Systemic therapy and attachment narratives
Arlene Veterea
and Rudi Dallosb
In this paper we describe how we bring together three major systems of
thought: attachment theory; family systems theories; and theories of
narrative development, in an integrated approach to systemic practice
that we call attachment narrative therapy (ANT). ANT provides a
four-stage framework for practice: creating a secure base; exploring
narratives and attachment experiences within a systemic framework;
considering alternatives and taking action; and the future and maintenance
of the therapeutic base. Formulation, especially regarding the role
of emotions and attachments in family dynamics and narratives, is at the
heart of our approach and helps us hold ourselves accountable for our
particular weave of theory and practice.
Introduction
We have been working for a number of years on a shared systemic
project that weaves together three major systems of thought:
attachment theory; family systems theories; and theories of narrative
development (Vetere and Dowling, 2005; Dallos, 2006; Vetere
and Dallos, 2007). We have articulated this project into an approach
to therapeutic practice called `attachment narrative therapy' (ANT).
ANT offers a four-stage framework for thinking about the processes
of couple and family therapy: creating a secure base (Byng-Hall,
1995, 2008); exploring narratives and attachments within a systemic
framework; de-escalating unhelpful patterns and considering
alternatives; working towards the future, and maintaining a
therapeutic base.
The narrative turn in attachment theory
Bowlby (1980, 1988) emphasized from the start that attachments
function through our `working models' ­ a system of meanings or set
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: 374­385
0163-4445 (print); 1467-6427 (online)
a
Deputy Director of the PsychD in Clinical Psychology, University of Surrey, UK.
b
Programme Director D. Clin. Psychology, University of Plymouth, UK.
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
of beliefs/expectations about how we see others and ourselves. These
attachment representations or internal working models are thought
to guide our actions, thoughts and feelings, and help us to
make predictions about behaviour in relationships. They are intimately
concerned with protection and safety in relationships ­ how
we keep our selves safe, close to, and loved by others. Importantly
our capacity to reflect on experience and our narrative ability to
tell coherent stories about our lives is shaped by our attachment
experiences in families (Main et al., 1985; Crittenden, 1998).
Consistent with this development, there has been a move towards
assessing attachments through both the content and the form
or structure of the stories people tell about their lives. For example,
a secure attachment is indicated by a positive view of others as
caring and our self as worthy of care. In addition, we are able to
develop coherent and reflective stories of our experiences that help us
to develop strategies to cope with conflicts, changes and challenges as
they arise in the future. We are free to be able to employ both our
thoughts and feelings to make sense of, anticipate and develop
solutions to deal with events. In contrast in avoidant attachment
narratives the person has learnt that her feelings will not be attended
to and ceases to be able to adequately employ emotional material
to help manage her life and relationships. In contrast, with an
ambivalent style she may have learnt that her parents' words
and actions are unpredictable and not to be trusted, and the child
may therefore develop an inability to make use of cognition ­ ideas of
when or why things happen, and come to overly rely on her
immediate feelings. Finally, if the main carer is frightened and/or
frightening, abusive and confusing, as we often see in homes where
domestic violence occurs, the child finds it difficult to develop and
predict a consistent pattern of responding, and learns to distrust
both words and feelings (Liotti, 2004). The key issue here is that
these response patterns emerge at times of attachment threat, and
attachment insecurity.
In all the insecure forms of attachment the person has difficulty in
being able to reflect on her own and others' experiences ­ to stand in
others' emotional shoes and to be able to reflect back to others that we
connect with how they are feeling and thinking. In addition, since we
shut down some of the richness of the information about experiences
with others we are less able to talk with others or to engage in internal
dialogues that help us to make sense of and continue to develop
strategies for coping.
Systemic therapy and attachment narratives 375
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Attachment styles as communicational patterns
We can describe attachment styles as `open' or `closeď communication
patterns in families, or as a set of family communicational rules
(Minuchin, 1974) about what can be discussed, with whom, and in
what way, and so on. For example, with secure attachment patterns,
the expression of both positive and negative feelings will likely be met
by acknowledgement, reflection and negotiation ­ a mix of semantic
and emotional responses. However, with more insecure attachment
patterns, communication may involve distortions in that we cannot say
openly or clearly how we feel and what we need. For example, with a
dismissing family pattern, the communication of feelings may be met
by an injunction to suppress feelings and `keep a stiff upper lip', or by
avoidance and/or rejection. Conversely, a more preoccupied style
could lead to a symmetrical escalation of feelings, characterized by
mutual accusations and blaming ­ an emotional responsiveness that is
so overwhelming that reflection is very difficult.
Thus the theoretical intersect between attachment and family
systems lies in the notion that insecurity can develop from family
interactional and communicational patterns. Periods of change, such
as family life cycle stages (see Figure 1), may offer both opportunities
and challenges For example, a new mother with a predominantly
avoidant style, and with the support of a more securely attached
partner, may find that her baby's demands for care and attention
trigger a more emotionally connected and secure strategy or alternatively
lead to deterioration.
Implications for therapy
Some of the key implications that flow for systemic therapy from our
amalgamation of ideas are as follows.
Naming feelings. In insecure patterns family members may have come
to ignore or oversimplify feelings, and helping them to be able to
identify, elaborate and communicate these feelings to each other and
themselves is an important step. For example, in work with trauma
and violence it is often the case that being able to identify alternative,
softer, more vulnerable feelings in each other and one's self is central.
Standing in the emotional shoes of the other. This goes beyond encouraging
and supporting empathic responding; it speaks to our capacity to
376 Arlene Vetere and Rudi Dallos
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
tolerate and bear negative emotion as expressed by the people we
love. We are not talking about abusive experiences, but, for example,
if a family member is upset, another family member is able to get
alongside their upset and listen, and offer support and comfort as
appropriate, rather than trying to distract too soon. Good listening is
soothing and calming, and helps a person feel deeply understood.
This involves both being able to experience each other's feelings but
also to feel safe to be able to show and communicate to each other
this understanding. For example, in families with an avoidant style
they may feel that if they resonate with and show each other their
sadness, and do not put a brave face on things, then everything may
fall apart.
Comfort and self-soothing. We meet many people who have not been
helped as children and young people to comfort, soothe and calm
themselves, and who have learned not to trust others to provide
comfort in a reliable way (Hazan and Shaver, 1987). They may have
Mother Child
Mother ­ may feel overwhelmed by her infanťs emotional demands,
her dismissive strategy may be thrown into turmoil: may reject the baby,
feel bad, inadequate, confused... withdraw further... abuse, neglect...
leading a consolidation of her dismissive pattern or even a crisis to a
more extreme disorganized pattern...
more secure one ...
leading to a reorganization of the dismissive attachment pattern to a
experience a new way of relating, giving and receiving affection,
if, e.g. supported by partner, parents, friends, she may be able to
Mother ­ may feel overwhelmed by her infanťs emotional demands but
Figure 1. Family life cycle points and reorganization of attachment patterns
Systemic therapy and attachment narratives 377
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
turned to psychoactive substances for emotional soothing or to
emotionally and physically risky or dangerous activities almost as if
to prove to themselves that they are still alive.
Information processing. Attachments are represented in multiple ways:
as embodied (pre-verbal experiences), visual images, episodes or
stories, semantic generalizations and reflective processes. When we
are afraid, in the face of actual or perceived loss, abandonment and
rejection, we may have learnt to cut out some of these different forms
of experience, or overemphasize some at the expense of others. In
avoidant patterns we may not be able to employ visual or embodied
information from others and either delay in responding emotionally,
or alternatively we may become preoccupied with regulating our own
arousal and become worse at reading relationship cues from others.
Creating a sense of security can help people to access more of the
information available in relationships and consequently to be able
to take relational risks; for example, being more able to see
others' expressions, postures, feelings, needs and vulnerabilities can
allow some different possibilities: helping to create some `news of
difference'.
Transformations in representational systems. Helping people develop more
secure and satisfying intimate interactions means that their response
repertoire expands, and their style shifts to become more inclusive of
other possibilities. For example, a partner who downplays the significance
of emotion and withdraws emotionally during conflict is
helped to take emotional risks and come forward. Similarly, partners
who become over-aroused and preoccupied during these moments,
perhaps pursuing and blaming, need to be helped to soften down, so
that they can be responsive when their more withdrawn partner
reaches out for them.
Therapy as scaffolding: supporting emotional and cognitive development
Therapy may be seen as helping people start to take small steps or
relational risks (Mason, 2005) within their zone of safety and narrative
ability (Vygotsky, 1962; Bateson, 1972). Therapy can encourage a
more secure emotional base for couples and family members, with its
emphasis on creating a trusting, non-judgemental and accepting
environment within which people can begin to illuminate their
emotional experiences, walk around in them with the support
378 Arlene Vetere and Rudi Dallos
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
and validation of the therapist and begin to process and reprocess
emotions. In this context unhelpful arousal can be understood
and processed, emotional safety can be developed, and critical and
blaming patterns of interaction can be de-escalated. People are
encouraged to notice and respond to each other's feelings, and to
offer words and phrases in identifying feelings. We may offer words
and phrases to talk about how others might feel, or to help discuss
how people might comfort each other, and support them in trying. We
assist with the integration of feelings and events, bringing together the
different representational systems of semantic memory, episodic
memory, sensory memory and procedural memory.
Formulation and hypothesizing from an ANT approach
We are aware that it is not simple, desirable or even possible to put
families into attachment style boxes, which interestingly bear resemblances
to early structural family therapy of disengaged or enmeshed
family patterns of relating (Minuchin, 1974; Minuchin et al., 1978;
Hillburn-Cobb, 1998). However, we suggest that by paying attention
to attachment styles we can make some helpful adjustments to how we
work with different families or how we adjust our stance with different
family members.
Dismissive patterns
Under conditions of attachment threat, someone who withdraws
emotionally in intimate relationships and downplays the significance
of emotion may need help to `warm up', to take emotional risks. In
our view, the systemic approaches and techniques that address
an avoidant and dismissing attachment style by encouraging the
expression of feelings include:
 enactments and role plays
 empathic questioning and coaching
 internalized other interviewing, and other Gestalt techniques, such
as the empty chair
 caring and comforting
 identifying areas of conflict, conflict management and de-escalating
unsafe patterns of interaction.
Systemic therapy and attachment narratives 379
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Preoccupied patterns
Similarly, people who become overly anxious and preoccupied in
intimate moments, such as over-arousal, have a reduced capacity to
reflect on their actions, think through consequences and `think
straighť, need to be helped to calm down and begin to think
reflectively. In our view, the systemic approaches and techniques
that address an ambivalent and preoccupied attachment style by
encouraging the expression of cognitions include:
 genograms and lifelines
 tracking circularities
 mapping relationships
 scaling questions
 circular questions
 identifying beliefs, punctuations and shared family beliefs.
We would note that reflecting team discussions may be adapted to
encourage the development of more emotional responsiveness, or the
development of reflectivity in family members' interactions.
Attachment narrative therapy with couples and families
We conceptualize our approach as having four interconnected phases.
Importantly, we see the creation of a secure base as an essential
first step and one that in itself contains the central ingredients of
therapeutic work. In fact for many families, creating this relationship
of trust and safety is the most demanding and skilled part of therapy.
Creating a secure base
John Byng-Hall (1995) first wrote of the importance of therapy as
creating a safe haven and secure base within which trust could
develop, and emotional risk-taking (Mason, 2005) could lead to
more positive and satisfying interactions. In creating a secure base
we engage warmly with all family members, listening carefully to their
concerns, identifying resilience and safety, mapping the context for
the work, and if necessary, having talks about talks. We reflect on
process, and our relationships in the room, trying to model open and
straightforward communication, and often use our own reflections
upon our own experiences. We may use an externalizing framework if
380 Arlene Vetere and Rudi Dallos
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
it helps avoid a perception of blaming, or helps family members soften
from a critical stance towards a more cooperative orientation in the
work. We often find that identifying unhelpful patterns of interaction
promotes non-blaming as family members see the pattern as the
problem, and not each other. We fully explore the problems that
trouble people, emotional responses and needs, beliefs and explanations,
triggers and stresses, and resources and sources of support,
while trying to support what is going well. If possible we try to expand
and illuminate emotional experience, and encourage people to walk
around in it with us, so that we may say we deeply understand their
concerns. This is often the first opportunity to process or reprocess
emotional experience and helps in de-escalating unhelpful patterns,
as family members witness each other's intentions, wishes, hopes and
fears unfolding with safety. We are not pushing for change as such,
rather encouraging the expansion of experience and experiencing.
Exploring narratives and attachment experiences within a systemic framework
Just as children are able to venture, play and explore from their sense
of the safety of their parents so families start to be able to explore their
explanations, narratives and feelings about each other. We pay most
attention initially to current attachments and invite family members to
move from descriptions, sometimes catalogues of the problems and
blaming to the more subjugated stories of vulnerabilities, needs,
hurts and the love that they want to feel for each other but is
currently so difficult. We may help to contextualize their stories
within transgenerational attachment traditions and subcultural
demands and expectations. Therapy may well move to healing within
earlier attachment patterns, or within the extended kin network,
addressing emotional cut-offs, for example, or addressing earlier
attachment traumas and their impact on relationships in the present.
We encourage sociality and empathy, and focus on good listening, and
its potential for soothing and calming. We encourage awareness of self
and others in interaction while tracking the impact of the `problem' on
family members and vice versa. We draw on practices of `gossiping'
and use circular questioning when people are at risk of finding
therapeutic exposure overwhelming.
Considering alternatives and taking action
We describe middle therapy as a process of working within and
between ­ moving between intra- and inter-personal narratives
Systemic therapy and attachment narratives 381
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
and emotional experiences. We contemplate alternative narratives
and emotional responses. Building on the suppressed and subjugated
stories of vulnerabilities and unmet needs that may have emerged
from the explorations above we may invite families to start to
take relational/attachment risks. Enactments, both spontaneous and
organized, are a feature of the work here, whereby people are
encouraged to take emotional risks ­ to become less withdrawn, and
to soften and become less critical and more responsive. Constant
support and validation from the therapist is crucial as people
are encouraged to open up to other, more satisfying and deeply
bonding possibilities in their relationships. This is not about teaching
problem-solving skills as such. It is more about helping to create an
environment where people can reconnect and take risks that had not
been possible at home. Family members may engage in `therapeutic
experiments' outside the therapy room. Exceptions and unique
outcomes are explored and supported in a constant process of
feedback between therapy sessions and family life (White, 1995).
Impasses in therapy often occur here, because an earlier, unspoken
family experience blocks avenues of change (Johnson and Best, 2003).
This may be something that occurred between a couple or a parent or
child, where one of them was hurt, or felt disappointed and let down
by the other, and yet the other did not know at the time. The hurt
person may have vowed at that time not to trust the other again.
These interactions can become `islands' of experience that inform
moments of intimacy and emotional risk-taking in ways that stall the
development of further intimacy and trust. In our experience these
moments need to be understood and processed, and all those
involved need to be helped to listen and to hear, and to find ways
through the experience that both address the hurt and offer healing
opportunities.
We add a word of caution here, in that we do not work in these ways
if people are living in abusive relationships and contexts. See Cooper
and Vetere (2005) for a description of how we pay attention to safety
first, before contemplating therapeutic work that potentially makes
people emotionally vulnerable to others.
The future and maintaining the therapeutic base
This stage is all about consolidating desired changes, promoting
more satisfying exchanges, and continuing to support what is going
well. We make plans to deal with setbacks and contemplate future
382 Arlene Vetere and Rudi Dallos
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
possibilities. We find that when people experience a stronger sense of
felt security in their close relationships, they turn more easily to attend
to issues of practical problem-solving. In this stage we focus more on
integration, and how people create a shared account, or narrative of
how they healed their relationship. The ability to narrate experiences
in clear and coherent ways draws on all our abilities within our
representational systems, rather than limiting our possibilities by
relying on one or two representational systems to the detriment of a
fuller lived experience in our intimate relationships.
In many ways, we anticipate the ending of therapy from the
beginning, preparing the ground for people to cope without
therapeutic support by creating and strengthening their mutual
bonds. Having said that, however, there are times when it is more
helpful to consider ways of continuing contact, through aftercare
approaches and follow-up meetings. We talk about endings, and
feelings around separations, and try to find ways to end that suit
the people we work with, for example, by extending the length of
time between sessions in this stage of the work. Above all we take care
not to pathologize dependency. For some families this may be the
first time they have been able to come to trust someone ­ us and
the therapy team. We do not want to replicate a sense that we will, as
others appear to have done, just abandon and forget about
them when they walk out of the door following their last formal
session with us.
Putting ANT into practice
To offer guidelines for working with families and attachments we have
developed a number of `Formats for exploration'. We outline two
below and a further set will be published in our forthcoming book,
Systemic Therapy and Attachment Narratives: Specific Applications. Two of
these formats (briefly) are:
1 Corrective and replicative scripts ­ where we invite family members
to think in a three-generational perspective about what they may have
tried to do similarly or differently with their children in contrast to
how their parents were with them. Either way this can allow us to
positively connote and validate their intentions to make things better
for the children. From this place of positive connotation it can be
easier to explore what is or is not working now and what emotions are
tied into their scripts.
Systemic therapy and attachment narratives 383
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
2 Exploring patterns of comforting ­ here we focus attention on the
emotional needs and softer emotions in families, their vulnerabilities
and hurts. The questions can be transgenerational in terms of
exploring how the grandparents comforted (or did not), the parents'
experiences, and how in turn they dealt with the children when they
were physically hurt, upset or ill.
Conclusion
We have attempted to present an overview of how we have brought
ideas from attachment, systemic and narrative approaches together to
work with families. We are not advocating this as a distinct new model
but as an approach that captures for us some combinations of ideas
that are useful from the different approaches. Interestingly this is
similar to Bowlby's own approach which was an attempt to integrate
what was best in theory and research in his time. In a way this
integrative framework also encapsulates what is at the heart of
attachment theory and our ANT approach; namely that we believe
that as family members become more able to be open and draw upon
all the information that is potentially available to them in their
relationships with others, they have the materials available to develop
an integrative and reflective stance on and in their lives. To use a
building metaphor, to have all the material available does not mean we
are able to build a palace. We also need plans, opportunities, and the
support to practice and to discuss with others how to go about this.
ANT therapy is about helping families to be able to both gather the
necessary materials together and to be able to discuss their plans with
us to start with and then among themselves.
References
Bateson, G. (1972) Steps to an Ecology of Mind. New York: Ballantine Books.
Bowlby, J. (1980) Attachment and Loss, Vol. 3. Loss. New York: Basic Books.
Bowlby, J. (1988) A Secure Base: Clinical Applications of Attachment Theory. London:
Tavistock.
Byng-Hall, J. (1995) Creating a secure base: some implications of attachment
theory for family therapy. Family Process, 34: 45­58.
Byng-Hall, J. (2008) The crucial roles of attachment in family therapy. Journal of
Family Therapy, 30: 129­146.
Cooper, J. and Vetere, A. (2005) Domestic Violence and Family Safety: Working
Systemically with Family Violence. Chichester: Wiley.
384 Arlene Vetere and Rudi Dallos
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice
Crittenden, P. (1998) Truth, error, omission, distortion, and deception: an
application of attachment theory to the assessment and treatment of psychological
disorder. In S. M. Clany Dollinger and L.F DiLalla (eds), Assessment and
Intervention Issues Across the Life Span. London: Lawrence Erlbaum.
Dallos, R. (2006) Attachment narrative therapy: integrating ideas from narrative
and attachment theory in systemic family therapy with eating disorders.
Journal of Family Therapy, 26: 40­66.
Hazan, C. and Shaver, P. (1987) Romantic love conceptualised as an attachment
process. Journal of Personality and Social Psychology, 52: 511­524.
Hillburn-Cobb, C. (1998) Adolescent­parent attachments and family problemsolving
styles. Family Process, 35: 57­82.
Johnson, M. and Best, M. (2003) A systemic approach to restructuring adult
attachment: the EFT model of couples therapy. In. P. Erdman and T. Caffrey
(eds), Attachment and Family Systems. New York: Brunnner-Routledge.
Liotti, G. (2004) Trauma, dissociation and disorganised attachments: three
strands of a single braid. Psychotherapy: Theory, Research Practice and Training,
41: 472­486.
Main, M., Kaplan, N. and Cassidy, J. (1985) Security in infancy, childhood and
adulthood: a move to the level of representation. In I. Bretherton and
E. Waters (eds), Growing Points of Attachment Theory and Research. Monographs
of the Society for Research in Child Development, 50: Serial No. 209.
Mason, B. (2005) Relational risk taking and the therapeutic relationship. Journal
of Family Therapy, 27: 183­305.
Minuchin, S. (1974) Families and Family Therapy. Harvard, MA: Harvard University
Press.
Minuchin, S., Rosman, B. and Baker, L. (1978) Psychosomatic Families: Anorexia
Nervosa in Context. Cambridge, MA: Harvard University Press.
Vetere, A. and Dallos, R. (2007) Attachment narratives and systemic therapy.
Context, 90: 5­9.
Vetere, A. and Dowling, E. (eds) (2005) Narrative Therapies with Children and Their
Families: A Practitioner's Guide to Concepts and Approaches. London: Routledge.
Vygotsky, L. (1962) Thought and Language. Cambridge, MA.: MIT Press.
White, M. (1995) Re-authoring Lives: Interviews and Essays. Adelaide, Australia:
Dulwich Centre Publications.
Systemic therapy and attachment narratives 385
r 2008 The Authors. Journal compilation r 2008 The Association for Family Therapy and Systemic Practice