The crucial roles of attachment in family therapy
John Byng-Halla
This paper's aim is to enable family therapists from whatever approach to
address family attachments during their work. It explores the role of
attachment in the family, and how to enable therapists to increase security
in the family so that family members can solve their own problems during
and after therapy. The article gives a brief overview of the nature of family
attachment relationships and the influence of secure and insecure
attachments within the family and their narrative styles. This is described
in language that a therapist might readily hold in mind and share the
ideas in dialogue with families. The paper discusses the interplay between
insecure attachments and other family problems, such as parental conflict
and disagreements over authority. It also discusses ways of establishing a
secure therapeutic base and the influence of the therapisťs own attachment
style. The implications for family therapy practice are described and
illustrated by work with a specific family.
Introduction
Over the past few years interest in attachment has grown, and the
appeal of this way of conceptualizing family interaction is finding a
place in the panoply of family therapy conceptualizations. Most family
therapists will now be familiar with the basic patterns of attachment.
However, the application of attachment ideas to everyday family
therapy practice has not yet been fully integrated.
Attachment is part of family life; family therapists are inevitably
relating to family attachments in every family and couple seen,
whether or not they expressly address this in their theory and practice
base. Attachment is in mutual influence with every other function
within the family ecology (Hill et al., 2003). A family therapist has to
deal with issues and dilemmas arising in any family interaction
including those interactions which are activated by attachment needs,
r The Association for Family Therapy 2008. Published by Blackwell Publishing, 9600 Garsington
Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
Journal of Family Therapy (2008) 30: 129­146
0163-4445 (print); 1467-6427 (online)
a
Child and Family Psychiatrist, Institute of Family Therapy. Address for correspondence:
27 Estelle Road, London NW3 2JX, UK. E-mail: john.byng-hall@blueyonder.co.uk.
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
and then the therapist has to make sense of how these internal systems
articulate with each other.
Attachment plays a key role in the processes of change in family
therapy. The family needs to develop a sufficiently secure attachment
to the therapist in order to feel safe enough to explore new ways of
relating. This in turn enables them to become more secure with each
other. In therapy they acquire the capacity to explore new ways of
solving problems. This capacity may then be applied to solving fresh
problems in the future.
Attachment has been researched extensively and there is an
expanding volume of research-based knowledge about it; for instance,
the Handbook of Attachment (1999 in revision, Crittenden and Claussen,
2000). Attachment has been linked to systemic practice (Byng-Hall,
1995, 1999a; special attachment edition of Family Process, 41:3, 2002;
Erdman and Caffery, 2003; Johnson and Whiffen, 2003, Akister and
Reibstein, 2004; special issue of Context, April 2007).
Some approaches to family therapy have integrated attachment
fully into another specific approach (Dallos, 2006). This paper aims to
describe family attachment relationships in a simple and clear way that
could enable a family therapist to use attachment ideas within the
intensity of ongoing interactions during therapy in whatever modality
they have been originally trained. The language used is also aimed to
make it easier to convey the ideas to families and hence make it
possible for the therapist to have a dialogue with the family about
attachments (Vetere, 2006; Bertrando and Arcelloni, 2006), and to coconstruct
attachment discourses. For this purpose it tries to avoid
using the jargon of research terminology as far as possible.
In this context attachment is making its way into family therapy
training. It may not yet be a standard part of the curriculum on all
courses, but family therapists can benefit from having an overall
systemic picture of attachment for their work. It is not yet possible to
assume that all family therapists are familiar with these ideas. Therefore
an overview of attachment patterns is included here.
Overview of attachment patterns
Patterns of attachment across the life span will be described here in
terms of relationships rather than using one term for children and
another for adults as is usual in most attachment research. The family
therapist needs to be able to recognize the predominant attachment
strategies used in the family's relationships, and be aware of any
130 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
incoherent patterns in their narratives about attachment issues.
Incoherent narratives are characterized by contradictions in
the story which are difficult to clarify or discuss without explicitly
addressing them. Young children have generally not developed
the capacity to ask for clarification in confusing adult messages
(Crittenden, 2006).
The attachment system is activated when fear or anxiety is present.
Reflecting on frightening times also invokes the attachment response
in children from preschool age upward so long as they have begun to
develop the ability to reflect on their experience. Indeed many of the
projective procedures such as Narrative Stem Stories (Hodges et al.,
2000) and the School Age Assessment of Attachment (Crittenden,
2006) are based precisely on the phenomenon of reflection activating
the attachment system.
Secure parent/child attachment relationships
The majority of attachments are secure and have been shown to be
associated with fewer problems and better social adjustments. The
parent is emotionally available, and sensitively responsive to the chilďs
needs. When the child is distressed or frightened this leads to parent
and child getting together for comfort or protection. This is called
attachment behaviour. The child comes to feel secure enough to
explore, knowing that the parent has her or him in mind. The parenťs
narrative is coherent which means that it is consistent, plausible and
capable of conveying feelings appropriate to the situation being discussed.
The child learns a similar way of talking which contributes to the
capacity for empathy with each other. Coherent narrative is associated
with secure attachments, but incoherent narrative is associated with
insecure attachments (Byng-Hall, 1999b; Hesse, 1999).
Insecure parent/child attachment relationships
1 Insecure avoidant. The parent avoids coping with uncomfortable
feelings about attachment experiences and so rejects a chilďs bids for
comfort when in distress, and tends to be dismissive of the importance
to the child of negative feelings. This parental preference for only
positive feelings is transmitted to the child who in turn is faced with
the dilemma of trying to ignore her own needs for comfort from
others and to give the appearance of being prematurely self-sufficient.
This self-reliance, while preserving the relationship with the parent, is
Crucial roles of attachment in family therapy 131
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
not thought to be the same as true self-reliance that comes when
children are more securely attached. The chilďs attachment behaviour
is underactivated in comparison with securely attached children.
The parent/child relationship may appear emotionally cool and
distant. Narrative is sparse and takes little account of feelings, being
concentrated instead on achievements or external objects.
2 Insecure ambivalent. The parent is only intermittently available
because the parent is often preoccupied with his or her own past
upsets. The child learns to cling on and keep close to the parent to
avoid being abandoned. In order to be soothed in situations of stress
or danger the child needs to make her needs highly apparent and
bring the parent back from preoccupation with the past. The relationship
is over-close but feelings are ambivalent. The child may appear to
reject the parent through moments of resistance to the closeness with
pushing away. The parent may then distance herself from the
apparent rejection. Thus a pattern of interaction can be established
in which the chilďs attachment behaviour becomes overactivated.
Stories are long, convoluted and without resolution.
3 Insecure disorganized/disorientated. The child may become disorientated
and/or disorganized when the parent has an unresolved trauma
and, at times, is frightened because the past trauma is being relived as
if in the present. The parenťs fear is frightening to the child who does
not know whether to approach or go away from the parent. A similar
approach/avoidance conflict arises when a child is maltreated by a
parent. Insecure disorganized/disorientated children may develop
strategies that are very controlling or compulsively caregiving in
response to the lack of organization and care.
In some situations the chilďs behaviour may trigger a parenťs
deeper conflict. In situations, not as uncommon in clinical practice as
might be thought, where the child is frightening to the parent, a
paradox comes into being in which the child turns for comfort to the
parent who is the person who is frightening them and hence activating
the chilďs attachment system.
In attachment research, unresolved trauma may be connected to
one of the other categories of attachment (Hesse et al., 2003). It is
therefore possible to have a secure attachment style and still be
`unresolveď with regard to a specific trauma. Where parents are
otherwise balanced in attachment terms, their narrative is coherent
except when talking about the trauma.
132 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Insecure attachments between parents
Romantic attachments include sexuality and shared parenting.
Evidence shows that these systems are connected to each other within
the attachment system (Hazan, 2003). Insecure couples' attachments
that are matching in style can create patterns not dissimilar to those
described above. Insecure avoidant couples are distant, while insecure
ambivalent are close, intense but with mixed feelings (Lowyck
et al., 2008). A relationship between an avoidant and an ambivalent
parent, however, creates distance conflicts; these are discussed below.
In clinical experience, couples with these patterns frequently present
for therapy (Johnson, 2002).
Family attachment patterns
Minuchin's (1974) overall categories of family patterns equate well
with patterns of attachment: adaptable families equate with the secure
pattern; disengaged families equate with insecure/avoidant patterns;
and enmeshed families equate with insecure/ambivalent patterns
(Marvin, 2003). Attachment patterns in families, however, are usually
more complex, with differing types of attachment in the various
relationships, including children who can have different patterns of
attachments with each parent.
There is research evidence to support the notion of a hierarchy of
importance of attachments within the family (Berlin and Cassidy,
1999; Howes, 1999).
Case study: the Brown family
The Brown family will be used to illustrate features of attachment
patterns as well as the process of therapy. They were referred to me
because their 6-year-old daughter Jean was anxious, asthmatic and
resisted going to school. The parents, Angela and George, were in
total disagreement about what to do in the family and could not make
any decisions. Fred, aged 10, tried to be helpful to his parents'
relationship by taking on a parenting role.
One of the most telling moments of family attachment patterns
occurs as the family comes into the therapy room which is strange to
them. The novelty and formality is likely to activate the attachment
system, and the strategies young children have developed may be
revealed. In the first session Fred came in first, glancing back at his
parents; father followed, looking at the therapist. Mother came in,
Crucial roles of attachment in family therapy 133
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
holding Jean's hand. Fred looked around the room and at the chairs.
He glanced up at his father and then looked down at a chair at one
end of the circle of chairs. Father was talking to me and, without
thinking, took his son's cue and sat in the chair signalled by Fred. Fred
sat next to him, thus ensuring that his mother would sit on the
opposite side of the room from his father. Jean sat on her mother's lap
and tried to pull her mother's coat over her face. Angela let her do
this. The parents did not communicate verbally with each other at all
during this time, nor did they look at each other. A few minutes later
George demanded that Jean get off her mother's lap and sit in a chair,
which she did reluctantly, and I could then see that she and her
mother appeared to resent this. I noticed myself feeling angry with
George and protective of Jean, but did not show it. This awareness of
my feelings gave me certain clues about some of the dynamics of the
family which are discussed later.
Conceptual frameworks for understanding family attachments
Family scripts and attachment scripts
Research has shown that the individuals' scripts for attachment can be
linked to behaviour in families (Vaughn et al., 2006). The systemic
concept of family scripts can provide a framework that encompasses
all family functions, including attachment, which may be called family
attachment scripts. The definition of a family script is the family's shared
expectations of how family roles are to be performed in various contexts (ByngHall,
1995, p. 4). The term `expectations' includes anticipation of what
will be said and done in family relationships, as well as involving
family pressures on members to perform the roles as expected. The
shared expectations are made possible, as Marvin and Stewart (1990)
suggest, by evolving a shared working model of how the family
functions. The various roles, such as comforting, frequently become
identified with particular family members, but if someone is absent or
a task not done, everyone is broadly aware of what is to be done, and
others may then be recruited to take it on, including a child doing
some parenting.
Some parents' expectations of how family roles are to be performed
are based on those experienced in their family of origin, thus
following what is called a replicative family script. In the Brown
family Angela, as a child, had been expected by her parents to take on
a parental role in the family, just as Fred was in his generation.
134 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Attachment research shows that role reversal in children takes on
different forms within the various insecure attachment strategies
(Byng-Hall, 2002, 2008). Research shows that the majority of attachment
patterns are replicated in the next generation. However, a
sizeable minority vary, thus showing that change is possible. There
are however attempts by parents not to repeat what they felt were
mistakes made by their parents. These are called corrective parenting
scripts. Angela was determined to be a better parent than her parents,
but slowly became aware in therapy of how she expected Fred, and to
some extent Jean, to take on parenting roles.
A secure family base
The family as a whole can provide a secure base for its members. The
definition of a secure enough family base (Byng-Hall, 1995, p. 104) is
a family that provides a reliable network of attachment relationships that
enables all family members, of whatever age, to feel secure enough to explore
their relationships with each other and with others outside the family. One
feature that is essential in achieving this state is the capacity for
collaboration, which is shown to be higher in secure families (Cobb,
1996). It is collaboration that makes it feel safe enough to explore
fresh ways of relating. The Brown parents were not able to collaborate
with each other.
Interactional awareness
The concept of interactional awareness, an essential part of collaboration,
is useful in thinking about how family members keep in touch
with what is going on in interaction in various situations, and so
maintain secure family relationships (Byng-Hall, 1995).
Interactional awareness is defined as the ability to be sensitive to how
each family member is likely to be experiencing the unfolding situation and the
role that each may be expected to play, and to be aware of one's own feelings and
how one's imminent behaviour might impact on the others. This involves
keeping the whole family, including oneself, in mind during the
interaction, while retaining the possibility of adjusting one's own
behaviour to influence what is evolving. A major element of this
capacity is empathy, which enables one to put oneself in others' shoes
within the current context. This can lead to tuning into the emotions
and feelings of others in a particular situation. Secure parents can
respond appropriately to their children's distress and upsets within
Crucial roles of attachment in family therapy 135
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
family scenarios. Children can expect this support, and feel safe and
know they are in their parents' minds. Another element of interactional
awareness is reflective functioning (Fonagy et al., 1991; Slade,
2005), which enables the understanding of underlying mental states
and hence the intentions of others, and of oneself. This makes
it possible for the individual to respond appropriately to others as
a family scenario unfolds.
Secure enough families can share how they feel about what happens
during and after the event. The narrative used is coherent (Byng-Hall,
1997, 1999b; Dallos, 2004), conveying an appreciation of how they all
felt about it, including hurt or uncomfortable feelings. In this way family
members can update their collective view of how each of them reacts to
particular contexts, and what each may have preferred to have
happened. This is useful when facing similar situations in the future.
It is then possible to collaborate in ensuring that everyone, especially
children, is kept in mind and their preferences known, at least by
someone in the family. The image of `family' can then provide members
with a sense of security. `They have me in mind.' For instance, a child
can rely on being handed over and handed back, within the family and
with reliable outsiders. Not every member of the family needs to be
secure for this to happen, as those who are more secure can collaborate.
Affect leading to action; exploration when calmer
Hill et al. (2003) observed and researched family interaction. Among
other things they describe two general modes beyond the routine
family functioning mode. The Affect/Action mode is driven by
heightened affect that leads to rapid automatic responses when
attachment behaviour is triggered by fear of danger. Other family
interactions can also evoke Affect/Action modes, including those such
as anger, clashes over power, or dangerous behaviour requiring
discipline. In any of these fraught situations attachment behaviour
within the family may also be evoked. This can create a complex
situation in which both attachment behaviour and other behavioural
systems are activated but are in opposition. A common clash is that
between moves to comfort and protect versus discipline. Emphatic
discipline can be misinterpreted as anger or rejection. Angela was
protective of the sick Jean from her husbanďs insensitive discipline.
George could see Jean's need for clear limits, and he became cross
when his wife did not support his authority, thus making her even
more protective.
136 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
The other mode is Exploratory, which can emerge when affect is
low and there is no urgency. Now an exploratory review of existing
assumptions and the revising of working models of how the family
functions can be started. In family script terms there are often
established cues for opening up a discussion, expressing curiosity,
and starting to improvise fresh ways of relating and so on. This
depends on family relationships feeling secure enough to disagree
and take risks even in uncertain situations.
Distance conflicts
Conflicts can arise when there are opposing strategies in the parents'
attachment patterns. For instance, George used an insecure avoidant
strategy to pre-empt being upset about possible rejections; Angela, on
the other hand, used an insecure ambivalent strategy of clinging on to
him to prevent being abandoned. These relationships produce what I
call a `too close/too far' conflict (Byng-Hall, 1995). For example, the
Browns' relationship would feel `too close' for George, but `too far' for
Angela. This led to George distancing, which made Angela seek even
more closeness, and so on until George stormed out. Systems theory
explains this by using the cybernetic concept of mutual positive
feedback amplification. Each move is intended to slow the escalation
down through negative feedback, but has precisely the opposite effect,
acting as positive feedback, which amplifies the other's behaviour
instead. This creates rapid escalations pushing the family into Affect/
Action mode, leading to highly disruptive pursuing/distancing cycles
which threaten the relationship.
Lynn Hoffman (1975) described these escalations in family life as
`runaways'. A chilďs intervention can halt pursuing/distancing runaways,
sometimes dramatically, for instance, by being in such danger
that it is given priority, such as Jean's asthma attacks. This triangulates
the child into the parents' relationship. This may then become the
chilďs role in the family script. In the long term a triangulated child
who remains a worrying problem can reduce the danger of further
runaways because both parents are shoulder to shoulder looking after
her. This keeps parents together, but not intimately. In this family it was
different. Jean's parents disagreed about how to help her, each believing
the other would make her worse, so they stayed together to protect Jean
from each other, both trying to help her in contradictory ways.
Another way to reduce the potential escalations is created by the
parent who is in desperate need of closeness, and so turns to a child
Crucial roles of attachment in family therapy 137
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
who then becomes `too close' to that parent but `too far' from the
other parent. Jean found herself in this position. She and her mother
became inseparable, excluding her father who disapproved of closeness.
A boundary infringement can also arise in parental power battles
when one parent recruits a child as an ally, thus creating a crossgenerational
coalition against the other. This undermines parental
authority and can interfere with the sibling group cohesion. Mother
and Jean were not only excluding father but also in a coalition against
him; while Fred being parental did not play with Jean, but bossed her
around and also tried to organize the parents.
Children can reduce the frequency of escalations by taking parental
roles such as peacekeeper, go-between or mediator, or just take over
the parental role that is not functioning; for instance, Freďs role in
guiding his parents to a decision. He was also a mediator. The
runaways did not happen as frequently as they might have done in
the Brown family because of the children's involvement.
Implications for family therapy
Creating a secure therapeutic base
A family's sense of urgency increases the likelihood of quickly establishing
a temporary attachment relationship to a therapist who is seen
as able to provide a solution in threatening situations. For this reason
it is important, if possible, for the therapist to respond quickly. It is
valuable for the therapist to ring the family home to make the
appointment. This is done preferably at a time in the early evening
when those at work might be at home. Ten minutes' warm and
supportive discussion about the problem and how it affects each
member of the family can lead to a relationship with a parent who
feels that she would like to bring her family. A useful practice is to
invite the whole family, and make attempts to find a time when all are
able to attend. This whole process also helps to widen the perspective
to include all members. Fred would probably have been left behind as
the good boy who did not need help. The family's attachment pattern
would not have been addressed adequately, and the consequences of
his taking on a parental role not tackled.
The first session could, if possible, last one and a half hours or
more. This allows time for painful or very difficult issues to emerge
with some intensity and have some chance of being addressed. During
this time the family will hopefully experience the therapist as someone
138 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
who is able to hear about, and be ready to tackle, the serious issues
facing them. The therapist is sensitive to family members' anxieties,
enabling each to feel that the therapist could be useful to them. Family
therapists need to deploy interactive awareness in their work. This
involves scanning the family, observing how each member is responding
to what is going on; occasionally asking how they are feeling or
what they think; and noting the interactions between each member. At
the same time the therapist is concerned about his own impact on each
member of the family. This sensitivity is likely to be noticed by
members, who then experience the therapist as being there for
each of them, as well as for the family as a whole. They feel the
therapist has the potential to hold the whole family in mind. Sessions
are then ideally held every two weeks to begin with, which leaves
family members in charge of themselves but with sufficient momentum
to continue exploring new ways of relating while holding the
therapy in their minds.
The interplay between therapist and family
Therapy should be a form of dialogue. The therapist, by sharing a
hypothesis with the family when it first comes to mind and before
feeling certain, can lead to the family discussing the idea. This allows
for a co-evolution of ideas about what is going on (Bertrando and
Arcelloni, 2006; Vetere, 2006). The therapist might say something like
this: `I've just had this idea. I have seen situations a bit like yours in
which . . . , some of this might fit your family, but some I expect won't.'
The therapisťs uncertainty is genuine and enquiring which can lead
to a mutual exploration of the idea and its implications (Mason, 1993).
It is also a mutual discussion which is complementary. The family
comes to someone who has had experience of many family problems
while the family knows about what goes on at home. The therapist is
thus constantly learning more about the rich variety of family life.
Attachment issues have the additional advantage in that the therapist
can, when appropriate, refer to patterns found through research that
was carried out on normal populations, which helps to allay the
family's shame about being uniquely bad.
The family's emotional state is likely to arouse affect in the therapist
in which she feels she has to take some action. If the therapist does
take action, however, she risks being drawn into the family's script; but
by remaining in reflective mode the therapist will acquire clues as to
what is going on in the family. My anger with the father, George, in the
Crucial roles of attachment in family therapy 139
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
first session when he demanded that Jean, the daughter, should sit on
her own chair, I came to realize, was directed towards protecting Jean
and wanting to stop her father exposing her to a frightening situation.
This was indicating an important family dynamic, involving both
attachment and discipline issues. If I had been drawn into such action
I would have been in danger of joining mother and Jean's crossgenerational
coalition against father, and would have reduced my
capacity to establish a secure therapeutic base for the whole family.
Therapists need to reflect on the significance of their impulses to act
and what that experience might tell them about the family.
At the same time therapists need to be aware of their own family
scripts and the sorts of scenarios that are likely to trigger their action,
or to draw them into adopting a particular attitude to what is going on
in the family; in other words, it is too close to home. It has been shown
through self-reports from therapists of many orientations that those
who have secure attachment styles have better therapeutic alliances
than insecure therapists (Black et al., 2005). Many therapists have
played parental roles in their families of origin (Jurkovic, 1997) and
have identified with the caring role in their families. This early
training is very valuable but may have been more appropriate to
the past rather than now. It is helpful for therapists to be aware of the
style in which their caring role is usually deployed, but which they can
then adapt to fit the family's particular needs. Mine is to be rather too
protective and positive towards family members and hesitant about
necessary challenges.
Discussing present and past episodes
I often start each session by asking `What has been happening?' which
can elicit stories about family scenes, rows or escalations. It is valuable
to hear the family's accounts, noting how stories are told, asking
questions about how they felt and extending the enquiry wider to
include how the children were involved if they had not been mentioned.
Were they trying to mediate, keep the peace or were they too
scared of the looming threat and so left the room? This review process
is an opportunity to make family members more aware of how they
affect each other as the saga unfolds. It also adds greater sensitivity to
each other's difficult dilemmas. This review also provides an opportunity
to explore the mutual influence between attachment and other
functions, such as discipline. We explored how George and Angela
could collaborate over discipline as opposed to George being firm
140 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
with Jean and Angela protecting her anxious child from what she saw
as bullying.
Drawing a family tree
I usually draw a family tree when the family is less caught up with
urgent issues, somewhere between the second and fourth sessions.
The process of drawing the family tree provides the family and the
therapist with an image of the family and the extended family as it is
now, and how this was influenced by past generations. It can also
provide an intense experience for family members who can see their
own story dramatically unfold, juxtaposing so many aspects of their
life together and revealing previously unseen patterns.
This also provides a valuable moment to explore attachment issues.
Questions about the relationships in the parents' upbringing and the
traumas and losses can evoke powerful feelings. The way these are
described can indicate whether or not the trauma was unresolved
(Zulueta, 2006). There are some invaluable routine questions, such as
`And were there any other pregnancies?' after naming a line of
siblings, thus revealing miscarriages, abortions and stillbirths, which
are potentially pivotal family events not discussed and possibly
unresolved. A routine question to ask when told about a trauma is
`And who looked after whom?' Frequently children look after their
grieving parents and can then become parental. The effect of the
traumas on attachments down the generations can be explored.
Collaborative interventions with parents
It is valuable to explore how the parents' attachment strategies
interact with each other. This approach enables collaboration, rather
than one member being identified as having the problem. When
working with parents or couples I label repeated escalations or rows as
`vicious circles', an idea they quickly recognize, and discuss the
situations that lead to these escalations. Even if the row was not about
attachment it can become so frightening that it can evoke attachment
responses, as the people involved seek safety. This may of course
include observers, such as children.
After reporting an episode, work can be done to help each family
member see why his or her own behaviour sets off the other's
reactions. It is equally valuable for each parent to understand the
origin of their partner's responses. One way this can be done is to turn
Crucial roles of attachment in family therapy 141
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
to the family tree to explore how the current situation resonates with
traumatic events from each of their pasts. This can start a discussion
on why they respond to each other as they do, and come to appreciate
the partner's vulnerabilities and where they came from. Each partner's
feeling of being understood is hugely important. A collaborative
intervention avoids blaming each other and establishes the idea of
`Here we go again' as opposed to `Here he/she goes again'.
One day the Brown parents came on their own. They reported
what they called a `bit of a blow-up'. Angela had been looking after a
frienďs baby. She got upset, once again, about not having any more
children. She wanted to be comforted, he got irritated, and she started
crying and went to sit on the sofa next to him. George shouted, `Will
you never stop this nonsense?' and stormed out. Angela told me that
he was reluctant to have another baby.
Looking at the family tree we could see that Angela had come from
a large family and a much-loved baby sister had died at age 10
months. Angela was 7 at the time and she had loved and cared for the
baby, and longed for a replacement that never arrived. When he was
13 George's mother had died after being ill for seven years. His rather
distant father then remarried quickly. His sister married and moved
out. Both his stepmother and his sister then had babies. George
recollected bitterly that `Then nobody had me in minď, and he
became determined to be even more self-sufficient. They could share
the knowledge that for him babies had been a disaster, for her not
having babies was a disaster. They were encouraged in future to share
together when either of them felt a rising sense of `Here we go again'
in an impending vicious circle. This could help them to move out of
Affect/Action to Exploratory mode so that they could discuss what was
going on instead. Appreciating each other's painful experiences
enabled them to become friendlier with each other, a significant
aspect of a more secure attachment.
Angela wanted to do further work on her losses. At the age of 2 she
had been taken to hospital for a month for a corrective operation.
Visiting times were limited and she was told that she was very clingy
for the rest of her childhood, `and I still am'. I pointed out that each of
them was dealing with losses and insecurity but were understandably
using opposite strategies: she was clinging on to avoid being left, while
he was making himself sufficiently independent to avoid being hurt.
He predictably played down the importance of all this, but was able to
acknowledge his upset. This reduced their polarization somewhat,
and they felt a bit more that they were at least on the same planet.
142 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
After this, Angela, surprisingly, became more withholding and
pushed him away. She did not like finding that she had a distancing
side of herself. George, however, started to spend more time in the
same room as her. She experienced feeling this as intrusive and stayed
out more herself. `Runaways' or escalations in conflictual situations
can be known to suddenly reverse (Hoffman, 1975). This was
happening here. Therapy enabled the couple's mutual projections
to become apparent. He had been scornful of her neediness which hid
his own longing for care for himself, while she had hated his withdrawing
and indifference which hid her own ambivalence about being
intimate. Angela was able to work with this idea while George seemed
to understand, but without seeming to take it on board.
Work with the children
During the therapy, work with the children had been going on, much
of it on Jean's asthma, and other issues not directly linked to
attachment. I did some work with the children to help to release
them from their role in their parents' relationship. This included
talking to them about how it felt when they tried to help their parents
when they became cross with each other. This was something of a
revelation to the parents. Time was spent working on how the parents
could collaborate over their way of looking after Jean. Establishing the
need for discipline, in order to get her to school, as well as the need
for comfort and protection when frightened by asthma, was important.
Jean became less anxious as she saw her parents collaborating
rather than squabbling. I started by alternating whole family sessions
and parental sessions, but increased the proportion of parental
sessions, partly because the children were settling down. When
parental children see that the therapist is looking after their parents
effectively they can start relaxing their vigil and be children. Fred
started playing with Jean rather than keeping an eye on his parents.
Finishing
Finishing therapy involves a steadily increasing length of time between
sessions, often ending with a three-month gap. This extends
the length of contact without increasing therapy time, enabling
families to establish ways of solving their own problems. I tell them
that they can come back to discuss situations and do not have to be
desperate before they do so. In this way I make myself available and
Crucial roles of attachment in family therapy 143
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
would be experienced as keeping them in mind. A sense of availability,
even if only in mind, is fundamental to autonomy in attachment. In
my clinical experience the encouragement to come back before
matters get too bad leads to fewer returns to therapy than final
discharges that can lead to re-referrals in a deteriorated state.
Therapy with the Browns finished once the parents were secure
enough to collaborate when needed and had found a way to make
decisions. They were still having disputes but these did not get out of
hand so readily. Jean's asthma was better controlled, she was going to
school without difficulty, and Fred was more relaxed. The parents
were more able to treat each other's differences as complementary,
rather than just opposing each other. He could now be seen as having
his feet on the ground, while she was the one who was in touch with
her emotions.
Conclusions
Family therapists need to know how to address the various subsystems
within the ecology of the family, including attachment. The attachment
subsystem plays a crucial role in family therapy because it
provides a secure base during therapy leading to a more secure
base for the family in the future. The importance of this should not
lead to an approach that only focuses on the attachment subsystem or
on any subsystem for that matter. Further research is needed to
explore the nature of the relationship between subsystems, thus
enabling the work to be truly systemic.
Acknowledgements
I would like to acknowledge the very valuable contributions Chip
Chimera has made. Her many creative ideas have greatly enhanced
this paper.
References
Akister, J. and Reibstein, J. (2004) Links between attachment theory and systemic
practice. Journal of Family Therapy, 26: 2­16.
Berlin, L. J. and Cassidy, J. (1999) Relations among relations. In J. Cassidy and
P. R. Shaver (eds) Handbook of Attachment: Theory, Research and Clinical Applications.
New York: Guilford Press.
Bertrando, P. and Arcelloni, T. (2006) Hypotheses are dialogues: sharing
hypotheses with clients. Journal of Family Therapy, 28: 370­387.
144 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Black, S., Hardy, G., Turpin, G. and Parry, G. (2005) Self-reported attachment
styles and therapeutic orientation of therapists and their relationship with
reported general alliance quality and problems in therapy. Psychology and
Psychotherapy: Theory, Research and Practice, 78: 363­377.
Byng-Hall, J. (1995) Rewriting Family Scripts: Improvisation and Systems Change. New
York and London: Guilford Press.
Byng-Hall, J. (1997) Toward a coherent story in illness and loss. In R. Papodopoulos
and J. Byng-Hall (eds) Multiple Voices: Narratives in Systemic Family
Psychotherapy. London: Duckworth.
Byng-Hall, J. (1999a) Family and couple therapy: toward greater security. In
J. Cassidy and P. R. Shaver (eds) Handbook of Attachment: Theory, Research and
Clinical Applications. New York: Guilford Press.
Byng-Hall, J. (1999b) Creating a coherent story in family therapy. In G. Roberts
and J. Holmes (eds) Narrative Approaches in Psychiatry and Psychotherapy. Oxford:
Oxford University Press.
Byng-Hall, J. (2002) Relieving parentified children's burdens in families with
insecure attachment patterns. Family Process, 41: 375­388.
Byng-Hall, J. (2008) The significance of children fulfilling parental roles:
implications for family therapy. Journal of Family Therapy, 30: 147­162.
Cobb, C. L. H. (1996) Adolescent­parent attachments and family problemsolving.
Family Process, 35: 57­82.
Crittenden, P. M. (2006) A dynamic­maturational model of attachment. Australia
and New Zealand Journal of Family Therapy, 27: 105­115.
Crittenden, P. M. and Claussen, A. H. (2000) The Organisation of Attachment
Relationships: Maturation, Culture and Context. Cambridge: Cambridge University
Press.
Dallos, R. (2004) Attachment narrative therapy, integrating ideas from narrative
and attachment theory in systemic family therapy with eating disorders.
Journal of Family Therapy, 26: 40­65.
Dallos, R. (2006) Attachment Narrative Therapy: Integrating Systemic and Narrative
Approaches. Maidenhead and New York: Open University Press.
Erdman, E. and Caffery, T. (2003) Attachment and Family Systems: Conceptual,
Empirical, and Therapeutic Relatedness. New York and Hove: Brunner-
Routledge.
Fonagy, P., Steele, M., Steele, H., Moran, G. S. and Higgitt, A. C. (1991) The
capacity for understanding mental states: the reflective self/parent in mother
and child and its significance for security of attachment. Infant Mental Health
Journal, 12: 201­218.
Hazan, C. (2003) The essential nature of couple relationships. In S. M. Johnson
and V. E. Whiffen (eds) Attachment Processes in Couple and Family Therapy. New
York and London: Guilford Press.
Hesse, E. (1999) The adult attachment interview: historical and current perspectives.
In J. Cassidy and P. R. Shaver (eds) Handbook of Attachment: Theory,
Research and Clinical Applications. New York: Guilford Press.
Hesse, E., Main, M., Abrams, K. Y. and Rifkin, A. (2003) Unresolved states
regarding loss or abuse can have `second-generation' effects: disorganization,
role inversion, and frightening ideation in the offspring of traumatised, nonCrucial
roles of attachment in family therapy 145
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
maltreating parents. In M. F. Solomon and D. J. Siegel (eds) Healing Trauma
Attachment, Mind, Body and Brain. New York and London: W. W. Norton.
Hill, J., Fonagy, P., Safier, E. and Sargent, J. (2003) The ecology of attachment in
the family. Family Process, 42: 205­222.
Hodges, J. and Steele, M. with Hillman, S., Henderson, K. and Neil, M. (2000)
Effects of abuse on attachment representations: narrative assessments of abused
children. Journal of Child Psychotherapy, 26: 433­455.
Hoffman, L. (1975) Enmeshment and the too richly cross-joined system. Family
Process, 14: 457­468.
Howes, C. (1999) Attachment relationships in the context of multiple caregivers.
In J. Cassidy and P. R. Shaver (eds) Handbook of Attachment: Theory, Research and
Clinical Applications. New York: Guilford Press.
Johnson, S. M. (2002) Emotionally Focused Couple Therapy with Trauma Survivors.
New York and London: Guilford Press.
Johnson, S. M. and Whiffen, V. E. (eds) (2003) Attachment Processes in Couple and
Family Therapy. New York and London: Guilford Press.
Jurkovic, G. J. (1997) Lost Childhoods: The Plight of the Parentified Child. New York:
Brunner/Mazel.
Lowyck, B., Luyten, P., Demyttenaere, K. and Corveleyn, J. (2008) The role of
romantic attachment and self-criticism and dependency for the relationship
satisfaction of community adults. Journal of Family Therapy, 30: 78­95.
Marvin, R. S. (2003) Implications of attachment research for the field of family
therapy. In P. Erdman and T. Caffrey (eds) Attachment and Family Systems:
Conceptual, Empirical and Therapeutic Relatedness. New York and Hove: Brunner-
Routledge.
Marvin, R. S. and Stewart, R. B. (1990) A family systems framework for the study
of attachment. In M. T. Greenberg, D. Cicchetti and E. M. Cunningham (eds)
Attachment in the Preschool Years: Theory, Research, and Intervention. Chicago, IL:
University of Chicago Press.
Mason, B. (1993) Towards positions of safe uncertainty. Human Systems, 4:
189­200.
Minuchin, S. (1974) Families and Family Therapy. Boston, MA: Harvard Press, and
London: Tavistock.
Slade, A. (2005) Parental reflective functioning. An introduction. Attachment and
Human Development, 7: 269­282.
Vaughn, B. E., Waters, H. S., Coppola, G., Cassidy, J., Bost, K. K. and Verissimo,
M. (2006) Script-like attachment representations and behaviour in families
and across cultures: studies of parental secure base narratives. Attachment and
Human Development, 8: 179­185.
Vetere, A. (2006) Commentary ­ The role of formulation in psychotherapy
practice. Journal of Family Therapy, 28: 388­382.
Zulueta, de F. (2006) The treatment of psychological trauma from the perspective
of attachment research. Journal of Family Therapy, 28: 334­351.
146 John Byng-Hall
r 2008 The Author. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice