), 1007 W. Harrison Street, Chicago, IL 60607-7137.
Electronic mail may be sent via Internet to
Religion and child abuse
Religious beliefs can foster, encourage, and justify child abuse, yet religious
motivations for child abuse and neglect have been virtually ignored in social science
research. In this paper, we compare victims' retrospective reports of religion-related child
physical abuse to other reported cases of child physical abuse. We describe in statistical
detail the nature and circumstances of the abuse, characteristics of victims and
perpetrators, and the spiritual and psychological impact of the abuse. Results indicate that
although the basic characteristics of religion-related physical abuse are similar to non-
religion-related physical abuse, religion-related abuse has significantly more negative
implications for its victims' long-term psychological well-being.
Religion and child abuse
Religion-related Child Physical Abuse:
Characteristics and Psychological Outcomes
Religion provides specific directives for positive moral action and the promotion of
human welfare. It may be difficult to realize, then, that religious beliefs can also foster,
encourage, and justify abusive behavior. The myriad connections between religion and child
abuse led Donald Capps, a past president of the Society for the Scientific Study of Religion,
to entitle his presidential address, "Religion and Child Abuse, Perfect Together." Although
religious himself, Capps (1992) sorrowfully traced the indisputable connection between
traditional religion and violence against children. Similar points have been made by Capps
(1995), Straus (1994), Pagelow and Johnson (1988), and by Greven (1991) in his sobering
book, Spare the Child: The Religious Roots of Punishment and the Psychological Impact of
Physical Abuse.
In this paper, we explore the complex role of religion in child abuse cases reported
retrospectively by young adults in a survey. We sought to understand the nature,
circumstances, and emotional, psychological, and spiritual outcomes of religion-related
physical abuse as compared to other physical abuses. We are among the first researchers
to conduct a systematic examination from the victims' perspective of religion-related child
physical abuse as compared to other forms of physical abuse. Herein, we review the limited
literature on the ways in which religious beliefs can be involved in child physical abuse and
the impact of this abuse on psychological well-being and spirituality. Then, we delineate our
hypotheses regarding the impact of religion-related abuse, then present the methods and
findings of our study.
Religion and Child Physical Abuse
Religious groups often play an active, positive role in prevention of child abuse and
treatment of abuse victims. Yet, as Greven (1991) points out, encouragement for violent,
physically abusive child-rearing techniques can be traced to Biblical passages such as, "He
Religion and child abuse
that spareth his rod hateth his son: but he that loveth him chasteneth him betimes" and
"Withhold no correction from the child: for if thou beatest him with the rod, he shall not die.
Thou shalt beat him with the rod, and shalt deliver his soul from hell" (Proverbs 13:24 and
23:13-14, respectively). Directives such as these and belief in a vengeful God who would
punish earthly pleasure with the ultimate torture of hell drive some parents to use corporal
punishment to enforce parental authority and to prohibit supposed sinful behavior (Bottoms,
Shaver, Goodman, & Qin, 1995; Ellison & Sherkat, 1993). Sin (e.g., disobedience) is
considered the vehicle to hell, and parents may believe that it is better to inflict temporary
pain than allow their children to burn in eternal hell. Some believers (and even mental
health professionals; Friesen, 1992) even equate children's misbehavior with the actual
activity of Satan or other evil spirits who literally possess the children and who must be
exorcized by beatings. Thus, physical abuse is sometimes perpetrated by parents or by
religious leaders or teachers who believe they are helping to deliver children from sin.
Because of media attention, the public is generally aware of cases in which fringe
religious groups and isolationalist cults practice beatings in the name of Godly discipline
(e.g., Malcarne & Burchard, 1992). When discovered, such cults' abusive practices and
even their particular religious beliefs are immediately highlighted in the news media, and
criticized and rejected by society with much self-righteousness. Yet cultbeliefs and
practices may differ only in degree from those of mainstream religious groups such as
Methodists, Baptists, and Catholics, who believe the physical punishment of children is
religiously sanctioned, especially those who are Biblical literalists (Ellison, Barthkowski, &
Segal, 1996a, 1996b; Wiehe, 1990).
Social scientists have contributed to a growing, but still conceptually and empirically
thin, literature that identifies ways in which religiosity and belief systems promote and justify
abuse, especially by increasing parentsrisk for perpetrating and rationalizing child abuse.
Capps (1992, 1995) articulates clearly how the values of fundamental Christianity can be
Religion and child abuse
seen as supportive of the physical and emotional abuse of children. In studies of parenting
practices, Jackson, Law, Thompson, Christiansen, Colman, and Wyatt (1999) found that
religious ideology was a critical factor in predicting proneness to abuse, and Neufeld (1979)
found similarities in parental attitudes between abusive parents and parents who hold
fundamentalist Christian values. Shor (1998) determined that religious values may be
related to child maltreatment in ultra-orthodox Jewish families. Others have drawn similar
conclusions about the connection between religion and beliefs in corporal punishment (e.g.,
Ellison et al., 1996a, 1996b; Ellison & Sherkat, 1993; Flynn, 1996; Maurer, 1982; Nelsen &
Kroliczak, 1984). Although there has been little research on this topic, the relation between
religion and abuse does not go unnoticed, even among the devout: A church-funded survey
of nearly 650 members of the Christian Reformed Church revealed that while church
attendance was inversely related to reported perpetration of child abuse, a majority of
members believed the church does little to prevent abuse,that Christians too often use
the Bible to justify abuse,and that Church leaders are not prepared to help members of
their churches who are victims of abuse(Rice & Annis, 1992).
Thus, child abuse is indeed occurring within the context of religion. But what are its
particular circumstances and what are its effects on child victims? Bottoms et al. (1995)
conducted one of the largest studies to date of religion-related child abuse, a survey of
mental health professionals (psychologists, psychiatrists, and social workers) who had
encountered the following types of cases in their clinical practices: child abuse perpetrated
by persons with religious authority, religiously motivated medical neglect, and severe forms
of abuse and even murder perpetrated by parents and religious groups who believed they
were ridding children of evil (see also Bottoms, Shaver, & Goodman, 1996; Goodman,
Bottoms, Redlich, Shaver, & Diviak, 1998). Sixty-six percent of their cases in the latter
category included physical abuse, but almost half of the cases were characterized by sexual
abuse and nearly a third involved neglect, either in addition to or instead of physical abuse.
Religion and child abuse
Also, many of their cases of physical abuse were quite extreme; for example, they reported
one case in which an "eyeball was plucked out of a youth's head during an exorcism
ceremony," and another in which a father performed an exorcism on his children by
dismembering and then boiling them.Thus, their category of physical abuse is not
precisely like the cases we focus on in the present paper; nonetheless, given the overlap
and the lack of other information in the literature, it is informative to consider a profile of
their cases. In particular, 43% of the cases involved fundamentalist or fringe Christian
religions, 38% involved other protestants, and 16% involved Catholics. The abuse was
usually perpetrated by parents (85% of cases). On average, victims were abused between
the ages of 5 and 11 years of age. Psychological sequellae included depression in half the
cases, with a third of the victims having suicidal ideation. Other symptoms included phobias,
social withdrawal, inappropriate aggression, and dissociative disorders. Even though the
abuses were often quite violent, corroborative evidence was present only about half of the
time, and trials (12%) and convictions (9%) were rare. Social service agencies performed
investigations in 56% of the cases, but less than a fifth of the cases involved police and 6%
involved prosecutors.
There are few other studies of the circumstances or psychological impact of religion-
related physical abuse, but some studies provide insights regarding the effects of abuse on
childrens spirituality. In a study of children in grades 4 through 8, Nelsen and Kroliczak
(1984), revealed that when parents use the threat, "God will punish you" as a disciplinary
tool, children tend to engage in more self-blame and are more obedient. Studies of children
sexually abused by priests have revealed that the trauma can have a negative impact upon
the victims religion and spirituality (e.g., Rossetti, 1995). Also, child abuse generally (not
necessarily abuse related to religion) influences victimsreligious and spiritual practices. For
example, Johnson and Eastburg (1992) found that although physically and sexually abused
children saw their parents as more wrathful, this malevolent perception did not extend to
Religion and child abuse
God. Even so, other research suggests that by the time victims reach adulthood, their
relationship with and perception of God is conflicted at best. For example, Ryan (1998)
concluded that the effect of child abuse on adult spirituality is complex, with some survivors
reporting that retaining their religious beliefs helped them to heal from abuse, while others
feel angry at God, abandoned by God, and view God as punitive. In a study of over 1000
veterans, Lawson, Drebing, Berg, Vincellette, and Penk (1998) found that men with a
history of sexual or other child abuse were more likely than nonabused men to engage in
prayer and spiritual activities, yet they also reported feeling greater spiritual injury and
instability in spiritual behaviors and experiences. Doxey, Jensen, and Jensen (1997)
concluded that sexual abuse was associated with a decrease in religiosity. Kane, Cheston,
and Greer (1993) found that female sexual abuse survivors feel that God is disapproving
and ashamed of them.
The results of these studies provide evidence that spiritual damage is linked to
abuse. Few studies, however, distinguish religion-related abuse from abuse without
religious connotations, so our understanding of the impact of religion-related abuse on
spirituality is incomplete. As Durkheim (1915/1965) speculated, one of the major functions
of religion is to provide us with meaning, the sense of which offers us comfort when we are
vulnerable. If abuse is perpetrated in the name of God, its victim may be robbed of the
meaning and comfort that spirituality can provide, which might intensify the trauma of the
abuse, and compromise coping efforts. Thus, any negative impact on spirituality or
religiosity might itself add to the psychological impact of the abuse. In our study, we
measured both current psychological functioning as well as spirituality.
A Profile of Non-religion-related Child Physical Abuse
Although little is known specifically about the circumstances and outcomes of
religion-related child physical abuse, much is known about child physical abuse in other
contexts. It is important to consider a profile of physical abuse as a basis for understanding
Religion and child abuse
the additive or exacerbating effects of abuse-linked religious ideology or context.
Physical abuse affects hundreds of thousands of children each year, and accounts
for about a quarter of all child abuse reports (U.S. Department of Health and Human
Services, 2000). A comprehensive review by Kolko (2002) indicates that rates of reported
abuse do not generally differ by age or gender, though younger children may sustain the
most severe injuries (Lung & Daro, 1996). Nearly all child physical abuse is perpetrated by
parents and other loved ones. Ecological models proposed to explain physically abusive
parenting (e.g., Belsky, 1993; Lynch & Cicchetti, 1998) stress the importance of family
interactional and contextual variables, such as cultural and community factors. Other
models focus on parent factors such as perceptions of and attributions for childrens
behaviors; expectations, developmental knowledge, and problem solving skills; beliefs
about the appropriateness of using physical discipline; and tolerance for stress and ability to
inhibit aggression, etc. (Azar & Gehl, 1999; Milner, 2000; Simons, Whitbeck, Conger, &
Chyi-In, 1991; Wolfe, 1999; see Kolko, 2002, for review). Researchers such as Kolko (2002)
have noted the need for a better understanding of the host of factors that contribute to
excessive parental negativity toward childrens misbehavior. We propose religious beliefs as
one important factor. Religion can contribute to the likelihood of abuse by providing
knowledge and attitudes that directly support abusive behaviors. This likelihood of abuse is
probably greatest when other risk factors are also present.
The detrimental effects of physical abuse on psychological health in childhood and
adulthood are well documented. Of course, how any one individual adjusts psychologically
after having been abused as a child depends on numerous factors (e.g., Winton & Mara,
2001), but generally, children who are battered can develop an assortment of physiological,
intellectual, psychological, and behavioral problems in response to the stress of their abuse
(for review, see Kolko, 2002). For example, abused children often have lower self-esteem
than their peers (Allen & Tarnowski, 1989; Oats, Forrest, & Peacock, 1985), although Kolko
Religion and child abuse
notes that the long-term impact on self-esteem needs further investigation. Externalizing
problems such as increased aggression are commonly associated with physical abuse, but
so too are some internalizing problems such as depression and hopelessness (e.g.,
Pelcovitz, Kaplan, Goldenberg, & Mandel, 1994). The limited longitudinal and retrospective
cross-sectional data that are available reveal that adult survivors of child abuse also differ
from non-abused adults. For example, childhood abuse leads to later life aggression,
violence, and criminal behavior (Dodge, Pettit, & Bates, 1997; Quas, Bottoms, & Nunez,
2002; Maxfield & Widom, 1996); antisocial personality disorder (Luntz & Widom, 1994);
anxiety and emotional behavioral problems (Silverman, Reinherz, & Giaconia, 1996);
depression (Brown, Cohen, Johnson, & Smailes, 1999; Levitan et al., 1998; Silverman et
al., 1996), and suicidal ideation (Brown et al., 1999; Bryant & Range, 1997; Levitan et al.,
1998; Silverman et al., 1996).
Thus, the factors that lead parents to abuse their children and the psychological
effects of physical abuse are many and varied. We propose that parentsreligious attitudes
and beliefs (e.g., spare the rod, spoil the child) are among the attitudinal factors that can
contribute to the likelihood of familial abuse. Further, we believe that for reasons outlined
next, the long-term emotional toll of religion-related child physical abuse will be similar in
type to that of non-religion-related physical abuse, but compounded, because of the added
stress and guilt associated with religious justifications for maltreatment.
Overview and Predictions
Given the fact that religion-related justifications for abuse abound and sometimes
manifest in cases of physical abuse, and that abuse related to religion may be particularly
detrimental to children's well-being, it is surprising that little research has been conducted to
explore its characteristics or consequences. Many questions about religion-related abuse
exist. What are the circumstances of this form of abuse who are its victims, its
perpetrators? What is its impact on religiosity and spiritual beliefs? Does this type of abuse
Religion and child abuse
have distinct psychological consequences? To address these important questions, we
undertook a retrospective survey study of three groups of young adults: (a) adult survivors
of religion-related child physical abuse, (b) adult survivors of non-religion-related child
abuse, and (c) a control group of individuals who had not suffered any form of childhood
abuse. We queried all individuals about their current psychological and emotional well-
being, including self-esteem, and we asked the two victim groups about the nature and
circumstances of their abuse experiences.
In keeping with the literatures reviewed above, we theorize that religious overtones
add an additional layer of complexity and harm to child abuse experiences. Victims of
religion-related abuse must deal with the trauma associated not only with parental betrayal
but also perhaps the additional despair related to perceived betrayal by God. This may
inhibit a child's ability to cope with maltreatment. Therefore, we hypothesized that
individuals in our study who experienced religion-related abuse would experience more
emotional and psychological distress and have lower self-esteem than victims of other
forms of physical abuse. We expected both groups of victims to exhibit more emotional and
psychological difficulties than the control group. Further, we predicted that religion-related
abuse victims would report greater spiritual injury and be less religious than individuals who
had suffered non-religion-related childhood abuse. We did not expect to find differences in
the basic circumstantial characteristics of the two types of physical abuse in terms of form
or frequency of abuse, its likelihood of being disclosed or otherwise discovered,
characteristics of victims and perpetrators, etc. Instead, we expected both forms of physical
abuse to fit the profile of physical abuses found in many other studies (for review, see
Kolko, 2002), with the exception that, as found in the Bottoms et al. (1995) work, we
expected more religion-related abuse to be perpetrated by religious leaders.
Religion and child abuse
Fifty-four men and 72 women from the University of Illinois at Chicago (a large,
racially and ethnically diverse, urban university) volunteered to serve as participants in
exchange for Introductory Psychology course credit. Twenty-six (54% women) had
experienced childhood physical abuse that was in some way related to religion as defined
below; 46 (61% women) experienced non-religion-related physical abuse, and 54 (57%
women) experienced no abuse of any kind and served as the control group. The groups
were statistically similar in terms of all demographic variables we measured (all ps > .20),
including age (M = 19.4 years), year in college (51% freshmen, 35% sophomores, 14%
juniors or seniors), marital status (99% single), race/ethnicity (16% African American, 26%
Asian American, 43% Caucasian, 13% Hispanic/Latino/Latina, 2% other), mothers
education (Mdn = at least one year of college), fathers education (Mdn = college degree),
and estimated parental income (Mdn = $60,000 - $69,000, on an 8-point scale ranging from
$0 to $80,000+ in increments of $10,000).
Demographic questionnaire. A demographic questionnaire requested the
information reported in the participantssection above.
Religiosity measures. Several items measured participantsinvolvement in religion
and their religious attitudes. First, participants indicated their religious affiliation, then rated
the importance of their religious beliefs on a scale ranging from 1 (Not at all important) to 5
(Extremely important). (Participants who considered themselves spiritualwithout believing
in a specific deity were asked to frame this and the other religious items in terms of
spirituality rather than religion.) Another item asked participants to rate the importance of
religion to their self-concept using a 1 (Not at all) to 5 (Extremely) scale. The frequency with
which participants attended religious services, and the frequency of their prayers, were
rated on separate scales ranging from 1 (Less than once a year) to 8 (Once a day). Finally,
Christian orthodoxy was assessed using the short version of the Christian Orthodoxy Scale
Religion and child abuse
(Hunsberger, 1989), which consists of three items that address core Christian doctrines
described in the Apostles Creed and Nicene Creed such as Jesus Christ is the divine Son
of God,and three that describe the rejection of those doctrines such as Despite what
many people believe, there is no such thing as a God who is aware of our actions.
Responses were given on a seven-point scale, ranging from 3 (Strongly disagree) to +3
(Strongly agree). Appropriate items were reverse-coded so that higher scale scores reflect
greater orthodoxy. Responses were recoded from 1 to 7, and item scores were summed,
producing a theoretical range from 6 to 42 (higher scores indicating greater orthodoxy).
Abuse Experiences Survey. As part of a larger survey study, experiences of
physical abuse were assessed with our Abuse Experience Survey.1 The question used to
determine whether participants had suffered childhood physical abuse was based on
definitions used by Straus and Gelles (1988), and was similar to a question used previously
to assess physical abuse with a similar university sample (Epstein & Bottoms, 2002): When
you were a child or teenager (age 17 years or younger), did any person who was at least 5
years older than you ever do any of the following types of things to you for any reason: (a)
hit, slapped, spanked, lashed, grabbed, shoved, choked, punched, kicked, bit, burned, beat,
stabbed, or whipped you and it left you with welts, bruises, bleeding, or other marks or
injuries; (b) punished or disciplined you with a belt, cord, or other hard object; or (c)
threatened you with a weapon?Participants were also asked, Do you consider your
experience abuse?We labeled respondents to be physical abuse victims even if they
failed to consider the experience to have beenabusethemselves. This method helps to
avoid underestimating the true number of abuse victims, because some individuals who
have had abusive experiences fail to self-label as victims (e.g., Silvern, Waelde, Baughan,
Karel, Kaersvang, 2000).
The participant then moved on to a series of specific questions about the
experience. As evident in our description of the results below and therefore not detailed
Religion and child abuse
here, a variety of survey questions tapped information about the nature and circumstances
of the abuse, characteristics of the victims and the perpetrator(s), etc. To classify cases as
religion-related, the survey included this question: Do you think this experience had
anything at all to do with religion?We considered their abuse to have been religion-related
if they chose any one or more of the following possible answers: (a) it happened in a
religious setting; (b) person(s) who did it to you had some religious authority or was in any
religious leadership position; (c) the person(s) told you God would punish you if you told
about it; (d) it was done to punish or discipline you in a way that was suggested by a
religious text or spiritual being; (e) person(s) who did it thought you were possessed by
devils or evil spirits; (f) person(s) who did it thought God or another spiritual figure or
religious text told him/her to do it for reasons other than discipline or possession; (g) you
were told God or another spiritual being would love you more because of the experience;
(h) other religious reason.(In a few cases, cases were classified as religion related when
respondents failed to choose abut later on the survey indicated a religious setting for the
abuse, or failed to choose bbut later indicated that the perpetrator had religious authority.)
Brief Symptom Inventory (BSI). The BSI is a well-validated 53_item inventory used
to assess psychological symptom patterns among patient and non_patient populations
(Derogatis & Spencer, 1982). It is a brief form of the SCL-90-R. Participants use a 5-point
Likert scale to rate the extent to which the BSI items (describing various emotional states,
feelings, and behavior) apply to themselves. The BSI provides a Global Severity Index
(average response per item), a Positive Symptom Total (the number of positive or
symptomatic responses overall), and a Positive Symptom Distress Index (ratio of the BSI
Global Severity Index score and the BSI Positive Symptom Total, which is intended to
indicate the severity of symptoms indicated). Component subscales measure 9 primary
symptom dimensions: hostility, depression, somatization, obsessive-compulsive tendencies,
phobic anxiety, interpersonal sensitivity, anxiety, paranoid ideation, and psychoticism.
Religion and child abuse
Center for Epidemiological Studies Depression Scale (CESD). The CESD
consists of 20 items designed to assess recent experience of depression (Radloff, 1977).
Participants respond to items in Likert-scale fashion, indicating the extent to which each
statement describes their experience during the past week. The test yields an overall score,
with a score of 16 indicating the possibility of depression and a score of 23 used to indicate
clinically significant levels of depression.
Rosenberg Self-Esteem Scale (Rosenberg, 1965). The Rosenberg Self-Esteem
Scale is one of the most popular and well-validated measures of self-esteem, with excellent
internal consistency and test-retest reliability. Respondents use a 4-point scale ranging from
1 (strongly agree) to 4 (strongly disagree) to answer items such as I feel that I am a person
of worth, at least on an equal basis with others.Items are coded such that higher numbers
indicate higher levels of self-esteem, and summed.
All measures were randomly ordered within one stapled packet, except that the
demographic form was always placed first. Participants completed the packet in small
groups during an experimental session in a large room where at least two chairs separated
them from each other. Signed informed consent forms were returned separately from
surveys to ensure complete anonymity.
We begin by presenting data regarding the nature of the abuses suffered by the two
groups of victims (religion-related physical abuse victims and non-religion-related physical
abuse victims), including the form of the physical maltreatment, the number of times the
abuse allegedly occurred, and whether the abuse was disclosed or otherwise discovered.
Then we examine characteristics of the alleged victims and perpetrators, such as the
relationship between them, their religious affiliation, the victims age, and the perpetrators
gender. Finally, we test our main hypotheses regarding the long-term impact of physical
Religion and child abuse
abuse: We examine the spiritual, emotional, and psychological impact of the abuse as
perceived by the victims, then the current psychological well-being as measured by our
standardized psychological tests. All analyses, except those of current psychological well-
being, were oneway analyses of variance (ANOVAs) or chi square analyses (as
appropriate) comparing the two abuse groups. (Controls were not included in those
analyses because they did not suffer abuse and therefore could not describe abuse
experiences.) Analyses of well-being compared the three groups of participants (victims of
religion-related physical abuse, victims of non-religion-related physical abuse, and non-
abused controls). Main effects of abuse type in these latter analyses were followed by
Tukey tests of pairwise comparisons of means or tests with Scheffe corrections, as
appropriate (Keppel, 1982).
Characteristics of the Alleged Abuses
Form of physical abuse. The specific forms of the physical maltreatment suffered
was statistically similar in both victim groups (see Table 1). That is, the two groups did not
differ with regard to their likelihood of having been: (a) hit, slapped, spanked, lashed,
grabbed, shoved, choked, punched, kicked, bit, burned, beat, stabbed, or whipped you and
it left you with welts, bruises, bleeding, or other marks or injuries,x2(1, N = 72) = .05, p =
.82; (b) punished or disciplined with a belt, cord, or other hard object,x2(1, N = 72) = 3.00,
p = .08; or (c) threatened with a weapon,x2(1, N = 72) = 2.78, p = .10.
Reflecting the degree to which our society accepts the physical abuse of children,
only about a quarter of the participants whom we labeled victims believed that their
experiences were abusive. Among those who experienced non-religion-related abuse, 22%
considered the actions to be abusive, compared to 35% of those in the religion-related
abuse group, a difference that was not statistically significant, x2(1, N = 72) = 1.42, p = .23.
Involvement of religion in the abuse. How was religion involved in the cases?
According to victimsresponses to the non-mutually exclusive items used to identify cases
Religion and child abuse
as religion-related, in 31% of cases, the perpetrator was a religious leader. (In light of recent
scandal and publicity, it is important to note that sexual abuse is not the only form of abuse
perpetrated on children by religious authorities.) In 31% of the religion-related abuse cases,
the abuse was punishment or discipline that the perpetrator justified with religious texts, and
in 12% of the cases, it was done for reasons other than punishment or discipline (but still
believed by perpetrators to be ordered by God or a religious text). In 20% of cases, the
perpetrator told the victim that God would punish her/him if the abuse was disclosed. Eight
per cent of the cases involved physical abuse done to rid the child of supposed devils or evil
spirits, a form of religion-related physical abuse specifically studied by Bottoms et al. (1995;
Bottoms et al., 1996). In 35% of the cases, the abuse occurred in a non-secular setting.
When asked in a separate question to identify the setting of abuse, 12% of victims specified
church, 23% indicated a parochial or church-affiliated school, and 8% named some other
religious meeting place. Like all forms of child abuse, however, it was still most likely to
happen in the home (73% of cases; more than one setting could have been indicated.)
Frequency of abuse. The physical abuse reported was not typically a one-time
occurrence (see Table 1). In fact, 40% of the religion-related abuse occurred over 20 times.
To gain clarity about the nature of differences in frequency that might exist between the two
types of abuse, we converted these data to a scale ranging from 1 (once) to 6 (21 or more
times). We acknowledge the imperfect nature of this conversion, which collapses across
many numbers at the higher end of the scale, but results from an analysis of this scale are
actually conservative in this regard. An ANOVA revealed that the religion-related cases
occurred, on average, with statistically similar frequency (M = 4.20, SD = 1.94) as other
abuse cases, (M = 3.76, SD = 1.27), F(1, 69) = 1.33, n.s.
Disclosure and case outcome. Not surprisingly, most abuse victims told no one
about their experiences, perhaps out of fear or simply because they did not believe that the
actions were abusive. There was no significant difference in the percentage of religion-
Religion and child abuse
related abuse victims (42%) and other victims (61%) who failed to disclose the abuse, x2(1,
N = 71) = 2.39, p = .12. Disclosures that were made by victims of religion-related or other
physical abuse, were made to parents (31% and 18%), another relative (15% and 11%), a
friend (27% and 21%), or a teacher (4% and 7%), respectively, all x2s (1, N = 71) < 1.47, ps
> .23.
In addition, most respondents (69% of religion-related abuse victims and 75% of
others) indicated that no one discovered the abuse (apart from disclosure), x2(1, N = 71) <
1, n.s. In only 27% of religion-related cases and 25% of other cases did someone try to stop
the abuse, x2(1, N = 70) < 1, n.s.
A formal investigation by the police, social services, or district attorneys office was
rare, occurring in only 4 of the religion-related cases and 3 of the other cases. In two of the
former, there was a formal determination that the abuse occurred. In one of the latter, a
conviction was obtained.
Characteristics of Victims and Perpetrators
Victim age. There was no significant difference in the age at which the abuse began
for victims of religion-related abuse (M = 6.46 years, SD = 3.43) and non-religion-related
abuse (M = 6.39 years, SD = 2.86), t(63) < 1, n.s. The abuse lasted around 6 years, on
average, ending by the age of 12.5 years (SD = 3.91) for victims of religion-related abuse
and by 12 years (SD = 3.43) for other victims, t(63) < 1, n.s., as in the study by Bottoms et
al. (1995).
Victimsreligious affiliation and religiosity. The victims from the two abuse
groups did not differ significantly in terms of denominational affiliation. As shown on Table 2,
over a third of the sample was Roman Catholic (whereas Catholics comprise around 25% of
the general population). The next most popular category was mainstream Protestant (20%
or less), then "Christian." The two groups also failed to differ with regard to perceived
importance of religious beliefs, importance of religion to self-concept, frequency of attending
Religion and child abuse
religious services, frequency of prayer, and scores on the Christian Orthodoxy Scale, all
ts(>66) < 1.31, n.s. (see Table 2). In short, the participants were much more similar than
dissimilar in terms of their religious beliefs and practices.
Relationship of perpetrator to victim. As true in most cases of child physical
abuse, parents were the most common perpetrators. This was more likely to be true,
however, in cases of non-religion-related abuse than in religion-related abuse, x2(1, N = 71)
= 4.09, p = .04 (see Table 2). In no cases was the abuse perpetrated by a stranger or mere
acquaintance. The two abuse groups did not differ significantly in the likelihood that the
abuser was another family member such as a grandparent, x2(1, N = 71) = 0.51, p = .48, or
other relative, x2(1, N = 71) = 2.55, p = .11. A teacher was more likely to have been the
abuser in religion-related abuse cases than in non-religion-related abuse cases, x2(1, N =
71) = 6.16, p = .01, probably reflecting the involvement of teachers at religious schools. We
were particularly interested in instances of abuse perpetrated by religious authorities. In one
religion-related abuse case the abuser was a Sunday School teacher, and in three cases
(12%) the abuser was a priest. No cases were specified as having been perpetrated by a
nun, minister, rabbi, or religious healer.
We were also interested in the perceived emotional relationship between the victim
and perpetrator, and so we asked the victims to rate the degree to which they felt
emotionally close to the abuser both before and after the abuse, on a scale ranging from 1
(Not at all close) to 5 (Extremely close). We performed a 2 (abuse type) x 2 (time: before vs.
after the abuse) ANOVA on these scale scores, with time as a within-subjects variable.
There was no significant difference in perceived closeness reported overall by victims of
religion-related (M = 3.67, SD = 0.26) and non-religion-related abuse (M = 3.97, SD = 0.20),
F(1, 69) < 1, n.s. As would be expected, however, there was a significant main effect for
time, with respondents feeling closer to the abuser before the abuse (M = 4.07, SD = 1.26)
than they did following the abuse (M = 3.65, SD = 1.58), F(1, 69) = 9.26, p = .003. There
Religion and child abuse
was no statistically significant interaction between these two factors, F(1, 69) < 1, n.s.
Perpetrator gender. There was no statistically greater likelihood of a man (77%) or
woman (62%) being a perpetrator in religion-related cases, as compared to the other cases
(61% and 64%, respectively), x2s (1, N = 70) < 1.79, ps > .18. These results are consistent
with our finding that parents were most likely to be the abusers, and suggests that the
parents worked together in performing the abuse.
Perpetrator religion. Abusers' religious affiliations did not differ significantly
between the two abuse categories, x2(7, N = 64) = 6.29, p = .51 (see Table 2). Unsurprising
for this particular midwest sample, the abuser was most likely to be Catholic in over half of
both kinds of cases. In light of previous writings (e.g., Capps, 1992, 1995), it is surprising
that no perpetrators were reported to have been fundamentalist.This might be a function
of inaccurate reporting tendencies. That is, fundamentalist might be a category unlikely to
be used by our young victims, who may have chosen to leave this item blank or use
identifiers like Christian.
Spiritual, Emotional, and Psychological Effects of the Abuse
Perceived effects on religiosity. Respondents indicated whether the abuse caused
them to change their religious beliefs. Victims of religion-related abuse (69%) were
significantly less likely than other victims (94%) to report that the abuse had no effect, x2(1,
N = 72) = 7.55, p < .01. Twelve per cent of the religion-related victims indicated that the
abuse made them more religious, 16% less religious or atheistic, and 4% indicated that it
made them switch religions. It is interesting that fully 28% of the religion-related abuse
victims told us that they felt that the physical abuse was acceptable or justified for religious
reasons.Perhaps this is a necessary coping mechanism for reducing the cognitive
dissonance that would otherwise result from identifying ones parents or trusted religious
leaders or teachers as abusers, or from believing that religious teachings are not helpful or
comforting. This finding warrants further examination in future research.
Religion and child abuse
Perceived emotional effects. We asked respondents to tell us how emotionally
upset they were by the abuse on a scale ranging from 1 (not at all upset) to 5 (highly upset).
There was no difference in the average ratings given by non-religion-related (M = 3.33, SD
= 1.49) and religion-related abuse victims (M = 3.04, SD = 1.45), t(70) < 1, n.s. Separately,
we asked respondents whether they experienced any lasting effects of the experience.
Most respondents (76% of non-religion-related and 54% of religion-related abuse victims)
reported none, a difference that only approached statistical significance x2(1, N = 71) =
3.55, p = .06. The most common type of lasting effect noted was emotional or
psychological harm,which was indicated by 17% of non_religious and 23% of religion-
related abuse victims, x2(1, N = 71) < 1, n.s. Only 2% of victims from the non-religion-
related group reported positive spiritual effects,as compared to 19% of religion-related
abuse victims, x2(1, N = 71) = 6.16, p = .01. None of those who reported positive spiritual
effects changed their religious affiliation; whatever positive effects that might have occurred
took place within their faith tradition, not as a result of changing their affiliation. The rate at
which victims reported negative spiritual effectsdid not differ between the non-religion-
related (4%) and religion-related (8%) abuse groups, x2(1, N = 71) < 1, n.s. No participants
reported lasting physical impairment.
Current self-esteem and psychological well-being. We analyzed scores on the
Rosenberg Self-Esteem scale, the Center for Epidemiological Studies Depression Scale,
and the Brief Symptom Inventory scales in separate ANOVAs comparing all three groups
(religion-related abuse victims, non-religion-related abuse victims, and the non-abused
control group).2
First, with regard to participantsgeneral self-esteem, there was only a non-
significant trend in the expected direction. Victims of religion-related abuse reported slightly
lower levels of self-esteem (M = 30.50, SD = 6.16) as compared to victims of non_religion-
related abuse (M = 33.13, SD = 5.65) and the control participants (M = 33.37, SD = 5.70),
Religion and child abuse
F(2, 123) = 2.37, p = .10.
Next, we examined participantscurrent level of depression. CES-D scores revealed
the effects we hypothesized: 42% of religion-related abuse victims, but only 15% of other
victims and 20% of the control group could be classified as clinically depressed using the
recommended criterion score of 23, x2(2, N = 126) = 7.25, p = .03. Follow-up chi square
analyses with Scheffe corrections (as recommended by Sheshkin, 2000) revealed that the
differences between the religion-related abuse group and each of the other two groups were
both statistically significant, x2s > 4.24, ps < .02. A similar but non-significant trend emerged
in level of depression as measured by the BSI-Depression subscale (see Table 3). The
religion-related abuse group reported slightly higher levels of depression (M = 1.15, SD =
0.85) than did the other abuse group (M = 0.80, SD = 0.88) and the control group (M = 0.85,
SD = 0.76), F(2, 123) = 1.72, p = .18.
Table 3 illustrates all means from the BSI measures. In terms of the component
subscales, as predicted, victims of religion-related abuse faired more poorly than other
victims and than control participants in terms of anxiety, F(2, 123) = 4.33, p = .02; phobic
anxiety, F(2, 123) = 3.87, p = .02; hostility F(2, 123) = 4.04, p = .02; psychoticism, F(2, 123)
= 3.79, p = .02; and paranoid ideation, F(2, 123) = 3.46, p = .04. A similar trend approached
significance for obsessive-compulsive tendencies, F(2, 123) = 2.34, p = .10. In fact, the
religion-related victimsmean score for obsessive-compulsiveness approached the mean
score of psychiatric outpatients (1.57; Derogatis & Spencer, 1982). For somatization, there
was also an overall group effect, F(2, 123) = 3.15, p = .05, with the religion-related group
mean for somatization approaching the mean score for psychiatric outpatients (.83;
Derogatis & Spencer, 1982). However, Tukey comparisons revealed that their score differed
significantly only from the non-religion-related group, not from the control group. Mean
differences for interpersonal sensitivity failed to reach statistical significance, F(2, 123) =
1.53, p = .22.
Religion and child abuse
Not surprisingly, given these component results, general symptomology, as
measured by the BSI Global Severity Index was, overall, greatest among people who had
suffered religion-related abuse than among the other two groups of people, F(2, 123) =
4.40, p = .01. There was a similar, but only marginally significant effect, on the Positive
Symptom Distress Index scores, F(2, 123) = 2.83, p =.06. Nevertheless, the analysis of
individual subscales on the BSI demonstrates that, above and beyond possible effects of
abuse frequency, religion-related abuse is associated with more negative effects in many
areas of mental health than is abuse that has no religious component.
In a 1988 review article, Gorsuch asked, "Is religion an important psychological
variable?" When considering the abuse of children, our data indicate that it is, and that
religion-related abuse should be considered distinct from other forms of child abuse. The
religion-related cases in our sample were defined as such because the perpetrator of the
physical abuse was a religious leader or religious teacher, believed he or she was directed
by God or a religious text, and/or perpetrated the abuse in a religious context, all of which
are circumstances that could lead victims to feel as though the abuse was sanctioned by a
church or even God.
Our hypotheses were largely supported by our data. In particular, as expected, we
found that abuse related in some way to religious beliefs or practices was quite similar to
non-religion-related abuse in terms of its basic form, frequency of occurrence, likelihood of
having been disclosed or otherwise discovered by anyone or acted upon by legal or social
service authorities (which rarely occurred), victim characteristics (age, religious affiliation,
and religiosity), and perpetrator characteristics (religious affiliation, gender). The only
reliable difference in the nature of the cases involved the relationship of victim to
perpetrator: Although all cases were more likely to be perpetrated by a parent than by
anyone else, this was somewhat more likely in non-religion-related cases, and teachers
Religion and child abuse
were more likely to be perpetrators in religion-related cases than in other cases.
We expected to find more spiritual injury and decreased religiosity among religion-
related abuse victims. These hypotheses were not supported: There were no significant
group differences in terms of perceived importance of religion generally and to the self-
concept, frequency of attending religious services and of praying, and religious beliefs as
measured by the Christian Orthodoxy Scale. Thus, findings of decreased spirituality or
religiosity noted in victims of child sexual abuse (e.g., Doxey et al., 1997; Lawson et al.,
1998) may not generalize to victims of child physical abuse.
Of most importance were our hypotheses regarding long-term psychological impact.
We found that victims from the two abuse groups did not differ in the extent to which they
reported having been emotionally upset by the abuse. In fact, victims of religion-related
abuse were more likely than victims of other physical abuse to say that their experience had
lasting positive spiritual effects. The two groups also failed to differ in terms of general self-
esteem, contrary to our expectations. However, other standard measures of psychological
well-being revealed significant differences such that victims of religion-related abuse had
greater psychological symptomology than did victims of other physical abuse or non-abused
controls. In particular, victims of religion-related abuse displayed more depression, anxiety,
hostility, psychoticism, phobic anxiety, and paranoid ideation (although the latter two effects
weakened after covarying for proxy measures of abuse severity). Those victims were also
higher in somatization than victims of non-religion-related abuse, but did not differ from the
control group in that regard. As noted in our introduction, several of these sequellae were
also found in the Bottoms et al. (1995) study of religion-related child abuse. Although prior
research has shown that many of these emotional difficulties, including hostility, anxiety,
depression, and suicidal ideation, result from child physical abuse without religious
components, the victims of non-religion-related abuse in our study did not differ significantly
from the control group on any dimension of well-being. (This is cause for some caution in
Religion and child abuse
interpreting our results, because it is possible that our non-religion-related abuse
comparison group was more highly functioning than such victims in the general population.
But, if so, then one would expect religion-related abuse group to also be proportionally
higher functioning.)
What could explain the poorer psychological outcomes for the religion-related victims
compared to the other two groups? As we theorized in the introduction, religious contexts
and justifications may add an additional layer of complexity and harm to the experience of
child physical abuse. Additional research is necessary to determine the specific nature of
this additional complexity, but like others (Bottoms et al., 1995; Goodman et al., 1998), we
speculate that there is an additional sense of betrayal involved, and much internal cognitive
dissonance and perhaps guilt as victims deal not only with the physically abusive actions,
but also with the confusing relation of the actions to religion, which they are taught to
believe and follow. This may interfere with normal coping mechanisms. Young victims may
come to believe (or are told to believe) that the abuse is parentally and/or supernaturally
sanctioned or required, or is a punishment for their own sins. Fully 28% of our victims had,
in fact, come to believe that the abusive experiences were justified for religious reasons.
Only 35% of them considered their experiences to have been abusive.
To understand the differential psychological effect of religion-related abuse, it is also
worthwhile to consider theories proposed by researchers such as Pargament and
Kirkpatrick. First, Pargament (1997) has demonstrated that, for some people, religion is a
source of meaning that helps them cope with daily and existential concerns. We suggest
that victims of religion-related abuse, whose relationship to or views of God could be
disturbed by the abuse experience, might be deprived of the benefits of viewing God as a
coping resource. Victims abused by a person with religious authority have also lost that
spiritual leader as a source of social support. Second, according to Kirkpatrick (1997, 1998)
God serves as an attachment figure for some individuals, especially those who are not
Religion and child abuse
securely attached to parents. In the typical religion-related abuse case in our sample, one or
both parents were the perpetrator, the perpetrators invoked religion or God to legitimize
their actions, and the victims felt more distant emotionally from the perpetrators after the
abuse. Thus, our victims might have felt deprived of secure attachments with parents and
with God, and might have had fewer social psychological resources for effective coping.
Although we failed to find decreased religiosity or religious practices among victims of
religion-related abuse, we did not directly measure the use of religion as a coping
mechanism or source of social support, nor did we measure feelings of attachment to
parents or God. Of interest, Kennedy and Drebing (2002) similarly found no relation
between current religious behaviors and past experiences of abuse (non-religion-related
abuse), but they did find that self-reported frequency of abuse was related to feelings of
alienation from God.
Strengths and Limitations
Our study is one of very few studies, if any, to use data reported directly by victims
(as opposed to data reported by their therapists or by the authorities, for example, as in
Bottoms et al., 1995 and Bottoms et al., 1996) to compare child physical abuse that is
religion-related with abuse that is not. Our methodology is solid in many ways: We used
appropriate comparison and control groups, well-conceived and validated measures, and a
sample that was diverse and therefore representative in terms of socioeconomic status and
race/ethnicity. Even so, there are many questions we were unable to address and our
findings should be considered in light of our studys limitations. For example, we
investigated only one form of religion-related abuse, ignoring other forms that are also of
importance such as religion-related medical neglect and sexual abuse perpetrated by
persons with religious authority (but these have been covered in some depth in other
articles, e.g., Bottoms et al., 1995). Further, even though our sample was diverse in some
ways, it was also small in size, young, and included only victims from a non-clinical, urban,
Religion and child abuse
midwest setting. Our findings might not represent the experiences of individuals who
suffered extremely severe abuse or emotional sequellae and who therefore are less likely to
appear in a sample of highly functioning college students (e.g., Duncan, 1999). Nor might
they be generalizable to other locales. We also have no way of moving from our data to
base rates in the general population, so we cannot provide any insight into the incidence or
prevalence of this form of abuse. Further, as is true for any study using retrospective self-
report methodology, people are not always accurate in reflecting upon their past
experiences, their motivations, or their cognitive processes (Azar, 1997). We have
necessarily accepted the accuracy of information provided by respondents, but victims
perceptions and recollections might not all be accurate. It is also worth noting that we
measured victimsperceptions and psychological well-being at only one point in time, when
the victims are relatively young; we do not know what longer term effects might be. Finally,
no one study can measure all important variables. Although we included several measures
of abuse severity, including one (frequency) which did differ between the two abuse groups,
we do not know if there are additional unmeasured variables which might help explain the
differences we observed in mental health status of the victims from the two abuse groups.
Obviously, random assignment to abused and nonabused groups is not, and never will be,
For all these reasons, our study is an important start in documenting that differences
exist in the mental health outcomes of religion-related and other forms of physical abuse,
but it is only a start. Future research must be conducted to determine if the differences we
observed can be replicated, and if so, additional research should use a theoretically
oriented approach to understand the specific reasons for those differences. Even so, we
believe our study makes an important contribution to a controversial field by providing
scientific findings that can lead to a better understanding of religion-related abuse.
Understanding necessarily precedes and informs identification, treatment, and prevention
Religion and child abuse
efforts, especially efforts to work with parents or others who need to learn non-abusive
methods for disciplining children. We are pleased that some churches are already engaging
in such intervention efforts.
A regrettable picture has been developed by our study: Physically abusive actions
taken against a child for religious reasons are generally not severe enough to attract the
attention of authorities (as indicated by our data on disclosure and discovery), and not often
considered by its victims to be abusive, but nonetheless might have long-lasting negative
consequences for psychological well-being.
As Greven (1991) notes, abusive parenting styles have been driven by mainstream
religious beliefs for centuries. They are part of our Euro-American heritage, and if religion-
related child abuse is not acknowledged now as a problem by our society, it will be our
legacy to the future. The freedom to choose religions and to practice them will, and should,
always be protected by our constitution. The freedom to abuse children in the course of
those practices ought to be curtailed.
Religion and child abuse
1The larger parent survey consisted of four parts, each addressing a different form of child
maltreatment (physical, sexual, emotional, and neglect). For the present paper, we
analyzed data from participants who suffered only child physical abuse (or, for the control
group, no abuse). Thus, we were able to examine the nature and effects of child physical
abuse specifically rather than in combination with other forms of abuse. That is, religious
beliefs can contribute to single or multiple forms of child abuse (Bottoms et al., 1995), and
the presence or absence of various combinations of abuse has differential effects on victims
(Goodman, Bottoms, Redlich, Shaver, & Diviak, 1998). Even so, we acknowledge different
and uncontrolled levels of emotional abuse among our cases, consistent with arguments
that all forms of child abuse involve some level of psychological maltreatment (Brassard,
Germain, & Hart, 1987; Claussen & Crittenden, 1991; Garbarino, 1980; Garbarino &
Vondra, 1987). It is also important to note that the distribution of respondents across our
three groups is not at all indicative of incidence or prevalence; it results from purposeful
sampling from among a much larger number of students who participated in the parent
2We also performed analyses of covariance (ANCOVA) comparing only the two abused
groups while statistically controlling for two variables that might reflect abuse severity: (a)
frequency of abuse, as measured by the 6-point frequency scale described in the results
section (and which did not differ significantly between the two groups), and (b) whether or
not a parent or parents were the perpetrators of the abuse (which did differ significantly
between the groups, with parents more likely to be the perpetrators in cases of non-religion-
related abuse). These ANCOVAs revealed few differences in the results presented herein:
After covarying for the two factors simultaneously, (a) the significant effects of abuse type
on phobic anxiety and on somatization weakened to marginal levels of significance, F(1, 66)
= 3.04, p = .09 and F(1, 66) = 3.07, p = .08, respectively; and (b) the significant effect on
Religion and child abuse
paranoid ideation became non-significant, F(1, 66) = 1.95, p = .17.
Religion and child abuse
Allen, D. M. & Tarnowski, K. G. (1989). Depressive characteristics of physically
abused children. Journal of Abnormal Child Psychology, 17, 1-11.
Azar, B. (1997). Poor recall mars research and treatment: Inaccurate self-reports
can lead to faulty research conclusions and inappropriate treatment. The APA Monitor, 28,
Azar, S. T., & Gehl, K. S. (1999). Physical abuse and neglect. In R. T. Ammerman &
M. Hersen (Eds.), Handbook of prescriptive treatments for children and adolescents
(pp.329-345). Boston: Allyn & Bacon.
Belsky, J. (1993). Etiology of child maltreatment: A developmental-ecological
analysis. Psychological Bulletin, 114, 413-434.
Bottoms, B. L., Shaver, P. R., Goodman, G. S., & Qin, J. (1995). In the name of God:
A profile of religion-related child abuse. Journal of Social Issues, 51 (2), 85-111.
Bottoms, B. L., Shaver, P. R., & Goodman, G. S. (1996). An analysis of ritualistic
and religion-related child abuse allegations. Law and Human Behavior, 20 (1), 1-34.
Brassard, M., Germain, R., & Hart, S. (1987). Psychological maltreatment of children
and youth. New York: Pergamon Books.
Brown, J., Cohen, P., Johnson, J. G., & Smailes, E. M. (1999). Childhood abuse and
neglect: Specificity of effects on adolescent and young adult depression and suicidality.
Journal of American Academy of Child and Adolescent Psychiatry, 38, 1490-1505.
Bryant, S. L., & Range, L. M. (1997). Type and severity of child abuse and college
studentslifetime suicidality. Child Abuse & Neglect, 21 (12), 1169-1176.
Capps, D. (1992). Religion and child abuse: Perfect together. Journal for the
Scientific Study of Religion, 31 (1), 1-14.
Capps. D. (1995). The childs song: The religious abuse of children. Louisville, KY:
Westminster John Knox Press.
Religion and child abuse
Claussen, A. H., & Crittenden, P. M. (1991). Physical and psychological
maltreatment: Relations among types of maltreatment. Child Abuse and Neglect, 15, 5-18.
Derogatis, L. R., & Spencer, P. M. (1982). The Brief Symptom Inventory (BSI):
Administration, scoring, & procedures manual - I. Johns Hopkins University School of
Dodge, K. A., Pettit, G. S., & Bates, J. E. (1997). How the experience of early
physical abuse leads children to become chronically aggressive. In Developmental
perspectives on trauma: Theory, research, and intervention (Vol. 8, pp. 263-288).
Rochester, NY: University of Rochester Press.
Doxey, C., Jensen, L., & Jensen, J. (1997). The influence of religion on victims of
childhood sexual abuse. International Journal for the Psychology of Religion, 7, 179-186.
Duncan, R. (1999). Childhood maltreatment and college drop-out rates: Implications
for researchers and educators. Presented at the 7th Annual Colloquium of the American
Professional Society on the Abuse of Children, San Antonio, TX.
Durkheim, E. (1915/1965). The elementary forms of the religious life. New York:
Free Press.
Ellison, C. G., & Sherkat, D. E. (1993). Conservative Protestantism and support for
corporal punishment. American Sociological Review, 58 (1), 131-145.
Ellison, C. G., Bartkowski, J. P., & Segal, M. L. (1996a). Do conservative Protestant
parents spank more often? Further evidence from the national survey of families and
households. Social Science Quarterly, 77, 663-673.
Ellison, C. G., Bartkowski, J. P., & Segal, M. L. (1996b). Conservative Protestantism
and the parental use of corporal punishment. Social Forces, 74 (3), 1003-1028.
Epstein, M. E., & Bottoms, B. L. (2002). Forgetting and recovery of abuse and
trauma memories: Possible mechanisms. Child Maltreatment, 7, 210-225.
Flynn, C. P. (1996). Normative support for corporal punishment: Attitudes,
Religion and child abuse
correlates, and implications. Aggression and Violent Behavior, 1 (1), 47-55.
Friesen, J. G., Jr. (1992). Ego-dystonic or ego-alien: Alternate personality or evil
spirit? Journal of Psychology and Theology, 20, 197-200.
Garbarino, J. (1980). Defining emotional maltreatment: The message is the
meaning. Journal of Psychiatric Treatment and Evaluation, 2, 105-110.
Garbarino, J., & Vondra, J (1987). Psychological maltreatment: Issues and
perspectives. In M. Brassard, R. Germain, & S. Hart (Eds.)., Psychological maltreatment of
children and youth (pp. 25-44). New York: Pergamon.
Goodman, G. S., Bottoms, B. L., Redlich, A., Shaver, P. R., & Diviak, K. (1998).
Correlates of multiple forms of victimization in religion-related child abuse cases. Journal of
Aggression, Maltreatment, and Trauma, 2, 273-295. [Reprinted in B. B. R. Rossman & M. S.
Rosenberg (Eds.) (1998). Multiple victimization of children: Conceptual, developmental,
research, and treatment issues (pp. 273-295). Binghamton, NY: Haworth.]
Gorsuch, R. L. (1988). Psychology of religion. Annual Review of Psychology, 39,
Greven, P. (1991). Spare the child: The religious roots of punishment and the
psychological impact of physical abuse. New York: Knopf.
Hunsberger, B. (1989). A short version of the Christian orthodoxy scale. Journal for
the Scientific Study of Religion, 28, 360-365.
Jackson, S., Law, L., Thompson, R.A., Christiansen, E. H., Colman, R. A., & Wyatt,
J. (1999). Predicting abuse-prone parental attitudes and discipline practices in a
nationally representative sample. Child Abuse & Neglect, 23 (1), 15-29.
Johnson, B. W., & Eastburg, M.C. (1992). God, parent and self concepts in abused
and nonabused children. Journal of Psychology and Christianity, 11 (3), 235-243.
Kane, D., Cheston, S. E., & Greer, J. (1993). Perceptions of God by survivors of
childhood sexual abuse: An exploratory study in an underresearched area. Journal of
Religion and child abuse
Psychology and Theology, 21 (3), 228-237.
Keppel, G. (1982). Design and analysis: A researcher's handbook. Englewood Cliffs,
NJ: Prentice-Hall.
Kirkpatrick, L. A. (1997). A longitudinal study of changes in religious belief and
behavior as a function of individual differences in adult attachment style. Journal for the
Scientific Study of Religion, 36, 207-217.
Kirkpatrick, L. A. (1998). God as a substitute attachment figure: A longitudinal study
of adult attachment style and religious change in college students. Personality and Social
Psychology Bulletin, 24, 961-973.
Kolko, D. J. (2002). Child physical abuse. In J. E. B. Myers, L. Berliner, J. Briere, C.
T. Hendrix, C. Jenny, & T. A. Reid. The APSAC handbook on child maltreatment (pp. 21 _
54). Thousand Oaks, CA: Sage.
Lawson, R., Drebing, C., Berg, G., Vincellette A., & Penk, W. (1998). The long term
impact of child abuse on religious behavior and spirituality in men. Child Abuse
& Neglect, 22 (5), 369-380.
Levitan, R. D., Parikh, S. V., Lesage, A. D., Hegadoren, K. M., Adams, M., Kennedy,
S. H., & Goering, P. N. (1998). Major depression in individuals with a history of childhood
physical or sexual abuse: Relationship to neurovegatative features, mania and gender.
American Journal of Psychiatry, 155 (12), 1746-1752.
Lung, C. T., & Daro, D. (1996). Current trends in child abuse reporting and fatalities:
The results of the 1995 annual fifty state survey. Chicago: National Committee to Prevent
Child Abuse.
Luntz, B., & Widom, C. S. (1994). Antisocial personality disorder in abused and
neglected children grown up. American Journal of Psychiatry, 151, 670-674.
Lynch, M., & Cicchetti, D. (1998). An ecological-transactional analysis of children
and contexts: The longitudinal interplay among child maltreatment, community violence, and
Religion and child abuse
childrens symptomatology. Development and Psychopathology, 10 (2), 235-257.
Maurer, A. (1982). Religious values and child abuse. Child & Youth Services, 4, 57-
Malcarne, V. L., & Burchard, J. D. (1992). Investigations of child abuse/neglect
allegations in religious cults: A case study in Vermont. Behavioral Sciences & the Law,
10(1), 75-88.
Maxfield, M. G., & Widom, C. S. (1996). The cycle of violence: Revisited six years
later. Archives of Pediatrics and Adolescent Medicine, 150, 390-395.
Milner, J. S. (2000). Social information processing and child physical abuse: Theory
and research. In D. J. Hansen (Ed.), Nebraska symposium on motivation, Vol. 46, 1998:
Motivation and child maltreatment (pp. 39-84). Lincoln: University of Nebraska Press.
Nelsen, H. M., & Kroliczak, A. (1984). Parental use of the threat God will punish:
Replication and extension. Journal for Scientific Study of Religion, 23 (3), 267-277.
Neufeld, K. (1979). Child-rearing, religion, abusive parents. Religious Education, 74
(3), 235-243.
Oats, R. K., Forrest, D., & Peacock, A. (1985). Self-esteem of abused children. Child
Abuse & Neglect, 9, 159-163.
Pagelow, M. D., & Johnson, P. (1998). Abuse in the American family: The role of
religion. In A. L. Horton & J. A. Williamson (Eds.), Abuse and religion: When praying isn't
enough. (pp. 1-12).
Pargament, K. I. (1997). The psychology of religious coping. New York: Guilford.
Pelcovitz, D., Kaplan, S., Goldenberg, B., & Mandel, F. (1994). Posttraumatic stress
disorder in physically abused adolescents. Journal of American Academy of Child and
Adolescent Psychiatry, 33 (3), 305-312.
Quas, J. A., Bottoms, B. L., & Nunez, N. (Eds.) (2002). Linking Juvenile Delinquency
and Child Maltreatment: Causes, Correlates, and Consequences. Special issue of
Religion and child abuse
Children's Services: Social Policy, Research, and Practice, 5(4).
Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research
in the general population. Journal of Applied Psychological Measurement, 1, 385-401.
Rice, R. R., & Annis, A. W. (1992). A survey of abuse in the Christian Reformed
Church. Grand Rapids, MI: Social Research Center of Calvin College.
Rosenberg, M. J. (1965). Society and the adolescent self-image. Princeton, NJ:
Princeton University Press.
Rossetti, S. J. (1995). The impact of child sexual abuse on attitudes toward God and
the Catholic Church. Child Abuse & Neglect, 19 (12), 1469-1481.
Ryan, P. L. (1998). Spirituality among adult survivors of childhood violence: A
literature review. The Journal of Transpersonal Psychology, 30 (1), 39-51.
Sheshkin, D. J. (2000). Handbook of parametric and nonparametric statistical
procedures (2nd ed.). Boca Raton, FL: Chapman & Hall/CRC.
Shor, R. (1998). The significance of religion in advancing a culturally sensitive
approach towards child maltreatment. Families in Society, 79 (4), 400-409.
Silvern, L., Waelde, L. C., Baughan, B. M., Karel, J., Kaersvang, L. J. (2000). Two
formats for eliciting retrospective reports of child sexual and physical abuse: Effects on
apparent prevalence and relationships to adjustment. Child Maltreatment, 5, 236-250.
Silverman, A. B., Reinherz, H. Z., & Giaconia, R. M. (1996). The long-term sequelae
of child and adolescent abuse: A longitudinal community study. Child Abuse & Neglect, 8
(8), 709-723.
Simons, R. L., Whitbeck, L. B., Conger, R. D., & Chyi-In, W. (1991).
Intergenerational transmission of harsh parenting. Developmental Psychology, 27, 159-171.
Straus, M. (1994). Beating the devil out of them: Corporal punishment in American
families. San Francisco: Jossey-Bass
U.S. Department of Health and Human Services. (2000). Child maltreatment 1998:
Religion and child abuse
Reports from the states to the National Child Abuse and Neglect Data System. Washington,
DC: Government Printing Office.
Wiehe, V. R. (1990). Religious influence on parental attitudes toward the use of
corporal punishment. Journal of Family Violence, 5, 173-186.
Winton, M. A., & Mara, B. A. (2001). Child abuse and neglect: Multidisciplinary
approaches. Needham Heights, MA: Allyn & Bacon.
Wolfe, D. (1999). Child Abuse: Implications for child development and
psychopathology (2nd ed.). Thousand Oaks, CA: Sage.
Religion and child abuse
Table 1.
Characteristics of the Abuse as a Function of Condition (Percentage of Cases)
Religion-related Non-religion-related
Specific form of the physical abuse
Hit, slapped, spanked, etc. 69% 72%
Punished with objects 77% 57%
Threatened with a weapon 12% 2%
Other 0% 2%
Frequency of abuse
Once 16% 4%
Twice 8% 4%
3 - 5 times 12% 39%
6 - 10 times 8% 30%
11 - 20 times 16% 7%
21 or more times 40% 15%
Religion and child abuse
Table 2.
Victim and Perpetrator Characteristics as a Function of Condition
Religion-related Non-religion-related
Victimsreligious affiliation (percentage of victims)1
Fundamental protestant 4% 2%
Protestant 17% 20%
Catholic 38% 35%
Christian 17% 13%
Measures of religiosity (mean ratings)
Importance of religious beliefs 3.80 (1.12) 3.42 (1.18)
Importance of religion to self-concept 3.04 (1.17) 2.77 (1.16)
Frequency of attending religious services 4.20 (1.83) 3.58 (2.10)
Frequency of prayer 5.96 (2.57) 5.42 (2.80)
Christian Orthodoxy Scale 33.31 (10.09) 31.80 (10.56)
Relationship of perpetrators to victims (percentage of cases)2
Parent or step-parent 73% 91%
Grandparent 12% 7%
Other relative 15% 4%
Teacher 19% 2%
Perpetratorsreligious affiliation (percentage of cases)3
Fundamental protestant 0% 0%
Protestant 23% 7%
Catholic 59% 52%
Christian 9% 17%
Note. Standard deviations are noted parenthetically. For religiosity measures, higher values indicate more
1 The two victim samples also included Hindu (N = 2), Islam (N = 1), Judaism (N = 1), atheist (N = 4), and
other (N = 6).
2 In addition, a sibling and some other person were each the abuser in 2.2% of the non-religion-related
abuse cases.
3 The two perpetrator samples also included Hindu (N = 2), Islam (N = 2), and atheist (N = 6).
Religion and child abuse
Table 3.
Mean Scores on the Brief Symptom Inventory as a Function of Condition
Religion-related Non-religion-related Control
abuse abuse
Global Severity Index 1.16a (.84) .71b (.62) .78b (.55)
Positive Symptom Total 28.88a (16.6) 20.65b (13.07) 23.26ab (12.25)
Positive Symptom Distress Index 1.96a (.51) 1.73a (.65) 1.64a (.48)
Primary Symptom Dimension Subscales
Depression 1.15a (.85) .80a (.88) .85a (.76)
Anxiety 1.13a (1.11) .59b (.68) .73b (.60)
Phobic anxiety .72a (.96) .29b (.54) .37b (.52)
Obsessive compulsiveness 1.54a (.96) 1.06a (.85) 1.26a (.93)
Somatization .81a (1.04) .41b (.51) .52ab (.53)
Hostility 1.39a (1.21) .82b (.75) .95b (.68)
Paranoid ideation 1.41a (1.06) .93b (.94) .89b (.67)
Psychoticism 1.06a (.88) .59b (.74) .64b (.67)
Interpersonal sensitivity 1.31a (1.08) .95a (.95) .95a (.86)
Note. Standard deviations are noted parenthetically. Means within a row that share letters in their
subscripts do not differ from each other at p < .05. Higher values indicate greater distress.
... Living with this additional psychological burden would have the putative impact of increasing self-isolation to avoid social embarrassment. Bottoms et al. (2004) examined the impact of religion-related child abuse using a control group design; the findings highlighted that victims/survivors of religion-related physical child abuse demonstrated greater psychological symptomology than victims/survivors of other physical abuse or a non-abused control group. The authors concluded that religion-related abuse adds "an additional layer of complexity and harm to the experience" (Bottoms et al., 2004, p. 107). ...
... Despite some participants reporting actively re-connecting with their faith in adulthood, the initial damage caused to their belief and practice was substantial. With faith recognised as providing meaning to life itself (Bottoms et al., 2004) and being integral to the Muslim identity, the damage was not only profound but also deeply rooted, causing individuals to leave the faith altogether for some time. This created an acute shift in their identity as Muslims, which further cast them outside their communities. ...
Full-text available
The abuse of individuals by religious authority figures has generated considerable political, civic and media attention. To date, much of this focus has been on Catholic and Anglican priests, although instances in the Buddhist community have also emerged. This paper presents an analysis of the experiences of individuals (n = 6) who were victims of abuse by Imams (Muslim leaders) and/or Muslim faith teachers. Participants were interviewed and their accounts analysed using Interpretative Phenomenological Analysis (IPA). This paper presents two superordinate themes that emerged from the rich data set: (i) Toxicity of silence and (ii) Barriers to the acknowledgement of abuse. Findings and implications of the research are discussed in relation to facilitating the reporting of abuse perpetrated by religious authority figures in the Muslim community. The authors argue that the difficulties for victims in reporting abuse have exacerbated the impact of the abuse, affecting their wellbeing and their relationships with their families and their faith.
... However, we can add a few issues drawn from our own experience of working many years as pastoral workers with survivors of sexual abuse (Demasure, 2007). To make the abuse possible by convincing the child, or to justify abuse, some perpetrators make use of religious arguments (Bottoms, nielsen, murray, Filipas, 2003;mahoney, 2006). We call this religion-related abuse. ...
... Religious groups, not only priests or bishops play an active role in abusive child-rearing techniques, as they are often misguided in the way in which they interpret biblical verses or passages (Bottoms, Nielsen, Murray & Filipas, 2003). These groups often believe that physical beatings, help or assist the deliverance of the child from sin and evil. ...
Full-text available
Child abuse is a worldwide phenomenon that has in recent years received much attention While research and advocacy surrounding child abuse has gained momentum, it can also be expressed in unique ways, such as African children accused of witchcraft. Most research on children accused of witchcraft in however, appears in unpublished literature and as such, the expression and understanding thereof is limited. Belief in supernatural entities, like witches and witchcraft exist in many religious and cultural systems and are believed to be the agents of misfortune. Generally, a person accused of witchcraft is likely to experience physical and psychological harm, with historic and contemporary evidence supporting this. This study aims to describe children accused of witchcraft, through an analysis of secondary data. The overall was to examine news reports, advocacy agencies and various commentaries on children accused of witchcraft toward providing firstly, a description of an expression of child abuse in Africa, which seems to be underreported; and secondly, the representation of religio-cultural beliefs in ways that justify the abuse of children. Data took the form of news reports, as well as information from web pages and advocacy agencies. A content analysis of secondary data was undertaken, and three themes identified are discussed namely, 1) protagonists of violence; 2) consequences of witchcraft accusations and 3) religio-cultural beliefs and justification of accusation. This research contributes to understanding an expression of child abuse, as well as indicates how religio-cultural beliefs may be represented in a way that is used as a justification to violence..
... For example, Falsetti, Resnick, and Davis (2003) suggest that those who experience multiple traumas have a greater struggle with grasping the meaning of what happened to them and the meaning of their lives. Furthermore, studies have shown that victims of religion-related abuse (i.e., perpetrated by a representative for the religious community) display more depressive symptoms, more anxiety, hostility, psychoticism, paranoid ideation, and somatization than individuals who suffered either no abuse or an abuse that was not religion-related (Bottoms, Nielsen, Murray, & Filipas, 2003). It seems that, in some cases, religiousness may be a moderator of the development of post-traumatic stress and other mood disorder symptoms and their associated disorders; found on the Axis I in diagnostic manual DSM (Walker, Reid, O'Neill, & Brown, 2009). ...
Every year thousands of people are subjected to sexual abuse and a substantial part of that abuse occurs within the family. Many victims of sexual abuse find it hard to disclose being a victim, especially if someone close, that the victim loves and trusts, is the perpetrator of the abuse. For the victim who has a religious faith, besides the psychological consequences suffered, the victim’s faith may also become traumatized. This chapter will outline the psychological consequences of being sexually abused as well as the potential consequences on the victims’ faith, highlighting the search for meaning and the effect of sexual abuse on an individual’s relation to God. Thereafter, commonalities and diversities in religious coping across religions will be outlined. Specific challenges will be discussed, highlighting reactions from faith communities, the role of family structure, sexual moral, shame, and lack of knowledge. Lastly, conclusions and practical implications will be outlined.
Full-text available
Child maltreatment is a poorly recognized phenomenon worldwide, and many pediatric healthcare professionals do not receive adequate training in child abuse and neglect evaluations. In response to the paucity of child maltreatment literature on low- to middle-income countries, this study aims to determine the education and training needs of medical professionals at the Centre Hospitalier Universitaire de Kigali (CHUK) with respect to child abuse and neglect identification, investigation, evaluation, and case management. Data collection took place in July and August 2019 at CHUK in Kigali, Rwanda. A mixed-methods approach was employed, including semi-structured qualitative interviews with sixteen healthcare professionals (3 pediatricians, 7 pediatric residents, 2 nurses, and 4 social workers) at CHUK and a systematic chart review of child maltreatment cases evaluated at the hospital from July 2015 to July 2019. Interviews with professionals revealed perceived gaps in training regarding child maltreatment case evaluation, reporting, and management. Study participants noted the need for standardized, hospital-wide protocols for the handling of confirmed abuse and neglect cases as well as expanded curriculum on the topic throughout professional education. Chart reviews demonstrated inconsistent and sparse documentation of maltreatment cases in hospital records. In conjunction with our findings, we provide informant-based suggestions for the improvement of child abuse and neglect case management at CHUK, including consistent training modules, inter-departmental collaboration, and systematic documentation. Beyond the hospital, participants widely agreed that child maltreatment awareness and prevention measures should be implemented at the community level with professionals from CHUK getting involved in local efforts.
Full-text available
While there are no official data and published studies on clergy-perpetrated sexual abuse (CPSA) from Ghana, local media reports continue to show worrying trends of the phenomenon. We drew on 73 media reports from January 2000 to March 2019, to describe the offence characteristics and profiles of the perpetrators and survivors of CPSA in Ghana. The findings showed females aged 10-19 as predominant survivors. The perpetrators were all males found guilty of lone rape, incest, defile-ment, indecent assault, sodomy, attempted rape, or gang rape. A preventive measure could involve streamlining the recruitment, training, and leadership structures of the church.
Full-text available
One-third of all children worldwide experience trauma. This includes domestic violence, serious illness, physical or mental abuse, and death of a parent, among other traumatic experiences. Children who experience trauma are more likely to experience other mental health disorders and often reject the religious and spiritual institutions of their upbringing. While cognitive-based therapy and other traditional psychological methods are crucial to the care of these children, incorporating their spirituality can be a helpful adjunct for their healing. Further, religious communities can play a critical role in reintroducing healthy forms of relationship building and community attentiveness, while providing a safe environment for coping.
Full-text available
The article addresses the responses of the government of Zimbabwe and its proxies to a letter issued by Catholic bishops on 14 August 2020, entitled ‘The march is not ended’. The response to the letter presents an ambivalent view of the nexus of the state, law and religion in Zimbabwe, which needs to be teased out and challenged in order to reinvent a democratic nation. This theoretical article taps into decoloniality theory to problematise state responses to the letter. The articles discuss responses by government actors, such as Monica Mutswanga and Nick Magwana, and regime enablers, such as Mutendi and Wutawunashe. The responses indicate the weaponization of religion and law to silence dissenting voices, and to enact a skewed nationalism. The article argues that, in the context of crisis, authoritarianism, and abuse of human rights, politicians and religious leaders should position their narratives to enact social justice, ontological density, peace and accountability, as a healing process to usher in sustainable development.
Full-text available
In this paper, I interrogated the Gabola church in terms of its origins, purpose and its distinctiveness as a postcolonial manifestation of freedom of religion in South Africa. I answered two questions, is Gabola church a representation of a decolonial church and could it be a manifestation of trajectories of the postcolonial ill-defined freedom of religion? In responding to these questions, I used decoloniality, a theory whose agenda among many others is geared to usher a future free from oppression, where all can participate in modernity and in postmodernity. Data was generated through participatory action research. The approach enabled us to unearth the theology of Gabola, philosophy and the gap they seek to fill in the religious space. Ten Gabola church members and five church members from a mainline Christian movement participated in this research. The findings indicated that Gabola church presents a new religious movement that is socially inclusive, that seeks to promote social justice and social transformation. On the other hand, the research revealed that the lack of a regulating body for religious movement is the reason for the rise of questionable movements such as Gabola, a serious threat in the praxis of the Christian faith. To this end, I concluded that while freedom of religion is a good idea in line with the decolonial move, there is a need for participative and collaborative regulation of religious movement to eliminate criminal elements that overshadowed the beauty of religion manifested through ‘unthinkable’ ethical irregularities.
Nunn (1964) presented data to give limited support to the thesis that in order to secure compliance from their children, powerless parents tell them "God will punish them if they misbehave." Youths who subscribe to this belief have higher levels of self-blame and greater feeling they should be obedient than youths who do not. Using data from Minnesota children in grades four through eight, we find little support for the hypothesis that it is powerless parents who use this technique. However, like Nunn, we find the belief that "God punishes youths who are bad" is positively associated with self-blame and obedience. We introduce an additional variable, i.e., subscribing in general to the image of God as punishing or angry. Using this variable in relation to the belief that "God punishes youths who are bad" gives four cells, one of which is the image of God as angry/punishing but rejecting the belief that "God punishes youths who are bad" (this combination is one of a malevolent but impersonal God.). Youths subscribing to this combination have the lowest levels of self-blame and feeling of obedience to parents. Our findings agree with Nunn's linkage of parental "coalition with God" and the child's benevolent image of God; the malevolent view is linked with a lower level of internalization.