87

Psychology and Psychotherapy: Theory, Research and Practice (2003), 76, 87–102
© 2003 The British Psychological Society
www.bps.org.uk

What is the prevalence of narcissistic injury

among trainee counselling psychologists?

Andrea Halewood and Rachel Tribe*

Department of Psychology, University of East London, UK

The purpose of this study was to ascertain the level of narcissistic injury among trainee
counselling psychologists using the Narcissistic Injury Scale (Slyter, 1991). This 38-item
Likert scale is based on Miller’s (1981) deŽ nition of narcissistic injury; a speciŽ c type of
psychological damage which focuses on feelings about the self and past relationships
related to self-development. Theorists suggest that if untreated, narcissistic issues can
interfere with client work and lead to a number of problems for trainees, in uencing
drop-out rates and increasing burnout. The results of the study indicate that a high
degree of narcissistic injury may be prevalent among trainee counselling psychologists
and furthermore, that narcissistic injury does seem to be related to the quality of the
perceived parent–child relationship. Consequently, the study suggests that therapeutic
work could be affected in those trainees who fail to address their own narcissism. The
study provides some tentative support for the utility of the Narcissistic Injury Scale.

The term ‘narcissism’ was inspired by Ovid’s myth of Narcissus in Metamorphoses

( AD 8), in which a young Greek man falls in love with his reection while gazing into a
pool. Finding himself unable to unite with this loved object, Narcissus dies broken-
hearted. Mollon (1993, p. 25) suggests that the themes of this myth include:
· A lack of self-knowledge
· A desire for reection and mirroring
· The fear of being possessed and taken over
· A turning away from objects.
Freud (1914) described narcissism as a normal part of human sexual development rather
than a perversion, seeing it as ‘the libidinal complement to the egoism of the instinct of
self-preservation’ (p. 66).
However, theorists argue that when normal infant narcissim is suppressed, damage
to the individual’s sense of self can arise. Miller (1981) argued in The Drama of Being a
Child that the legend of Narcissus fully illustrates all the elements of the tragedy of

* Requests for reprints should be addressed to Dr Rachel Tribe, Department of Psychology, University of East London, Romford
Road, London E15 4L2, UK (e-mail: [email protected]).
88 Andrea Halewood and Rachel Tribe
narcissistic disturbance and viewed narcissim as a loss of a sense of self. This loss and the
extent to which it is prevalent in trainee counselling psychologists are the research
questions with which this study is concerned.

DeŽ nition of narcissism

Narcissism is an extremely complex phenomenon, which in everyday terms tends to
describe positive feelings about the self. However, in terms of psychological function-
ing, it refers to disorders in individuation and in the formation of the self. Johnson (1987)
argues that the narcissistic character style and narcissistic character or personality
disorder (NPD) form a continuum of ego functioning, ranging from mild to severe.
Consequently, the distinction between healthy and pathological narcissism is
difŽcult to specify. According to the Oxford Textbook of Psychiatry (Gelder et al.,
1983) narcissistic traits are characterized by a grandiose façade which hides a deep
sense of inadequacy caused by disturbances in the experience of the self.
The more severe narcissistic personality disorder is deŽned in DSM-IV (American
Psychiatric Association, 1994) as follows ‘A pervasive pattern of grandiosity (in fantasy
or behaviour), need for admiration and lack of empathy, beginning in early adulthood
and present in a variety of contexts.’ (p. 282).
Indications of this disorder also include a preoccupation with fantasies of unlimited
success, power, brilliance, beauty, or ideal love, a sense of ‘specialness’ or entitlement,
and envy.
This paper is mainly concerned with ‘narcissistic injury’ as deŽned by Miller (1981).
However, in order to avoid confusion between narcissistic traits and narcissistic
personality disorder, the narcissistic personality disorder will be referred to as NPD.
‘Narcissistic injury’ describes damage to the individuals’ experience of their ‘real
self’. In its more extreme forms, individuals are left with no awareness at all of who
they really are. In the less extreme variations of this disorder, there is often a vague
comprehension of the real self but also a rejection of it. Johnson (1987) suggests that
narcissists have learnt to deny their true self-expression in response to an early
rejection of this by their caregivers, and have replaced it with a highly developed
‘false self’.
Winnicott’s (1960/1976) model of the ‘false self’ suggests that people with a false self
are more aware of how they must present themselves to others than they are of their
own inner feelings. According to this theory, because the ‘true self’ was not recognized
and responded to by the mother or caregiver, these individuals suffer a profound lack of
conŽdence in their actions and feelings. Phillips (1988) elucidates Winnicott’s argument
as follows
The mother is the constitutive witness of the True Self. If she violates the infant’s initial
omnipotence she insults the infant’s self and drives it into hiding (p. 130)

Similarly Lichtenstein (1977) suggests that when the mother recognizes and
encourages only certain aspects of the infant’s emerging personality she imparts an
‘identity theme’ quoted in Mollon (1993, p. 64). This view emphasizes the mother’s
role in the development of the sense of self. Johnson (1987) suggests that much of
the pathology of the narcissist is caused by the individual’s rejection of the parts of
him/herself that were not supported and mirrored. Consequently, the narcissist will
try hard to conceal these parts and will work towards presenting a protective false
self instead.
 Narcisstic injury among trainee psychologists 89
Theoretical models of narcissism
Since Freud, several psychological theories have attempted to explain the narcissistic
phenomenon; the most comprehensive of these argue that developmental problems in
childhood lead to the development of narcissistic disorders.
The psychodynamic model of narcissism is dominated by two overlapping schools
of thought; the object-relations school and the self-psychology school represented by
Kohut.
Kohut suggests that we all have a component of narcissism in our psyche, that we are
all born as narcissists but gradually our infantile narcissism matures into a healthy adult
version. Anarcissistic disorder results when this process is disrupted. Kohut argued that
the narcissistic experience begins when the infant’s relationship with his/her mother is
undermined by her inevitable failure to respond to the infant’s every whim. The infant
attempts to compensate for this disappointment by attempting to imbue an outside
‘other’ with absolute power and perfection instead, an image which Kohut refers to as
the ‘idealised parental imago’. Attachment to this other is the infant’s way of restoring
the early sense of wholeness. Alternatively, the infant might attempt to create instead a
perfect ‘self’, which contains everything good and excludes everything bad. Kohut
(1977) refers to this as the conŽguration of the ‘narcissistic self’. These conŽgurations
make up Kohut’s concept of the ‘bipolar self ’ with narcissism on the one side and
grandiosity on the other.

Kernberg—The object relations approach

Kernberg, in contrast, representing the object relations school, describes a more
pathological form of narcissism. This consists of a fusion of the inner representations
of actual self, ideal self and ideal other, which creates an illusion of self-sufŽciency as a
defence against neediness and envy. Kernberg (1975) suggests that it is as if the
narcissist is saying that;

(the) ideal person and my ideal image of that person and my real self are all one, and better
than the ideal person whom I wanted to love me, so that I do not need anybody else
anymore. (Quoted in Mollon, 1993, p. 78)

Kernberg (1975) argues that the narcissist is one who failed to successfully master the
rapprochement subphase, the stage between the ages of 14 and 24 months, at which
children discover that there are limits to what they can do, that they are not omnipotent.
His view of narcissistic psychopathology is that of a turning away from relatedness to
others. In this sense, it is more representative of broader trends in psychoanalytic
theorizing, whereas Kohut’s formulation represented a more radical theoretical
innovation.
Johnson (1987) suggests that although the culture reinforces pathological narcis-
sism, it originates within the family, and argues that the key to understanding narcissism
is through an understanding of the rapprochement subphase and narcissistic injury.

The origin of the narcissistic injury

When infants are a few months old their psychological requirements include a healthy
narcissistic need to be noticed, admired, understood, taken seriously, and respected by
their mothers or caregivers. If these needs are satisŽed a healthy sense of self will begin
to establish itself in the growing child.
90 Andrea Halewood and Rachel Tribe
However, a mother who has not had her narcissistic needs met as a child may not be
able to allow herself to be ‘used’ in this way. On the contrary, she may unconsciously do
the opposite and use the infant to satisfy her own needs. As a result the child’s emotional
development may be hampered. The infant may adapt in order to satisfy his/her
mother’s needs, thus remaining needy his/herself. Miller (1981) speculates that narcis-
sistically disturbed mothers may go on to bring up narcissistically disturbed children,
and that these disturbances may therefore be transmitted through several generations.
However, she points out that this does not mean that the mother does not love her child,
but rather that she loves the child as an extension of herself, rather than for his/herself.
In The Drama of Being a Child, Miller explains the conditions she believes to be
necessary for healthy narcissistic development. Following Kohut she argued that a child
has a basic need for respect and understanding, and that healthy development takes
place when the mother (or primary caregiver) responds to the child by adapting to the
child’s needs. Miller argued that narcissistically disturbed parents cannot offer this
degree of exibility because their own problems with self-esteem and depression do not
allow them to. Consequently, they place emotional demands on the child, which require
the child to attend to and respond to the parent’s needs. Miller suggested that children
used in this way often develop an unusual sensitivity to the narcissistic needs of others.
Consequently, as Glickauf-Hughes and Mehlman (1995) suggest, the child whose
narcissistic development has been obstructed will often lack awareness of their real
feelings—their ‘true self ’. This makes it difŽcult for them to develop an authentic sense
of themselves and they fall back on their accommodating, but inauthentic, ‘false’ self.
Miller suggested that this abandonment of authenticity results in either depression or
grandiosity as a defence against depression. It can also lead to difŽculties in coping and
to problems in relationships.
Russell (1985) suggests that the ability of some narcissistic clients to win admiration
may be due to a sense of ‘specialness’ that they experienced as a child. This could have
been due to an ability or talent, real or attributed, which aroused admiration and/or envy
in others. Whether this was felt to be a positive thing or not, being seen as special could
have offered a refuge from feelings of being unloved. Furthermore, the mother may have
imposed her own narcissistic needs on the child: for example, requiring him or her to
fulŽl her own ambitions and in this way using the child as a narcissistic extension of
herself.
Under these circumstances children learn that they will only be responded to when
they behave in the way required by others. As the child’s whole image is governed by
other’s expectations, the child can never be spontaneous and is therefore forced to
perform on the other’s terms in order to receive any narcissistic rewards for him/herself.
Miller argues that if narcissistic needs are not fulŽlled the individual will feel
compelled to gratify them through substitute means, such as their own children. She
suggests that the therapist is likely to have been raised by a parent who used his or her
children in order to gratify his or her own unfulŽlled narcissistic needs—a parent who is
looking for what their own parents could not give them.
Glickauf-Hughes and Mehlman (1995, p. 218) argue that children who have experi-
enced this type of parenting learn to develop ‘emotional antennae’ in order to respond
to the emotional needs of their parents. This role, they suggest, may well attract them to
the therapeutic professions.
Storr (1979, p. 177) argues that those attracted to the therapeutic professions,
‘ . . . often have some personal knowledge of what it is like to feel insulted and
injured . . . which actually extends the range of their compassion’ and suggests that
 Narcisstic injury among trainee psychologists 91
many of these individuals seem to have had depressed mothers. He believes that
sensitivity to the feelings of others can generally be traced to the circumstances of
the therapist’s childhood and is partly due to the child’s anxiety not to upset, anger or
distress one or both parents. Furthermore, if anxiety over the effects that their
behaviour may have on their parents governs their actions, the child has in effect
learnt to put him/herself second. The child may then have a tendency towards self-
abnegation as an adult. Mollon (1993) describes these clients as compulsively
accommodating, sensing what is required of them and behaving accordingly.
Menninger (1957) argues that therapists tend to have experienced some form of
rejection from their parents, which he suggests could explain their interest in the
emotionally vulnerable. Essentially, these individuals project their own needs into
others, rather than deal with their needs directly. He suggested that this pain is
repressed, and professional functioning provides a form of self-healing. Similarly, Ford
(1963) suggested that trainees were often attracted to the profession by a striving for
self-realization and self-identity, noting that student’s mothers tended to have a ‘pivotal
controlling role’ in the family, whereas fathers tended to be revealed as passive men who
failed to interact. Mollon (1993), argues that if the mother denigrates the father he
cannot function effectively as an ‘Oedipal rival’ or third person who could intervene in
this identiŽcation and help separate mother and child.
Ford points out that his psychotherapy trainees tended to view their mothers as
understanding, and had recollections of them helping others. This idea persisted despite
a growing awareness that this may not have been the case, and that she was in fact
pressing them to achieve. Ford inferred that such childhood experiences present severe
threats to the therapist’s ego integrity and require many years of working through during
the therapist’s training.
Burton (1970, 1972, 1975 in Glickauf-Hughes & Mehlman, 1995) suggested that
family life, in sensitizing the therapist to emotional pain, provides a powerful motivation
for career choice. Racusin, Abramowitz, and Winter (1981) used both qualitative and
quantitative data to look at the impact of therapist’s early family experiences on career
choice. They found that three-quarters of therapists interviewed were involved in
caretaking within the family in one form or another, either ‘parenting’ or ‘counselling’.
These individuals reported themselves to have played roles designed to fulŽl the needs
of others. As children they were selected at an early age because of their ability to satisfy
emotional needs of family members.
Glickauf-Hughes and Mehlman (1995) suggest that good therapists manifest sensi-
tivity, empathy, and awareness of the needs and feelings of others, the type of individuals
that parents with intense narcissistic needs often misuse. From childhood these
individuals are being prepared for a profession that involves listening to the messages
and unspoken needs of clients. However, therapists with this background often struggle
with their own unresolved narcissistic issues, which include audience sensitivity,
perfectionism, imposter feelings, and unstable self-esteem.
Half of the therapists in Racusin et al.’s study felt that their primary role had been to
provide parenting in the form of responsibility for family functions, or nurturing. Most of
the therapists felt they had Žlled this role for at least one parent. Others played the role
of counsellor or mediator, which entailed being aware of the emotional life of the family
and being responsible for reducing family tensions and resolving arguments.
‘ParentiŽcation’ is the term used by Glickauf-Hughes and Mehlman (1995) to
describe a situation in which the child is assigned an adult role in the family, or takes
on adult responsibilities which the parents may have abdicated. There are several
92 Andrea Halewood and Rachel Tribe
problems associated with this, including the belief that you are loved for what you do
not for who you are, difŽculties with boundaries and with identifying needs and getting
them met.
Perfectionism also becomes increasingly problematic for the trainee therapist owing
to the demands of the work. As Glickauf-Hughes and Mehlman (1995) point out,
becoming a therapist involves many feelings of doubt and uncertainty. The work lacks
structure and has been referred to as ‘the impossible profession’. As Hinshelwood
(1985) argues, the work is frustrating for those who need a concrete sense of
achievement. In addition, as Mollon (1989) suggests, the nature of the transference
and projection processes may cause a therapist to feel ineffective.

The ‘imposter phenomenon’

Therapists with narcissistic issues may also suffer from the imposter phenomenon
due to the presence of a false self (Langford & Clance, 1993). ‘Imposters’ often attribute
their achievements to personal charm and the ability to read others’ expectations, rather
than to ability.
If, as Mollon (1989) suggests, the lack of feeling a sense of genuine self-acceptance
and regard can make a psychotherapy training difŽcult for clinical psychologists because
of the narcissistic vulnerability it provokes, then the same could also be anticipated
for trainee counselling psychologists. As Mollon (1989, p. 120) argues: ‘Trainees
inevitably suffer injuries to their self-esteem and self-image when Žnding that they are
oundering.’
Another problem as Ford (1963) has pointed out is that the therapist’s self is
constantly probed and provoked by his or her clients. Gluckauf-Hughes and Mehlmon
(1995) counsel that as the vast majority of clients utilize defences and projections, and
that therapists do have a propensity to internalize these, counsellors will frequently
struggle with doubts and insecurities about being ‘good enough’.
These authors caution that ‘The very qualities that may enable an individual to . . .
excel as a therapist may make being a therapist difŽcult’ (Gluckauf-Hughes and Mehl-
man, 1995, p. 218).
These issues can lead to a number of problems for both counsellor and client, the
more so for trainee therapists with unstable self-esteem who have problems with setting
boundaries. When therapists fail to satisfy their own needs while exceeding their own
capacities, stress and burnout may occur. Symptoms of burnout include anxiety,
depression, exhaustion, irritability, emotional detachment, cynicism, boredom, and a
desire to withdraw from clients (Guy, 1987).
Corey (1991) suggests that certain personality traits and characteristics can increase
the risk of burnout, for example, a strong need for approval in the therapist or an
inordinate desire to be needed.
Studies suggest that the prevalence of distress and impairment among psychologists
is at least as high and often higher than in the general population (Laliotis & Grayson,
1985). Cushway (1992) studied the levels of stress in trainee clinical psychologists and
found that 59%of this group suffered from psychological distress.
Glickauf-Hughes and Mehlman (1995) suggest that narcissistically injured therapists
risk turning to their clients to meet their emotional needs in the same way that their
parents turned to them as children. A number of problems may emerge as a result. First,
the therapist may infantilize their client. Hardy (1979, quoted in Glickauf-Hughes
& Mehlman, 1995) suggests that by unconsciously attempting to rescue clients,
 Narcisstic injury among trainee psychologists 93
counsellors are assuming a dominant role, thereby allowing their clients less
independence. Second, counsellors may project their own needs onto clients, thus
misunderstanding the client. Finally, the counsellor may subtly discourage the client’s
negative transference by redirecting it to outside targets, or may be reluctant to
challenge the client for fear of breaking the narcissistic collusion.

Personal therapy
Guy and Liaboe (1986) recommend that more attention be drawn to the beneŽts of
personal therapy, not just as a means to enhance therapeutic skills, but also as a way to
minimize any harmful consequences of clinical practice on therapists.
The British Psychological Society (BPS) course requirement for personal therapy for
part one of the Diploma in Counselling Psychology is as follows
Candidates must undertake at least 40 hours of personal counselling psychology/therapy,
normally on an individual basis. Personal therapy should ideally be provided by an
experienced counselling psychologist . . . Under normal circumstances it is expected
that the candidate remains with the same therapist throughout. (British Psychological
Society, 1999, p. 8).

For part 2 the BPS states that ‘Candidates are encouraged to pursue a course of
personal therapy although they are not required to do so’ (British Psychological Society,
1999, p. 11).
Personal therapy is thought to be important for a number of reasons: to relieve the
stress of practising therapy, to improve the therapist’s awareness of their own problems
and areas of conict, and to help recognize and deal with countertransference issues.
The results of a study by MacDevitt (1987) indicate that awareness of counter-
transference issues is highly signiŽcantly and positively related to the amount of
personal therapy a therapist has received. Many theorists have argued that therapists’
awareness of countertransference reactions is viewed as essential for successful therapy
outcomes (Singer & Luborsky, 1977; Strupp, 1980; Winnicott, 1949 all as quoted in
MacDevitt 1987).
Freud noted in 1910 that each therapist’s complexes and resistances will limit his or
her achievements (Freud 1910/1916). Or, as Corey (1991) argues, therapists cannot take
their clients any further than they have been prepared to go.
The debate continues to rage around the statutory requirement that trainee
counsellors and counselling psychologists engage in personal therapy. The effects of
personal therapy on client or therapist outcomes are not clearly established, and the
evidence is not unequivocal for or against the beneŽts. (For a more in depth discussion
of this issue the reader is referred to Feltham, 1999; Grimmer & Tribe, 2001; Macran &
Shapiro, 1998; Macran, Stiles, & Smith, 1999.)
However, a study of narcissistic problems in analysts suggested that narcissistic
issues should be an area of concern (Schumacher Finell, 1985). Schumacher Finell
questions whether the narcissism of therapists is something of a blind spot, too sensitive
to be dealt with. Certainly this is a subject largely neglected in the research literature,
although there is a huge amount of research on narcissism in everybody else.
However, personal therapy may not resolve the problem if the narcissistic individual
has split off his or her aggression. They may idealize the therapist so much so that the
negative transference needs to be split off and projected into others. Glickauf-Hughes
and Mehlman (1995) suggest that personal therapy for the narcissistically injured
counsellor will be difŽcult for a number of reasons. Clients in this group tend to be
94 Andrea Halewood and Rachel Tribe
highly responsive to their therapist’s feelings and will be constantly monitoring
themselves and the therapist to ensure that they are not too burdensome. They may
be unaware of their own needs and therefore unlikely to ask that these needs be met,
and may also have great difŽculty in acknowledging any form of vulnerability. These
authors suggest that it is essential that the therapist’s therapist has adequately resolved
his or her own narcissistic and dependency issues before working with this client group.
Narcissism in the therapist may not, therefore, always be a reaction to the client’s
narcissism. It may exist independently and quite separately. As Shumacher Finell points
out, if the therapist fails to work through their own grandiosity, exhibitionism,
aggression, and power, the potential for countertransferenc e over and above a speciŽc
reaction to a narcissistic transference in clients is enormous. Furthermore, an over-
emphasis on using a particular label such as narcissistic personality disorder (NPD) may
reect projected narcissistic tendencies that can be defensively disowned by Žnding
them in one’s clients rather than in oneself. As Hinshelwood (1985) points out, trainees
may disown unwanted parts of themselves by projecting them into patients.
Given the dangers outlined above of unresolved narcissism, this study looks at the
degree of narcissism in trainee counselling psychologists to ascertain whether this
should be an area of concern.

The Narcissistic Injury Scale

In order to measure narcissistic injury the Narcissistic Injury Scale (NIS) was used. This
was developed by Slyter (1991) based on Miller’s (1981) book, The Drama of Being a
Child. Slyter took many of Miller’s descriptions of adults with narcissistic parenting and
developed corresponding items for the NIS. The scale was designed to measure the key
aspects of narcissistic injury:
· Restriction of emotional affect
· Lack of understanding of the self and the presence of a ‘false self’
· The need for mirroring
· Problems with setting boundaries
· Grandiosity
· Depression
· Perceptions of the parent–child relationship
· Perfectionism
· Feelings about the self.

Research hypotheses
The research hypotheses were as follows:
1. There will be a higher degree of narcissistic injury among trainee counselling
psychologists than among controls.
2. There will be a relationship between the parent–child relationship and narcissistic
injury.

Method
Design
The study had a between-groups quasi-experimental design. The experimental group
was composed of trainee postgraduate counselling psychology students (N = 36) and
 Narcisstic injury among trainee psychologists 95
the control group was composed of mature postgraduate students (N = 34). The groups
were matched for gender, age, and nationality.

Participants
The participants for this study were 36 counselling psychology students enrolled on the
second year of the BPS diploma all of whom had completed the 40 h of personal therapy
required by the BPS with a mean of 42 sessions. Participants were assessed using The
Narcissistic Injury Scale (Slyter, 1991). Forty-two percent of the counselling psychology
students were aged 26–35, 50%were aged 36–45, and 8%were aged 46–55. The sample
was predominantly female (60%).
The control group was composed of 34 postgraduate mature students. Six percent
were aged 19–25, 41%were aged 26–35, 41%were aged 36–45, and 12%were aged
46–55. None of the control group had undertaken any form of personal therapy. All
participants were recruited using the non-random selection technique of snowball
sampling among postgraduate students at three London universities. Students were told
that the study was concerned with the ‘developmental factors which may inuence the
choice of counselling psychology as a profession’. The questionnaire was administered
between March and April 2000.

Instruments
Narcissistic injury was measured by the Narcissistic Injury Scale (© Slyter, 1991). The
items for the instrument are scored by summing the ratings, which yields a possible
range of 38–228.
The total scale score for the NIS is the sum of 38 individual Likert ratings. Twelve items
are Žller items and are not used to compute the total but are included in the scale to
reduce the response set. The minimum total for the NIS is 38, and the maximum is 228.
Rios and Hill (1993) presented evidence for the construct validity of the NIS. These
authors found that the NIS correlated positively with depression, grandiosity, and
difŽculties in coping, and negatively with self-esteem and parental bonding.

Results
Hypothesis 1
The mean narcissistic injury scores are given in Table 1. A linear regression was used in
order to predict which of the independent variables (age, sex, and grouping) could
predict high scores on the NIS (dependent variable). This demonstrated that age and sex
were not signiŽcant variables, but that grouping was signiŽcant at the p > .001 level. A
t test was then used to test whether the difference between the two sets of scores was
statistically signiŽcant. This demonstrated a signiŽcant difference between the narcis-
sistic injury scores of the experimental group and the control group, with the
experimental group gaining signiŽcantly higher scores (t = 8.1; df = 68; p > .001,
one-tailed test).

Hypothesis 2
An additional area of interest was the relationship between perceived parent–child
relationship and score on the NIS (Table 2). Three of the ‘Žller’ items (‘my parents
96 Andrea Halewood and Rachel Tribe
Table 1. Mean of narcissistic injury scores on narcissistic injury scale (standard deviations in
parentheses)

Counselling psychologist
trainees Control group All

Female 133 (24) 91 (20) 112 (31)

N = 21 N = 20 N = 41
Male 149 (29) 100 (14) 126 (34)
N = 15 N = 14 N = 29
All 140 (27) 95 (18) 118 (32)
N = 36 N = 34 N = 70

Note. A t test for unrelated samples showed that the experimental group obtained signiŽ cantly higher
scores than the control group (t = 8.1; df = 68; p > .001, one-tailed test).

understood me’, ‘my parents respected my feelings’, and ‘my parents saw me as the
person I really was’) were used to assess this. Pearson’s product– moment correlation
coefŽcient was used to explore this relationship, and a negative correlation was found to
be signiŽcant at the .05 level (r = ± .258; df = 68; p > 0.05; two-tailed test).

Table 2. Correlation matrix showing relationship between parent–child relationship and narcissistic
injury score (N = 70)

NIS score Good relationship

NIS score 1.000 ± .258*

Good relationship ± .258* 1.000

* Correlation is signiŽ cant at the .05 level (two-tailed test).

Discussion
Evaluation of results
Hypothesis 1
A highly signiŽcant difference was found between the mean NIS scores of the trainee
counselling psychologists and that of the control group (t = 8.1; df = 68; p > .001; one-
tailed test), indicating a higher degree of narcissistic injury among trainee counselling
psychologists than among controls, as predicted.
The 38 items of the Narcissistic Injury Scale were then divided into subscales in order
to bring the variables together for analysis. Only those items that quite clearly pertained
to one of the subscales were included. Mean scores on all subscales of the NIS between
groups were highly signiŽcant, with trainee counselling psychologists scoring higher
than controls on the restriction of emotional affect (t = 6.1; p < .001), lack of under-
standing of the self and the presence of a false self (t = 4.1; p < .001), the need for
mirroring and understanding (t = 4.3; p < .001), problems with setting boundaries
(t = 4.3; p < .001), grandiosity (t = 3.9; p < .001), depression (t = 5.5; p < .001),
perceptions of the parent–child relationship (t = 7.0; p < .001), and perfectionism
(t = 4.5; p < .001).
 Narcisstic injury among trainee psychologists 97
These Žndings, suggesting the possible presence of narcissistic injury among trainee
counselling psychologists, may therefore support the contention that those with a high
degree of narcissistic injury are attracted to the therapeutic Želd. If this is so, trainees
may well struggle with the narcissistic issues outlined above, and may need to consider
how to resolve these. Kernberg (1975) commented on the difŽculties faced by
therapists in training whose narcissistic resistances were not resolved, and these results
may indicate cause for concern, certainly for further research, given that the sample was
taken from trainees who have already completed the requisite 40 h of personal therapy
demanded by the BPS.
As Schumacher Finell (1985) argues, narcissim is notoriously difŽcult to treat.
Narcissistic defences involve splitting, denial, and projection and therefore make insight
very difŽcult. She suggests that teachers and supervisors should attempt to identify the
narcissistic problems that are not picked up and worked through during the trainee’s
personal therapy. Having done so, they should then suggest that these issues be taken up
in personal therapy. She goes on to say that if the trainee counsellor cannot work
through these issues, only partial results will be achieved, and the student’s ability to
deal with these issues with clients will be limited.

Hypothesis 2
The second research hypothesis was that there would be a relationship between the
perceived parent–child relationship and the degree of narcissistic injury. This was
demonstrated in two ways. First, by computing the three ‘Žller’ items on the NIS which
related to a positive parent–child relationship. Slyter (1991) added these items to the
scale in order to reduce the response set, and consequently they were reverse scored
and so not included in the overall NIS score. However, they provided useful information
about perceptions of the parent–child relationship and so were retained for the purpose
of this study.
The items were as follows
1. My parents understood me.
2. My parents respected my feelings.
3. My parents saw me as the person I really was.
A negative correlation was found to be signiŽcant (r = ± .258; df = 68; p < .05;
two-tailed test). High scores on these questions therefore correlated with low scores
on the NIS, and therefore a lower degree of narcissistic injury. This could indicate that
those who enjoyed a parental relationship characterized by feeling understood,
respected, and mirrored, seem to have a far lower degree of narcissistic injury than
those for whom these elements were lacking. However, this Žnding would need to be
interpreted with some caution: narcissistically injured therapists may be reluctant to
criticize parents towards whom they may still adopt a protective role. As Ford (1963)
found in his study, his trainees entertained fantasies of their mothers as being
understanding and caring, an idea which persisted despite later awareness that this
may not have been the case.
Perceptions of the parent–child relationship were then explored in terms of the
items on the NIS which pertained to perceived parent–child relationship. The 38 items
of the NIS were divided into subscales in order to bring all the variables together for
analysis, and also to clarify the relative contributions of each dimension of narcissistic
98 Andrea Halewood and Rachel Tribe
injury to the total score. One of the subscales contained the following questions which
pertained to less positive perceptions of the parent–child relationship;
1. I must not show any dissatisfaction or disappointment with my parents, since this
would lead to their withdrawal and loss of affection.
2. I always had to be strong for my parents.
3. My parents were critical whenever anyone displayed weakness.
4. My parents reacted negatively to any expression of anger on my part.
5. As a child I had ‘been good’, suffering quietly and without crying.
A highly signiŽcant difference was found between the scores of the counselling
psychologists and the control group on this dimension (t = 7; df = 68; p < .001; two-
tailed test), indicating that the counselling psychologists perceptions of the parent–
child relationship were more negative than those of the control group; they perceived
their parents as less supportive. These relationships were characterized by a with-
holding of the self, a focus on good behaviour, and a need for the child to support the
parents.

Evaluation of the scale

Certain problems were encountered with the administration of the NIS. The language
and concepts used require a degree of psychological awareness and self-knowledge,
which was understandably less prevalent in the control group; therefore some
questions, particularly 12, 20, 32, and 42, caused confusion among controls. The
experimental group, in contrast, had all completed the 40 h of personal therapy
required by the BPS in addition to a training in counselling psychology, which
meant that they were far more familiar with the psychological concepts presented in
the questionnaire. This may well have inuenced the degree of narcissistic
injury recorded compared with that of the control group. There are also cultural
and linguistic problems involved in using an American scale on mainly British
participants.
Afurther theoretical problem is with the difŽculties around self-reported narcissism.
Arespondent who has completely split off his or her narcissim will deny it completely. It
could, therefore, be argued that the scale might not be powerful enough to detect more
extreme forms of narcissistic injury. Furthermore, the questionnaire suffers from having
obvious face validity.
In addition, a number of theorists have pointed out the difŽculties associated with
measuring self-reported narcissism (e.g. Watson & Biderman, 1994). The variables are
self-reported and are being studied retrospectively, and thus reect the potential biases
of such measures. The degree to which responses may have been affected by defen-
siveness and the desire to present well, both narcissistic traits, need to be considered.
The fact that the questionnaire guaranteed anonymity would not prevent this —
narcissists have a desire to think well of themselves.
However, methodological considerations aside, it is suggested that a greater
understanding of narcissistic issues and vulnerabilities would still beneŽt counselling
psychologists in a number of ways.

Motivation
Counselling psychologists may beneŽt from an understanding of why they were drawn
to the work in the Žrst place. Racusin et al. (1981) suggest that therapists’ awareness of
 Narcisstic injury among trainee psychologists 99
their covert motivations for career choice should facilitate professional growth and
maximize effective therapeutic functioning.

Countertransference awareness
Identifying and understanding his or her own need to receive narcissistic supplies
through working with clients can increase the counselling psychologist’s counter-
transference awareness which could prevent the counselling psychologist from enter-
ing into a narcissistic collusion with clients. Glickauf-Hughes and Mehlman (1995) warn
that in order to improve the quality of treatment provided to others with similar
difŽculties, the therapist should be aware of how these issues may manifest themselves
in their personal lives and, furthermore, that it is essential that the therapist is able to
differentiate between narcissistic clients and those who have been victims of some form
of narcissistic abuse.

Working with narcissistic clients

Awareness of their own latent narcissism may enable counselling psychologists to
increase their ability to work with clients who present with narcissistic issues and who
may, therefore, activate the counselling psychologist’s own narcissism. Johnson (1987)
warns that therapists who have been narcissistically injured themselves may resent
being ‘used’ by their clients as a mirror and sounding board, having previously had
experience of being used by their parent or parents. They may unconsciously resent
giving that which they so desperately want for themselves.

ParentiŽ cation
An understanding of their own role within the family and of how they may have been
‘parentiŽed’ may increase the counselling psychologist’s own understanding of their
family dynamics. Learning how their boundaries were not respected when they were a
child, how they were not able to set limits, may enable the therapist to improve these
abilities both with their clients and in their interpersonal relationships.

Ethical issues
Failure to develop self-awareness in counselling psychologists raises a number of
important ethical issues. Corey (1991) suggests that unless counsellors develop self-
awareness, they are in danger of obstructing the client’s progress or, worse still, they
may use their clients in various ways to satisfy their own needs. The British Association
for Counselling (1993) Code of Ethics and Practice for Counsellors stresses that
counsellors are expected to monitor their own personal functioning, and to seek
help if their personal resources are sufŽciently depleted to require this.

Burnout
Without an adequate examination of their own narcissistic issues, counselling psychol-
ogists are failing to observe these guidelines, with potential costs to their clients and to
themselves in terms of burnout. Nichols (1988) suggests that there may come a time
when, if psychologists fail to practise good self-care and use support systems, they will
be seen as guilty of irresponsibility in their professional behaviour.
100 Andrea Halewood and Rachel Tribe
It is therefore important that counselling psychologists are self-aware, honest, and
able to be self-reective and self-critical. As Shillito-Clarke (1996) points out, one of the
hallmarks of counselling psychology is the emphasis on self-awareness.
The current study has a number of implications for counselling psychology trainees
and trainers. If, as these data suggest, trainee counselling psychologists are more likely
to suffer from some form of narcissistic injury, then this should be an area of some
concern. The operationalization of narcissistic injury presented in this investigation
(NIS; Slyter, 1991) might be of value in identifying individuals who are at risk of various
psychological difŽculties resulting in burnout or dropping out of training altogether.
One problem in the therapy of these individuals is that they often separate themselves
from these experiences through defences such as denial, repression, and splitting.
Furthermore, if anger, hostility, and rage are central to the life of the narcissist as
Raskin, Novacek, and Hogon (1991) suggest, then this should also be an area of concern
for both the trainee counselling psychologist and the client.
As the research hypotheses were conŽrmed, this study adds some tentative support
to the research literature on the causes and manifestations of narcissistic injury.
However, the design of the study would be improved with a far larger sample size,
and would beneŽt from combining the NIS with other measures of narcissism and a
measure of global functioning and/or a depression inventory. This is an area that would
beneŽt from further research.

References
American Psychiatric Association. (1994). Diagnostic and statistical manua l of mental disorders
(4th ed.). Washington, DC: Author.
British Association for Counselling. (1993). Code of ethics and practice for counsellors. Rugby,
UK: British Association for Counselling.
British Psychological Society. (1999). Regulations and syllabus for the diploma in counselling
psychology April 1999 to March 2000. Leicester, UK: British Psychological Society.
Buss, D. M., & Chiodo, L. M. (1991). Narcissistic arts in everyday life. Journal of Personality, 59(2),
179–215.
Corey, G. (1991). Theory and practice of counselling and psychotherapy (4th ed.). PaciŽc Grove,
CA: Brooks/Cole.
Cushway, D. (1992). Stress in clinical psychology trainees. British Journal of Clinical Psychology,
31, 169–179.
Feltham, C. (1999). Controversies in psychotherapy and counselling. London: Sage.
Ford, E. S. C. (1963). Being and becoming a psychotherapist: The search for identity. American
Journal of Psychotherapy, 17, 472–482.
Freud, S. (1914). On narcissism: An introduction on metapsychology: The theory of
psychoana lysis, The Penguin Freud Library Vol. II. London: Penguin Books.
Freud, S. (1986). The future prospects of psychoanalytic psychotherapy (standard ed.). In
J. Klauber (Ed.), DifŽculties in the ana lytic encounter. London: Free Association Books
MaresŽeld Library. (Original published 1910).
Gelder, X., Gath, D., & Mayou, R. (1983). Oxford textbook of psychiatry (3rd ed.). Oxford, UK:
Oxford University Press.
Glickauf-Hughes, C., & Mehlman, E. (1995). Narcissistic issues in therapists: Diagnostic and
treatment considerations. Psychotherapy, 32(2), 213–221.
Grimmer, A., & Tribe, R. (2001). Counselling psychologists perceptions of the impact of
mandatory personal therapy on professional development—An exploratory study.
Counselling Psychology Quarterly, 14(4), 287–301.
 Narcisstic injury among trainee psychologists 101
Guy, J., & Liaboe, G. (1986). Personal therapy for the experienced psychotherapist: Adiscussion of
its usefulness and utilisation. The Clinical Psychologist, Winter, 20–23.
Guy, J. D. (1987). The personal life of the psychotherapist. New York: Wiley.
Hinshelwood, R. D. (1985). Questions of training. Free Associations: Psychoanalysis, Groups,
Politics, Culture, 2, 7–18.
Johnson, S. M. (1987). Humanizing the narcissistic style. New York: Norton.
Kernberg, O. (1975). Borderline conditions and pathological narcissism. Northvale, NJ:
Aronson.
Kohut, H. (1971). The analysis of the self. New York: International University Press.
Kohut, H. (1977). The restoration of the self. New York: International University Press.
Kohut, H. (1978). Forms and transformations of narcissism. In P. Ornstein (Ed.), The search for the
self (Vol. 1, pp. 427–460). New York: International Universities Press. (Original published
1966)
Laliotis, D. A., & Grayson, J. H. (1985). Psychologist heal thyself: What is available for the impaired
psychologist? American Psychologist, 40(1), 84–96.
Langford, J., & Clance, P. (1993). The impostor phenomena: Recent research Žndings regarding
dynamics, personality and familypatterns and their implications for treatment. Psychotherapy,
113(3), 495–501.
Lichtenstein, H. (1977). Identity conŽguration and developmental alternatives. The dilemma of
hum an identity. New York: Aronson.
MacDevitt, J. (1987). Therapists’ personal therapy and professional self-awareness. Psycho-
therapy, 24(4), 693–703.
Macran, S., & Shapiro, D. A. (1998). The role of personal therapy for therapists: A review. British
Journal of Medical Psychology, 71, 13 –25.
Macran, S., Stiles, W. B., & Smith, J. A. (1999). How does personal therapy affect therapists’
practice? Journal of Counselling Psychology, 46(4), 419–431.
Menninger, K. (1957). Psychological factors in the choice of medicine as a profession. Bulletin of
the Menninger Clinic, 21, 51–58, 99–106.
Miller, A. (1981). The drama of being a child. London: Virago.
Mollon, P. (1989). Anxiety, supervision and a space for thinking: Some narcissistic perils for clinical
psychologists in learning psychotherapy. British Journal of Medical Psychology, 62, 113–122.
Mollon, P. (1993). The fragile self—The structure of narcissistic disturbance. London: Whurr.
Nichols, K. A. (1988). Practising what we preach. The Psychologist, February, 50–51.
Ornstein, P. (1991). From narcissism to ego psychology to self psychology. In J. Sandler (Ed.),
Freud’s on narcissism: An introduction (pp. xx –xx). London: Yale University Press.
Phillips, A. (1988). In F. Kermode (Ed.), Winnicott (pp. xx –xx). Location: Fontana Modern
Masters.
Racusin, G. R., Abramowitz, S. I., & Winter, W. D. (1981). Becoming a therapist: Family dynamics
and career choice. Professional Psychology, 12(2), 271–279.
Raskin, R., Novacek, J., & Hogan, R. (1991). Narcissism, self-esteem and defensive self-
enhancement. Journal of Personality, 59(1), 19 –38.
Rios, X., & Hill, X. (1993). The correlates of narcissistic injury: Parental bonding, depression,
grandiosity and the impact of divorce. Master’s thesis research, University of Maryland.
Rowan, J. (1989). A late developer. In W. Dryden & L. Spurling (Eds.), On becoming a
psychotherapist (pp. xx –xx). London: Tavistock/Routledge.
Russell, G. A. (1985). Narcissism and the narcissistic personality disorder: A comparison of the
theories of Kernberg and Kohut. British Journal of Medical Psychology, 58, 137–148.
Schumacher Finell, J. (1985). Narcissistic problems in analysts. International Journal of
Psycho-Analysis, 66, 433–445.
Shillito-Clarke, C. (1996). Ethical issues in counselling psychology. In The handbook of
counselling psychology (pp. xx –xx). London: Sage.
Slyter, S. (1991). The na rcissistic injury scale © (Rev. version). Unpublished instrument.
Storr, A. (1979). The art of psychotherapy. Oxford, UK: Butterworth Heinemann.
102 Andrea Halewood and Rachel Tribe
Watson, P. J., & Biderman, M. D. (1994). Narcissistic Traits Scale: Validity evidence and sex
differences in narcissism. Persona lity and Individua l Differences, 16(3), 501–504.
Winnicott, D. W. (1976). Ego distortion in terms of true and false self. In The maturational
processes and the facilitating environment (pp. xx –xx). London: Hogarth. (Original pub-
lished 1960)
Woolfe, R. (1996). The nature of counselling psychology. In X. Woolfe & X. Dryden (Eds.), The
handbook of counselling psychology (pp. xx –xx). London: Sage.
Zamostny, K. P., Slyter, S. L., & Rios, P. (1993). Narcissistic injury and its relationship to early
trauma, early resources, and adjustment to college. Journal of Counselling Psychology, 40(4),
501–510.

Received 26 March 2001; revised version received 15 July 2002