Definitions of psychological abuse are reviewed and a new definition
proposed, operationalized as an extension of an existing measure of
childhood, the Childhood Experience of Care and Abuse (CECA). This
semistructured, investigator-based interview is designed for use with
adults to collect retrospective accounts of childhood adverse
experience. The CECA extension identifies nine subtypes of psychological
abuse, with a single global severity rating. The definition is clearly
differentiated from other adverse experiences with emotionally abusive
elements such as parental antipathy (hostile parenting),
neglect, and role reversal. Examples are given. A
community-based series of 301 women were interviewed using the extended
CECA to gather a range of experiences of childhood maltreatment on which
to devise the new measure. Interrater reliability was satisfactory, and
several features of psychological abuse were examined, including its
prevalence, frequency, and the characteristics of the abused child and
perpetrator. The new scale is discussed in terms of its potential use
not only by researchers but also by practitioners such as clinicians and
social workers in the child protection field in order to more accurately
identify and assess multiples of abuse in childhood. A companion article
(Bifulco, Moran, Baines, Bunn, & Stanford, 2002) examines the
relationship of psychological abuse to adult major depression. (Bulletin
of the Menninger Clinic, 66,
negative effects of psychological abuse in childhood has increasingly
emerged in recent years. Psychological abuse defined in terms of abuse
of children's mental and emotional processes has been linked with
childhood outcomes such as impaired emotional, social, and cognitive
development, including helplessness and lowered self-esteem (Cerezo &
Frias, 1994), aggression, emotional unresponsiveness and neuroticism
(Black, Smith Slep, & Heyman, 2001), dependency, incompetence, and
educational failure (see Skuse & Bentovim, 1994, for a summary). It has
also been linked with psychiatric disorders in childhood such as major
depression and dysthymia (Kaufman, 1991; Stone, 1993), and has been
implicated in impaired physical development (Skuse, 1989). Of the few
studies documenting its long-term effects, Mullen and colleagues report
that childhood psychological abuse relates to low self-esteem, eating
disorders, depressive illness, and suicidal behaviors in adulthood
(Mullen, Martin, Anderson, Romans, & Herbison, 1996), psychological
distress (Pitzner, McGarry-Long, & Drummond, 2000), and poor adult
relationships (Varia & Abidin, 1999).
Yet, until most recent times, psychological abuse has rarely formed the
reason for referral to child protection services, with children who are
subjected to neglect, physical, or sexual abuse being far
more likely to be registered (Creighton, 1992). This is partly due to
the enormous pressure on child protection staff to focus on aspects of
child maltreatment such as gross neglect or extreme
physical abuse, which lead directly to child fatalities (Reder, Duncan,
& Gray, 1993). Prevalence rates are thus likely to be underestimates,
but they suggest that psychological abuse varies from 113 to 257 per
1000 children, depending on the definition applied (Fortin & Chamberland,
Another reason why psychological abuse remains a relatively neglected
area of childhood experience is the problem of its definition. The term
mental injury first appeared in child protection legislation in the
as early as 1974 (Edmundson & Collier, 1993). However, the concept has
remained elusive despite greater awareness of it as a form of
maltreatment (Black et al., 2001).
(1987) suggests that research in the area of psychological abuse has
been hindered by a lack of consensus regarding an operational
definition. Consequently, few if any satisfactory measures of
psychological abuse have been developed. The current article sets out to
develop an operational definition of psychological abuse and describes a
new interview scale of psychological abuse for use in retrospective
studies with adults. A companion article examines the long-term
consequences of such abuse in terms of depressive disorder (Bifulco,
Moran, Baines, Bunn, & Stanford, 2002).
Some authors prefer to distinguish psychological abuse from emotional
abuse based on the type of damage the abuse is thought to inflict (e.g.,
O'Hagan, 1993, 1995). However, it is clear from the literature that many
use the term interchangeably, with American reports typically using the
former term and British reports the latter (Edmundson & Collier, 1993).
In the present discussion the two labels are not distinguished, but the
term psychological abuse preferred.
survey of psychological abuse by Thompson and Kaplan (1996) identified
four key features common to definitions of psychological abuse: adverse
parental behavior, a sustained pattern of negative interaction, child
vulnerabilities, and damage in terms of emotional and psychological
functioning. These features provide a useful framework for discussion of
the components required for an adequate definition of psychological
abuse. Each is discussed, and the implications in terms of the
development of a new measure of psychological abuse are outlined.
following types of parental behavior are viewed as constituting
psychological abuse: rejecting, isolating, terrorizing, ignoring, and
corrupting a child or youth (Garbarino, Guttman, & Seeley, 1986). To
this list Baily and Baily (1986) have added excessive threats, refusal
of psychological treatment, sexual exposure and exploitation, denial of
opportunities to grow socially and emotionally, singling out one child
in the family to punish or criticize, and unrealistic expectations.
Glaser's scheme includes persistent negative attributions or
misattributions to the child, failure to recognize the child's
individuality, or inconsistent developmental expectations (Glaser &
Prior, 1997). Combining these and other definitions, Burnett (1993)
offered the following nine parental acts as constituting types of
psychological abuse: confining to a small space, public humiliation,
Cinderella syndrome, severe verbal abuse, coercing into delinquency,
threatening a child, refusal of psychiatric treatment, not allowing
social and emotional growth, and not providing a loving home.
Many of these adverse behaviors overlap extensively with alternative
categories of child maltreatment, and the distinction between them
requires clarification. For example, ignoring a child, denial of social
opportunities, or refusal of psychiatric treatment cannot readily be
distinguished from neglect as defined by other taxonomies
of childhood maltreatment (e.g., Childhood Experience of Care and Abuse
[CECA]; Bifulco, Brown, & Harris, 1994; Bifulco & Moran, 1998).
Similarly, ongoing negative interaction involving unrealistic
expectations, rejecting, or singling out one child for criticism cannot
readily be distinguished from parental antipathy. In the same vein,
Cinderella syndrome has clear overlaps with the exploitation involved in
role reversal, where this child is pressured to take on adult roles. The
extent to which the distinction between psychological abuse and other
forms of maltreatment is one of severity/intensity or of the specific
type of parental behavior needs to be clarified. The failure to reliably
distinguish specific types of abuse can lead to difficulties when
attempting to establish specificity of outcome of such experiences
(Crittenden, 1985). In addition, some combinations of types of abusive
experiences may be more damaging than others, and estimating the precise
number of multiples of abusive experience may prove critical in
explaining outcome (Claussen & Crittenden, 1991; Ney, Fung, & Wickett,
1994). In addition, precision about multiplicity of abuse may prove
critical in child protection settings or in legal contexts. Hence a
system of measurement is needed that assesses psychological abuse
together with each other form of maltreatment separately in terms of
perpetrator behaviors--this while recognizing that psychological and
other forms of abuse commonly co-occur in the same settings and may at
times exist as features of the same actions.
This point is developed by Barnett, Manly, and Ciccetti (1991) in
seeking to define psychological abuse: "an adequate operational
definition cannot be developed separately from definitions of other
forms of maltreatment" (p. 20). They identify six types of child
maltreatment, including psychological maltreatment, and allow for acts
of maltreatment to be categorized more than once rather than forcing
classification into a single category. This latter point is significant
given that, for example, an instance of neglect that has
psychologically abusive elements might prove to be more harmful in
combination than an act that is solely one of "pure" neglect
or "pure" psychological abuse. The ability to capture the difference
between simple and compound abuses in a measurement system is therefore
important. The new interview scale to be described allows for the
classification of acts of psychological abuse separately, and in
addition to, a range of other forms of maltreatment. It also allows for
the same abusive act at times to be cross-referenced between categories
addition to controversy surrounding the types of acts to be included in
an operational definition of psychological abuse, there is also much
debate concerning the significance of the intentions of perpetrators.
Some researchers argue that motive is highly relevant to the
categorization of such abuse, whereas others take the view that it may
be of clinical importance, but is probably not relevant to the severity
of maltreatment (e.g., Hart & Brassard, 1991). McGee and Wolfe (1991)
suggest that psychologically abusive behavior should be defined
irrespective of parental motive. They suggest that parental behavior be
measured on a continuum ranging from mild to dangerous forms, and that
parental intent will typically influence the severity along this
continuum. In addition, they suggest that the child's perceptions of
parental intent will influence the impact of the parental act rather
than its nature, and that such perceptions should be measured separately
from the parental behavior itself.
O'Hagan (1993) also recommends that psychological abuse be defined
independently of the perpetrator's intentions. He suggests that failure
to do so could lead to the dangerous position whereby if perpetrators
deny an intention to abuse or awareness of the abuse (e.g., if drunk),
their actions could be argued to be nonabusive. Yet the same argument is
rarely applied to physical abuse, where the practice of harsh physical
discipline is condemned whether or not the perpetrators construe it as
beneficial to the child, or whether injuries inflicted on the child
occur while the parent is drunk. The physical risks to the child hold
equally in both scenarios, and the same could be concluded with
psychological abuse. In the proposed new measure, therefore, perpetrator
intention is not regarded as a necessary characteristic of psychological
abuse. However, evidence of perpetrator behavioral strategies that have
all the appearance of premeditation and hence intent can be used in
deciding on inclusion of incidents as psychological abuse.
Many definitions of psychological abuse also limit the inclusion of
abusive acts to those committed by parents. Limiting the category of
those included as perpetrators omits a host of potential perpetrators,
such as other caregivers, relatives, teachers, siblings, and even peers
(O'Hagan, 1995; Shaver, Goodman, Rosenberg, & Orcutt, 1991; Whipple &
Finton, 1995). Avoiding this definitional limitation has proved critical
in the field of sexual abuse, where abuse from nonrelated adult males is
common (Bifulco & Moran, 1998). Although the perpetrators most likely to
be influential are primary caregivers such as parents, it is possible
that significant short- and long-term damage may occur at the hands of
others in the wider social network. Hence the proposed new measure of
psychological abuse, while routinely assessing abuse from parents and
surrogate parents, also considers abusive acts from other sources.
Another feature of the definition under debate is whether the adverse
parental behavior typically involves a sustained pattern of interaction
(Thompson & Kaplan, 1996). Some studies stress that repeated exposure to
instances of adverse parental acts is required before an act can be
defined or included as psychologically abusive (e.g., Burnett, 1993;
O'Hagan, 1993, Glaser & Prior, 1997). Others allow single instances of
adverse behavior to qualify as psychological abuse (e.g., Baily & Bally,
1986) and suggest that the isolated nature of these instances should be
taken into account in rating its severity (McGee & Wolfe, 1991). Rather
than decide on a priori grounds where the frequency threshold for
inclusion lies, the new measure includes single adverse acts in addition
to those of a more sustained nature. However, frequency (in addition to
intensity and variety of modes of psychological abuse) is implicated in
third element common to psychological abuse definitions involves
features of the child that may make the child particularly vulnerable to
damage: "low IQ, absence of nurturing adults, a child's attribution of
parental misdeeds as being malevolent and a child's particular
developmental level" (Thompson & Kaplan, 1996, p. 144). These are indeed
factors that might render a child more vulnerable to the effects of
psychologically abusive acts. However, Garbarino (1991) argues that
psychological abuse, as with other abuses such as sexual abuse, should
be defined independently of child characteristics. In defining sexual
abuse, standards of behavior that are deemed appropriate or
inappropriate are delineated by society. Thus, performing sexual acts
with a minor is seen as inappropriate adult behavior, regardless of the
child's characteristics. Garbarino argues that taking into account the
child's vulnerabilities in a definition of abuse implies that a hardier
child subjected to the same abusive act as a more vulnerable child would
be considered less abused. He (1991) finds such a position untenable:
"The logical (and ethically appalling) implication is that the standards
for dealing with a resilient child offer less protection than the
standards of care for a vulnerable child (the former can absorb more
psychic assault than the latter)" (p. 47). The new measure to be
described in the present report therefore makes no reference to specific
child vulnerabilities and limits its definition to a description of
perpetrators' acts rather than child characteristics.
Definitions that incorporate outcome factors such as emotional or
psychological damage to the child are problematic and potentially
tautological (McGee & Wolfe, 1991). An operational definition useful in
empirical research should distinguish the predictor (i.e. parental
behavior) from the criterion (i.e., psychological harm). Therefore there
is a need to develop an operational definition independent of outcome so
that the effects of the abuse can be established empirically. A useful
distinction used in deriving definitions of childhood maltreatment that
avoids such circularity is the "consequences" model versus the "standard
of care" model (Garbarino, 1991). Definitions of abuse that incorporate
outcomes for the child are examples of consequences models, as in the
case of definitions of physical abuse that depend on evidence of
physical injury. These are perhaps more relevant to child protection
practice than research criteria. For example, in the
mandated intervention requires evidence of "significant harm," which
includes both parental action and observable harm to the child (Glaser &
Prior, 1997). In contrast, a standard of care model defines abuse in
terms of what society deems to be unacceptable treatment of a child,
whether evidence of damage to the child is present or not: "The
definition, while informed by theory and research, must be formulated
according to societal standards of appropriate parental behavior"
(Barnett et al., 1991, p. 24). In this vein, McGee and Wolfe (1991)
argue for a probabilistic model of psychological abuse, based on the
potential for harm to a child rather than the observable harm. In the
new measure of psychological abuse outlined in this report, actual or
observable harm is irrelevant to the judgment of whether an act is one
of abuse or not: it is the likelihood of damage that is assessed.
first aim of this article is to develop an operational definition of
psychological abuse. Given some of the difficulties just described, this
might be considered "working" rather than "definitive" with results
contributing to a later "firming up" of the definition and replications
in new series. In sympathy with Haugaard's position (1991), we agree
that it may be worth delaying the formulation of a final definition of
psychological abuse until results from empirical studies are sufficient
to clarify which parent-child interactions result in psychological harm.
The definition we offered here is derived in part from existing
literature on the topic, but also from scrutiny of psychological abuse
examples gleaned inductively from hundreds of retrospective accounts of
childhood gathered in the course of a program of research stretching
back over many years (Bifulco & Moran, 1998). The proposed working
definition is as follows:
Psychological abuse is concerned with cruelty demonstrated by verbal and
nonverbal acts, repeated or singular, intended or not, from a close
other in a position of power or responsibility over the child. These
have potential for damaging the social, cognitive, emotional, or
physical development of the child and are demonstrated by behaviors
which are humiliating/degrading, terrorizing, extremely rejecting,
depriving of basic needs or valued objects, inflicting marked
distress/discomfort, corrupting/exploiting, cognitively disorientating,
or emotionally blackmailing. The perpetrator behaviors involved in
psychological abuse exclude physical or sexual attack, although
psychological abuse may accompany these. They also exclude those forms
of maltreatment identified as neglect, antipathy, role
reversal, high discipline, or lax supervision, as these fall into
terms of differentiation from existing abuse categories, the most
controversial may be that of antipathy--a negative ongoing relationship
with the parent where the latter is critical, cold, hostile, or
rejecting on a day-to-day basis. Although antipathy is identified as
emotional abuse in other categorizations, it is contrasted by its
established definition within the lack of care domain (Andrews & Brown
1988; Bifulco & Moran, 1998; Parker, Tupling, & Brown, 1979), its
overlap with more normalized family behavior, and its rare inclusion as
behavior relevant for mandated intervention and child protection, which
typically does not reach the identifiable abusive peaks and categories
of psychological abuse outlined in the literature and in the present
report. Therefore verbally aggressive parental behavior and other
negative interaction with parents is differentiated from psychological
abuse in this report.
second aim of this article is to introduce a new scale designed to study
psychological abuse in conjunction with other forms of neglect,
abuse, and poor parenting. There are relatively few
published measures of psychological abuse, and nearly all are
self-report questionnaires. An exception is the Psychological
Maltreatment Rating Scale (PMRS; Brassard, Hart, & Hardy, 1993), which
is designed for rating mother-child interaction during a 15-minute
videotaped teaching task. Other alternatives to self-report include
scrutiny of case files (McGee & Wolfe, 1991) and the multiclassification
of Barnett and colleagues (1991). A self-report measure that soleley
assesses psychological abuse is the Psychological Maltreatment Scales
(PMS; Briere & Runtz, 1988; Varia & Abidin, 1999). This measures uses a
Likert scale to endorse seven statements of parental behavior toward a
child under age 16 for an average year in childhood, including "yelled
at you," "criticized you," "made you feel guilty," and "ridiculed or
humiliated you." Most other measures are extensions to existing
childhood measures such as the Child Abuse and Trauma Scale (CATS;
Sanders & Becker-Lausen, 1995). This is a 38-item questionnaire for use
with adults that assesses various aspects of adversity in childhood.
Seven items have been incorporated, aimed at tapping psychological abuse
in childhood (Kent & Waller, 1998). Another measure modified to
incorporate childhood psychological abuse is the Child Abuse
Questionnaire (CAQ) as used by Gross and Keller (1992) to examine
psychological abuse in relation to depression, self-esteem, and
attribution styles. An Australian measure, the Comprehensive Child
Maltreatment Scales (CCMS; Higgins & McCabe, 2001), is a self-report
scale that covers five types of childhood maltreatment, including
psychological abuse. However, in addition to these extensions of
self-report measures adding specific items to reflect psychological
abuse, other investigators have used existing measures of neglect
or physical abuse to answer questions about psychological abuse. Thus a
widely used measure of care and control in childhood (Parental Bonding
Instrument) has been relabeled in terms of emotional abuse for some
analyses with no increase of items (Mullen et al., 1996). Similarly, the
Conflict Tactics Scale (CTS; Straus, 1979) has been used in terms of
subscales relating to parent-to-child verbal aggression in relation to
psychosocial problems in the children, but again without broader
psychologically abusive categories (Vissing, Straus, Gelles, & Harrop,
Examination of the items that have been added to existing measures of
childhood experience suggests that they may often be assessing other
related but previously categorized forms of maltreatment, such as
parental neglect or antipathy, which may lead to an
overinclusiveness in prevalence figures and to unreliable assessments of
the relationship of psychological abuse to other forms of abuse and
neglect. It may also confound results concerning the
differential effects of psychological and other abuse in relation to
outcomes such as depression in adulthood or suicidal behavior, as
discussed in a companion article (Bifulco et al., 2002).
proposed new measure differs from existing measures of psychological
abuse in its administration. It is a semistructured interview designed
for use with adults, and involves investigator-based judgements of
childhood history gathered retrospectively. Narrative, chronological
accounts of childhood experience are collected on the basis of extensive
probing questioning, and the classification and severity assessment are
determined by trained researchers according to predetermined criteria.
This has many similarities to a diagnostic clinical interview. The new
assessment is an extension of an existing measure of childhood, the CECA
(Bifulco, Brown, & Harris, 1994), a semistructured interviewing
instrument originally developed in the 1980s to assess experiences such
as physical and sexual abuse, neglect, antipathy, role
reversal, supervision, and discipline. As an addition to an existing
measure, the psychological abuse component has the benefit of being
assessed in relation to, and distinct from, other forms of abuse using
the same scaling procedures. As with the other forms of maltreatment
that the CECA covers, the new measure treats psychological abuse as a
continuous variable, as recommended by authors such as McGee and Wolfe
(1991). However, it also identifies categories of perpetrator behavior
to aid in identification of relevant items.
use of an interview rather than a questionnaire for the assessment of
psychological abuse has the advantage of flexibility in identifying
psychologically abusive behaviors which are notorious not only for being
varied but also at times idiosyncratic. The format of a semistructured
interview with open questions thus allows for exploration of both
typical and atypical parental behaviors, which may not lend themselves
easily to itemization on a questionnaire. The use of investigator or
"expert" judgment rather than respondent judgment in assessing inclusion
and severity of psychological abuse has advantages of avoiding
underassessment due to respondent unawareness of its definition or
normalization of "odd" parental behavior. Investigator-based measurement
allows the investigator to decide on the inclusion of an item as
psychological abuse, using anchoring examples as a guide. Thus the same
thresholds for inclusion and severity of examples can then be applied
consistently across samples.
CECA interview involves detailed questioning regarding relationship
with, and behavior of, biological parents and surrogate parents who have
responsibility for the participant before age 17, for any period lasting
12 months or more. To facilitate recall, experiences are asked about in
chronological order. Most of the relevant experiences are rated on a
4-point scale of severity ranging from "1-marked" for the most severe to
"4-little/none" for the least negative experience. Ratings rely on
"objective" details of behavior impinging on the child based on
collecting details of specific incidents, their frequency, age at which
they occurred, who was involved, and so on, regardless of the
individual's responses to those experiences or current feelings about
them. A manual containing hundreds of benchmark examples is available to
interviewers to assist rating and regular researcher trainings held by
research team (Bifulco, Brown, Neubauer, Moran, & Harris, 1994).
Reliability of the instrument is high, with interrater agreement
reaching above 0.78 (Kw, weighted Kappa; Cohen 1968) on all scales (Bifulco,
Brown, & Harris, 1994). Validity in terms of interrespondent agreement
(e.g., sisters reporting on each others' childhood experience) is also
good, reaching an average of 0.60 Kw across numerous scales (Bifulco,
Brown, Lillie, & Jarvis, 1997). The following types of childhood
experience are included in the CECA and are here differentiated from
Parental neglect. Neglect reflects parental
lack of interest in the child's material care, health, friendships,
schoolwork, or career plans, or being emotionally unavailable when the
child is distressed. Severity of neglect is determined on
the range of such indicators being present and rated on a 4-point scale
from "marked" through "little/none." Each parent is rated independently
and an overall assessment is made.
Parental antipathy. The amount of dislike, rejection, irritation, and
coldness shown by parents and surrogate parents is rated as antipathy.
Information regarding critical and rejecting comments and verbal
aggression directed at the subject is taken into account, as are
behavioral instances of rejection or "scapegoating" the child. Severity
is again reflected on a 4-point scale.
Role reversal. The extent to which parents require the child to takeover
a parental or adult role is identified as role reversal. This includes
taking on responsibility in the household for running the home (washing,
cooking, ironing) and for looking after younger siblings. It also
includes the parent using the child as confidant or support figure,
bearer of family secrets, and source of comfort for the parent when
distressed. Severity is determined by the number of areas in which such
role reversal occurs and its frequency. It is rated on a 4-point scale.
Parental discipline. Discipline imposed by parents or surrogate parents
is rated using information regarding the presence and enforcement of
rules about manners and behavior, socializing, clothing, smoking, and
drinking. The scaling differs from those just described, in that
moderate is regarded as the most acceptable form of discipline and is
distinguished from high, low, or variable levels. Thus the scale points
used are "1-marked," "2-moderate," "3-1ax," "4-variable."
Parental supervision. The extent to which parents monitor the child's
behavior and maintain safety is defined as supervision. This includes
the child being left home alone, being allowed out unsupervised or late
at night, or in dangerous surroundings. The scaling is the same as for
the discipline scale.
Physical abuse. Physical abuse from any household member, including
mothers, fathers, surrogate parents, older siblings, or other adults
living in the home, is assessed. Each instance of abuse by a different
perpetrator is rated independently. A number of features of assaults are
asked about, including frequency of attacks, use of implements or
weapons, number of hits, age at time of attack, and injuries sustained.
This information was used to derive a rating of severity of physical
abuse on a 4-point scale.
Sexual abuse. Unlike other forms of maltreatment, incidents of sexual
abuse are included regardless of whether the perpetrator was a household
member or not. Each abuse by a different perpetrator is rated
separately. As with physical abuse, a rating of severity on a 4-point
scale is derived based on a number of features of the assault, such as
frequency and duration, age of victim at time of abuse, relationship to
perpetrator, and intrusiveness of the sexual contact.
Felt shame. In addition to objective characteristics of maltreatment, a
rating of subjective feeling of shame in childhood was included. This
involves feelings of shame, embarrassment, or stigma for any aspect of
childhood, including appearance (e.g., shabbiness, unfashionable
clothes, uncleanness), family characteristics (e.g., parents drinking,
fighting, causing scenes at school), and social deprivation (e.g.,
poverty, deprived home conditions). This is rated on a 4-point scale of
origins of the CECA interview lie in a protocol first designed in
in the early 1980s when the effects of early maternal loss and lack of
care on adult depression were investigated (Harris, Brown, & Bifulco,
1986). The basic measure at that stage included care and control scales
such as neglect, role reversal, supervision, and
discipline from birth or surrogate parents. The measure was subsequently
used on a large representative series of inner-London mothers where
scales for abuse (physical and sexual) and antipathy (hostile
parenting) were added. The interviews involved collecting
extensive narratives about individual's childhood experiences before age
17. The biographical information was transcribed and then ratings were
made of the level of severity of each different childhood experience.
During the course of using the measure over 10 years on subsequent
series of women, it became apparent that some experiences of abuse were
not sufficiently covered in the existing scales, particularly
psychological abusive experience involving coercive control and sadistic
behavior. For example, while it was common to hear about parents who
were critical, disrespectful, scapegoating, cold, or rejecting, the CECA
investigators also learned about a child who had confessed to a fear of
the dark being locked in a room with no light bulb, about another whose
pet was destroyed in front of her for some minor misdemeanor, and about
a child whose sole photograph of her dead mother was torn to shreds by a
cold and controlling stepmother. Such incidents seemed to raise the
level of cruelty and control to a new level.
assess incidents as psychologically abusive, the trained CECA raters
considered for inclusion any communications to the child that were
cruel, overcontrolling, or manipulative, including both verbal and
nonverbal behavior from the perpetrator. It was decided to set a high
threshold for inclusion of abusive behaviors, relevant to long-term
effects and to psychopathological outcomes, and to differentiate them
from those already covered in related scales. Evidence of the
perpetrator's apparent premeditated strategy in "designing" a punishment
or means of control specifically tailored to the child's fears and
vulnerabilities would make inclusion as psychological abuse easier to
determine. One woman we interviewed claimed that her stepfather would
spend the day thinking of new ways to torment her in the evenings when
he returned home, as evidenced by the many new varieties of abuse he
would impose. Malevolent intent often appeared evident in the detailed
descriptions of childhood experiences, as, for example, in the case of a
father who told his young daughter that he would derive great pleasure
from withholding her life-saving medication and watching her die slowly.
However, perpetrator strategy was at times unclear, even though the
nature of the abusive behavior toward the child in terms of categories
identified qualified: Explicit malevolent perpetrator strategy was
regarded as a sufficient but not a necessary feature for defining such
Based on the examples to emerge from the participants' accounts of
childhood recounted in the course of the CECA interview, an overall
assessment of severity was determined and a taxonomy of nine
psychologically abusive behaviors was generated.
Instances of psychological abuse were rated in terms of overall severity
using the same 4-point scale used for the majority of the CECA scales
(i.e., 1, marked; 2, moderate; 3, some/mild, or 4, little/none). The
overall severity was influenced by the number of subcategories of
psychological abuse rated but equally by the intensity and frequency of
single forms of such abuse. Judgments were made by considering a number
of features of the abuse in the child's particular context, consistent
with predefined rating criteria. As with other CECA scales, severity
ratings reflected the potential for long-term psychological and
emotional damage rather than actual damage to the subject, and raters
were always blind to the psychiatric outcome of the subject or his or
her reported reactions to the abuse.
following are examples of overall severity of psychological abuse
determined from the cases examined and typical of those used as rating
Marked psychological abuse. "I went through a phase of having a bag by
my bed at night because quite often Dad would come home late at night
and order us all out of the house. My mother had to take us children
out, whatever the time or weather, to go and stay with friends. There
was no warning, you never knew when it would happen" (subcategory of
"deprivation of basic needs"). "Also, I had asthma as a child and I used
to keep my tablets and my inhaler and a few personal bits and pieces
with me just in case I needed them at night. One night when Dad came in
drunk, he came into my room and woke me up and threatened to take my
medication away. He said that nothing would give him more pleasure than
to take my tablets away and watch me die slowly" (subcategory of
Moderate psychological abuse. The girl was brought up by her aunt after
being separated from her parents at age 5. Her aunt was cold and
critical throughout the relationship. When the girl went on a school
trip lasting a few days, her aunt offered to look after her pet dog, to
feed and walk the dog, usually the girl's responsibility. The girl
describes her return from the trip: "I went on a school trip, I was
about 9. My aunt picked me up from the school at the end of the trip.
She handed me the dog's collar and lead and said, "You won't be needing
these any more. The dog's dead. I had it put down." "It was the first
thing she said as I got off the school bus." No explanation was given of
this action. The dog had been perfectly healthy, and there were no
practical reasons for disposing of it. There was no prior discussion of
the aunt's decision (subcategory of "deprivation of valued object").
Mild psychological abuse. The woman reported that as a child she had a
kidney complaint, which was never diagnosed or treated, and as a result
had poor bladder control. She used to wet the bed at night and wet her
knickers regularly by day. "I was still wetting at the age of 7 or 8,
and at primary school. At home we didn't have a washing machine. Washing
was a problem--there were no radiators or drying facilities. I can
remember mother saying to me once: "If you wet your knickers again, I'm
going to rub them in your face," and I came home from school with wet
knickers and so she did" (subcategory of "humiliation").
the basis of examples such as those just given, careful scrutiny of 300
cases of childhood experience, and following the definitions in the
psychological abuse literature, the following subcategories of
psychological abuse were derived to help in the identification and
scoring of the severity of the experience.
Humiliation/degradation. The key feature of this category is those
actions or comments that degrade and humiliate the child and have the
potential for invoking shame. The rating is based on the likelihood of
shame being induced in most children subjected to such treatment rather
than the actual degree of shame reported. Public viewing of such
humiliation would add to the severity. (See the example of bedwetting
under mild abuse.)
Terrorizing. This involves attempts to invoke extreme fear or dread in
the child in a calculated way, excluding physical attacks. This
includes, for example, deliberately playing on a fear that the child is
known to have, such as forcing the child to sweep insects out of the
basement when the child is known to be terrified of insects. As with the
humiliation category, the rating is based on the likely response by the
child in such a situation, rather than the actual response invoked.
Cognitive disorientation. This type of abuse includes a number of
techniques aimed at confusing and disorienting the child in terms of (1)
his or her belief in the evidence of his or her senses (e.g., repeatedly
telling the child she had misunderstood a command, which had in fact
been correctly followed), (2)
memory (e.g., enforcing a belief that the could not recall valued
experiences in the past) or (3)
sense of identity (e.g., convincing the child that a biological parent
was not the child's parent or that a separated parent was dead). In
extreme instances, strategies akin to brain washing are utilized. One
woman described her stepfather's behavior: "I was on house duty, where
everything had to be scrubbed and then when you'd finished, you had to
start it all again. And it was like mental torture ... 'Hang the washing
out,' and you'd put it out and then he'd say, 'No, I didn't want it out,
I wanted it in,' and you'd go and bring it all back in and he'd put it
on the table and say, 'I thought I'd told you to hang that washing out.'
... He was making you feel as though you'd misunderstood the first
command or you'd done the opposite to what he'd asked. Until you
wondered if you were going mad."
Deprivation of basic needs. This involves depriving the child of basic
needs such as light, sleep, food, or the company of others. This type of
abuse is closely related to neglect, but it may be
distinguished from the latter by the way in which the resource is
controlled and rationed by the parent/perpetrator. This is not typically
present in neglect, where depriving a child is often a
by-product of mismanagement and poor coping. In the following example,
the child was deprived of social contact with her siblings. The woman
talks about when her father remarried when she was 6: "I wasn't allowed
to speak to my half-siblings. It was sort of like "us" (me and my
sister) and "them" (my stepmother's children). My stepmother always
favored her own children. She wouldn't feed me like her own children. It
was different when my father was at home. Then I was allowed to sit and
eat at the table with the others. When he wasn't there, I used to have
to stand in the corner when they ate."
Deprivation of valued objects. This form of abuse involves depriving the
child of a specific object that the he or she values or treasures. This
may be an inanimate object, such as a toy or precious memento, or it may
be animate, as with a pet, but it may also occasionally extend to
significant others to whom the child is forbidden to talk (such as
separated parents or relatives). It may also be more abstract in terms
of a valued aim or achievement (e.g., taking up educational or sports
opportunities). Frequently the object is the only source of comfort
available to the child, for example, a toy or a special present, and it
is typically removed or destroyed in a deliberate, calculating fashion
to maximize distress and disappointment. (See the example of the pet
being destroyed under moderate abuse).
Extreme rejection. This involves peaks of rejection that indicate
abandonment or wishing the child was dead. Examples include locking the
child out of the house when he or she is distressed or hurt for a long
period of time, or abandoning a young child in an unfamiliar place, then
returning some time later without explanation. The example of the wish
for the child's death when undergoing an asthma attack described earlier
would also fit this criterion.
Inflicting marked distress or discomfort. Although distress and
discomfort are features accompanying most types of abuse, notably
physical and sexual abuse, particular elements distinct from these are
required for psychological abuse. One is force feeding, not only at
mealtime but also with noxious substances such as shoe polish or feces.
Vomiting often ensues, and physical pain, discomfort, or revulsion are
involved. One woman related: "If I didn't want to eat what she'd cooked,
I'd have my nose held to force me to eat. If she didn't manage to get it
all into me, it would be presented cold at a later meal, then again at
another. Eventually the food made me vomit."
Emotional blackmail. The key feature of this category of abuse is the
use of serious threats to close others to ensure control and compliance.
Emotional blackmail may also include threats to reveal stigmatizing
information about the child or family, which will have negative
consequences for the child. This form of psychological abuse often
accompanies sexual abuse in order to induce compliance by threats to
harm others (e.g., siblings) or to tell others of the child's "guilt."
(See the example of telling others the child was a liar to prepare for
possible disclosure of sexual abuse.)
Corruption/exploitation. This type of abuse involves forcing the child
to take part in (usually) illicit activities such as stealing or drug
taking. For example, one 10-year-old child was repeatedly given
amphetamine by her mother and told it was sherbet. Incidents of
exploitative sexual activity are also included in this category, for
example, taking pornographic photographs of the child for distribution,
or involvement in pedophile groups. The latter would rate on both sexual
and psychological abuse. Thus child prostitution and ritualized sexual
abuse would also be rated as both sexual and psychological abuse.
last two categories indicate how psychological abuse can integrally
co-occur with other abuse such as sexual abuse. An abuse "complex" was
rated only when psychological abuse formed an integral part of the same
incident with either physical or sexual abuse. For example, a
14-year-old girl was punished by her father for a minor infraction by
pulling down her underwear and belting her on her bare bottom in front
of her family and relatives. Here the psychological abuse subcategory of
humiliation was rated in addition to physical abuse. Abuse complexes
were distinguished from pure forms of psychological abuse, where sexual
or physical abuse did not form part of the same incident as the
psychological abuse example. However, it was also possible for the
perpetrator of pure psychological abuse to be engaging in other forms of
abuse or neglect on separate occasions with the same
victim. For example, if the 14-year-old girl was hit by her father on
other occasions with no publicly humiliating element, then she would
rate as having recurrent physical abuse in addition to the specific
psychological/physical abuse complex just described.
Similarly, psychological abuse could be a complex with sexual abuse. One
woman described sexual abuse from her father: "It was full intercourse.
And if I refused, I would get a hiding [beating] from him. If, for
instance, if I bunked off [skipped] school and he found out, he would
threaten to tell my mother--and I was more frightened of my mother funny
enough--but if I let him screw me, he wouldn't say nothing about the
school thing. He always said that if I got pregnant, I must say that it
was my boyfriend who did it. He also made me out to be a liar ....It was
like preparing in case I ever did say anything [about the abuse]."
Psychological abuse accompanying sexual abuse usually related to
compliance or secrecy, such as campaigns of terrorizing the child or
threatening the safety of close others to increase compliance and
prevent the child from disclosing. Elements intrinsic to the sexual
abuse, such as the betrayal of trust by a perpetrator who is a close
other and the inappropriateness of the sexual behavior, were reflected
purely in the severity rating of the sexual abuse. Helplessness on the
part of one parent in relation to any abuse from the other parent (e.g.,
"bystander apathy") did not in itself qualify as psychological abuse.
Greater collusion or participation in tormenting the child was required
in order to rate psychological abuse; only the figure active in the
abuse rated as the perpetrator. For example, a mother who did nothing to
stop sexual abuse of her child by the father was not rated as
perpetrating psychological abuse, although her behavior was considered
an indicator of neglect.
CECA, together with the psychological abuse scales, was tested on a
community-based series of women interviewed between 1990 and 1994. The
participants were selected from responses to screening questionnaires
sent to women in the 18 to 50 age group who were registered with general
practitioner physicians in Islington, North London, UK. The sample is
discussed at length in the companion article (Bifulco et al., 2002). The
series consisted of 105 women selected by screening questionnaire for
adult depressive vulnerability such as poor support and low self-esteem
and 59 selected by screening questionnaire for adverse childhood,
particularly maternal neglect and antipathy. Forty
consecutive questionnaire responders formed a comparison group. The
latter two groups were also used for a validation of childhood
experience and for each a sister within 5 years of age was interviewed
(N = 99). In the process of developing a reliable scale of psychological
abuse, it was decided to include all 304 women in the series, including
sisters, in order to maximize the number of examples obtained. (In the
companion article, the sisters were excluded because of possible
familial confounds in the association with adult disorder.)
Overall, 45% responded to the initial screening questionnaire, 64% of
suitable women agreed to complete the full interview, 23% refused, and
16% proved unobtainable. In terms of demographics, the average age of
the women was 35 (range 19-53), more than (60%) were middle class, 67%
were working full or part time, 61% were married or cohabiting, and 77%
the women were questioned about using the full CECA. The psychological
abuse categories were determined on the basis of reading the transcribed
childhood material, blind to depressive outcome. This was conducted by a
team of six researchers trained and experienced in the CECA but who were
not responsible for the original interviews and who were blind to
psychiatric outcome. The scale was devised on the basis of full
discussion on positive examples for half the series and then reliability
was undertaken on cases from the second half of the series.
new scale was devised on the basis of examples in half the series, where
rules for inclusion and severity were determined, and then examples from
the second half were used to determine interrater reliability. Twenty
examples of psychological abuse were rated independently for severity by
two trained raters, and a correlation of.68 (Kw; Cohen, 1968) was
initially achieved. When the severity scale was dichotomized about its
midpoint (marked/moderate vs. mild/little-none) as used in the
quantitative analysis, inter-rater reliability rose to.80 (Kw). This
latter figure is comparable with figures for interrater reliability of
other CECA scales (see Bifulco, Brown, & Harris, 1994).
Sixteen per cent (49/301) of women in the total series experienced
marked to mild levels of psychological abuse from at least one
perpetrator. This is similar, but lower than, the levels of
neglect (27%), sexual abuse (24%), role reversal (24%), and
antipathy from father (19%) in this particular highly selected series.
However, it is substantially lower than physical abuse (40%) and
antipathy from mother (51%), the latter being one of the sample
selection criteria. Given the high threshold for inclusion of
psychological abuse, its relatively low apparent prevalence might be
expected in community-based series. It was extremely rare in the
comparison series, where only 4% (3/80) were found.
total, there were 56 instances of psychological abuse when those from
mother, father, and others were included. In terms of the frequency with
which abuse occurred, 66% (37)
of these cases involved repeated psychological abuse, occurring at least
weekly in most instances. The remaining 34% (19)
were isolated instances. The age at which the psychological abuse began
was evenly distributed across three age ranges: beginning at age 5 or
below, 38% (20);
between age 6 and 10, 25% (13);
and between age 11 and 17, 36% (19).
There was no distinct pattern regarding the severity of abuse in
relation to the age at which it began, with similar proportions of mild
(55%), moderate (75%) and marked (64%) abuse examples beginning before
age 11 (p = ns). Most perpetrators (90%, 52) were parents or surrogate
parents. The few nonparental perpetrators involved teachers, other
relatives, boyfriends and strangers. Perpetrators were as likely to be
female as male: 50% (29)
of of the instances of psychological abuse were committed by women
compared with 45% (26)
by men, and the remaining 5% (3)
were committed by men and women jointly. Although the proportions of
male and female perpetrators were similar, cases rated "marked" in
severity were twice as likely to be committed by men: 67% (10/15)
compared with 33% (5/15) by women. However when male and female
perpetrator cases were compared, there was no significant difference in
the proportion rated mild, moderate, or marked.
Most cases of childhood psychological abuse (76%, 44) were rated on only
one of the nine subcategories. A further 17% (10)
rated on two subcategories, 5% (3)
rated on three subcategories, and the remaining 2% (1)
rated on a total of five subcategories. In terms of frequency,
humiliation was the subcategory most often rated, featuring in 38% of
the cases, followed by deprivation of valued object (24%), deprivation
of basic needs (16%) and extreme rejection (16%). Inflicting
distress/discomfort occurred in 14% of cases. Less common categories
were emotional blackmail (9%), terrorizing (7%), and cognitive
disorientation (7%) corruption was the least-rated category, featuring
in only 5% of cases. The subcategory of emotional blackmail occurred
only as a feature of abuse complexes, in particular in relation to
number of subcategories rated was related to the overall severity of the
abuse, producing a correlation of .48 (Pearson's r, p < .001). To
elaborate this relationship further, a score was derived based on the
individual severity ratings for the subcategories of psychological abuse
and summed for each case of abuse. This score was correlated with the
overall rating for severity of abuse and was found to be highly related,
producing a correlation of .69 (Pearson's r, p < .001).
Psychological abuse was examined in relation to reports of the
subjective state of shame. This was assessed for the 198 women in the
child risk and comparison series, with 43% rated as having marked or
moderate levels of shame in childhood. For those with any level of
psychological abuse rating 72% (21/29) reported high levels (marked or
moderate) of shame in childhood compared with 39% (65/168) of those
without any psychological abuse (p < 0.001).
Given the lack of consensus regarding the definition of childhood
psychological abuse, it is not surprising that relatively little
empirical research has specifically addressed this topic. The current
article has presented a working definition based on a standard of care
rather than a consequences model (Garbarino, 1991). This avoids the
tautology so often featured in definitions of this form of abuse,
namely, that psychological abuse is that which causes psychological
damage. The definition presented here specified nine ways in which
psychological abuse can be demonstrated and derived a scheme for rating
overall severity in the childhood histories of just over 300
community-based women. Such examples were specifically distinguished
from related experiences of parental neglect, antipathy,
role reversal, discipline, and supervision and rules for co-occurrence
with physical and sexual abuse were determined. The prevalence of
psychological abuse in this series was lower than for related
experiences such as neglect, antipathy, or physical abuse.
It also proved rare in the unselected comparison series. Although
parents were by far the most frequent perpetrators in this series
(consistent with the family-based focus of the measure) examples of
psychological abuse were identified from other relatives and teachers.
There was no specific patterning of severity of psychological abuse in
relation to age of child or gender of perpetrator. The objective
assessment of psychologically abusive experience was highly related to
feelings of shame in childhood.
Limitations of this study should be noted:
The analysis of
psychological abuse was essentially exploratory, with the scale
designed after the narrative data were collected. It may therefore be
an underestimate of actual experience of psychological abuse. This
needs to be reexamined using specific probing questions now
The sample on
which the new scale was tested was highly selected, so the prevalence
of this psychological abuse assessment in the representative community
is unknown. In a similar vein, the sample was only of women. Thus the
scale needs to be assessed on a mixed-gender series.
nature of the measurement may have involved underreporting of
psychologically abusive experience.
the literature, categories of psychological abuse identified were
varied and the numbers in subcategories were small in this series. It
is possible that on larger numbers, it would be possible to group the
psychological abuse subcategories further to extend the taxonomy in
terms of the components of negative care versus high control.
abuse was by parents, a small proportion was by other adults (such as
teachers) with responsibility for the child. This indicates the
importance of considering the wider social context when assessing
childhood abuse. In future investigations, greater exploration of such
abuse in school, leisure, and peer settings needs to be undertaken.
There are, however, advantages to using a retrospective, semistructured,
investigator-based interview. The retrospective approach to measurement
allows for the exploration of outcomes that may be associated with abuse
in the long term as well as the short term. Such an approach is
advocated by other researchers (e.g., Burnett, 1993), and there is now
good evidence to suggest that retrospective accounts are reliable when
the material is carefully collected (Bifulco et al., 1997; Brewin,
Andrews, & Gotlib, 1993). Also, the use of a semistructured approach in
preference to a structured questionnaire or interview allows for the
possibility of eliciting new categories of psychological abuse, thus
extending and clarifying the list of abusive perpetrator behaviors used
in the definition. Hence the definition of psychological abuse may
continue to evolve as a result of this method of data collection. The
list of categories of psychological abuse presented here is not
exhaustive, but is a summary of the types of abusive behavior to emerge
inductively from accounts of childhood supplied by a community-based
sample. Childhood histories obtained from other cohorts such as clinical
populations might involve more complex or perverse forms of
psychological abuse, thus requiring extension of the categorization.
Another advantage of the current measurement system is that it offers
the ability to grade examples of abuse in terms of severity. It may
therefore be possible to specify a level of abuse requiring intervention
by child protection agencies, a point raised by Edmundson and Collier
(1993) in a discussion of the role of educationalists in the
identification and prevention of abuse. In the present study, it was
both the number and the severity of abuse strategies used by the
perpetrator that related to the overall severity level of the abuse,
although lack of complete correspondence still required interviewer
judgment to weigh the degree of cruelty involved, the potential for
short- and long-term damage to the child, and the extent to which the
abuse was tailored to the child's characteristics in assessing an
overall rating. Indeed, Garbarino (1991) states that it is not practical
to specify in concrete detail what constitutes psychological
maltreatment in the way that might be possible for physical or sexual
abuse. The present study also indicates the importance of considering
single attacks in addition to repeated examples of psychological abuse,
because as many as one third of examples rated marked in severity
involved such single attacks. In addition, such single attacks might
well have formed part of a more sustained pattern of abuse that the
subjects were unable to report or recall, and thus might constitute the
tip of the iceberg in relation to ongoing interaction.
likely impact and meaning of psychological abuse needs to be explored in
future studies to develop theoretical understanding. For example, to
search for specific effects, the short-term response to psychological
abuse needs to be contrasted with responses to other forms of
maltreatment. The presence of psychological abuse in this series was
highly related to feelings of shame in childhood. This might be expected
because subcategories of psychological abuse such as humiliation can be
seen as forms of shame induction. For example, denying a child access to
basic facilities for washing can potentially be shaming to the point
where the child is reluctant to socialize with peers, and experiences
great isolation (Bifulco & Moran, 1998). Similarly, aberrant parental
behaviors would be expected to make the child feel stigmatized and
embarrassed about allowing friends to observe the child's family life.
Shame about appearance has been shown to act as a mediator between other
types of abuse and depression using the CECA approach (Andrews,1995). It
has also been suggested that the shame induced by humiliation is closely
associated with subordination and defeat (Gilbert, 1989). Indeed, in the
most extreme cases seen in the present series, the perpetrator strategy
appeared to be directed toward subjugation of the victim. The coercive
control that formed the central feature of some of the most severe
examples of psychological abuse has much in common with the process of
domination seen in other contexts, such as women in situations of
domestic violence, and even that of political hostages (Herman, 1994).
Goddard and Stanley (1994) suggest that an understanding of child abuse
can be increased if the abusive parent and the abused child are viewed
as captor and hostage. The perpetrator at times has the power to distort
the child's cognitions and emotions over substantial periods of
childhood. This not only has obvious implications for self-esteem and
helplessness, which are known to be associated with adult disorders such
as depression, but it also has a high potential for more enduring
personality damage, with implications for personality disorder outcomes.
some instances, psychological abuse may arise as a function of
overzealous parenting. An example is the case of a father
who, on discovering that his 10-year-old daughter had been smoking, made
her smoke 40 cigarettes in succession to induce vomiting in order to try
to extinguish her desire for cigarettes. Similar examples of mothers
force-feeding their children were set in a context of trying to "train"
them to eat a full meal at each sitting. This is consistent with
reported findings of a series of borderline personality disorder
patients who had psychologically dominant and abusive parents who had
grandiose ideas of their own competence as parents. The abuse was
demonstrated in tactics of control rather than in long-term designs of
deliberate harm (Park, Imboden, Park, Hulse & Unger, 1992).
Psychological abuse, often not directly observable in its effects, is in
danger of being overlooked in favor of more physically tangible forms of
abuse such as physical or sexual abuse: "With society more likely to
denounce blatant forms of abuse (Ney, 1987) psychological abuse may pass
less conspicuously" (Fortin & Chamberland, 1995, p. 281). In the United
Kingdom, statutory guidelines for registering a child as "at risk"
specify that if other abuse is present, then psychological abuse should
not be listed as the reason for registering. Thus the true rate of
psychological abuse among children on risk registers is still largely
unknown. These guidelines also imply that psychological abuse is
subsidiary, posing less danger to the well-being of the child than other
forms of neglect or abuse. However, the damage inflicted
as a result of psychological abuse may prove to be just as far-reaching
and may have its own specific damaging effects. Thompson and Kaplan
(1996) suggest that it has implications for development of insecure
attachment, delay or damage to psychological development, and failure of
physical growth. Outcomes involving adult depression and suicide
attempts are explored in the companion article (Bifulco et al., 2002)
and highlight the importance of acting to reduce the prevalence of
hope that the present discussion has advanced knowledge of psychological
abuse not only by offering an operational definition, but also by
providing a system by which it can be measured. The definition and
measurement of psychological abuse are still evolving and are open to
adjustment as new examples come to light. In the meantime, the current
approach to definition and measurement may advance theory and research
into this neglected area of childhood and will, we hope, ultimately
contribute to the development of child protection policy. Arriving at a
consensus regarding the definition of psychological abuse is of
paramount importance. Not only does a definition aid future research
into the prevalence and damaging effects of such ill-treatment, but it
is also of use to child welfare agencies and judicial services aimed at
child protection, where there is a pressing need for such a definition (Edmundson
& Collier, 1993). However, what constitutes "actionable psychological
abuse" (Garbarino, 1978, 1987) in legal and child protection terms may
differ from the definition required for research purposes (Glaser &
Prior, 1997). The definition presented here may offer a starting point
for definitions required for use in other contexts, such as courts of
law and social service child protection departments.
This study was funded by the Medical Research Council (program grant
G702833; principal investigator, Professor George W. Brown). Additional
funding was provided by the John T. and Catherine D. MacArthur
Development and Psychopathology Network, which enabled the development
of the measure and a training package. We would like to thank Ann
Brackenridge, Julie Jarvis, Amanda Lillie, Amy Neubauer, Hedy Wax, and
Caroline Campbell for the data collection and Laurence Letchford for
computer analysis. Finally, we would like to express our gratitude to
the general practitioners who allowed us access to their patients, and
to the women themselves who participated in the study.
Andrews, B. (1995). Bodily shame as a mediator between abusive
experiences and depression. Journal of Abnormal Psychology, 104,
Andrews, B. & Brown, G. W. (1988). Social support, onset of depression
and personality: An exploratory analysis. Social Psychiatry, 23, 99-108.
T.F., & Baily, W.H. (1986). Operational definitions of child emotional
maltreatment: Final report (National
on Child Abuse and Neglect, DHHS 90-CA-0956).
Government Printing Office.
Barnett, D., Manly, J. T., & Cicchetti, D. (1991) Continuing towards an
operational definition of psychological maltreatment. Development and
Psychopathology, 3, 19-29.
Bifulco, A., Brown, G. W., & Harris, T. (1994) Childhood experiences of
care and abuse (CECA): A retrospective interview measure. Journal of
Child Psychology and SPsychiatry, 35, 1419-1435.
Bifulco, A., Brown, G. W., Lillie, A., & Jarvis, J. (1997). Memories of
childhood neglect and abuse: Corroboration in a series of
sisters. Journal of Child Psychology and Psychiatry, 38, 365-374.
Bifulco, A, Brown, G. W., Neubauer, A., Moran, P. M., & Harris, T.
(1994). The Childhood Experience of Care and Abuse (CECA) interview:
Bifulco, A., & Moran, P. (1998). Wednesday's child: Research into
women's experience of neglect and abuse in childhood and
depression in adulthood.
Bifulco, A., Moran, P. M., Baines, R, Bunn, A., & Stanford, K., (2000).
Exploring psychological abuse in childhood: II. Association with other
abuse and adult clinical depression. Bulletin of the Menninger Clinic,
D. A., Smith Slep, A., Heyman, R. E. (2001) Risk factors for child
psychological abuse. Aggression and Violent Behavior, 6, 189-20l.
Brassard, M. R., Hart, S.N., & Hardy, D. B. (1993). The Psychological
Maltreatment Rating Scales. Child Abuse and Neglect, 17, 715-729.
C. R., Andrews, B., & Gotlib,
H. (1993). Psychopathology and early experience: A reappraisal of
retrospective reports. Psychological Bulletin, 113, 82-98.
J., & Runtz, M. (1988). Multivariate correlates of childhood
psychological and physical maltreatment among university women. Child
Abuse and Neglect, 12, 331-341.
Burnett, B.B. (1993). The psychological abuse of latency age children: A
survey. Child Abuse and Neglect, 17, 441-454.
J. (1968). Weighted kappa. Psychological Bulletin, 70, 213-220.
M. A., & Frias, D. (1994). Emotional and cognitive adjustment in abused
children. Child Abuse and Neglect, 18, 923-932.
Claussen, A. H., & Crittenden, P. M. (1991). Physical and psychological
maltreatment: Relations among types of maltreatment. Child Abuse and
Neglect, 15, 5-18.
Creighton, S. J. (1992). Child abuse trends in
National Society for the Prevention of Cruelty to Children.
Crittenden, P. M. (1985). Maltreated infants: Vulnerability and
resilience. Journal of Child Psychiatry, 26, 85-96.
Edmundson, S. E., & Collier, P. (1993). Child protection and emotional
abuse: Definition, identification and usefulness within an educational
setting. Educational Psychology in Practice, 8, 198-206.
Fortin, A., & Chamberland, C. (1995). Preventing the psychological
maltreatment of children. Journal of Interpersonal Violence, 10,
Garbarino, J. (1978). The elusive 'crime' of emotional abuse. Child
Abuse and Neglect, 2, 89-99.
Garbarino, J. (1987). What can the school do on behalf of the
psychologically maltreated child and the community? School Psychology
Review, 16, 181-187.
Garbarino, J. (1991). Not all bad outcomes are the result of child
abuse. Development and Psychopathology, 3, 45-50.
Garbarino, J. Guttman, E., & Seeley, J. (1986). The psychologically
Gilbert, P. (1989). Human nature and suffering.
Glaser, D., & Prior, V. (1997). Is the term child protection applicable
to emotional abuse? Child Abuse Review, 6, 315-329
Goddard, C., & Stanley, J. (1994). Viewing the abusive parent and the
abused child as captor and hostage. Journal of Interpersonal Violence,
A., & Keller, H. (1992). Long-term consequences of childhood physical
and psychological maltreatment. Aggressive Behavior, 18, 171-185.
Harris, T., Brown, G. W., & Bifulco, A. (1986). Loss of parent in
childhood and adult psychiatric disorder: The role of lack of adequate
parental care. Psychological Medicine, 16, 641-659.
S. N., & Brassard, M. R. (1991). Psychological maltreatment: Progress
achieved. Development and Psychopathology, 3, 61-70.
Haugaard, J. J. (1991). Defining psychological maltreatment: A prelude
to research or an outcome of research? Development and Psychopathology,
Herman, J. (1994). Trauma and recovery.
Higgins, D., & McCabe, M. P. (2001). The development of the
Comprehensive Child Maltreatment Scale. Journal of Family Studies, 7,
Kaufman, J. (1991). Depressive disorders in maltreated children. Journal
of Child and Adolescent Psychiatry, 30, 257-265.
A., & Waller, G. (1998). The impact of childhood emotional abuse: An
extension of the child abuse and trauma scale. Child Abuse and
Neglect, 22, 393-399.
R. A., & Wolfe, D. A. (1991). Psychological maltreatment: Toward an
operational definition. Development and Psychopathology, 3, 3-18.
Mullen, P. E., Martin, J. L., Anderson, J. C., Romans, S. E., & Herbison,
G. P. (1996). The long-term impact of the physical, emotional, and
sexual abuse of children: A community study. Child Abuse and Neglect,
P. G. (1987). Does verbal abuse leave deeper scars? Canadian Journal of
Psychiatry, 34, 371-378.
P. G., Fung, T., & Wickett, A. R. (1994). The worst combinations of
child abuse and neglect. Child Abuse and Neglect, 18,
O'Hagan, K. P. (1993). Emotional and psychological abuse of children.
Open University Press.
O'Hagan, K. P. (1995). Emotional and psychological abuse: Problems of
definition. Child Abuse and Neglect, 19, 449-461.
L. C., Imboden, J. B., Park, T. J., Hulse, S. H., & Unger, H. T. (1992).
Giftedness and psychological abuse in borderline personality disorder:
Their relevance to genesis and treatment. Journal of Personality
Disorders, 6, 226-240.
Parker, G., Tupling, H., & Brown, L. B. (1979). A Parental Bonding
Instrument. British Journal of Medical Psychology, 52, 1-10.
Pitzner, J. K., McGarry-Long, J., Drummond, P. D. (2000). A history of
abuse and negative life events in patients with a sexually transmitted
disease and in a community sample. Child Abuse and Neglect, 24,
P., Duncan, S., & Gray, M. (1993). Beyond blame: Child abuse tragedies
Rosenberg, M.S. (1987). New directions for research on the psychological
maltreatment of children. American Psychologist, 42, 166-171.
Sanders, B., & Becker-Lausen, E. (1995). The measurement of
psychological maltreatment: Early data on the Child Abuse and Trauma
Scale. Child Abuse and Neglect, 19, 315-323.
Shaver, P.R., Goodman, G.S., Rosenberg, M.S., & Orcutt, H. (1991). The
search for a definition of psychological maltreatment. Development and
Psychopathology, 3, 79-86.
D. (1989). Emotional abuse and delay in growth. British Medical Journal,
D., & Bentovim, A. (1994). Physical and emotional maltreatment. In M.
& L. Hersov (Eds.), Child and adolescent psychiatry: Modern approaches
(3rd ed., pp. 213-216).
N. (1993). Parental abuse as a precursor to childhood onset of
depression and suicidality. Child Psychiatry and Human Development, 24,
Straus, M. (1979). Measuring intrafamily conflicts and violence: The
Conflict Tactics (CT) Scales. Journal of Marriage and the Family, 41,
Thompson, A. E., & Kaplan, C. A. (1996). Childhood emotional abuse.
British Journal of Psychiatry, 168, 143-148.
R., & Abidin, R. (1999). The minimizing style: Perceptions of
psychological abuse and quality of past and current relationships. Child
Abuse and Neglect, 23, 1041-1055.
Vissing, Y. M., Straus, M. A., Gelles, R. J., & Harrop, J. W. (1991).
Verbal aggression by parents and psychosocial problems of children.
Child Abuse and Neglect, 15, 223-238.
Whipple, E., & Finton, S. (1995). Psychological maltreatment by
siblings: An unrecognised form of abuse. Child and Adolescent Social
Work, 12, 135-146.
Patricia M. Moran, PhD; Antonia Bifulco, PhD; Caroline Ball, BSc;
Catherine Jacobs, BSc and Kate Benaim, BA
authors are with the Lifespan Research Group, Department of Health and
Social Care, University of London, United Kingdom. Correspondence may be
sent to Dr. Moran at the Lifespan Research Group, Department of Health
and Social Care, Royal Holloway, University of London, 11 Bedford
Square, London WC1, United Kingdom; e-mail: email@example.com.
(Copyright © 2002 The Menninger Foundation)