Compassion Focused Therapy

Research into the beneficial effect of developing compassion has advanced enormously in the last ten years,
with the development of inner compassion being an important therapeutic focus and goal. This book explains
how Compassion Focused Therapy (CFT)—a process of developing compassion for the self and others to
increase well-being and aid recovery—varies from other forms of Cognitive Behaviour Therapy.

Comprising 30 key points this book explores the founding principles of CFT and outlines the detailed
aspects of compassion in the CFT approach. Divided into two parts—Theory and Compassion Practice—this
concise book provides a clear guide to the distinctive characteristics of CFT.

Compassion Focused Therapy will be a valuable source for students and professionals in training as well as
practising therapists who want to learn more about the distinctive features of CFT.

Paul Gilbert is Professor of Clinical Psychology, University of Derby and has been actively involved in
research and treating people with shame-based and mood disorders for over 30 years. He is a past President
of the British Association for Cognitive and Behavioural Psychotherapy and a fellow of the British
Psychological Society and has been developing CFT for twenty years.
Cognitive-behavioural therapy (CBT) occupies a central position in the move towards evidence-based
practice and is frequently used in the clinical environment. Yet there is no one universal approach to CBT
and clinicians speak of first-, second-, and even third-wave approaches.

This series provides straightforward, accessible guides to a number of CBT methods, clarifying the
distinctive features of each approach. The series editor, Windy Dryden, successfully brings together experts
from each discipline to summarise the 30 main aspects of their approach divided into theoretical and
practical features.

The CBT Distinctive Features Series will be essential reading for psychotherapists, counsellors, and
psychologists of all orientations who want to learn more about the range of new and developing cognitive-
behavioural approaches.

Titles in the series:

Acceptance and Commitment Therapy by Paul Flaxman and J.T.Blackledge

Beck’s Cognitive Therapy by Frank Wills

Behavioral Activation by Jonathan Kanter, Andrew Busch and Laura Rusch

Compassion Focused Therapy by Paul Gilbert

Constructivist Psychotherapy by Robert A.Neimeyer

Dialectical Behaviour Therapy by Michaela Swales and Heidi Heard

Metacognitive Therapy by Peter Fisher and Adrian Wells

Mindfulness-Based Cognitive Therapy by Rebecca Crane

Rational Emotive Behaviour Therapy by Windy Dryden

Schema Therapy by Eshkol Rafaeli, David P.Bernstein and Jeffrey Young

For further information about this series please visit www.routledgementalhealth.com/cbt-distinctive-

features
Compassion Focused Therapy

Distinctive Features

Paul Gilbert

LONDON AND NEW YORK

First published 2010 by Routledge
27 Church Road, Hove, East Sussex BN3 2FA

Simultaneously published in the USA and Canada

by Routledge
270 Madison Avenue, New York NY 10016

Routledge is an imprint of the Taylor & Francis Group, an Informa business

This edition published in the Taylor & Francis e-Library, 2010.

To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of
eBooks please go to www.eBookstore.tandf.co.uk.

© 2010 Paul Gilbert

Cover design by Sandra Heath

All rights reserved. No part of this book may be reprinted or

reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.

This publication has been produced with paper manufactured to

strict environmental standards and with pulp derived from
sustainable forests.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication Data

Gilbert, Paul, 1951 June 20–
Compassion focused therapy: distinctive features/Paul Gilbert.
p. cm.
ISBN 978-0-415-44806-2 (hbk.)—ISBN 978-0-415-44807-9 (pbk.)
1. Emotion-focused therapy. 2. Compassion-Psychological aspects.
3. Cognitive therapy. I. Title.
RC489.F62G55 2010
616.89’1425–dc22
2009046045

ISBN 0-203-85119-6 Master e-book ISBN

ISBN: 978-0-415-44806-2 (hbk)

ISBN: 978-0-415-44807-9 (pbk)

Contents

Preface and acknowledgements vii

Part 1 THEORY: UNDERSTANDING THE MODEL 1

1 Some basics 3

2 A personal journey 13

3 The evolved mind and Compassion Focused Therapy 19

4 Multi-mind 29

5 Attachment and the importance of affection 39

6 Affect regulation: The three affect-regulation systems, caring and CFT 43

7 Affiliation, warmth and affection 53

8 Clarifying the CFT approach 59

9 Formulation 67

10 Shame 83

11 Self-criticism 93

12 Distinguishing shame, guilt and humiliation: Responsibility vs. self-critical blaming 101

13 Distinguishing compassionate self-correction from shame-based self-attacking 105

14 Threat and the compensations of achievement 109

Part 2 COMPASSION PRACTICE 115

15 Understanding soothing: The wider context of balancing affect-regulation systems 117

16 The nature of compassion 125

 p

17 Preparing and training one’s mind: Mindfulness and soothing breathing rhythm 137

18 Introducing imagery 145

19 Creating a safe place 151

20 Varieties of compassion focused imagery 155

21 Developing the compassionate self 159

22 Compassionate chair work 167

23 Focusing of the compassionate self 171

24 Compassion flowing out 177

25 Compassion flowing into oneself: Using memory 181

26 Compassion flowing into oneself: Compassionate images 185

27 Compassion letter writing 195

28 Compassion and well-being enhancing 197

29 Fear of compassion 199

30 Last thoughts 209

References 213

Index 231
Preface and acknowledgements
I would like to thank Windy Dryden for putting this excellent series together, inviting me to contribute, and
his patience with my various efforts to do so. I found it daunting because to date there is no major text on
Compassion Focused Therapy (CFT), so there was a need for some background and evidence for the value of
a compassion approach. In consequence this volume is a smidgen longer and more referenced than others in
the series—so many thanks to Windy, Joanne Forshaw and Jane Harris at Routledge for all their support.

I have tried to indicate the distinctive features while at the same time recognizing the huge debt and
borrowing from other approaches. Many thanks to all who have supported CFT especially my current
research team: Corinne Gale, Kirsten McEwan and Jean Gilbert; the board members of the Compassionate
Mind Foundation: Chris Gillespie, Chris Irons, Ken Goss, Mary Welford, Ian Lowens, Deborah Lee, Thomas
Schroder and Jean Gilbert; clinical colleagues who have also been working on this approach, Michelle Cree,
Sharon Pallant and Andrew Rayner, who have all shared their knowledge, insight and learning, greatly
enriching our understanding of CFT. Gratitude goes to Giovanni Liotti who guided me in the importance of
mentalizing and how to link that with social mentalities; to Andrew Gumley for his support, interest and
leadership in the field of psychosis and to Sophie Mayhew and Christine Braehler for their inspiring CFT
work with people with psychosis. Compassionate focused therapy is supported by the Compassionate Mind
Foundation which offers links to other compassion focused websites and updates, downloads and training in
CFT (see www.compassionatemind.co.uk). Thanks also to all those who contribute to the compassionate
mind discussion list (also see www.compassionatemind.co.uk). My gratitude goes to Diane Woollands for
her wonderful support in running the Compassionate Mind Foundation and Kelly Sims for her enthusiasm,
secretarial work and reference checking—no small feat.

This book is dedicated with thanks to all the clients who over many years have shared their tragedies and
triumphs and guided the development of CFT with their honesty of what helps and what doesn’t. To all I owe
a great debt.
Part 1
THEORY: UNDERSTANDING THE MODEL
1
Some basics
All psychotherapies believe that therapy should be conducted in a compassionate way that is respectful,
supportive and generally kind to people (Gilbert, 2007a; Glasser, 2005). Rogers (1957) articulated core
aspects of the therapeutic relationship involving positive regard, genuineness and empathy—which can be
seen as “compassionate”. More recently, helping people develop self-compassion has received research
attention (Gilbert & Procter, 2006; Leary, Tate, Adams, Allen, & Hancock, 2007; Neff, 2003a, 2003b) and
become a focus for self-help (Germer, 2009; Gilbert, 2009a, 2009b; Rubin, 1975/1998; Salzberg, 1995).
Developing compassion for self and others, as a way to enhance well-being, has also been central to
Buddhist practice for the enhancement of well-being for thousands of years (Dalai Lama, 1995; Leighton,
2003; Vessantara, 1993).

After exploring the background principles for developing Compassion Focused Therapy (CFT), Point 16
outlines the detailed aspects of compassion in the CFT approach. We can make a preliminary note, however,
that different models of compassion are emerging based on different theories, traditions and research (Fehr,
Sprecher, & Underwood, 2009). The word “compassion” comes from the Latin word compati, which means
“to suffer with”. Probably the best-known definition is that of the Dalai Lama who defined compassion as “a
sensitivity to the suffering of self and others, with a deep commitment to try to relieve it”, i.e., sensitive
attention-awareness plus motivation. In the Buddhist model true compassion arises for insight into the
illusory nature of a separate self and the grasping to maintain its boundaries—from what is called an
enlightened or awake mind. Kristin Neff (2003a, 2003b; see www.self-compassion.org), a pioneer in the
research on self-compassion, derived her model and self-report measures from Theravada Buddhism. Her
approach to self-compassion involves three main components:

1 being mindful and open to one’s own suffering;

2 being kind, and non self-condemning; and

3 an awareness of sharing experiences of suffering with others rather than feeling ashamed and alone—an
openness to our common humanity.

In contrast, CFT was developed with and for people who have chronic and complex mental-health problems
linked to shame and self-criticism, and who often come from difficult (e.g., neglectful or abusive)
backgrounds. The CFT approach to compassion borrows from many Buddhist teachings (especially the roles
of sensitivity to and motivation to relieve suffering) but its roots are derived from an evolutionary,
neuroscience and social psychology approach, linked to the psychology and neurophysiology of caring—
both giving and receiving (Gilbert, 1989, 2000a, 2005a, 2009a). Feeling cared for, accepted and having a
sense of belonging and affiliation with others is fundamental to our physiological maturation and well-being
(Cozolino, 2007; Siegel, 2001, 2007). These are linked to particular types of positive affect that are
associated with well-being (Depue & Morrone-Strupinsky, 2005; Mikulincer & Shaver, 2007; Panksepp,
1998), and a neuro-hormonal profile of increased endorphins and oxytocin (Carter, 1998; Panksepp, 1998).
These calm, peaceful types of positive feelings can be distinguished from those psychomotor activating
emotions associated with achievement, excitement and resource seeking (Depue & Morrone-Strupinsky,
2005). Feeling a positive sense of well-being, contentment and safeness, in contrast to feeling excited or
achievement focused, can now be distinguished on self-report (Gilbert et al., 2008). In that study, we found
that emotions of contentment and safeness were more strongly associated with lower depression, anxiety and
stress, than were positive emotions of excitement or feeling energized.

So, if there are different types of positive emotions—and there are different brain systems underpinning these
positive emotions—then it makes sense that psychotherapists could focus on how to stimulate capacities for
the positive emotions associated with calming and well-being. As we will see, this involves helping clients
(become motivated to) develop compassion for themselves, compassion for others and the ability to be
sensitive to the compassion from others. There are compassionate (and non-compassionate) ways to engage
with painful experiences, frightening feelings or traumatic memories. CFT is not about avoidance of the
painful, or trying to “soothe it away”, but rather is a way of engaging with the painful. In Point 29 we’ll note
that many clients are fearful of compassionate feelings from others, and for the self, and it is working with
that fear that can constitute the major focus of the work.

A second aspect of the CFT evolutionary approach suggests that self-evaluative systems operate through the
same processing systems that we use when evaluating social and interpersonal processes (Gilbert, 1989,
2000a). So, for example, as behaviourists have long noted, whether we see something sexual or fantasise
about something sexual, the sexual arousal system is the same—there aren’t different systems for internal
and external stimuli. Similarly, self-criticism and self-compassion can operate through similar brain
processes that are stimulated when other people are critical of or compassionate to us. Increasing evidence
for this view has come from the study of empathy and mirror neurons (Decety & Jackson, 2004) and our
own recent fMRI study on self-criticism and self-compassion (Longe et al., 2010).

Interventions
CFT is a multimodal therapy that builds on a range of cognitive-behavioural (CBT) and other therapies and
interventions. Hence, it focuses on attention, reasoning and rumination, behaviour, emotions, motives and
imagery. It utilizes: the therapeutic relationship (see below); Socratic dialogues, guided discovery, psycho-
education (of the CFT model); structured formulations; thought, emotion, behaviour and “body” monitoring;
inference chaining; functional analysis; behavioural experiments; exposure, graded tasks; compassion
focused imagery; chair work; enactment of different selves; mindfulness; learning emotional tolerance,
learning to understand and cope with emotional complexities and conflicts, making commitments for effort
and practice, illuminating safety strategies; mentalizing; expressive (letter) writing, forgiveness,
distinguishing shame-criticizing from compassionate self-correction and out-of-session work and guided
practice—to name a few!

Feeling the change

CFT adds distinctive features in its compassion focus and use of compassion imagery to traditional CBT-type
approaches. As with many of the recent developments in therapy, special attention is given to mindfulness in
both client and therapist (Siegel, 2010). In the formulation CFT is focused on the affect-regulation model
outlined in Point 6, and interventions are used to develop specific patterns of affect regulation, brain states
and self-experiences that underpin change processes. This is particularly important when it comes to working
with self-criticism and shame in people from harsh backgrounds. Such individuals may not have experienced
much in the way of caring or affiliative behaviour from others and therefore the (soothing) emotion-
regulation system is less accessible to them. These are individuals who are likely to say, “I understand the
logic of [say] CBT, but I can’t feel any different”. To feel different requires the ability to access affect
systems (a specific neurophysiology) that give rise to our feelings of reassurance and safeness. This is a well-
known issue in CBT (Leahy, 2001; Stott, 2007; Wills, 2009, p. 57).

Over twenty years ago I explored why “alternative thoughts” were not “experienced” as helpful. This
revealed that the emotional tone, and the way that such clients “heard” alternative thoughts in their head, was
often analytical, cold, detached or even aggressive. Alternative thoughts to feeling a failure, like: “Come on,
the evidence does not support this negative view; remember how much you achieved last week!” will have a
very different impact if said to oneself (experienced) aggressively and with irritation than if said slowly and
with kindness and warmth. It was the same with exposures or home-works— the way they are done
(bullying and forcing oneself verses encouraging and being kind to oneself) can be as important as what is
done. So, it seemed clear that we needed to focus far more on the feelings of alternatives not just the content
— indeed, an over focus on content often was not helpful. So, my first steps into CFT simply tried to
encourage clients to imagine a warm, kind voice offering them the alternatives; or working with them in their
behavioural tasks. By the time of the second edition of Counselling for Depression (Gilbert, 2000b) a whole
focus had become concentrated on “developing inner warmth” (see also Gilbert, 2000a). So, CFT progressed
from doing CBT and emotion work with a compassion (kindness) focus and, then, as the evidence for the
model developed and more specific exercises proved helpful, on to CFT.

The therapeutic relationship

The therapeutic relationship plays a key role in CFT (Gilbert, 2007c; Gilbert & Leahy, 2007), paying
particular attention to the micro-skills of therapeutic engagement (Ivey & Ivey, 2003), issues of
transference/countertransference (Miranda & Andersen, 2007), expression, amplification, inhibition and/or
fear of emotion (Elliott, Watson, Goldman, & Greenberg, 2003; Leahy, 2001), shame (Gilbert, 2007c),
validation (Leahy, 2005), and mindfulness of the therapist (Siegel, 2010). When training people from other
approaches, particularly CBT, we find that we have to slow them down; to allow spaces, and silences for
reflection, and experiencing within the therapy rather than a series of
Socratic questions or “target setting”. We teach how to use one’s voice speed and tone, nonverbal
communication, the pacing of the therapy, being mindful (Katzow & Safran, 2007; Siegel, 2010) and the
reflective process in the service of creating “safeness” to explore, discover, experiment and develop. Key is
to provide emotional contexts where the client can experience (and internalize) therapists as
“compassionately alongside them”—no easy task because as we will discuss below (see Point 10) shame
often involves clients having emotional experiences (transference) of being misunderstood, getting things
wrong, trying to work out what the other person wants them to do and intense aloneness. The emotional tone
in the therapy is created partly by the whole manner and pacing of the therapist and is important in this
process of experiencing “togetherness”. CF therapists are sensitive to how clients can actually find it hard to
experience “togetherness” or “being cared about”, and wrap themselves in safety strategies of sealing the self
off from “the feelings of togetherness and connectedness” (see Point 29; Gilbert, 1997, 2007a, especially
Chapters 5 and 6, 2007c).

CBT focuses on collaboration, where the therapist and client focus on the problem together—as a team. CFT
also focuses on (mind) “sharing”. The evolution of sharing (and motives to share), e.g., not only objects but
also our thoughts, ideas and feelings, is one of humans’ most important adaptations and we excel at wanting
to share. As an especially social species, humans have an innate desire to share—not only material things but
also their knowledge, values and the content of their minds—to be known, understood and validated. Thus,
issues of motivation to share versus fear of sharing (shame), empathy and theory of mind are important
evolved motives and competencies. It is the felt barriers to this “flow of minds” that can be problematic for
some people and the way that the therapist “unblocks” this flow that can be therapeutic.

Dialectical Behaviour Therapy (DBT; Linehan, 1993) addresses the key issue of therapy-interfering
behaviours. CFT, like any other therapy, needs to be able to set clear boundaries, and use authority as a
containing process. Some clients can be “emotional bullies”, threatening the therapist (e.g., with litigation or
suicide) and are demanding. Frightened therapists may submit or back off. The client, at some level, is
frightened of their own capacity to force others away from them. For other clients, during painful moments,
therapists might try to rescue rather than be silent. So, clarification of the therapeutic relationship is very
important. This is why DBT wisely recommends a support group for therapists working with these kinds of
clients.

Research has shown that compassion can become a genuine part of self-identity but it can also be linked to
self-image goals where people are compassionate in order to be liked (Crocker & Canevello, 2008).
Compassion focused self-image goals are problematic in many ways. Researchers are also beginning to
explore attachment style and therapeutic relationships with evidence that securely attached therapists
develop therapeutic alliances easier and with less problems than therapists with an insecure attachment style
(Black, Hardy, Turpin, & Parry, 2005; see also Liotti, 2007). Leahy (2007) has also outlined how the
personality and schema organization of the therapist can play a huge role in the therapeutic relationship—for
example, autocratic therapists with dependent patients, or dependent therapists with autocratic patients. So,
compassion is not about submissive “niceness”—it can be tough, setting boundaries, being honest and not
giving clients what they want but what they need. An alcoholic wants another drink—that is not what they
need; many people want to avoid pain and may try to do so in a variety of ways—but (kind) clarity, exposure
and acceptance may be what actually facilitates change and growth (Siegel, 2010).

Evidence for the benefits of compassion

Although CFT is rooted in an evolutionary, neuro- and psychological science model, it is important to
recognize its heavy borrowing from Buddhist influences. For over 2500 years Buddhism has focused on
compassion and mindfulness as central to enlightenment and “healing our mind”. While Theravada
Buddhism focuses on mindfulness and loving-(friendly)-kindness, Mahayana practices are specifically
compassion focused (Leighton, 2003; Vessantara, 1993). At the end of his life the Buddha said that his main
teachings were mindfulness and compassion—to do no harm to self or others. The Buddha outlined an eight-
fold path for practice and training one’s mind to avoid harming and promote compassion. This includes:
compassionate meditations and imagery, compassionate behaviour, compassionate thinking, compassionate
attention, compassionate feeling, compassion speech and compassionate livelihood. It is these multimodal
components that lead to a compassionate mind. We now know that the practice of various aspects of
compassion increases well-being and affects brain functioning, especially in areas of emotional regulation
(Begley, 2007; Davidson et al., 2003).

The last 10 years have seen a major upsurge in exploring the benefits of cultivating compassion (Fehr et al.,
2009) In an early study Rein Atkinson and McCraty (1995) found that directing people in compassion
2009). In an early study Rein, Atkinson and McCraty (1995) found that directing people in compassion
imagery had positive effects on an indictor of immune functioning (S-IgA) while anger imagery had negative
effects. Practices of imagining compassion for others, produce changes in the frontal cortex, immune system
and wellbeing (Lutz, Brefczynski-Lewis, Johnstone, & Davidson, 2008). Hutcherson, Seppala and Gross
(2008) found that a brief loving-kindness meditation increased feelings of social connectedness and
affiliation towards strangers. Fredrickson, Cohn, Coffey, Pek and Finkel (2008) allocated 67 Compuware
employees to a loving-kindness meditation group and 72 to waiting-list control. They found that six 60-
minute weekly group sessions with home practice based on a CD of loving kindness meditations
(compassion directed to self, then others, then strangers) increased positive emotions, mindfulness, feelings
of purpose in life and social support, and decreased illness symptoms. Pace, Negi and Adame (2008) found
that compassion meditation (for six weeks) improved immune function and neuroendocrine and behavioural
responses to stress. Rockliff, Gilbert, McEwan, Lightman and Glover (2008) found that compassionate
imagery increased heart rate variability and reduced cortisol in low self-critics, but not in high self-critics. In
our recent fMRI study we found that self-criticism and self-reassurance to imagined threatening events (e.g.,
a job rejection) stimulated different brain areas, with self-compassion but not self-criticism stimulating the
insula—a brain area associated with empathy (Longe et al., 2010). Viewing sad faces, neutrally or with a
compassionate attitude, influences neurophysiological responses to faces (Ji-Woong et al., 2009).

In a small uncontrolled study of people with chronic mentalhealth problems, compassion training
significantly reduced shame, self-criticism, depression and anxiety (Gilbert & Procter, 2006). Compassion
training has also been found to be helpful for psychotic voice hearers (Mayhew & Gilbert, 2008). In a study
of group-based CFT for 19 clients in a high-security psychiatric setting, Laithwaite et al. (2009) found “…a
large magnitude of change for levels of depression and self-esteem…. A moderate magnitude of change was
found for the social comparison scale and general psychopathology, with a small magnitude of change for
shame,…. These changes were maintained at 6-week follow-up” (p. 521).

In the field of relationships and well-being, there is now good evidence that caring for others, showing
appreciation and gratitude, having empathic and mentalizing skills, does much to build positive
relationships, which significantly influence well-being and mental and physical health (Cacioppo, Berston,
Sheridan, & McClintock, 2000; Cozolino, 2007, 2008). There is increasing evidence that the kind of “self”
we try to become will influence our well-being and social relationships, and compassionate rather than self-
focused self-identities are associated with the better outcomes (Crocker & Canevello, 2008). Taken together
there are good grounds for the further development of and research into CFT.

Neff (2003a, 2003b) has been a pioneer in studies of self-compassion (see pages 3–4). She has shown that
self-compassion can be distinguished from self-esteem and predicts some aspects of well-being better than
self-esteem (Neff & Vonk, 2009), and that self-compassion aids in coping with academic failure (Neff,
Hsieh, & Dejitterat, 2005; Neely, Schallert, Mohammed, Roberts, & Chen, 2009). Compassionate letter
writing to oneself, improves coping with life events and reduces depression (Leary et al., 2007). As noted,
however, Neff’s concepts of compassion are different from the evolutionary and attachment-rooted model
outlined here and, as yet, there is no agreed definition of compassion—indeed, the word compassion can
have slightly (but important) different meanings in different languages. So, here compassion will be defined
as a “mind set”, a basic mentality, and explored in detail in Point 16.
2
A personal journey
My interest in developing people’s capacities for compassion and self-compassion was fuelled by a number
of issues:

• First, was a long interest in evolutionary approaches to human behaviour, suffering and growth (Gilbert,
1984, 1989, 1995, 2001a, 2001b, 2005a, 2005b, 2007a, 2007b, 2009a). The idea that cognitive systems tap
underlying evolved motivation and emotional mechanisms has also been central to Beck’s cognitive
approach (Beck, 1987, 1996; Beck, Emery, & Greenberg, 1985), with a special edition dedicated to
exploring the evolutionary-cognitive interface (Gilbert, 2002, 2004).

• Second, evolutionary psychology has focused significantly on the issue of altruism and caring (Gilbert,
2005a) with increasing recognition of just how important these have been in our evolution (Bowlby, 1969;
Hrdy, 2009) and now are to our physical and psychological development (Cozolino, 2007) and well-being
(Cozolino, 2008; Gilbert, 2009a; Siegel, 2007).

• Third, people with chronic mental-health problems often come from backgrounds of high stress and/or low
altruism and caring (Bifulco & Moran, 1998), backgrounds that significantly affect physical and
psychological development (Cozolino, 2007; Gerhardt, 2004; Teicher, 2002).

• Fourth, partly as a consequence of these life experiences, people with chronic and complex problems can
be especially, deeply troubled by shame and self-criticism and/or self-hatred and find it enormously difficult
to be open to the kindness of others or to be kind to themselves (Gilbert, 1992, 2000a, 2007a, 2007c; Gilbert
& Procter, 2006).

• Fifth, as noted on page 6, when using CBT they would typically say, “I can see the logic of alternative
thoughts but I still feel X, or Y. I can understand why I wasn’t to blame for my abuse but I still feel I’m to
blame”, or, “I still feel there is something bad about me”.

• Sixth, there is increasing awareness that the way clients are able to think about and reflect on the contents
of their own minds (e.g., competencies to mentalize in contrast to being alexithymic) has major implications
for the process and focus of therapy (Bateman & Fonagy, 2006; Choi-Kain & Gunderson, 2008; Liotti &
Gilbert, in press; Liotti & Prunetti, 2010).

• Last, but not least, is a long personal interest in the philosophies and practices of Buddhism—although I do
not regard myself as a Buddhist as such. Compassion practices, such as becoming the compassionate self
(see Part 2), may create a sense of safeness that aides the development of mindfulness and mentalizing. In
Buddhist psychology compassion “transforms” the mind.

Logic and emotion

It has been known for a long time that logic and emotion can be in conflict. Indeed, since the 1980s research
has shown that we have quite different processing systems in our minds. One is linked to what is called
implicit (automatic) processing, which is non-conscious, fast, emotional, requires little effort, is subject to
classical conditioning and self-identify functions, and may generate feelings and fantasies even against
conscious desires. This is the system which gives that “felt sense of something”. This can be contrasted with
an explicit (controlled) processing system, which is slower, consciously focused, reflective, verbal and
effortful (Haidt, 2001; Hassin, Uleman, & Bargh, 2005). These findings have been usefully formulated for
clinical work (e.g., Power & Dalgleish, 1997) with more complex models being offered by Teasdale and
Barnard (1993). But the basic point is that there is no simple connection of cognition to emotion, and there
are different neurophysiological systems underpinning them (Panksepp, 1998). So, one of the problems
linking thinking and feeling (“I know it but I don’t feel it”) can be attributed to (different) implicit and
explicit systems coming up with different processing strategies and conclusions. Cognitive, and many other,
therapists and psychologists have not helped matters by using the concept of cognition and information
processing interchangeably as if they are the same thing. They are not. Your computer and DNA—indeed
every cell in your body—are information processing mechanisms but I don’t think that they have
“cognitions”. This failure to define what is and is not “a cognition” or “cognitive” in contrast to a motive or
an emotion has caused difficulties in this area of research.
Various solutions have been offered to work with the problems of feelings not following cognitions or logical
reasoning, such as: needing more time to practise; most change is slow and hard work; more exposure to
problematic emotions; identifying “roadblocks” and their functions (Leahy, 2001); a need for a particular
therapeutic relationship (Wallin, 2007); or developing mindfulness and acceptance (Hayes, Follette, &
Linehan, 2004; Liotti & Prunetti, 2010). CFT offers an additional position. CFT suggests that there can be a
fundamental problem in an implicit emotional system that evolved with mammalian and human caring
systems and which gives rise to feelings of reassurance, safeness and connectedness (see Point 6). The
inability to access that affect system is what underpins this problem. Indeed, as noted (page 6), some people
can cognitively (logically) generate “alternative thoughts” but hear them in their head as cold, detached or
aggressive. There is no warmth or encouragement in their alternative thoughts—the emotional tone is more
like cold instruction. I have found that the idea of feeling (inner) kindness and supportiveness as part of
generating alternative “thoughts” is an anathema to them. So, they just cannot “feel” their alternative
thoughts and images.

These clients also have a deep-seated sense of “being to blame for their problems”; it is “their fault”; there is
something “fundamentally flawed or damaged” about them—which blocks feelings of compassion and self-
acceptance. It becomes clear, then, that we need to “warm up” people’s ability to stand back from their
thoughts, feelings and problems and treat themselves with more compassion and kindness, as well as using
insight, logic, problem solving exposures and “mind training”.

Case example

Over twenty years ago I worked with Jane, who suffered from a particularly chronic bipolar depression with
borderline features and suicidality. She had been adopted early in childhood and had a long-lived feeling of
not fitting in anywhere. She became good at generating very reasonable alternative thoughts to the idea that
she was a failure and not really wanted—yet this did not shift her mood much. When I asked about the
emotional tone of her alternative thoughts, whether she experienced them as reassuring, helpful and kind, she
was puzzled, “Of course not,” she said, “I just need to be logical. Why would I want to be kind to myself?
Being kind to myself seems like a weakness and a self-indulgence! I just need to get a grip on these
thoughts.” Now, she was married with children and had a supportive family, so I pondered on this, “But isn’t
the whole point of exploring these alternative thoughts and ideas—to help you feel loved and wanted—to
feel that support, belonging, kindness and acceptance that you seek?” I put that to her. From distant recall she
said, “Yes but I don’t want to be compassionate to myself—there is just too much I don’t like about me!” It
took some time before she recognized that she was: (1) actually closed (and dismissive) to the kindness of
others because of her own self-dislike; (2) which undermined her abilities to be assertive; which then (3) led
on to submissive resentment and feelings of powerlessness; and (4) her secret inner resenting and hating led
to further feelings of self-dislike.

So, with Jane we pondered how she might generate more emotional warmth, “acceptance” and “kindness in
her alternative thoughts”; the idea being that when she thought of and (maybe) wrote down alternative
thoughts (to her depressing ones), she might focus on the feelings of warmth and reassurance in them. This
was to reveal another key aspect of working with compassion—at first Jane was contemptuous and
frightened of developing warmth for herself. She was also fearful of allowing herself to really pay attention,
and feel care from others; fearful of intense emotional closeness; that others getting close would sooner or
later end in shame and rejection. Typical are beliefs of “if you really got close and really knew what goes on
inside me you would not like me; you’d find out something that would turn you against me (e.g., my hateful
thoughts)” or fear of dependency, e.g., “if I get close to someone then I will need them, become dependent,
needy, weak and vulnerable”. So, exposure to a type of positive (kind and affiliative) affect, and working
with the fear of feeling this positive affect, was to prove helpful for her. Jane was one of the first to use
“imagining an ideal compassionate image” (see Point 26), which she saw as a Buddha dressed as an earth
Goddess! It was very tough going but she has not had a major depression relapse in over fifteen years now.
Working with Jane opened up the idea that some people are very fearful of positive emotions because of their
close association with negative emotions and outcomes—and so desensitizing to, and activating of, positive
and affiliative emotions requires as much work in terms of desensitization as threat-based emotions (see
Point 29).

So, at its simplest, it seems to me that people with high shame and self-criticism have difficulty in accessing
feelings of (self) warmth, compassion and reassurance. This key affect-regulation system seems “off-line”.
As an analogy, you can have all the sexual cues, thoughts and fantasies you like but if the system in the
pituitary that flushes the body with hormones is not working, these cues and fantasies will not have any
physiological impact—so will not be “felt”. Thus, CFT was originally developed for, and with, people who
suffer from high levels of shame and self-criticism, and who find it difficult to self-sooth or generate feelings
of inner warmth and self-reassurance (Gilbert, 2000a, 2000b, 2007a; Gilbert & Irons, 2005).

CFT is process rather than disorder focused because shame and self-criticism are transdiagnostic processes
that have been linked to a range of psychological disorders (Gilbert & Irons, 2005; Zuroff, Santor, &
Mongrain, 2005). CFT is a de-pathologizing approach that focuses more on people’s (phenotypic variations
in) adaptive responses to difficult environments. For example, we all have an attachment system but whether
we develop the phenotypes of trusting, openness and affiliativeness or mistrust, avoidance and exploitation
of others, depends upon whether we have experiences of love and care in our early life, or of neglect,
hostility and abuse, and our current social contexts. The phenotype for affiliative behaviour takes on different
patterns according to the context in which it develops.
3
The evolved mind and Compassion Focused Therapy
The Buddha and early Greek philosophers understood well that our minds are chaotic, subject to conflicts
and being taken over by powerful emotions, which can throw us into problems of anxiety, depression,
paranoia and violence. What they could not know is why. The beginnings of an answer came with the
publication in 1859 of Darwin’s Origin of Species, which revealed that our minds and brains are the result of
natural selection. Slow changes occur as species adapt to changing environments; environments are therefore
challenges that favour some individual variations within a population over others. Importantly, evolution
cannot go back to the drawing board but rather builds on previous designs. This is why all animals have the
same basic blueprint of four limbs, a cardiovascular system, a digestive system, sense organs, etc. Brains,
too, have basic functions, which are shared across species. This has huge implications for understanding how
our minds are designed and came to be the way they are (Buss, 2003, 2009; Gilbert, 1989, 2002, 2009a;
Panksepp, 1998).

Darwin’s profound insights had a major impact on psychology and psychotherapy (Ellenberger, 1970).
Sigmund Freud (1856–1939), for example, recognized that the mind contains many basic instincts and
motives (e.g., for sex, aggression and power), which need to be regulated (lest we all just act out our desires)
in a whole host of ways. So, we have various ways of keeping our lusts, passions and destructive urges under
control—such as with defence mechanisms like denial, projection, dissociation and sublimation. Freud made
a distinction between primary (id thinking generated by innate desires) and secondary (ego-based and
reality-based) thinking. In his model the mind is inherently in conflict between desires and control. For
Freud these conflicts could be overwhelming to ego consciousness and so become unconscious and a source
of mental disorder. The role of the therapist was to make conscious these conflicts and help the person work
through them.

Today there is much evidence that the brain does, indeed, have different systems that are linked to our
passions and motives (e.g., implicit vs. explicit; Quirin, Kazen, & Kuhl, 2009) located in old brain systems
such as the limbic system (MacLean, 1985), and to the regulation of motives and emotions—primarily
through the frontal cortex (Panksepp, 1998). Damage these areas of the brain and impulsiveness and
aggressiveness are often the major symptoms. Numerous studies using subliminal processing have shown
that unconscious processing can have a major impact on emotions and behaviours (e.g., Baldwin, 2005)—in
fact, consciousness is quite a late stage in information processing (Hassin et al., 2005). We also know that the
mind is riddled with conflicting motives and emotions (see Point 4). Today there is also scientific study on
the nature of defence mechanisms such as repression, projection and dissociation and how these affect
psychological functioning, self-constructions, social relating and therapy (Miranda & Andersen, 2007).

Archetypes, motives and meanings

Few now accept the tabula rasa view of human psychology. Rather, it is recognized that the human infant
comes into the world prepared to become a viable representative of its species (Knox, 2003; Schore, 1994).
If all goes well the child will form attachments to his/her care givers, acquire language, develop cognitive
competencies, form peer and sexual relationships and so forth. In other words, there are innate aspects to our
motives and meaning making. This is not a new idea as its origins can be traced back to Plato and Kant. A
person closely associated with attempts to illuminate the innate nature of the human capacity to create
different types of meaning in psychotherapy was Jung (1875–1961).

Jung called our innate guiding systems (e.g., to seek and form early attachments to parents/carers, to belong
to groups, to seek status, to discern and seek out sexual partners) archetypes. Archetypes influence the
unfolding of development (e.g., to seek care, to become a member of a group, to find a sexual partner and
become a parent, and to come to terms with death; Stevens, 1999). So, Jung postulated that humans, as an
evolved species, inherit specific predispositions for thought, feeling and action. These predispositions exist
as foci within the collective unconscious and serve to guide behaviour, thoughts and emotions.

Jung noted that we can see these themes of: parent-child caring, family and group loyalties and betrayal,
seeking romance and love, seeking status and social position in heroic endeavour, self-sacrifice and so on—
in all of the cultures, literatures and stories stretching back thousands of years. These are themes that will
play out time and again in mental-health difficulties, because they are part of us—they are archetypal.

J l t d th t th h t t f ti d bl d ith th h t i
Jung also suggested that the way an archetype matures, functions and blends with other archetypes is
affected by both our personality (genes) and our experience. For example, although we have an archetype
that inspires and guides us towards love and comfort in the arms of our mothers when we are infants, if this
relationship does not work well we can have a stunted mother archetype. Stevens (1999) refers to this as
thwarted archetypal intent. In this case, as adults, we might spend a lot of our lives searching for a mother or
father figure—trying to find someone who will love and protect us like a parent, or we can close down our
need for care and love completely and shun close caring. Researchers studying these early relationships and
what is called “attachment behaviour” have found that children (and adults) can indeed behave in such ways:
While some are open to love and care, others are anxious about losing love and need much and constant
reassurance, while yet others avoid close relationships because they are frightened, or are contemptuous and
dismissive of closeness (Mikulincer & Shaver, 2007).

Jung also suggested that because our inner archetypes are designed to do different things and pursue different
goals they can be in conflict with each other and this often causes mental-health difficulties. For Jung it is the
way that these archetypal processes mature, develop and become integrated, are thwarted, or are in conflict
within the self, that is the source of mental-health problems.

Social mentalities
Gilbert (1989, 1995, 2005b, 2009a) combined archetype theory with modern evolutionary, social and
developmental psychology and suggested that humans have a number of “social mentalities” that enabled
them to seek out and form certain types of relationship (e.g., sexual, tribal, dominant-subordinate, caring of-
cared for). The basic idea is that to pursue “species general, evolved biosocial goals and motives”—such as
seeking out sexual partners and forming sexual bonds, looking after one’s offspring, forming friendships and
alliances, developing a sense of group belonging, operating as a group member, and competing for status—
brain patterns are organized in different and particular ways. A social mentality can be defined as “the
organization of various psychological competencies and modules (e.g., for attention, ways of thinking, and
action tendencies) guided by motives to secure specific types of social relationship”. For example, when we
are in care-giving mentalities we focus our attention on the distress or needs of the other, feel concern for
them, work out how to provide them with what they need, engage in behaviours to do so, and feel rewarded
by their recovery or prosperity. In humans this may even become linked to self-identity, e.g., “I would like to
be a caring person”. In a care-seeking mentality we are seeking inputs from others that will relieve distress or
help us grow and develop. We turn our attention to those who are potentially helpful, signal our needs or
distress and orientate our behaviours to approach others who seem to be able to offer what we think we need.
We feel good if those inputs are achieved, but may feel angry, anxious or depressed if we can’t find the
caring inputs sought. We might then feel that others are deficient in what we need or are withholding.

In contrast, if we are competing with others we socially compare ourselves with them on relative strengths
and weaknesses. We make decisions to try harder or give up. We might increase aggressive feelings or
actions towards them and turn off concerns and feelings for any distress they might have. In this mentality
our thoughts about ourselves are in terms of inferior-superior or winner-loser. If we win we might feel a buzz
of good feelings, but be mildly depressed if we lose or feel inferior or defeated (Gilbert, 1984, 1992, 2007a).
Linked to a self-identity, a social rank mentality can become a need to achieve (more than others), with
status recognition, or a depressive sense of being a subordinate and lacking in certain qualities. There are,
however, different types of achievement motivation (see Point 14). Seeing others prosper might actually
make us feel envious and annoyed or bad, while seeing them fail or drop behind makes us feel good—which
is, of course, quite different to being in a caring mentality!

In contrast again, when we are in co-operative alliance-building mentality our attention is focused on seeking
and linking with people who are like us, who will co-operate and support us and/or pursue joint goals
(playing in an orchestra, working on a team). This is linked to the evolution of our intense desires to share.
We feel good when getting along with others but bad if rejected, marginalized or we feel people are cheating
in some way. “Getting along” versus “getting ahead” has long been recognized to involve very different
psychologies (Wolfe, Lennox, & Cutler, 1986). Lanzetta and Englis (1989) showed that priming people for
co-operative or competitive relationships produced major differences in skin conductance, heart rate and
EMG; with co-operation promoting empathy but competitiveness “counter-empathy”. So, different social
mentalities can organize a whole suite of psychological and physiological processes in different ways,
turning some aspects (e.g., care, sympathy or aggression) on but others off.

A simple comparison diagram for competing versus caring mentalities is offered in Figure 1.
So, according to the CFT approach our minds are organized and motivated for different goal pursuits and to
create different mentalities according to the biosocial goal(s) being pursued. Clearly, mentalities overlap,
some are more conscious than others, some are compensations (e.g., competing for status could be because
we want affection; see Point 14), and people switch between them. Indeed, the ability to switch between
them is a mark of health (Gilbert, 1989), e.g., such as the man who can compete in the job market but be a
loving father at home, rather than also trying to compete with his children for his wife’s affection and time.
Individuals who get trapped in a particular mentality, for example, who are competitive or submissive all the
time and really struggle with being co-operative or care giving or care receiving, can be disadvantaged in
many ways. Paranoid patients, for example, find care receiving extremely threatening because of basic
mistrust. Psychopaths find care giving and having empathic concern very difficult, but might mentalize very
well in a competitive situation. So, the argument is that we have evolved brains that pursue certain social
strategies, roles and relationships, and in order to pursue them different aspects of our minds are turned on
and off. If we are in a (say) tribal mentality and see the other group as enemies, empathic concern and care
giving (for the suffering we cause them) is firmly turned off, enabling us to behave aggressively without
concern or guilt for the harm we do. There is much in psychotherapy of complex cases where we’re trying to
tone down some mentalities and activate others. Beck, Freeman and Davis (2003) took a somewhat similar
evolutionary view in their approach to personality disorders.

Figure 1 A comparison of competitive and caring mentalities

The way that social mentalities develop, mature, blend and are activated is linked to genes, background and
current social demands. This is important because CFT takes an interactional and compensatory view of the
way the mind is organized. For example, children who are abused or neglected learn that care eliciting,
turning to others for help and being soothed by them is unlikely to happen, is unhelpful or even
threatening/dangerous. Rather, the need to pay attention to the power and potential for being harmed/shamed
by others is needed. This will shift their development into a threat-focused, social rank mentality
(competitive system), which will orientate them to be very attentive to cues of aggressiveness/rejection.
Liotti (2000, 2002, 2007) outlined how children can become disorganized in their attachment behaviour
especially when the care giver becomes the source of both safety and threat. In such cases children can
become disorganized in their social mentality coherence, switching between submissiveness, aggressiveness
withdrawal and closeness seeking in ways that are difficult for them or others around them to understand.

CFT also takes an interactional view to the extent that work on one social mentality, such as care eliciting or
care providing (through compassion), can have profound effects on the organization of other social
mentalities (Gilbert, 1989). Again, this is not a new view. Buddhism has long argued that compassion
transforms and reorganizes the mind; and Jung argued that the process of individuation was a process of
organizing and reorganizing our archetypal potentials.

The bottom line is that we need to understand that the brain is an evolved organ that is designed to function
in certain ways and change its patterns in different contexts and goal pursuits. It seeks out certain inputs
(e.g., affiliative relations with other minds), responds to those inputs, and shows defensive, developmental
deviations if those inputs are not forthcoming. Some psychotherapies and most psychiatric classifications do
not address this and are content to rely on how things look from the outside. CFT sits within the tradition of
understanding our
minds in terms of their evolved design and particularly our human biosocial goals and needs, e.g., for
affection, care, protection, belonging; and human competencies such as mentalizing, theory of mind,
empathy capabilities for fantasy and imagination as we now explore
empathy, capabilities for fantasy and imagination, as we now explore.
4
Multi-mind
We know that different psychologies (motives, emotions and cognitive competencies) have been laid down at
different times in evolution. For example, capacities for sex, fighting, hunting and gaining and defending a
territory can be traced back to the reptiles over 500 million years ago and before them. It was not until the
evolution of the mammals (about 120 million years ago) that psychologies (supported by emotions and
motivational systems) for infant-caring, alliance formation, play, and status hierarchies came into the world.
It was not until about 2 million years ago that the competencies for complex thinking, reflection, theory of
mind and having a sense of self and self-identity, began to emerge. So, brains have evolved in a series of
stages, making our minds full of a variety of different motives and emotions originating at different times,
many of which can conflict.

Old and new brain

One way to think about this, and to talk to clients about it, is that we have an old brain that has various
emotions and motivations that we share with many other animals, but that we also have the abilities of
thinking, reflecting, observing, and forming a self-identity (Gilbert, 2009a). Problems can arise in the way
our old brain and new brain interact. For example, bodily sensations when linked to the new brain abilities
that can think, reason and give explanations may conclude that “my increased heart rate means I am going to
have a heart attack and die”, which fosters panic attacks. Our reflection that “this mistake means I am a
failure, unlovable with no future”, can accentuate depression. Animals don’t get stressed out by worrying
about paying the mortgage, what the future might hold, what will happen to the children if they don’t study
hard, whether this lump or pain means an oncoming cancer—these are all created because of our ability to
think in certain “meta” ways (Wells, 2000).

A variety of mental states

The fact that we have many different motives, archetypal potentials and the like, that have evolved over
many millions of years, and these different potentials can give rise to serious problems, is well known now,
even noted in most introductory psychology undergraduate textbooks. For example, Coon (1992) opened his
introductory undergraduate text on psychology with this graphic depiction:

You are a universe, a collection of worlds within worlds. Your brain is possibly the most complicated and
amazing device in existence. Through its action you are capable of music, art, science, and war. Your
potential for love and compassion coexists with your potential for aggression, hatred… Murder?

(p. 1)

What Coon and other researchers suggest is that we are not unified selves, despite our experience of being
so. Rather, we are made up of many different possibilities for the creation of meaning and generating brain
patterns and states of mind. As Ornstein (1986) put it over twenty years ago:

The long progression in our self-understanding has been from a simple and usually “intellectual” view to the
view that the mind is a mixed structure, for it contains a complex set of “talents”, “modules” and “policies”
within…. All these general components of the mind can act independently of each other; they may well have
different priorities.

The discovery of increased complexity and differentiation has occurred in many different areas of
research,…in the study of brain functions and localization; in the conceptions of the nature of intelligence; in
personality testing; and in theories of the general characteristics of the mind.

(p. 9)

In fact, the idea of there being complex subsystems and programmes in our mind, and how they interact, has
been a focus of theorizing in CBT. For example, Beck (1996) posited the existence of a variety of different
“modes”, where each mode represented integrated motivational, emotional and cognitive systems—a view
that has some similarity to archetype and social mentality ideas. In the same tradition Teasdale and Barnard
(1993) posited a range of cognitive, emotional and motivational processing subsystems that interact.
In some ways both Coon and Ornstein are offering a modern exposition of the archetypal nature of the
human mind, and they echo another Jungian idea—that although we often think of ourselves as somehow
whole and integrated individuals this is an illusion. In fact, Jung suggested that integration and wholeness are
psychological feats—maturational accomplishments. We are made up of many different talents, abilities,
social motives, emotions, and so on, and coping with their various pushes and pulls is no easy matter.

This mixed bag of motives and meaning-creating modules (archetypes) can give rise to the experience of not
one self but a variety of selves (e.g., Rowan, 1990). These possible selves or subpersonalities can feel
different things and play different parts when we are in different states of mind. In therapy we can even learn
to name these different selves and speak with them. We can recognize the bullying self, the perfectionist self,
the vengeful-sadist self, the sexual self, the forgiving self, and so on. Another way to think of this is as the
potential to enact different social roles requiring different social mentalities (and brain patterns; Gilbert,
1989, 1992; Gilbert & Irons, 2005; Gilbert & McGuire, 1998).

Mentalizing and the sense of self

CBT was originally developed for those who are aware of their thoughts and emotions, and can articulate
them fairly easily— as captured by various assessment scales (e.g., Safran & Segal, 1990). Therapists then
use this skill to direct clients’ attention to their styles of thinking or core beliefs and schema. However, we
now know that the competencies that underpin our capacities to become aware of our motives, emotions and
thoughts, and our ability to articulate, think about and reflect on them, are very complex and follow a
developmental process. Evolutionists have long pointed out that some motivational processes may be almost
impossible to bring to consciousness. These are complex facets of our multi-minds.

Over the last 10 years there has been accelerated interest in the way that humans link self-understanding and
self-identity to attending and reasoning about internal emotions and thoughts. We know, for example, that the
brain is capable of generating a huge array of competing and complex emotions and (and at times bizarre and
unpleasant) thoughts/fantasies, which have to be organized in a coherent fashion for individuals to have a
coherent sense of self (Gilbert, 2005a; McGregor & Marigold, 2003). People can be deeply troubled and
overwhelmed by some of the contents of their mind. New information (feelings, fantasies, thoughts) that
threatens the sense of self and self-identity, even if it promises to produce positive change, can be actively
resisted. Swann, Rentfrow and Guinn (2003) suggested that people seek coherence, familiarity and
predictability in their self-identities, not just self-enhancement. Thus, submissive or aggressive self-identities
can resist change (see Leary & Tangney, 2003, for a detailed discussion of these issues, and Point 29 in this
volume).

One way coherence is achieved is through the ability to reflect on and understand the contents of one’s own
mind and those of others—and the ability to use this information to navigate in and out of different types of
social role and social relationship. Mentalizing clearly aids the use and organization of our social mentalities
(Allen, Fonagy, & Bateman, 2008). This takes us to the heart of how our newly evolved brains, which give
rise to types of reasoning and competencies to generate a self-identity, interact with old brain emotions and
motives—those basic strategies for survival and reproduction that have evolved over a very long time
(Gilbert, 1989, 2009a).

Unfortunately (as with so much in psychotherapy research and therapy) one of the problems with this area of
study (the competencies underpinning capacities for reflection on mental states) is that it is riddled with
different approaches, concepts and theories, many of which significantly overlap and research is yet to
clarify. For example, being able to pay attention to, think about, reflect on and contain different feelings,
along with how we attribute different causes and meanings to our and other people’s mental states, has been
linked to: emotional schemas, alexithymia, mentalization, theory of mind, empathy and sympathy, emotional
intelligence, experiential avoidance, mindfulness, the use of projective defences and Asperger syndrome
(e.g., Choi-Kain & Gunderson, 2008), to name just a few!

Taking just the one example of alexithymia—this describes difficulties in recognizing and identifying
feelings at a subjective level, difficulties in describing and reflecting on feelings, especially difficulties in
describing ambivalence and conflicts of feelings, and tendencies to focus on external events rather than
internal ones (e.g., Meins, Harris-Waller, & Lloyd, 2008). These authors also found evidence that some of
these difficulties are linked to (insecure) attachment history. There is increasing evidence that alexithymic-
type difficulties are prevalent in a wide range of mental-health difficulties—especially those associated with
trauma (Liotti & Prunetti 2010) It is also likely that there are many individuals perhaps successful in
trauma (Liotti & Prunetti, 2010). It is also likely that there are many individuals, perhaps successful in
business or politics, who have alexithymic traits but do not manifest current definitions of mental-health
difficulties— although their insensitivities may create them in others. We know, too, that some people simply
justify their own positions, thoughts and behaviours in all kinds of ways without reflection or apparent doubt
—and helping people get beyond that, by being more open, reflective and accepting of responsibility, and
entering into genuine dialogue, can be difficult. Justifiers and externalizers, however, often don’t attend
therapy—because they see no reason to, “there’s nothing wrong with them”—and may be highly shame
avoidant. So, they simply act out in all kinds of non-reflective ways.

Meins et al. (2008) suggested that the alexithymic trait of externally oriented thinking can arise from a
conscious decision not to explore the contents of one’s mind (motives, thoughts and feelings) and this should
be distinguished from an inability to do so. Koren-Karie, Oppenheim, Dolev, Sher and Etzion-Carasso
(2002) found that while some mothers were able to think about the minds of their babies, other (disengaged)
mothers found this distressing and actively avoided doing so. These mothers were likely to have insecurely
attached children.

Add into this complexity processes by which we use projection and projective identification (Miranda &
Andersen, 2007) to understand our own minds and those of others (including how we create our religious
thoughts and fantasies; Bering, 2002) and you can see that our “new brain” can get into some right tangles
with emotions and motives emerging from the old brain!

Development of psychological competencies

In CFT competencies such as empathy and mentalizing enable social mentalities to function at increasingly
complex levels. Competencies mature and unfold over time. So, in addition to the aspects, such as
alexithymia and mentalizing, that influence how we think about our states of mind, we also need to think
about developmental processes—these are not static skills. CBT therapists have drawn attention to the fact
that clients will have different cognitive abilities linked to Piagetian stages of cognitive development (Rosen,
1993). For example, some will be much more preoperational (difficulties in thinking in abstract terms and
reflecting on inner states) than others. So, people’s abilities to reflect on their inner feelings and those of
others is clearly linked to these cognitive abilities (see also Kegan, 1982, for an important discussion). We
also know that people’s theory of mind competencies develop over time—and may be linked to other
cognitive abilities. We further know that people’s abilities to think and reason morally develop over time
moving from simple concepts of right and wrong, according to what adults punish or reward, to much more
abstract thinking but are also influenced by social context (Gilbert, 1989).

Piaget also drew a distinction between assimilation (how new information is made to fit with current beliefs)
and accommodation (new information causes an internal change and transformation in knowledge
organization and insight). Another aspect here is “openness” to new possibilities and to change, which in turn
butts up against concepts of motivation and “readiness to change”.

Another related area of importance to CFT is that of emotional maturation. For example, Lane and Schwartz
(1987) suggested that affect complexity relates to the ability to differentiate different emotions and
experience. They suggest a set of stages that may parallel Piaget’s cognitive stages. These are awareness of:

1 bodily sensations;

2 the body in action;

3 individual feelings;

4 blends of feelings; and

5 blends of blends of feelings.

At the lowest level, the baby is only aware of experiences in body sensations and these are generally crudely
differentiated in some pleasure-pain dimension. Subsequently comes the awareness of feelings derived from
actions. Later comes the more differentiated affects of sadness, anger, anxiety, joy, etc. Later comes the
capacity to experience blends of these feelings and the capacity to cope with ambivalence. Later still comes
blends of blends of feelings. The way this emotional maturation takes place is still not well understood but
we know that the mentalizing abilities of the parent, and their affection, are crucial. Clients who struggle
with understanding their emotions and get stuck at lower levels of emotional awareness are certainly going
with understanding their emotions and get stuck at lower levels of emotional awareness are certainly going
to struggle with compassion and it’s not surprising that many of them will not be able to understand what
compassion feels like—because
emotions in general are problematic to them (see Point 29).

Increasingly, therapists are realizing that part of their role (not only with the more severe mental-health
problems) is to advance the patient’s capacity for mindfulness and mentalizing, to be less avoidant,
judgemental and self-critical, becoming able to reflect on their feelings, to better understand blends of
feelings, conflicts between feelings, conditioned emotions (and emotional memories) and positive and
negative beliefs about feelings (see also Gilbert, 1992, Chapter 4). The benefits of this is that clients become
tolerant and more accepting of their feelings, rather than fighting or avoiding them, and more open to
relationships based on genuine efforts to understand the mind of others.

Social mentalities and mentalizing

The link between these kinds of awareness and competencies like mentalizing—and those linked to the
pursuit of social roles, social mentalities and self-identity, is complex. Recently, Liotti and Prunetti (2010)
suggested that the way we mentalize is linked into social mentalities. For example, individuals who feel safe
in a particular role, such as talking co-operatively with their therapist, may be able to mentalize reasonably
well; but if their care-eliciting mentality is activated (and thus their attachment systems) this may be very
threatening to the client and overwhelm their ability to mentalize. Liotti (December 2009, personal
communication) suggests that we shouldn’t see mentalizing as an all-or-nothing phenomenon but rather
dependent on a number of other conditions—especially the degree of safeness that individuals feel. Threat
activation can turn off mentalizing because mentalizing is a high-level functioning competency but threat
activation is designed to select from a menu of attentional and response systems for rapid action (Liotti &
Gilbert, in press). Since some mentalities and motives or needs are more threatening than others there can be
differences in mentalizing about and within each—which can catch out the therapist.

I suggest that compassion exercises can create a sense of safeness that facilitates mentalizing. This is
especially true when we teach the “compassionate self” exercise because it focuses on holding the client in a
soothing and soothed position (see Point 21). We are still developing the research methodology for exploring
this, but indirect evidence noted above for the value of compassion (pages 9–12) suggests that this is a useful
avenue of research. One can also help clients understand mentalizing by helping them slow down and
recognize that there are many different parts to the self, with different thoughts and feelings from each (and
not to identify with any specific part)—we call this multi-mind.

Obviously, I have only touched on this complex area here, but CFT therapists need to be aware of these
different degrees of abilities (competencies)—clients’ (developmental) abilities to think about, and reflect on,
“that which arises from within one’s own mind”. This is going to be key to compassion work because we
teach clients to take a compassionate look at the contents of their minds—in part because much of it is
archetypal and was not designed by us but by evolution and personal life experiences (so is not our fault).

So, given our multi-minds, one thing is clear, we are not an integrated coherent whole self, but are rather
made up of a variety of motives and competencies that combine and interact in complex ways. We have the
illusion of being a single self because having a multitude of desires and possible roles and selves, not having
a sense of a single self-identity, would create just too much flexibility, competing possibilities for thought and
actions and incoherence (McGregor & Marigold, 2003). Many psychologists now see “the self” and “self-
identity” as organizing processes that co-ordinate memories, emotions, beliefs and other processes to create a
cohesive sense of self that allows a feeling of consistency or continuation and enables social relationships.
Hence, if we require development and a reorganization of our basic mentalities (e.g., to become more
assertive and competitive, recognize and process anger, or become more caring and compassionate) this can
threaten self-identity and our sense of cohesion.

So, our sense of self is an organizational feat, and when the organization of our potentials within us breaks
down (or e.g., we become overwhelmed by anxiety, rage, intrusions of trauma memories, a sense of
loneliness or submissive withdrawal) we can have mental-health problems (see, e.g., Leary & Tangney,
2003, for reviews). Indeed, people talk about losing control of “parts of themselves” or “falling apart”.

This turns us to the question of what enables us to cope with our multiple (helpful or destructive) potentials
for thoughts, feelings and deeds, and transitions of states of mind. Also, what enables us to open up and
explore new potentials within us and integrate them into our sense of self in what Jung would call an
i di id i i b i di i h f ili
“individuation process”. One answer is compassion, because compassion creates conditions that facilitate
openness, caring, safeness and integration (Gilbert, 2005a, 2005c). Compassion from without and self-
compassion from within, facilitate acceptance and tolerance, thus creating brain patterns that make exploring
and integrating different elements of our mind easier.
5
Attachment and the importance of affection
The importance of affection
CFT is a distinctively physiology-informed therapy that is concerned with the nature, evolved functions and
structure of our brains. In CFT, compassion is rooted in the evolved and socially shaped motives and
mentalities that underpin altruism and care giving. Gilbert (1989, 2005a) suggested that two sources of
altruism (with the motivating and processing systems) underpin compassion: One is kin altruism and
attachment systems of caring, the other is based on reciprocal altruism and desires for moral and fair/just
social relationships. More complex models are now developing (e.g., Hrdy, 2009).

When we take an evolutionary approach, one of the most central mammalian qualities, what stands out
above those of sex, fighting and status seeking, is the huge importance of caring. We now know that the
evolution of caring came with profound changes in central and peripheral nervous systems— one major
adaptation being the regulation of fight and flight to enable closeness, and for physical closeness to have
“soothing properties”. For example, in the brain the endorphins and the hormone oxytocin evolved to
regulate threat processing (inhibiting fight/flight) and promote social interest and caring (Bell, 2001; Wang,
2005). Porges (2003, 2007) detailed how the evolution of the myelinated vagus nerve (to the heart) has
supported interpersonal approach behaviours that enable social affiliations, caring and sharing. The
myelinated vagus nerve evolved with attachment and the ability for infants to be calmed by parental caring
behaviours (Carter, 1998; Depue & Morrone-Strupinsky, 2005). This addition to the autonomic nervous
system can inhibit sympathetically driven threat-defensive behaviours (e.g., fight/flight) and hypothalamic-
pituitary-adrenal (HPA) axis activity, and promote a calm physiological state, conducive to interpersonal
closeness and social affiliation. In general, the safer people feel, the more open and flexible they can be in
response to their environment (Porges, 2003, 2007). This is reflected in the dynamic balancing of the
sympathetic and parasympathetic nervous systems that give rise to the variability in heart rate (HRV; Porges,
2007). Hence, feeling interpersonally safe is linked to HRV, and higher HRV is linked to a greater ability to
self-sooth (quick to be able to tone down threat processing) when stressed (Porges, 2007).

Leaving aside the neurophysiology of caring, one of the most important evolutionary models of the last 40
years is that of the attachment theory of John Bowlby (1907–1990) (Mikulincer & Shaver, 2007); with
somewhat similar concerns being taken up by acceptance-rejection theory (Rohner, 1986, 2004). (See also
Hrdy, 2009, for the importance of multiple care givers in human evolution.) These profoundly important
theories have stimulated considerable research into the way in which children and parents interact such that
the interactions have major effects on a child’s brain and psychological competencies (Cozolino, 2007;
Siegel, 2001). For many mammals affection has evolved as a key to the regulation of motives and emotions.
For humans there is now considerable evidence that if one feels cared for, as opposed to neglected or
rejected, this has major effects on our physiological states and states of mind (Cozolino, 2007; Gerhardt,
2004; Porges, 2007). As we will see later, CFT is rooted in this evolutionary understanding of the huge
significance of the evolution of affection and affiliation.

For species without attachment systems, such as turtles, their life expectancy is short. The mother may lay
hundreds of eggs and only 1–2% survive to reproduce. However, with mammals, providing a safe, secure
base from external threats, and being responsive to distress calls (MacLean, 1985) was central to Bowlby’s
attachment theory (1969, 1973, 1980), making attachment primarily a threat-protection regulation system
(MacDonald, 1992). Bowlby (1969, 1973, 1980) was one of the first to consider the implications of evolved
security providing for infants and its effects on infant development, emotional regulation and internalization
of “working models of self and others” (Mikulincer & Shaver, 2007).

Multiple domains of caring

Over time, mammalian parenting has evolved, becoming increasingly elaborate, so that human parenting is
now highly multi-faceted with: protection (keeping infants out of harm’s way and coming to the rescue),
provisioning (e.g., milk with antibodies, food, warmth, and other resources necessary for development),
soothing (calming distress in the infant and “containing”), stimulating emotion systems (presenting facial
expressions and play opportunities),
mediating (the infant’s exposure to the world), teaching and socializing (teaching about the world and social
relating rules, setting boundaries), validation and “mind sharing” (emotional coaching and mentalizing;
relating rules, setting boundaries), validation and mind sharing (emotional coaching and mentalizing;
Hrdy, 2009). All these influence a range of brain-maturation processes and psychological competencies
(Cozolino, 2007; Gerhadt, 2004; Gilbert, 1989, 2005a; Siegel, 2001, 2007; Wallin, 2007). Over time (in
humans) there evolve potentials for other kin (e.g., siblings and aunts) to play a role in infant care. Indeed,
humans seem particularly adapted to share caring, so that the infant/child interacted with many others who
would care for him/her. This may have facilitated the psychological motives to seek out caring from
(different) others and the psychological desires and competencies to read the (caring) intentions of others
(Hrdy, 2009). In addition, throughout life humans are highly motivated to be liked, desired, valued and
approved of (Barkow, 1989; see Gilbert, 1992, especially Chapter 7, 1997, 2007a, especially Chapter 5,
2007c).

We now know that affectionate caring, associated with feeling wanted and valued, has far-reaching and
multiple effects on the physiological maturation of the infant (Cozolino, 2007; Hofer, 1994). For primates,
and humans especially, the power of nurturing relationships, from a variety of individuals (parents, friends
and lovers), to impact on physiologies throughout life, has become profound (Cacioppo et al., 2000;
Cozolino, 2008; Schore, 1994). Because CFT concepts of compassion are based on these processes, the
complexities of caring relationships should be kept in mind when considering compassion and CFT (see
Point 16). This evolutionary and physiologically based approach, rather than (say) Buddhist, is the
foundation for CFT.
6
Affect regulation: The three affect-regulation systems, caring and
CFT
Our biosocial goals and motivations for (say) sex, status, attachments and achievements are guided by
emotions. When we are successfully pursuing our biosocial goals and motivations we get a flush of positive
emotions, whereas when there are obstacles and threats we experience threat-based emotions. Research in
emotional processing has revealed a number of integrated circuits in the brain that give rise to different types
of emotion that regulate motivation (Panksepp, 1998).

A useful, if simplified, model derived from recent research (Depue & Morrone-Strupinsky, 2005; LeDoux,
1998; Panksepp, 1998) has revealed that our brains contain at least three types of major emotion-regulation
systems. Each one is designed to do different things. These three interacting systems are depicted in Figure
2.

Figure 2 The interaction between our three major emotion-regulation systems. First published in Gilbert, P.
(2009a) The Compassionate Mind. London: Constable & Robinson and Oaklands, CA: New Harbinger.
Reproduced with permission

1 The threat and self-protection system

The function of this system is to detect and pick up on threats quickly and select a response, e.g., fight,
flight, freeze or some other coping efforts, and give us bursts of emotions such as anxiety, anger or disgust.
These feelings will ripple through our bodies alerting us and urging us to take action against the threat. The
system will also be activated if there are threats to people we love, our friends or our group. Although it is a
source of painful and difficult feelings (e.g., anxiety, anger, disgust), keep in mind and explain to clients that
it evolved as a protection system. In fact, our brain gives more priority to dealing with threat than to
pleasurable things (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001). The threat system operates with
particular brain systems, such as the amygdala and the hypothalamic-pituitary-adrenal (HPA) axis (LeDoux,
1998). When in threat mode, attention, thinking and reasoning, behaving, emotions and motives, and images
and fantasies, can all be threat focused, with each aspect of our minds orientated to the goal of protection and
safety. So, we can call this “threat mind” because different faculties of our mind are organized in a particular
way (see Point 4). Once animals and humans are happy with their coping with a specific threat, there may be
little arousal in the presence of threat cues, and it is only when those safety strategies are blocked that “threat
mind” is reactivated. As we will note later, safety behaviours and strategies can reduce arousal in the short
term, but may have longer-term, unintended and very unhelpful consequences (Gilbert, 1993; Salkovskis,
1996; Thwaites & Freeston, 2005).

So the threat system picks up on threats quickly (focusing and biasing attention) and then gives us bursts of
feeling such as anxiety, anger or disgust. These feelings ripple through our bodies, alerting and urging us to
take action to do something about the threat—to self-protect. The behavioural outputs include those of fight,
flight and submission (Gilbert, 200 1a, 2001b; Marks, 1987). Partly because the system is orientated for
 g ( , , ; , ) y y
“better safe than sorry” (Gilbert, 1998) it is easily conditioned (Rosen & Schulkin, 1998). The genetic and
synaptic regulation of serotonin plays a role in the functioning of the threat-protection system (Caspi &
Moffitt, 2006). Problems with the threat system are linked to:

1 the type of triggers that activate the threat-protection system, which are linked to the nature of the threat,
conditioned emotional responses, and personal meanings;

2 the type and forms of the threat-protection response, e.g., anger or anxiety, fight or flight, heart rate,
nausea, sweating, flushing, racing thoughts and attentional focusing;

3 the rapidity and intensity of the threat-protection response;

4 the duration of the threat-protection response and the means and ways of calming aversive threat arousal;

5 the frequency of activation of the threat-protection system linked to external contextual cues (e.g., living in
a violent household) and internal cues (self-criticism, rumination, worry);

6 the way different forms of coping (e.g., experiential avoidance, and unhelpful safety strategies) or the
collapse in mentalizing abilities accentuate a sense of threat, thus further activating elements 1 to 5.

As pointed out elsewhere (Gilbert, 1989, 1993), nearly all psychotherapies focus to a greater or lesser degree
on the complexities of threat self-protection but with different types of theories and methods for engaging
with problems in this system.

We also know that threat processing and responding are complex. For example, emotional memories,
including threat memories, are stored in different systems such as sensory (amygdala) and event-memory
(hippocampal) systems, which can conflict; as in the case of someone who has an experience of fear in a
flashback and feels as if the trauma is happening again (Brewin, 2006; Lee, 2005). Certain types of fear can
interfere with the ability to mentalize (Liotti & Prunetti, 2010). We also know that different defensive
emotions and action tendencies can conflict; one can’t be in fight and flight at the same time— although we
can feel both anxious and angry in the context of a social conflict, or we may even wish to cry (Dixon,
1998). Conflicts between different emotions and “what to do?” increase with stress. We can become anxious
of losing control to anger, or angry at being inhibited and anxious. Stress can also be created through
approach-avoidance conflicts, such as hating one’s job but needing to stay there because of the money to pay
a mortgage—which leads to feelings of entrapment (Gilbert, 2001a, 2001b, 2007a).

Emotional complexity

It is useful to help clients recognize that in any distressed state of mind there can be multiple feelings and
conflicts. For example, Kim suffered from depression and anxiety. Following an argument with her partner
she felt overwhelmed with feelings. So, we wondered if this could be because of multiple feelings, conflicts
between feelings, and conflicts and dilemmas within the relationship. We could then draw a circle and put
“feel terrible” at the centre and around it brain stormed the feelings that were part of that. They included:
anger, anxiety, hopelessness, feeling alone and misunderstood, wanting to stay and wanting to leave, blaming
him and blaming me, powerlessness, doubt, sadness, tearfulness, etc. It is also possible to help clients
explore if they tend to focus on one type of feeling (e.g., anger) to block out and avoid (or an inability to
process) other feelings (e.g., of powerlessness, or sadness-grief). It might then be possible to reflect and
mentalize on each individual element of that distressed mind state—helping clients to do this for themselves
rather than being overwhelmed with complexity and competing feelings.

So, threat processing is complex and it helps to reveal this to clients, clarifying with them how their brains
work, why we go into threat states, and working out the precise difficulties in the threat system for that
person.

2 The incentive and resource-seeking, drive-excitement system

The function of this system is to give us positive feelings that guide, motivate and encourage us to seek out
resources that we (and those we love and care about) will need in order to survive and prosper (Depue &
Morrone-Strupinsky, 2005). We are motivated and pleasured by seeking out, consuming and achieving nice
things (e.g., food, sex, comforts, friendships, status and recognition). If we win a competition, pass an exam
or get to go out with a desired person we can have feelings of excitement and pleasure If you win the lottery
or get to go out with a desired person, we can have feelings of excitement and pleasure. If you win the lottery
and become a millionaire you might feel a mild hypomania—feel so energized that it may be difficult to
sleep, your mind will be racing and you may want to party all the time: the drive-excitement system gets out
of balance. People with manic depression can have problems with this system because it can shift from too
high to too low activation. When balanced with the other two systems, this system guides us towards
important life goals. When blocks to our wants and goals become “a threat”, the threat system kicks in with
anxiety or frustration-anger.

This system is primarily an activating and “go getting” system. A substance in our brain called dopamine is
important for our drives. People who take amphetamine or cocaine try to get “the dopamine” energized and
hyped-up good feeling. The come down, however, is of course the opposite. We will refer to this as the drive-
excitement system for short and to help us keep in mind its focus on activated positive feelings and motives.
As we will see, though, some achievement-focused drive is defensive (see Point 14).

Thwarting incentives and goals

However, as pointed out some time ago, blocks and thwarting of our drives, goals and incentives, typically
activates the threat system (e.g., anxiety, frustration, anger)—until we either overcome the block or
“disengage” from the goal (Klinger, 1977). Disengaging and giving up a goal or aspiration can underpin a
dip in mood (e.g., sadness), and the greater the implications for the self and network of other goals of giving
up are, the bigger the dip in mood. Some depressions are linked to continuing to pursue goals that cannot be
achieved (craving for) and a failure of disengagements and realignment of (achievable) goals (Gilbert, 1984;
Klinger, 1977), e.g., the person who can’t come to terms with losses (e.g., of a relationship), set backs,
illnesses and injuries. There is much in Klinger’s incentive disengagement approach that fits well with
Acceptance Commitment Therapy (Hayes et al., 2004). In CFT goal pursuits are examined for their functions
—especially those linked to different forms of achievement (see Point 14).

3 The soothing, contentment and the safeness system

This system enables us to bring a certain soothing, quiescence and peacefulness to the self, which helps to
restore our balance. When animals aren’t defending themselves against threats and problems, and don’t need
to achieve or do anything (they have sufficient or enough), they can be content (Depue & Morrone-
Strupinsky, 2005). Contentment is a form of being happy with the way things are and feeling safe; not
striving or wanting; an inner peacefulness that is a quite different positive feeling from the hyped-up,
excitement or “striving and succeeding” feeling of the drive-excitement systems. It is also different from just
low threat, which can be associated with boredom or a kind of emptiness. When people practice meditation
and “slowing down”, these are the feelings they report; not-wanting or striving, feeling calmer inside and
connected to others.

What complicates this system, but is of great importance for our exploration of compassion, is that it is also
linked to affection and kindness. For example, when a baby or child is distressed, the love of the parent
soothes and calms the infant. Affection and kindness from others helps sooth us adults too when we’re
distressed, and gives us feelings of safeness in our everyday lives. These feelings of soothing and safeness
work through brain systems similar to those that produce peaceful feelings associated with fulfilment and
contentment such as the endorphins. The hormone oxytocin is also linked to our feelings of social safeness
and (along with the endorphins) gives us feelings of well-being that flow from feeling loved, wanted and
safe with others (Carter, 1998; Wang, 2005). This system is a central focus in compassion training because it
is vital to our sense of well-being. I will refer to it as a soothing and contentment system.

Depue and Morrone-Strupinsky (2005) linked the two positive affect-regulating systems to different types of
social behaviour. They distinguished affiliation from agency and sociability. Agency and sociability are
linked to control and achievement seeking, social dominance and the (threat-focused) avoidance of rejection
and isolation. Warm and affiliative interactions, however, are linked to social connectedness and safeness as
conferred by the presence and support of others. Affiliative social relationships calm participants, alter pain
thresholds, the immune and digestive systems, and operate via the oxytocin-opiate system (Depue &
Morrone-Strupinsky, 2005). There is increasing evidence that oxytocin is linked to social support and buffers
stress; those with lower oxytocin having higher stress responsiveness (Heinrichs, Baumgartner, Kirschbaum,
& Ehlert, 2003). Oxytocin also impacts on threat processing in the amygdala.

CFT makes a big distinction between safety seeking and safeness. Safety seeking is linked to the threat
system and is about preventing or coping with threats Safeness is a state of mind that enables individuals to
system and is about preventing or coping with threats. Safeness is a state of mind that enables individuals to
be content and at peace with themselves and the world with relaxed attention and the ability to explore
(Gilbert, 1993). Safeness in not the same as low activity—when we feel safe we can be active and energized.
If some individuals try to create states of safeness by, say, isolation and keeping their distance from others
then we see this as more safety seeking. The problem with this is that the brain can read
isolation/disconnection itself as a threat and, in addition, this safety behaviour cuts them off from a natural
regulator of threat—the endorphin-oxytocin system for affiliation. So, although avoidance and isolation may
work to a degree, it’s difficult to know how this action affects well-being. Certainly, research has shown that
social anhedonia—the (in)ability to experience pleasure from social relationships—is linked to a range of
psychological difficulties.

Evolutionary functional analysis

A number of key points arise from understanding the nature and origins of our emotions and what they are
designed for; what their functions are. The first is that many negative emotions such as anxiety, anger,
disgust and sadness are a normal part of our emotional repertoire. Like diarrhoea and vomiting, they are
unpleasant but have basic protective functions, and even when having harmful effects are not necessarily
pathologies (Nesse & Ellsworth, 2009). So important are our protection emotions that they are the big,
emotional players in our brains and can easily override positive emotions (Baumeister et al., 2001). We can
also explain to our clients that our brains did not evolve for happiness but for survival and reproduction, so
sometimes we need to learn how to accept, tolerate and work with difficult emotions or low moods; these are
not evidence of “something wrong” with us but can be quite normal responses to things in our life.

Second, it follows that our emotion systems may be working perfectly normally but the inputs are
problematic. Someone who feels trapped in, say, a loveless relationship may become depressed, which can
be accentuated by their focus, beliefs and ruminations, of course, but the depression can also partly be a
normal consequence of this situation in which they find themselves. It is very hard to control your “stress-
cortisol output” if you are being bullied. Some people’s lives are stressful—or they may have experienced
tragedies and losses and that’s why they feel sad or are grieving. Of course, there are things we can do to
amplify threat-based feelings by how we think (Wills, 2009; Dryden, 2009) or engage in various forms of
avoidance (Hayes et al., 2004) but it is important to help people normalize and recognize that some of our
unpleasant feelings and reactions are not abnormal but need to be engaged with—compassionately.

The third issue is to recognize that modern societies are, in a whole variety of ways, over stimulating both
our threat (unstable employment, house repossession, poverty) and drive (“want more and need to do more”)
systems and playing havoc with our needs for connectedness and focus of social comparison (Gilbert, 2009a;
Pani, 2000; Wilkinson & Pickett, 2009). Twenge, Gentile, DeWall, Ma, Lacefield and Schurtz (2010)
produced a major review of evidence that mental health difficulties have been increasing at an alarming rate,
especially in younger people. They attribute much of the cause for this to cultural shifts towards extrinsic
goals such as individualism and materialism with competitive and rank-focused self-evaluation, and away
from intrinsic goals such as cooperation, community and sharing. So, we need to socially contextualize
shame and mental-health issues.
7
Affiliation, warmth and affection
We have seen that the evolution of attachment is one of the most fundamental aspects of the mammalian
mind. Also, the evolution of attachment had a major impact on the evolution of emotional-regulation systems
and in particular the importance of a social soothing system that operates with endorphins and oxytocin. This
system and these neuro-hormones play a special role in regulating threat and threat arousal.

However, different types of caring will impact and stimulate the soothing system in different ways. For
example, caring can be with or without warmth and affection (MacDonald, 1992) but it is warmth that is
probably most associated with soothing and endorphin release (Wang, 2005). Warmth involves a number of
qualities such as tenderness, gentleness, kindness and concern, and also playfulness. We can distinguish
warmth from protection giving. Also, one can have attachment in the absence of warmth-affection, and
provide affectionate care for others (e.g., care of the dying) in the absence of (a desire for) attachment.
Dominant animals and humans may be able to protect their infants from dangers and threats in a way that
subordinates cannot (Suomi, 1999) but this does not mean that they provide more warmth. People may form
attachments based on submission/appeasement to “not very warm others”, if they see them as best bets for
protection (Gilbert, 2005a). Indeed, anxiously attached people use appeasement as an attachment/security
device (Sloman, 2000). Now, warmth is an important attribute of compassion—so what do we mean by
“warmth”?

Warmth
There are a number of models that posit warmth as an important personality construct. It is linked to the
agreeableness dimension in the “Big Five” personality trait classification, and the central love-hate
dimension in the interpersonal circumplex model of personality (McCrae & Costa, 1989). Many studies have
shown that “agreeable-warmth” attributes are associated with prosocial behaviour, better academic
performance and well-being (e.g., Laursen, Pulkkinen, & Adams, 2002). It makes sense, then, that “warmth”
should be a focus for therapeutic research and work.

Warmth appears to have at least three key attributes. First, warmth provides verbal and nonverbal signals of
interest, caring and kindness that are soothing. Second, warmth can involve a sharing of positive affect
between individuals that stimulates liking, affection and feelings of connectedness (in contrast to
indifference, withdrawing or attacking). Third, warmth is more likely when individuals feel safe with each
other and are trusting. Individuals who are easily threatened and become defensive may struggle to feel or
express warmth.

Warmth underpins the positive feelings of soothing, calming and being soothed—it moderates defensive
emotions (anger, anxiety, sadness) and behaviours (e.g., aggression and flight), and can also turn off seeking,
doing, achieving and acquiring. Part of the positive feelings of soothing may come from declines in, or
regulations of, negative affect (e.g., feelings of relief; Gray, 1987). However, the positive affects of social
safeness-creating soothing, via the exchange of social signals, that impact on the mind of the other, need
consideration in their own right. This is because they have far-reaching organizing effects on the brain and
are associated with specific social (safeness-conferring) signals (not just the removal of threat signals) such
as nonverbal communication, facial expressions, voice tones and verbal content, and touch.

Field (2000) reviewed the evidence on the beneficial effects of holding, stroking and touching during
development—interac tions of affectionate care. Even laboratory rats grow up calmer if they are regularly
stroked. These signals are soothing in their own right, and influence the release of endorphins. As Sapolsky
(1994) observed:

We readily think of stressors as consisting of various unpleasant things that can be done to an organism.
Sometimes a stressor can be the failure to provide something to an organism, and the absence of touch is
seemingly one of the most marked of developmental stressors that we can suffer.

(p. 92)

So, from the first days of life, safeness-via-warmth is not simply the absence of threat but is conferred and
stimulated by others with the soothing, care-giver signals of touching, stroking and holding (Field, 2000),
voice tone the “musicality” of the way a mother speaks to her child positive/affectionate facial expressions
voice tone, the musicality of the way a mother speaks to her child, positive/affectionate facial expressions,
feeding and mutually rewarding interchanges that form the basis for the attachment bond (Trevarthen &
Aitken, 2001). These signals stimulate endorphins that give rise to feelings of safeness, connectedness and
well-being, and the infant’s physiological systems are attuned to them (Carter, 1998; Wang, 2005). Thus,
from birth, infants are highly sensitive to interpersonal communications and particular signals that regulate
arousal, help organize physiological systems and set strategies onto developmental trajectories.

When we think about a “compassionate other” we normally imagine them as having qualities of low
negative affect and generating warmth and expressing soothing signals (e.g., friendly facial expression and
soothing voice with feelings in-them-about-us of acceptance). A basic co-regulating partnership, that
involves genes and physiologies in one person communicating with genes and physiologies in another, and
minds communicating with other minds, thus emerges. The very sense of self is being sculptured in
interaction, and at the root of this are brains that need others to love and care for them.

Hence, to “be socially soothed and feel safe” requires certain social signals, while to be “soothing”
necessitates providing these signals to others (Gilbert, 1993, 2009a). As noted by Bowlby (1969, 1973) and
Porges (2003, 2007), once soothed, individuals either may relax (passive safeness) or may redirect attention
to explore the environment (active safeness). It has been this reasoning that has stimulated therapeutic work
to try to teach people how to generate soothing (compassionate) images, feelings and thoughts (Gilbert,
2000a, 2009a; Gilbert & Irons, 2005). Interestingly, the same idea, of developing inner images of soothing
others (e.g., bringing images of the compassionate Buddha to mind) is central to Buddhist compassion
imagery (Vessantara, 1993). Importantly, for clinical work though, many high-shame clients find warmth and
soothing difficult to do and are fearful and resistant of it (see Point 29)— as if this system is toned down and
suppressed.

Validation seeking
Validation for emotions and personal reactions to (early) life events can be sought or avoided in therapy.
Linehan (1993) and Leahy (2005) note the importance of emotion acceptance and validation of the child’s
emotions—a key process for compassion. Leahy outlined a useful model depicting how validation of
emotions and early emotional coaching can link to different emotion strategies and relating styles. This is
given in Figure 3. These styles will emerge in therapy and the therapeutic relationship.

When we develop the “the compassionate self” and engage in chair work (Points 21 and 22) we pay a lot of
attention to the ability of the clients to be compassionately validating of their feelings and conflicts of
feelings—which, for many, takes time to learn.

Figure 3 Relation of parenting, attachment styles, beliefs about validation and interpersonal strategies.
Reproduced with kind permission from Gilbert, P. (ed.) (2005c) Compassion: Conceptualisations, Research
and Use in Psychotherapy. London: Routledge

Cognitive abilities
Love and affection are like vitamins to the brain, but these go beyond physical interactions. There is
increasing work showing that how a mother understands and responds to her child’s emotions and needs is
crucial to the child’s ability to understand and regulate his/her own mind (Cozolino, 2007; Siegel, 2001;
Wallin, 2007). So, in the context of validating, loving relationships, we come to feel safe and able to explore
our own minds, to understand our emotions, to feel soothed and contained, and in the process are able to
understand the minds of others. These cognitive abilities are vital for metacognitive abilities to be able
reflect on our emotions, stand back from them and not be overwhelmed by them (Allen, Fonagy, & Bateman,
2008; Wallin, 2007; see Point 4).
8
Clarifying the CFT approach
As is evident from the above, CFT is rooted in the science of mind and basic psychological research in
contrast to, say, a special “therapy focus”. We also take a physiotherapy approach to mental health in the
sense that the manifestation of a problem may not be the source of the problem. For example, I have had a
problem in my lower back. Efforts to work on my lower back were only minimally helpful. Then a
practitioner noticed that I was very flat-footed and my knees turn inwards, which affected my hips and put
pressure on muscles in my back (what a wreck!). Without touching my back, but working on my flat-footed
problem, he resolved the back difficulty. Sometimes psychotherapy can be like this too. If we only focus on
trying to change the threat system we may fail to realize that part of the problem is in the balance of the other
affect-regulation systems, and in particular the poor output from the soothing system. When that system is
developed the others can settle down.

Philosophical position
The philosophical position of CFT arises from various observations on the nature of life. We call this a
“reality check” and explore the challenges of life (Gilbert, 2009a). The “reality check” is used to offset
pathologizing and is key to therapist training. These “reality checks” are as follows:

Evolved mind

We are an emergent species in the flow of life—part of mammalian and primate evolution. Our bodies,
brains and minds evolved to function in specific ways with capacities for certain emotions (e.g., anger,
anxiety, disgust, joy and lust), a range of defences (e.g., fight, flight and submission) and archetypal
motivational systems (e.g., to form attachments, seek status, belong to groups, and desire and seek sexual
partners). These are old brain motives and competences. They are the driving forces of much of what we do
and think. Our “new brain” competencies and talents (for complex thinking, reflection and self-awareness)
can interact with old brain motives and passions leading to the best and worst in ourselves. In different states
of mind, different elements of our minds are turned on and off. The threat- or vengeance-focused mind often
turns off motives and competencies for compassion to the targets of one’s vengeance. In contrast, a
“compassionate mind” tones down threat-focused feelings, thoughts and behaviours.

CFT begins, then, with the recognition that our brains are actually difficult and tricky; they are not well
designed; our thoughts, emotions and behaviours can be captured by primitive emotions, motives—and
terrors. In different mind states we think and feel quite differently. The Buddha took a similar view,
suggesting that our minds are chaotic and craving, and that only through training our minds could we
achieve some harmony within and take responsibility for ourselves and our actions.

Tragic mind

The second reality check is that our lives are relatively short (25,000–30,000 days, if we are lucky). We are
caught in a genetic lottery, which determines the length of our lives and the kind of illnesses we will suffer.
Young daughters, wives and mothers may die early due to breast-cancer genes. Leukaemia, cystic fibrosis or
malaria can rob families of their children. We are subject to a huge range of illnesses that can cripple us, rob
us of our capacities for hearing or sight or kill us slowly (e.g., AIDS, dementia). Humans have understood
for a long time that in many ways we live a tragic life with much pain and suffering. Indeed, trying to
explain why this should be has been at the root of many philosophical and spiritual traditions.

Compassion focused therapists work from a position of awareness of the “tragic mind”. Indeed, it is the very
real tragedies of our lives that summons up the importance of compassion; our common humanity and what
we are all caught up in and are struggling with.

Social mind

The third reality check is the fact that we live in a world of immense injustice and suffering. We know that
the social circumstances of our lives play a huge role in how our brains mature, the values that we develop,
our motivations and the self-identities we grow into (Schore, 1994). Even our cognitive and mentalizing
 g ( , ) g g
abilities are aided or stunted by the relationships that we grow up in (Allen et al., 2008; Cozolino, 2007). If I
had been born into a Mexican drug cartel, or somehow there had been a baby swap at birth, the chances are I
could now be dead, would have killed others, may be addicted to drugs myself and/or be living the rest of
my life in prison. There would be no chance of the potential of what was in me (to become a professor of
psychology and a clinician) ever coming to life in that environment. It is important to recognize and reflect
on the fact that this “you” in “this life” is only “one version” of many that could have emerged the day you
were born. When we look at our clients we are only seeing one aspect of their potential, only one version of
that self—can we (see and) help them recognize and develop others?

From “not our fault” to taking responsibility

This shared understanding, about the predicament of our human minds and brains (and lives) leads to a
recognition that much of what goes on in our minds is not of our design and is therefore not our fault.
Clarifying this aspect can play a key role in undermining a person’s sense of being worthless, useless and no
good when they become depressed or feel out of control of their minds. I have worked with groups of people
diagnosed with severe borderline personality disorder and introduce them to the CFT model, explaining why
much of what goes on in their brains is not their fault. We all just find ourselves here, with this difficult
brain, created over millions of years of evolution, genetic dispositions and with a sense of ourselves and
various emotional memories obtained through the social circumstances of our lives—none of which we
chose. Although many of the people have undergone various types of therapy before, they usually say that no
one had outlined this. We also discuss how “our coming to be ourselves” is the result of multiple factors
(Gilbert, 2009b). Clients find this very de-shaming and reassuring at the beginning of therapy because many
of them felt they were either bad or mad or had odd brains; that there was something very wrong with them
for having the problems they had and this diagnosis. We say to clients, “We are not so interested in your
diagnosis but we are very interested in how your threat, excitement and soothing systems are working for
you”. Time spent really helping people understand the “not our fault” aspect of our minds is time well spent
because it orientates the person to approach their difficulties in a particular, more objective (compassionate)
way.

However, this is only the beginning because it is easy to confuse “causality” with “responsibility”, and so it
is very important to distinguish these two. Therefore, we suggest to people that, “You might not be to blame
for how your mind is, the passions, terrors and rages that can flow through it, but only you can take
responsibility for training it for your and others’ happiness. It is like a garden. You can leave your garden to
grow and it will grow; weeds and flowers will grow but you may not like the tangles that emerge if you
simply leave it to its own devices. Same with our minds. So, cultivating, practising and focusing on those
elements of our minds that we wish to enhance is key if we are to take control over our minds”. In many
ways this is no different from saying to people that it may not be their fault that they are flat-footed, are
short-sighted or suffer from a variety of complaints—but it is up to them to do something about it.

It can be useful, therefore, to encourage guided discovery and guided reflection on the unintended
consequences of “not taking responsibility for change”. However, keep in sight that that, too, is only one step
because people are less likely to change just from threat, e.g., the threat of cancer may not stop people
smoking. Rather, one must also build realistic images and pictures of the outcomes of change.

Key also in CFT is the de-shaming approach and ensuring that interventions stimulate particular systems. So,
again, while it is important for clients to work at taking more and more responsibility for change, and to
learn to tolerate setbacks without shaming, building the positive coping and compassionate self (see Point
21) gives a positive non-shaming focus for responsibility taking.

The interactions of new brain with old brain

In the standard CBTs the therapist tries to identify unhelpful thoughts, beliefs and schemas and then seeks to
change the content of them. More recent therapies, such as Metacognitive Therapy (Fisher & Wells, 2009;
Wells, 2000), Mindfulness (MBCT; Segal, Williams, & Teasdale, 2002), Dialectical Behaviour Therapy
(DBT; Linehan, 1993) and Acceptance Commitment Therapy (ACT; Hayes et al., 2004), have all raised
questions about the adequacy of this approach. They have focused instead on the nature of rumination and
worry, and avoidance, and less on the content of cognitions. CFT suggests that both are important depending
on case and context. For example, helping people with the contents of their beliefs, including those related to
what is (un)necessary to stay safe, can be very helpful. However, we would add a shift to compassion
focusing first because it is not only the accuracy of alternatives that is key but also emotional experiences of
focusing first because it is not only the accuracy of alternatives that is key but also emotional experiences of
them (see pages 164–165).

Clearly, too, attention to the ongoing processes of thoughts, linked to rumination, worry and self-criticism,
are important. In CFT we argue that these processes operate as complex stimulators of the three (old brain)
affect-regulation systems, activate physiological systems and fuel certain mentalities. For example, feeling
put down or disrespected and having angry and vengeful rumination maintains the threat system and the
competitive mentality; ruminating on lack of love or affiliative bonds maintains the threat system and the
care-seeking mentality.

In Metacognitive Therapy, Wells (2000; Fisher & Wells, 2009) makes clear that rumination (looking back)
and worry (looking forward) exert their impact when they constantly stimulate the threat system, and
interfere with normal recovery processes. In this approach, problems arise because attention is overly
focused on the self and threat to the self from within and without; this sense of threat is maintained through
rumination and worry; and self-regulation strategies fail to modify this processing loop and the self-
experiences (e.g., inferiority or vulnerability beliefs) that give rise to it. In CFT we would suggest that self-
regulation strategies (also) fail because they’re not linked into the appropriate affect-regulation system. Thus,
it is not so much the content of cognition but the repetition of threat-focused thinking that is harmful.
Attention reallocation is therefore a key element of this intervention. CFT suggests that it is also useful to try
to stimulate the natural regulator of the threat-protection system—the soothing and affiliative system—using
compassionate refocusing, imagery and attention (see Part 2).

The key point in Wells’ approach, and in those that focus on mentalizing, is that problematic emotions and
difficulties are maintained because of the creation of an internal feedback loop between the content and
focus of one’s thoughts and the threat system. Keep in mind that the threat system is not designed for

Figure 4 Stimulating different systems

complex thinking but for rapid actions and, therefore, CFT suggests it makes sense to try to shift to a
different affect system in order to facilitate new processing.

This can be depicted simply, as in Figure 4. Arrows go both ways because the more threatened we feel the
more our attention and thinking becomes threat focused, and the more threat-linked intrusions we might
experience. The point is that self-criticisms, worries and ruminations that are constantly circulating in one’s
mind are also constantly stimulating threatbased central (brain) and peripheral systems (e.g., the sympathetic
and parasympathetic nervous systems). Over time these pathways become stronger.

In this context, therefore, CFT sees the mechanisms of change as:

1 to disengage from the (inner) stimulators of threat, e.g., ruminative, self-criticalness or anger (shared with
metacognitive- and mindfulness-based therapies), and refocus on compassion insights and feelings;

2 to be able to compassionately “stand back” from one’s inner storms of emotion and become more
“observant and watching” of one’s thoughts and feelings “as they arise” rather than caught up in them (as in
mindfulness and acceptance based therapies) having a compassion base can help that difficult process;
mindfulness- and acceptance-based therapies)—having a compassion base can help that difficult process;

3 to activate the natural threat regulator in the brain—the soothing system—by switching to compassionate
refocusing and imagery;

4 to be able to engage with aversive inner experiences, such as trauma memory, or avoided emotions, by first
developing an inner compassion base.

Rumination and goals

Rumination is often linked to goals and feeling thwarted (Klinger, 1977), so in CFT rumination often
requires functional analysis on goals. For example, ruminating after making a mistake can be linked to a
threat-fear of being seen as incompetent and rejected; so rumination can be linked to activity in the incentive
and resource-seeking system (wanting to succeed). CFT helps people recognize that they may have to make
changes in their lives, give up things, come to terms with losses, realign their goals and sense of self, engage
with things that frighten them—all of which require courage. In CFT we talk a lot about the development of
courage—and that this is more likely if we can create kind, helpful and supportive voices in our heads rather
than coldly logical, bullying, critical or pushy ones.

So, in a whole range of interventions a compassion focus textures that intervention, but the interventions
themselves are taken from a range of different evidence-based therapies.
9
Formulation
Case formulation is an individualized process that seeks to understand the nature, sources, maintaining
factors and alleviation factors for people’s difficulties (Eells, 2007; Tarrier, 2006). Many therapists recognize
that case formulation needs to be embedded within the person’s past and current context and should be based
on an understanding of the functions of that person’s emotions, behaviour and thoughts (Cullen & Combes,
2006). Two people may take an overdose—one was depressed, the other drunk and lost count of the
painkillers taken for a hangover. The meaning of a depression in someone who, on the surface at least,
appears to have a relatively good marriage and job is quite different from a depression in the context of
poverty and domestic violence or previous child sexual abuse. Even though both people may have similar
symptoms and beliefs of “being weak and a failure” when depressed, their beliefs may have very different
origins and functions.

Nearly all psychotherapies believe that symptoms of mental-health difficulties revolve around threat, and
mechanisms of defence and protection (Gilbert, 1993). Psychoanalysis derives a complex set of internal
defences, such as repression, projection, denial and sublimation. CBT also focuses on issues of threat and
defence but uses behavioural frameworks, where avoidance is a primary defensive behaviour. In a number of
writings Salkovskis (e.g., 1996) outlined various behaviours linked to the avoidance of aversive outcomes
(Thwaites & Freeston, 2005). He makes clear that (counter to how it is sometimes portrayed) CBT is not
about showing people that they are being irrational or erroneous in their thinking, but about investigating
how and where they have got stuck and trapped in understandable but unhelpful ways of trying, as best they
can, to make sense of their problems and to get (or stay) safe. He points out that people often develop beliefs
around safety behaviours that strengthen rather than test out or weaken those same beliefs. For example, a
panic client may sit down because they think they could have a heart attack. When the heart attack does not
come they do not attribute its absence to their erroneous belief but to the fact that they sat down. Safety
behaviours therefore become entrenched by: (1) experienced short-term benefits/reinforcers (e.g., relief); and
(2) beliefs that maintain them. Safety behaviours and strategies are aimed at avoiding both external and
internal threats/harms and are therefore key threat-self-protection (system) regulators. For example, a
socially anxious person may monitor their verbal output, speak little in order to avoid appearing stupid, and
constantly try to work out how they appear in the minds of other people. This is to avoid the external threat
of rejection, exclusion or humiliation. In order to control their internal threat of anxiety arising, and just
feeling bad, they may drink alcohol.

CFT only has three basic emotion-regulation systems to work with and so formulation is around the
organization of these systems with a particular focus on threat and safety strategy development. Moreover,
CFT suggests that there are a number of innate and evolved potential safety strategies that can be activated
and texture self-evaluative systems (Gilbert, 200 1a). For example, a child who we will call Ann who is
regularly threatened by her parents may come to monitor the aggressive state of mind of the parent and
quickly adopt submissive or avoidance strategies if she picks up cues of threat. The submissive and
avoidance strategies are a normal part of an innate repertoire of social defences. As Ann grows up, with
increasing cognitive competencies for self-other awareness coming on line, theses experiences, of the
activation of safety strategies, form part of her self-experience and identity. So, for example, when
confronted with a powerful other, Ann will automatically monitor their state of mind and herself to make
sure she is not doing things to stir up their anger. If anger is directed at her, she will then try to work out what
it was she did (self-blame) and adopt submissive strategies to minimize the threat. So, automatic defensive
strategies become linked to cognitive systems such that people quickly self-monitor, self-blame, and behave
submissively. In addition, they will, of course, have beliefs about the self to match these strategies, such as
seeing themselves as weak, blameworthy and so forth. In CFT we would make clear these are developed
safety protection strategies rather than use the language of cognitive distortions, and spend time helping the
person understand their function and the fear of changing. It is these unique profiles of safety strategies,
which arise from threat and unmet needs, that are important in CFT, rather than a more diagnostic approach
of trying to identify specific symptoms, core beliefs or schemas.

Key elements of a CFT formulation

The CFT formulation for high-shame and self-critical people integrates cognitive, behavioural and
attachment models and focuses on four key domains:
1 innate and historical influences that give rise to—

2 key external and internal threats and fears that give rise to—

3 externally focused and internally focused safety strategies that give rise to—

4 unintended consequences—that fuel more distress, safety strategies and difficulties—including self-
criticism.

The unintended problems/consequences can be related to symptoms, which then give rise to a fifth aspect
(say) of fear, anger and various metacognitions about how one is currently (not) coping and is suffering (e.g.,
depressed or anxious).

Background and historical influences

Here the therapist explores for basic early relating and attachment styles, life events and emotional
memories, that illuminate issues of feeling cared for or about, or neglected, unmet needs, experiences of
feeling threatened and forms of abuse. The latter may only emerge slowly and depends on the person feeling
safe enough to be able to tell of these events. We know, for example, that many people can go through
therapy, and due to shame, do not reveal key issues for them (e.g., MacDonald & Morley, 2001; Swan &
Andrews, 2003). Given this, therapists who rely on “problem lists” in therapy may be at most risk of missing
shame problems.

Early life experiences will have patterned various neurophysiological systems and the co-ordination of
various affectregulation systems (Cozolino, 2007; Ogden, Minton, & Pain, 2006). We now know, for
example, that early care (or a lack of it, or abuse) affects brain maturation, affect regulation (Gerhardt, 2004;
Schore, 1994, 2001; Siegel, 2001), cognitive abilities, and abilities to mentalize and understand other
people’s thoughts and feelings (Allen et al., 2008; Siegel, 2001). The therapist explores for key emotional
memories that act as a focus for self-experience and can be triggered by life events (Brewin, 2006).

Some people, however, may have poor recall of negative events, go blank or find revealing and “going into
the history” stressful. Others will tell that “all was wonderful” and it is only later that you find out it was not.
The importance of the coherence of a narrative of one’s background has been illuminated by research on
attachment using the Adult Attachment Interview (Mikulincer & Shaver, 2007). So we know that it is
important to ask specific questions. “You felt that mum/dad loved you—how did she/he show this? How did
she/he comfort you; how did she/he talk to you about your feelings? In what ways were they physically
affectionate; if you were distressed how would they help you?” Commonly, in shame-based problems you
will find the absence of feelings of closeness, validation and support, and not uncommonly experiences of
distance and threats/harms.

Careful history taking emerges over time because it is not just “fact finding” and identifying “hot spots” (that
were a spur for safety strategies and developing personal meanings) but it also offers key opportunities to
compassionately empathize and validate people’s life experiences (Leahy, 2005; Linehan, 1993). Creating
“safeness” in these ways, enables people to develop a coherent story and narrative of their difficulties.
Therapy may be the first time people have experienced another person’s mind orientated to them in this
interested, non-judgemental, containing, empathic and caring way and it helps them to create a coherent (de-
shaming) narrative (Gilbert, 2007b). This begins the process of people beginning to understand their
problems in terms of phenotypic development, a mind that has been orientated to “better safe than sorry” and
“safety first” life strategies, with a sense of self being textured by these experiences.

Emotional memories of the self

It is important to explore the felt sense of self and understand how this is linked to emotional memories. For
example, Gilbert (2003) suggested:

Consider early experiences of how a child experiences the emotions of others in an interaction and these
become the foundations for self-beliefs. A positive belief of, “I am a lovable competent person” is really
shorthand for, “in my memory systems are many emotionally textured experiences of having elicited positive
emotions in others and being treated in a loving way, and as competent—therefore I am lovable”. Suppose
parents are often angry towards a child. This child develops beliefs that others do not see her positively,
which is shorthand for “in my memory systems are emotionally textured experiences of
which is shorthand for, “in my memory systems are emotionally textured experiences of
having elicited anger in others and being treated as bad—therefore I am bad”. Suppose parents always show
contempt or withdraw their love and turn away from the child. It is not anger that is internalised but loss or
contempt. This child develops beliefs that others see her as someone to turn away from and believes she is
unlovable. This is shorthand for “in my memory systems are emotionally textured experiences of having
elicited withdrawal in others and being treated as undesirable—therefore I am undesirable”. Consider the
child who is sexually abused. This can become, “in my memory systems are emotionally textured
experiences of fear and disgust—therefore I am, disgusting and bad”. Tomkins (1987) argued that shame
(and other self-conscious emotions) are laid down in memory as scenes and fragments of images of self in
relationships. These encoded scenes can then become “mini co-ordinators” of attention, thinking, feeling and
behaviour—giving rise to what Jung called “complexes”.

(pp. 1221–1222)

Psychodynamic therapists would see these processes as “self-objects”, cognitive therapists as “self-
schemas”. The key point, though, is that schemas of self emerge from self-other interactions and are rooted
in emotional memories, which can affect body memory and the “felt sense of self” (Brewin, 2006; Ogden et
al., 2006). This is why, when working with shame, CFT often works directly with the core experiences (e.g.,
of threat and aloneness-sadness, see Points 10 and 11), revisiting threat memories and developing new
compassionate-safe emotional experiences (Hackmann, 2005; Lee, 2005). Working only on the explicit
processing systems by trying to develop rational alternatives to shame may not be sufficient to influence
emotion laden memories that fuel and maintain a defensive orientation (Brewin, 2006).

Key threats, fears and unmet needs

Early background experiences can enable us to feel safe and secure or easily threatened and insecure
(Mikulincer & Shaver, 2007). As Gilbert (1989), Beck et al. (2003) and other therapies note, key fears
arising in childhood, which are going to have long-term influence, are often around archetypal and innate
themes of abandonment, disengagement, rejection, shame and abuse/harm. In CFT we distinguish between
external threats and internal threats. External threats pertain to what the world or others might do; whereas
internal threats are related to what emerges, or is recreated inside oneself. For example, a person might be
frightened of the external threats of rejection, exploitation or harm from others; and/or of the internal threats
of losing control or becoming overwhelmed by anxiety, anger or depression (internal threat). Indeed, it can
be the fear of becoming depressed (again) that can set in motion rumination, avoidance, fear and dread of the
future and even suicide (Gilbert, 2007a). Helping people articulate and reflect on what key fears and
concerns they might be carrying from childhood can be immensely helpful—especially when we look at the
next element, which is how they will have tried to protect themselves from childhood onwards—setting in
place their protective strategies and shields. Indeed, all the time we talk in terms of what difficult
backgrounds generate in us, then the need to develop “safety strategies”, “efforts at self-protection”, “what
did you need to get good at”, and how one ends up “doing the best one could at the time”.

Safety and compensatory strategies

From the first days of life our brains automatically develop a range of strategies to seek safety, self-protect
and self-soothe. There are genetic differences in the disposition of infants and children to engage in certain
strategies. For example, some children are more easy to soothe and comfort with physical affection than
others. Shy children are more wary and avoidant than more explorative children. Some children are able to
use their parents as a safe base whereas for others the parents are themselves a source of threat (Liotti, 2000).
There is also now much research on how children respond to threats by using their parents as sources of
reference (is this safe or not?—as in the visual-cliff experiments), protection and calming (Mikulincer &
Shaver, 2007). While some children turn easily and expectantly to their parents for calming when threatened,
others are far more avoidant, with a third type being orientated to the parents but not able to calm down
when held or soothed. All of this speaks to the regulation of the threat system.

Parents are so important to a child’s safety that children will try to influence the mind of their parents (Liotti,
2000, 2002; Wallin, 2007). They may become very submissive, or competitive and striving, or caring and
rescuing, polite and well mannered. These safety strategies have a hope that others can be encouraged to be
helpful and value them. If more externalizing safety strategies, such as aggression and impulsiveness,
develop there seems to be a turning away (of hope) of building affiliative relationships in favour of more
self-reliant, protective strategies and of ensuring that one’s potential to be harmful registers in the mind of
others so that they remain wary of one. This is a strategy (of suddenly being aggressive) that some dominant
monkeys use to maintain fear in subordinates (Gilbert & McGuire, 1998).

Humans evolved to require caring attachments, to be looked after and soothed when stressed. Deviations
from these developmental trajectories have consequences. Unmet needs can constitute a threat in that the
individual is not able to develop optimally without specific inputs. This can lead to a certain yearning and
seeking and various fears that certain types of relationship cannot be obtained or maintained (Knox, 2003).
Along with this go complex safety and compensatory strategies. For example, a girl who has a poor
relationship with her father, or turns to him to protect her from (say) a cold mother, may have a yearning to
form a close relationship with a father figure and find herself attracted to older males who may or may not be
helpful attachment figures. Given the functions parents have, such as protection, provisioning, soothing,
validation, encouraging, “delighting in the achievements of”, these can all be sought in the therapeutic
relationship—or indeed from other people. When the person finds they are not forthcoming in the way that
they want, they can feel thwarted again, threatened and disappointed.

Unintended consequences
It is basic to CBT that safety strategies nearly always have unintended and often undesirable consequences;
“symptoms” being one of them (Salkovskis, 1996; Stott, 2007; Thwaites & Freeston, 2005). These
consequences can also either maintain the problem or make it much worse. For example, individuals who are
frightened of their emotions (e.g., some emotions trigger internal threat) may engage in experiential
avoidance—a key focus in various exposure therapies, such as ACT, and also Mindfulness (Hayes et al.,
2004). Fear of the impact that revealing or expressing their emotions might have on others can lead to
avoiding openness and honesty in relationships. As a result the relationship lacks self-correcting interactions
(e.g., sharing dissatisfactions) and the person begins to ruminate more and more on the unspoken
resentments. These ruminations are depressing and usually people become poor at identifying the real
difficulties and solutions or, on the other hand, acknowledging what they like and appreciate. In an effort to
try to maintain a positive relationship by being submissive and hiding negative feelings, unintended
consequences have actually produced the feared event. The person may then conclude that they are not
lovable because of some characteristic and may not recognize that it’s their safety behaviours that are
causing trouble.

The link between threats, safety strategies and unintended consequences can be complex. For example,
David grew up in an emotionally deprived background linked to the death of a young sibling. He grew up in
an atmosphere where “terrible things just happen”. His mother was frequently distressed and aggressive. It
took us some time to work out that he ran a strategy of “better never to hope of feeling positive about oneself
because it can all be dashed”. He could recall many times when he might feel positive about himself but then
the anger and criticism of his mother would give him “a dreadful heart-sinking feeling”, and a sense that
he’d been “completely wrong to feel positive”. He developed a life set of strategies of deliberately trying to
avoid stimulating his positive affect system and simply keep his head down and out of trouble. It took some
time for him to recognize that unless he stimulated his positive affect systems it would be near impossible to
get out of depression. He demonstrated a clear fear of positive emotion, positive feelings about the self, and
an inability to turn to others. In his work he was known as a very reliable person who would always “step
in”. He became aware, however, that this was not because he wanted to take responsibility but felt that if he
didn’t, he couldn’t trust other people to do a good job. He wasn’t frightened of being abandoned or rejected;
he was frightened that there was no one good enough to take responsibility, sort things out and it would “all
end in disaster”. You can imagine the serious transference issues this gave us!

Clearly, in this short book we can’t go into the details of these complex linkages, but if therapists focus with
a very clear Socratic method on guided discovery and look to the individual and unique patterns of threat
and self-protection in a person’s life, such themes will emerge. That’s why in CFT we encourage people not
to try to identify specific issues or core beliefs in advance as if ticking off on a checklist, but rather have
unique formulations based on complex functional analysis and reflective narratives—all around the three
circles (Point 6).

Although various therapists talk about helping clients to recognize that they have often been doing
“unhelpful things for good reasons”, it is important to have a very clear emotionally connected
understanding of the link between specific safety strategies and early life events. It is when people make
emotional connections that change can occur—and sometimes these connections can be very moving. In
other words, don’t “just say things”; ensure your client emotionally connects to it. Consider Susan who is
very angry with herself:
very angry with herself:

Paul: Susan, as I understand it, when your mother was taking drugs and alcohol, that was frightening for you
and when frightened you’d hide in your room, feeling very alone [pause and space]—but at the same time
you wanted to be closer to mum. Drifting into drugs yourself makes perfect sense because it seems to me
you were trying to find a way to connect with people, and feel better yourself—and of course get some
release from the lonely despair you had felt for so long [Pause and space]. Getting into drugs was not your
fault. It was the part of you that was trying to work it out, how to feel better and connect with people—to do
the best you could.

Susan: Yes but I should have known. I saw the mess drugs had made of her f***ing life. I was stupid—so
f***ing stupid.

Paul: [Very slowly and gently] Look at me Susan, this was not your fault, you felt so alone didn’t you. It was
not your fault. Had you been born into another family you would not have walked that path. [Goes quiet a
long time so I repeat slowly] It was not your fault.

Susan: [Susan is now tearful] Oh, I guess so, part of me didn’t want to go down that road yet another part
was desperate to do so; I just wanted to join in and connect with some people—anyone. I had been around
druggies all my life so they felt like my people in a way, people who would accept me and I was like them.

Susan both loved and was very angry with her mother, but to work on that we first moved a step to giving up
self-anger and towards how to change. In CFT, once we get hold of the anger and shame then we are much
more likely to be able to compassionately work on change. It will be a back and forth process though. In
shame it is important to emotionally connect with blaming and self-condemnation, and not be vague about
“good reasons” for problematic behaviour.

Formulation on those four basic aspects, can naturally lead to insight into how one’s symptoms have
emerged as natural consequences of early background and safety strategies. We are out of the domain of
pathologizing and labelling and into the domain of understanding and recognizing why much of what’s
happening inside the people we are working with “is not their fault”—how people have got unintentionally
trapped in their ways of thinking and behaviour (Salkovkis, 1996). This view, when clearly communicated,
along with the evolutionary model, can do much to eat away at the roots of shame and self-criticism.

Formulation and the circles

In addition, formulation can also be discussed in terms of the three circles (see Point 6). Here the client is
invited to think about each system/circle and reflect on how each works for them. How big are the circles
relative to each other? How does each work? What life factors may have influenced their development? How
do they regulate each other? What kinds of things would they put into them? Commonly clients feel that
their threat system seems to be bigger than their drive-pleasure or contentment/affiliative and soothing
systems. They may immediately see that they use achievements and striving to compensate for threats—or
that they are not open to kindness from others, and are more critical than kind to themselves. So, you can ask
people to draw each of their three circles in terms of how much they operate within them, how powerful they
are for them or how well developed they are. These pictorial representations can be useful for offering
insight, planning therapeutic interventions and independent practice. It also helps to explore and reflect
together that perhaps the threat and self-protection system has become overdeveloped for understandable
reasons.

Formulation is not one process

In CFT there are ongoing and different formulation processes that unfold in a series of steps. These can be
outlined as follows:

First formulation

• Presentation of current problems and symptoms.

• Validating and making sense of current difficulties.

• Establishing the therapeutic relationship noting potential difficulties

• Establishing the therapeutic relationship—noting potential difficulties.

Second formulation

• Exploring cultural and historical context.

• Narrating life history and story.

• Gaining insight into key emotional memories of self and other (self-other schema).

Third formulation

• Structured formulating in the context of the model’s four domains of: background; threats, fears, concerns
and unmet needs; internal and external safety and compensatory strategies; and unintended consequences.

• Identifying particularly problematic safety and affect-regulation strategies, such as avoidance, rumination,
substance use or self-harm.

Fourth formulation

• Explaining the model (evolved mind).

• Distinguishing between “not your fault” and “responsibility taking”.

• Explaining the “three circles”—and reformulating in those terms.

• Outlining the “brain diagram” (see page 146).

Fifth formulation

• Formulating the therapeutic tasks, e.g., thought monitoring and generating alternatives, behavioural
experiments or graded tasks, developing “the compassionate self”, compassionate imagery and letter writing.

• Exploring blocks and difficulties.

Sixth formulation

• Revisiting earlier formulations in light of progress on tasks and new information.

• Developing and adjusting therapeutic tasks.

• Future work together and then beyond therapy.

• Life practice.

• Preparing for ending.

These are obviously for general guidance only, not necessarily unfolding in this order in some linear
mechanistic manner, and there may be a number of other subformulations. So, although these are outlined as
a linear sequence, things may not be as neat as that. People may move through the different steps at different
speeds and go back and forth. The point is, formulation is an ongoing process that focuses on different
elements of the therapeutic journey (Eells, 2007). Keep in mind, too, that through all this process there will
be variations in the client’s ability to mentalize and stand back from and reflect on what’s going on in their
minds and in the therapy. These variations in mentalizing competencies will impact on your therapy. So
some assessment of these abilities is important for your formulation and treatment plan.

In addition, formulation will include and clarify “treatment goals” and outcomes. Understanding, working
and shaping together the client’s goals and objectives, including those linked to self-identity (e.g., to develop
one’s compassionate self), act as a key focus for intervention. This is different to only being symptom
focused.
10
Shame
Concerns with self-evaluation and self-feelings are central to many psychological therapies but few locate
their therapy within the shame literature and science of self-conscious emotion—about which we now know
a fair amount (Tracy, Robins, & Tangney, 2007). Understanding and working with the complexities of shame
plays a major role in CFT. The CFT model of shame (see Figure 5) is linked to the fact that humans have
evolved to want to create positive feelings about the self in the mind of others (see Gilbert, 2007c). It
suggests that:

1 We are all born with the need to connect to other minds and feel cared for. This blossoms into desires to
socially connect in one’s group; to find acceptance and social belonging to facilitate helpful relationships; to
be wanted, appreciated and valued (Gilbert, 1989; Hrdy, 2009). If we achieve this then our worlds are much
safer (and our threat systems settle) in contrast to not being valued or wanted, rejected or struggling alone.
Helpful relationships are physiologically regulating (Baumeister & Leary, 1995).

2 The way that we experience our intimate relationships—as either caring or neglectful and abusive; and our
peer relationships—as either caring and accepting or rejecting and abusive, has a major impact on how we
experience “ourselves” as living in the minds of others. To be vulnerable to external shame is to be sensitive
to negative feelings and thoughts about the self in the mind of others. So, working out how one exists for
others is fundamental to our feelings of safeness in the world. Hence, external shame is at the centre of this
model (see Figure 5).

3 There are two major defences (safety strategies) to external shame. One is the internalized shaming
response where one adopts a subordinate, submissive strategy associated with self-monitoring and self-
blaming. The other is an externalizing, humiliated response where one adopts a more dominant aggressive,
attacking response—one tries to create a sense of personal security via one’s ability to overpower or bully
potential attackers/rejecters. These are not consciously chosen strategies but reflect phenotypic variations,
and they can be context dependent.

4 Reflected shame is related to the shame others can bring on you by your association with them and shame
you bring to others. This can be especially important in some cultural

Figure 5 An evolutionary and biopsychosocial model for shame Adapted from Gilbert P (2002)
Figure 5 An evolutionary and biopsychosocial model for shame. Adapted from Gilbert, P. (2002)
“Evolutionary approaches to psychopathology and cognitive therapy”, in P.Gilbert (ed.) Special Edition:
Evolutionary Psychology and Cognitive Therapy, Cognitive Psychotherapy: An International Quarterly,
16:263–294

contexts, for example it is associated with honour killings (Gilbert, Gilbert, & Sanghera, 2004c).

Like other models, CFT distinguishes between fears and beliefs about the external social world (what others
think about and can do to the self) and internal fears and beliefs (e.g., fear of one’s own “inadequacies”,
failures or losing control to one’s own emotions, fantasies or thoughts). Although there is an obvious overlap
and interaction between “externally” focused threat and “internally” focused threat, CFT clarifies this
distinction with clients repeatedly because both the coping behaviours (protection-safety strategies) and
interventions vary as to the focus of the threat. So, it is useful to clearly separate out the two types of
thinking. For example, the break up of a romantic relationship may result in much grief but when shame is
involved there are two other streams of thoughts. Here is an example of Tim:

External shame Internal shame

Attention and reasoning focused on what is Attention and reasoning focused inwardly on one’s
going on in the minds of others about the self experience of self as subject.
as object.

This relationship broke up because Sally went I often feel anxious and uncertain what to say or do. I
off me. become confused and vulnerable. I worry that my anxiety
will stop me doing what I want to do.

She sees me as boring and anxious. She was This is rather pathetic. I am pathetic.
nice but she was probably thinking for some
time of how to end it.

My key fear is: My key fear is:

That in the minds of others I exist as a boring, Being unable to change or cope with anxiety and being
wallpaper person; not able to elicit love and stuck as an undesirable self. Stuck with feelings of being
affection—destined to be alone. alone and miserable; an unwanted nobody.

(Reactivation of childhood memory of (Reactivation of childhood memory of inner experience of

experience of others.) self.)

There is usually a link between external and internal shame and when written down separately people can
often see this—i.e., that the way they think others think about them is often how they think/feel about
themselves. One can explain that this is linked to projection and how, when we feel threatened, projection is
more likely, because of “better safe than sorry” thinking (Gilbert, 1998), creating a vicious circle.

Note also that we focus on key fears/threats. We use the term “key fears/threats” rather than, say, “bottom
line” (which is sometimes used in CBT) because we want to use language that links directly to the threat
system. Also, we want to link this to basic, evolved and archetypal fears of rejection and fears of “being
unable to influence the minds of others in one’s favour”. We can then explore a functional analysis of the
self-critical inner dialogue because, as related to threat, the external focus will have a different function to
the internal focus. The functions of thoughts related to external threat are often about warnings and
explanations, e.g., “People won’t like you because…; you are vulnerable to…; you are not making good
impressions in the minds of others…; if you don’t change/stop/control…then…; you must stop being
miserable because people won’t like you ” These are common involuntary subordinate concerns
miserable because people won t like you… . These are common, involuntary subordinate concerns.

Note that the protection-safety behaviours can be similar or quite different for external and internal shame.
To control his anxiety a man might drink alcohol before he goes to see his girlfriend. To impress her,
however, he might buy a car that he cannot really afford. Thus, the way we cope with fears emanating from
the inner and outer world can be different.

There are two different types of shamed-linked trauma that can activate the threat system. The first is the
obvious one of abuse. These are traumas of intrusion where another person (or people) has violated the
control and boundaries of the person and been highly threatening/hurtful. The other traumas are those of
being too distant from others, commonly where others have been hard to please or elicit affection from and
there has been an absence of affection; the shame of not being good enough to be selected or wanted (see
Dugnan, Trower, & Gilbert, 2002, for concepts and a scale to measure them).

Shame of exclusion
The shame here is related to feeling one is rarely noticed or wanted—it is not so much active rejection as
passive ignoring. One client said, “I loved mum, but she worked hard and there were always more important
things for her to do. I guess I just wasn’t important enough for her”. These folk can come to constantly feel
not “interesting or attractive or good enough”, can become strivers in an effort to find ways to feel connected
to others—but their successes rarely satisfy those searches (see Point 14). The example of Tim above could
be linked to an exclusion type of shame and anxiety.

Intrusion and violation shame

For those who suffer the intrusions of others, they can feel powerless to stop or defend against “the other”
doing things to them and are rendered small, powerless and frightened. One can experience the self as an
object “to be used” by the other. Verbal abuse and shaming is injecting negative meanings/labels into
experiences of the self. In evolutionary terms such indoctrinations are regarded as memes (basic ideas and
beliefs) that act like viruses/infections, replicating themselves inside the person and indeed in relationships
(Blackmore, 1996). There is some evidence that verbal abuse and having one “self” defined by others in
negative terms can be as powerful and pathogenic as physical and sexual abuse (Teicher, Samson, Polcari, &
McGreenery, 2006). Indeed, one sometimes hears stories like, “Well I could cope with the beatings if I knew
mum/dad loved me—it was when they called me stupid and ‘a useless bastard’ and I felt they did not like, let
alone love me, that really dug deep into me. If your own flesh and blood don’t love or want you then I guess
you are pretty useless”. Peer shaming can also have a significant impact on one’s sense of self as a social
agent, and vulnerability to shame and self-criticism (Gibb, Abramson, & Alloy, 2004).

Shame memory
There is increasing evidence that shame memories can act like trauma ones, involving intrusiveness,
hyperarousal, efforts to avoid shame feeling (Matos & Pinto-Gouveia, in press), and, of course, shame can
have a major impact on our sense of ourselves and who and how we engage socially (Gilbert, 2007c). To
explore the complexities of shame and trauma memory we can look at a case example.

Case study

Sara came from a poor family, whose mother had “a number of partners” and suffered from alcohol abuse.
Her mother was unpredictable and often verbally and physically aggressive. Sara had many trauma memories
but one was from about seven or eight. She recalls feeling relatively happy, playing with her friends in the
corridor of the house, making “giggly” noises. Her mother was drunk and “came flying out of the room”
where she had been sleeping and hit Sara about the head making her nose bleed and cut her lip. She also
screamed at Sara that she was a bad selfish girl for “waking mother up with her stupid, f***ing games”. Her
friends were alarmed and fled the scene immediately leaving Sara completely alone, overwhelmed with fear,
terror and sadness, and the pain of being hit. She recalls her body “crouched and shaking”.

The conditioning implications of this are clear. Tomkins (1987) pointed out that shame memories are scenes
in our minds, an interlinked set of body-based feelings and events that are sources of emotional conditioning.
It is vital to engage with this complexity.
In Sara’s, emotional memory will be:

1 the inner cues of having fun (inner positive affect);

2 associated with an intense raging (external) attack;

3 associated with self-defining verbal labels (stupid, selfish);

4 associated with friends running away and being left completely alone;

5 associated with pain and shock at being hit;

6 associated with her own defence system automatically creating in her body patterns of submissive, fearful/
terrorized, crouching and shaking.

After the attack Sara is sent to her room to be alone. At the very time that a child needs comfort and care
(when they are very distressed and frightened) Sara has been isolated. It is no wonder that she is
overwhelmed by feelings of aloneness when she becomes distressed because of conditioned emotional
memories. In addition, there are feelings of entrapment because there is no way she can get away from her
mother—go and live

Figure 6 Associations of threat “meanings” in shame-traumas

somewhere else. And even those desires might be internally alarming for her. In CFT we try to capture these
complexities of emotional memory.

An example of these interlocking processes is offered in Figure 6.

This diagram is simply to help you work with your clients so that you can see the awareness of emotions in
“the other”, what arises in the self including verbal labels and fears, and the experience in these cases of
feeling trapped and very alone. Not only did no one come to rescue your client from the abuse (and this may
include sexual abuse), but afterwards she/he was left alone with no soothing from the outside.

As an aside, I grew up in Africa in the outback and these things wouldn’t happen. If the parent was angry
with the child or even hit them, the child would run off to a grandmother or an aunt—they wouldn’t be left
alone in a room! (see also Hrdy, 2009). One can also see how and why Sara may become very attentive to
certain feelings in herself (e.g., fun) that will automatically trigger a fear of punishment and a sense of doing
something wrong, a sense of being alone in a frightening world and wanting to curl up and hide. Classical
conditioning is the model CFT refers to here
conditioning is the model CFT refers to here.

In CFT the experience of aloneness, being cut off from a source of affiliative soothing when distressed, is a
central experience to work with. You might also draw out another set of circles looking at what could not be
processed in that (frightened) brain state, e.g., rage at mother, or a desire to heal mother or abandon her, or
have her locked away, and the possible fears of acknowledging these possibilities or sense of betrayal
(Gilbert & Irons, 2005). Indeed, Sara wanted her mother arrested by the police, but felt very frightened of
acknowledging these feelings because at times her mother could be loving. Hence, she was a feared and
wanted object—which really scrambles the attachment system—wanting to be soothed by someone who has
threatened you. These issues, and their relation to disorganized attachment and mentalizing difficulties, have
been explored by Liotti and Prunetti (2010).

CFT might also tap grief for the mother/parent one so dearly wanted—the personal archetypal mother/parent.
This is not commonly considered in CBT therapies but can be extremely important—even though in one
sense one is grieving for a fantasy (Gilbert & Irons, 2005). Indeed, the abilities to grieve and work through
complex grief-linked emotions might aid the development of mentalizing.

Hence, rather than tackle this at the level of a belief (in, say, inferiority, vulnerability or defectiveness) or
automatic thoughts, one would unpack these experiences coded in the threat system; break them down into
their core components. These are fear of the other, arousal of inner fear, a sense of aloneness, and verbal
labels in the experience—of being stupid. After clarifying the different components in this memory (that
might fuel a core belief), one can then: (1) explore how Sara’s brain will (for self-protective reasons)
automatically try to develop defensive strategies; and (2) work out a compassionate intervention for each
aspect.

The many influences of shame

Shame can be focused on various aspects of self, such as one’s body, feelings, fantasises and desires, past
behaviours, and personal characteristics. Shame can also play a major role in:

1 the acquisition of vulnerability to emotional distress;

2 the development of sense of self;

3 coping behaviours and safety strategies;

4 lack of openness to others, unable to identify with others (the only one), avoid help seeking—including
whether to come for therapy or not;

5 coping with in-session feelings and process (e.g., being overwhelmed with tears, losing control, or
revealing abuse) and whether to drop out or not;

6 how they react in therapy overall, including shaming that might be happening at home as clients try to
change; and

7 what they reveal or keep hidden.

Anyone working with complex cases would do well to understand the many complexities of shame. Another
complexity is self-criticism.
11
Self-criticism
Self-criticism is very common in shame. Self-criticism has been associated with a range of mental-health
difficulties and in psychosis about 70% of voices are hostile and critical (Gilbert & Irons, 2005). Self-
criticism has various origins and functions. Figure 7 offers a brief model of self-criticism and areas that
clinicians can explore. To explore you can ask: “When was the first time you became aware of being self-
critical? What was happening? What was the frustration, disappointment or dashed hope/dream/goal? Why
was that hope/dream/goal of importance to you?” In other words, you are looking at the threats that were
around when self-criticism started.

Figure 7 Self-criticism: a self-critical mind is also a threat-focused mind

Self-criticism can typically kick in when involved in situations linked to the original threat. For example,
John’s teacher could be very contemptuous and shaming of his poor English and told him he was lazy and
would struggle in life because he was not very bright—only hard work would get him anywhere. John would
feel “terrible heart sink” at these accusations. To avoid such shame feelings John did work harder. Later in
life he became something of a workaholic to prove that he was “up to it” and to avoid those “isolating, heart
sink” and “being no-good” feelings. When he had to write reports, if people were mildly critical of them this
would reactivate the shame memories, of anger, self-criticism and feelings of being alone (recall that shame
usually comes with feelings of being alone). He would then be highly self-critical. This was partly linked to
hearing the voice of the teacher and partly out of panic, “Oh God, now what have I done!” So it’s useful to
link early threats with the origins and memories of self-criticism and possible triggers. John also believed
that he needed to stay self-critical to keep himself working hard and not be a failure.

In our own work we found that there are different forms and functions of self-criticism (Gilbert, Clarke,
Kempel, Miles, & Irons, 2004a). Attention to form is important. Whelton and Greenberg (2005) showed that
it was not the cognitive content of self-criticism as much as the emotions of anger and contempt that were
important in the pathogenic effects of self-criticism. One form we identified focuses on feeling inadequate.
This type of self-criticism is usually about disappointment and feeling inferior. However, there is another
form that is linked to hatred of the self. These are quite different and should be distinguished in your therapy.
Self-hating tends to be low in nonclinical populations but high(er) in populations who come from difficult
backgrounds. We suspect that self-hating is linked to a more abusive past but have no clear evidence as yet
(see Andrews, 1998).

Self-hating can be directed at parts of the self such as one’s body (“I hate my fat shape” or “I hate these
feelings in me). So, when exploring self-criticism ask, “How does the critical part of you typically feel about
you; what emotions does it direct at you?” I have to say that some current scales that supposedly measure
self-criticism don’t do so; they are more likely to measure things like social comparison or low self-esteem.
One needs to think of self-criticism very much in terms of critical comments, dialogues and feelings within
the self.
People can hate things about themselves but not necessarily blame themselves or feel responsible, e.g., for
their looks or a birth mark or disfigurement, or lacking talent. So, issues of causality, blaming and disliking
need to be distinguished. Note also that when people dislike things about themselves they can also be
envious of others (Gilbert, 1992, pp. 246–252).

It can sometimes be useful to ask people to imagine their self-criticism. Ask, “If you could take your self-
critic out of your head and look at it, what does it appear as (e.g., human or nonhuman)? What facial
expressions does it have (assuming it has a face)? What emotions is it directing at you? What is its greatest
fear/threat?” To understand the critic you can engage in chair work, putting the critic in a specific chair and
exploring its thoughts and feelings. You can then move the person to a facing chair and explore ways of
engaging with the critic. This Gestalt therapeutic intervention was further developed and popularized by
Leslie Greenberg (e.g., see Whelton & Greenberg, 2005). When working with the self-critic you can develop
a compassionate self first and then teach compassion for the critic (see Point 22). It is engaging with the
narratives, meanings and emotions of self-criticism, not just the thoughts, that is central. For example, if a
person imagines their self-critic as an angry, dominant person, you can ask them what would happen if they
went to their inner critic (angry-dominant) with alternative and more rational thoughts. The client quickly
gets insight that such would be dismissed and why cognitive interventions can run aground. This is why it’s
important to build up a part of the self that feels strong enough, and containing enough, to be able to work
with a self-critical side (see Point 21).

Functional analysis
Functional analysis of self-criticism is very important. Sometimes there is no particular function—people
just get angry with themselves when they make mistakes (e.g., dropping the ball in a ball game or forgetting
something they had to do). However, you can look at how self-criticism guides a person by asking, “What
would be your greatest fear in giving up self-criticism?” Typically you will find people fear becoming
arrogant, lazy or uncontrolled. So, people may see self-criticism as having a range of functions, like making
sure they pay attention to errors and to prevent them making errors in the future, to keep them on their toes.
Self-criticism can act as a warning (e.g., if you don’t lose weight nobody will like you). Therefore we teach
how compassion offers a different way (is rooted in a different emotional system) for self-correction and
improvement (see Point 13).

The German philosopher Friedrich Nietzsche said, “no one blames themselves without a secret wish for
vengeance”. Freud borrowed this for his theory of depression and thought that anger at oneself is really
internalized anger towards others one is dependent on or ambivalent about: you are frightened of being angry
with them so you take it out on yourself (Ellenberger, 1970). There is evidence that some people do
experience rage but are fearful of it and inhibit their anger (Gilbert, Gilbert, & Irons, 2004b). So it is always
worth exploring this possibility—to see if serious self-criticism and self-hatred are linked to unprocessed and
feared hostile feelings towards others. The moment someone tells me “I’m not an angry person” I get
suspicious. Occasionally it is true, but commonly this is a pointer to unprocessed and feared emotion.
Commonly, people like to see themselves as “nice caring people” where anger is not part of that identity. We
all have a capacity of anger—it’s how we recognize it and work with it that is the issue. I have come across
many chronically depressed people who, when they begin to work with and acknowledge their rage, without
being ashamed and frightened of it, recover. If rage is experienced as frightening or shameful and not
acknowledged as part of a repertoire of possible feelings, people can stay ashamed and frightened of it—
thus maintaining a sense of impotent rage and powerlessness, and shame of feelings.

Self-monitoring and self-blame as safety strategies

CFT sees self-criticism in terms of safety strategies, with complex forms and functions that require
exploring. One of these is power. As I’ve noted a number of times (e.g., Gilbert, 2007a, 2009a; Gilbert &
Irons, 2005), people in religious contexts will blame themselves rather than their gods for setbacks and
misfortunes. “What did we do to upset you, God of the Land, that you sent the famines?” History is full of
self-deprecation and efforts to appease and sacrifice to powerful gods because we are frightened of them.
Indeed, the concept of sin itself is a blaming process.

The link between self-criticism and powerful others, which Freud touched on, is particularly true in the case
of self-criticism that arises in the context of abuse or trauma. This is because trauma, when perpetrated by
powerful others, can automatically turn the victim to self-monitoring and self-regulating. So, we can also
l th dd f f th lf t ti t d h th t t t l d ith lf
explore the responses and defences of the self-protection system and how that gets entangled with self-
criticism. Consider that the first task of a child is to keep safe and stay out of trouble. If a parent’s or bully’s
behaviour is unpredictable, it is like creeping around a sleeping tiger; you might be cross with yourself if you
step on a twig that makes a sound that could wake the tiger to attack you. So, one must attend to one’s own
behaviour and self-monitor very carefully not to stir up the (tiger) bully and keep out of his/her way. This is
how the self-monitoring and self-blaming system gets linked to aversive outcomes. Because one’s own
behaviour is
the only possible source of control/protection (you can’t change the bully), self-blame inevitably follows
from doing things that (seem to) trigger the bully’s aggression/rejection.

CFT spends time with clients explaining these classical conditioning models, and the functional value of
self-monitoring and self-blame. The more of a framework clients have for understanding their self-criticism
—as linked to safety strategies, the more reflective they can be and the more collaborative in engaging with
these memories and developing self-compassion. Moreover, it helps people to understand why they “feel”
(from the amygdala) to blame even through they logically (cortex) know they are not. There is a mismatch
between two different systems. The amygdala does not listen to logic very much. I have found clients feel
de-shamed by this explanation, e.g., “well it makes perfect sense because your amygdala and threat system
cares only about keeping you safe and therefore it will self-monitor and self-blame as ways to do that
(sleeping tiger example). So your feelings are about safety strategies not about the person you actually are”.

In working with these memories we would first engage in developing the compassionate self that is able to
be “wise, kind and parent-like” (see Point 21). When the client feels this part of them has been developed we
might then explore re-engaging the memory which (say for Sara, see pages 88–92) might involve crouching
and then gradually standing, focusing on the “then and now” differences; re-scripting, assertive work with
“mother” in imagery or with an empty chair, or expressing her feelings in a letter (that may not be sent); or
explore rage expression to desensitize to the “fear of rage”. In other words, we are acting against the
submissive-fearful strategy, and engaging in a more dominant self-protective strategy. Compassion or
forgiveness to mother would come much later.

Note, too, that you can discuss with your client what the key elements are that they want to talk about and
work on. For example, it might be to work with a sense of powerlessness and so sessions of assertive body-
work could be helpful. Or it might be a sense of aloneness and therefore focusing on experiencing the
therapist in that moment might be helpful; or it might be a sense of entrapment so the person may practice
being assertive and then leaving the room; or it might be a sense of betrayal as in wanting to “get rid of the
parent or break contact with the parent”. One needs to break down the key issues and work on them
individually. As with other therapies various forms of imagery re-scripting can be helpful (Lee, 2005;
Wheatley et al., 2007).

The client collaborates to see what feels helpful and/or soothing and what does not. Because the client
understands the three circles model then, like Yoga or physiotherapy, the client and therapist work together to
try different things to bring the soothing system on line. The “bringing forth” of compassion is the explicit
goal of the therapy and the client understands this, and of course one never goes further or more intensely
than the person feels okay with.

Another key element of trauma and high levels of stress is that they can set up conflicting defences. For
example, to a shame event one might feel angry, want to run away, and sad all at the same time. Because
multiple emotions and defensive action tendencies can be aroused to the same event, some people become
highly disorganized in affect regulation. Sara, for example (see pages 88–92, could have a lot of anger at
herself and could be self-harming to try to manage her anger. She often suffered from a loss of drive.
Feelings of self-soothing and connectedness to others were almost nonexistent. At times she would go into
high arrested anger states, sitting in the chair in therapy hardly able to speak but the room would feel like a
volcano of anger. Inside her, however, she would just ruminate on escape, “Why don’t I just kill myself and
get it over with—what am I doing here?” In the system I have outlined, her threat system was bouncing from
arrested anger to arrested flight (Gilbert, 2007a; Gilbert et al., 2004b). The amygdala can generate a number
of conflicting defences—rage attacks vs. fearful escape vs. submissive withdrawal vs. tearful care-eliciting.
To act on one, the others are suppressed (see Dixon, 1998, for an animal model of conflicts of defences).
12
Distinguishing shame, guilt and humiliation: Responsibility vs. self-
critical blaming
It is important to clarify the distinctions between different types of self-conscious emotion (Tracy et al.,
2007). There are three major forms that are especially important: shame, guilt and humiliation. Shame and
humiliation are very much to do with self and defending the self (see Figure 5) whereas guilt is about one’s
behaviour, awareness of harm to the other, and motivated desires to repair harm (Gilbert, 2007c, 2009a,
2009b; Tangney & Dearing, 2002). As a rough guide, keep in mind that shame, humiliation and guilt will
each have a different attention focus, a different way of thinking and different behaviours. As a rough guide
see Table 1.

Generally speaking, while shamelessness is not desirable, and like other emotions we need to learn how to
face it and tolerate it, evidence suggests that shame and humiliation feelings (which are about feeling
attacked or vulnerable in some way) are often unhelpful and can lead to destructive, defensive behaviours
(Tangney & Dearing, 2002). In contrast, guilt is an emotion of behavioural responsibility, including
responsibility to make amends. Consider two men who have had affairs and their wives discover them. A
shame-focused response would be thinking about how bad one was and that other people would turn against
him, and that he needs to make amends so that other people will like him again. Shame is (bad) “me, me,
me”. A humiliation response would be to become angry and (say) blame his wife for not being attractive
enough such that he needed to have affairs, or be angry with the person who revealed his secret. Other
responses might be denial or minimization and so forth. A

Table 1 Rule of thumb distinctions between shame, humiliation and guilt

Internal shame rank mentality Humiliation rank mentality Guilt care mentality

Inwardly directed attention on Externally directed attention Externally directed

damage to self and reputation is the threat or damage done attention on hurt caused
to the self by the other with empathy for the other

Feelings of anxiety paralysis, Feelings of anger, injustice, Feelings of

heart-sinking, confusion, vengeance sorrow,
emptiness, self-directed anger sadness and
remorse

Thoughts focused on negative Thoughts focused on Thoughts focused on the

judgements of the “whole self” unfairness of any negative “harm to the other”,
judgements or behaviours of sympathy and empathy
others

Behaviours focused on submissive- Behaviours focused on Behaviours focused on

appeasing, wanting to be liked vengeance and silencing the trying to repair harm,
again, or closing down and moving other— having power over offering genuine apologies,
away, avoidance, displacement, the other, belittling and making amends for the
denial, self-harm humiliating back benefit of others

guilt response, however, would be a recognition of the betrayal, the harm done, genuine concern for the upset
he’s caused his wife, with a genuine desire to help his wife feel better and repair the damage.

Sometimes if people have been unkind or harmful to others it is important to change their shame (bad me)
based experiences to guilt (harm awareness) based ones, that is to help them tolerate aversive feelings when
they do hurtful things, without it becoming (just) an attack on the self. Sometimes, to be able to process the
t ey do u t u t gs, w t out t beco g (just) a attac o t e se . So et es, to be ab e to p ocess t e
harm they do to others, they must first process the harm that was done to them.

There are also disorders that look like problems of over-responsibility, as occurs in obsessional compulsive
disorder (OCD) or excessive caring (Wroe & Salkovskis, 2000). It is unclear, however, how much OCD
people feel personally bad and are focused on “a bad self” rather than genuine concern for “the other” but it
does seem as if some experience “guilt attacks”, which are not so different from panic attacks.

So, as a rule of thumb you can look at where people are allocating their attention, what they are focused on
in their thoughts and ruminations, and what they want to do. Keep in mind these are not mutually exclusive,
and various blends of each can ebb and flow in therapy and in general.

It is important, then, that people learn to understand guilt, not turn harmful acts into self-focused shame or
just become angry if somebody stimulates this in them. Guilt is helpful to building relationships in a way that
shame is not (Baumeister, Stillwell, & Heatherton, 1994). Guilt is something we must think about and
tolerate. While there is much discussion on tolerating sadness, anxiety and anger there is very little research
on learning to tolerate guilt and how to behave appropriately— but this is important in CFT. Guilt can be
normal and natural and is to be expected because no one can go through life without being hurtful, damaging
and plain unkind.
13
Distinguishing compassionate self-correction from shame-based self-
attacking
When working with self-shame and self-criticism (and shame vs. guilt) it is useful to distinguish two
different emotion systems underpinning self-correction. For example, if you ask people to completely give
up self-criticism they often worry that they will become arrogant or lazy or uncontrolled. This is true, even of
therapists in a workshop! However, as noted above, if I ask you to imagine the part of you that is critical, to
see it looking back at you as it were, and what emotions it is directing at you, the chances are these emotions
will be ones of frustration, anger, contempt or disappointment. These are all emotions of the threat system.
Therefore, our self-criticism of this type stimulates the threat system. It is not focused on our well-being.

The key, then, is to align with a different motivational system for orientating us towards self-correction. This
comes when we realise that we would actually like to do a good job, e.g., to treat our clients well and see
them improve. In other words, we are positively motivated to be at our best once we learn to stimulate that
motivational system. So, CFT builds on other cognitive and positive psychology principles of focusing on
our strengths. This approach has become increasingly important in the cognitive-behavioural
psychotherapies (Synder & Ingram, 2006). In this spirit we focus on the three circles model, helping people
to contrast the distinction between compassionate self-improvement and shame/fear-based self-criticism.
This is outlined in Table 2.

Again, this is fairlystraight forward. Imagine a child who is struggling at school with one of two teachers.
The first is very

Table 2 Distinguishing compassionate self-correction from shame-based self-criticism and attacking

Compassionate self-correction: Shame-based self-attacking:

• Focuses on the desire to improve • Focuses on the desire to condemn and

punish

• Focuses on growth and enhancement • Focuses on punishing past errors

• Is forward-looking • Is often backward

looking

• Is given with encouragement, support • Is given with anger, frustration,

and kindness contempt, disappointment

• Builds on positives (e.g., seeing what • Focuses on deficits and fear of

one did well and then considering exposure
learning points)

• Focuses on attributes and specific • Focuses on a global sense of self

qualities of self

• Focuses on and hopes for success • Focuses on high fear of failure

• Increases chances of engaging • Increases chances of avoidance and

ithd l
 withdrawal

For transgression/mistakes: For transgression/mistakes:

• Guilt, engage with • Shame, avoidance, fear

• Sorrow, remorse • Heart-sink, lowered mood

• Reparation • Aggression

• Use the example of the encouraging • Use the example of the

supportive teacher with the child who is critical teacher with the
struggling child who is struggling

Adapted from Gilbert, P. (2009a) The Compassionate Mind. London: Constable

& Robinson and Oaklands, CA: New Harbinger.

good at spotting errors and is quite critical and maybe slightly frustrated with the child’s mistakes. He or she
believes that it is necessary to keep children on their toes otherwise, without a degree of fear, they will be
lazy. In contrast, the second teacher has concerns for the well-being of the child, and is able to encourage the
child to accept and learn from their mistakes in a compassionate way. Which teacher would you prefer to
teach your child? In his passionately written book How Children Fail John Holt (1990) argued that much of
Western education teaches us to be frightened of our mistakes. Therapists may thus be working hard against
the social grain.

Another way that you can help people recognize the harm of self-criticism (recall that you can elicit images
of people’s own self-criticism) is to explore the origins of self-criticism. Commonly, it will be the voice of a
parent or teacher and you can ask, “Did those individuals criticize you out of a genuine care for your well-
being? Did you feel care for you when criticized (or more likely anger)? Did they have your best interests at
heart? Does your self-critical side have your best interests at heart or is it just rather angry and
disappointed?” You might also ask, “If you have your best interests at heart how would you really like to talk
to yourself? How would you help somebody you really cared about see this issue? How would you help a
child?” In these examples you are helping people recognize that they probably would not be harshly critical
to others, that self-criticism doesn’t really serve their best interests, whereas compassionate self-correction
does. Of course, they’ll think critical things of other people at times, but it is recognizing that they still
would not voice them because they see that would be harmful not helpful. So, it’s not the correction and the
awareness of errors that is the key but it’s the emotions that we bring to the process of trying to improve.

Working with self-criticism

Because self-criticism is a complex process, often associated with self-identity and safety
behaviours/strategies, we do not try to dismantle it head on. Rather, we are likely to say to clients, “Feel free
to maintain your self-criticism if you find it helpful. However, we are going to teach new ways to think about
and treat yourself including compassionate self-correction. You may find, as you practice these, that they
serve you better than self-criticism.” Trying to work with self-criticism simply by looking at the evidence for
and against rarely works in high-self-criticism and shame clients. Once clients don’t feel under any threat to
“change or get rid of safety strategies” until they have new strategies, they feel more confident at trying
things.
14
Threat and the compensations of achievement
While shame can typically lead to the safety strategies of self-criticism, avoidance, closing down, hiding and
various unhelpful ways of trying to regulate emotions, shame can also lead to invigorated drive behaviours in
the form of achievement seeking—linked to the “musts” and “have tos”—as in Rational Emotive Behaviour
Therapy (REBT; Dryden, 2009). Alfred Adler (1870–1937) argued that people who feel inferior (have an
inferiority complex) may strive to compensate and prove themselves to others—a view now well accepted by
most psychotherapies. CFT therefore links to the research on compensation and achievement. Some years
ago, McClelland, Atkinson, Clark and Lowell (1953) made a distinction in motivation theory between value
achievers and need achievers. Value achievers set their achievements to bring pleasures and stretch
themselves, whereas need achievers set their standards to try to impress others. These themes have been
taken up by other researchers. For example, Dykman (1998) suggested that there are two main motivations
behind achievement, which he called growth seeking versus validation seeking. Growth seekers enjoy
challenges and their ability to learn and mature through challenges/mistakes. Validation seekers, however,
feel under constant pressure to prove themselves as likeable and acceptable to others. He also suggested that
validation seeking is a defensive coping strategy that develops in the context of critical and perfectionist
parenting. In a series of studies Dunkley and colleagues (e.g., Dunkley, Zuroff, & Blankstein, 2006) explored
various measures of perfectionism and suggested two underlying factors: the first is setting and striving for
personal standards; the second is striving to avoid criticism/rejection from others—labelled “evaluative
concerns”. Dunkley et al. (2006) found that it is the evaluative concerns dimension that is linked to various
psychopathological indicators. Our research has also shown that insecure striving, to avoid inferiority (which
is different from seeking superiority) can be distinguished from secure non-striving. Insecure striving is to
avoid the social consequences of rejection, exclusion and shame (Gilbert et al., 2007). Secure non-strivers
think they are accepted whether they succeed or fail.

Goss and Gilbert (2002) suggested that this was particularly true of people with eating disorders. Anorexic
people focus on feelings of pride in their control of their weight and impulses to eat. When they lose control
of their weight or impulses to eat this activates high threat and alarm. For these individuals it is the drive
system that they are using to try to regulate the threat system. Figure 8, which Ken Goss and I developed to
depict eating-disorder problems, offers a simplistic view of this.

Figure 8 Types of affect regulator systems (developed with Ken Goss, 2009)

We can leave out the arrows linking the contentment and soothing system because we don’t know how they
actually work in these cases. This is a model that can be used for a whole range of individuals who have
become competitive and striving in order to stave off feelings of threat or to feel connected. This is an
increasing problem in our Western society (Gilbert, 2009a; Pani, 2000) and may be one of the reasons why
we have growing problems with adolescent depression, anxiety, drug taking and self-harm (Twenge et al.,
 g gp p , y, g g ( g ,
2010). When you talk to these adolescents the sense of disconnectedness (especially from adult society),
inferiority, aloneness and the struggle they have to really feel valued by others is sad. Schools, businesses
and Western governments over the last 20 years have deliberately infected our societies with beliefs that only
the competitive and able can make good—we must all have a “competitive edge” and prove ourselves in the
market. It is problematic because it affects our brains (Pani, 2000) and creates high levels of inequality that
are known to be pathogenic and unhelpful to our well-being (Wilkinson & Pickett, 2009).

If you draw out the three circles for your clients, and explain this to them, they very quickly see what’s
happening and why. As to whether they will then form a contract with you, to try to balance their systems by
becoming more compassionate and soothing, that is another matter. They will often hold strong beliefs that
only if they are successful will people like them— one client told me that she had been told, “Those who
come second are the first losers”, and “No one remembers who came second”. So, it is useful to distinguish
value-based achievement striving from threat-based attachment striving—the latter being strongly linked to
shame proneness and fears of social exclusion and rejection. There can also be a fear of positive emotions
associated with contentment and affiliation (see Point 29), and striving can be built into the sense of self-
identity. You can use behavioural experiments, planning non-achievement fun things, playfulness and
desensitization to positive effect, and help people think about their striving in different ways.

What emerges from this research on achievement motivation is that problems in people’s ability to feel
connected to others, able to rely on them and feel safe with them, generate desires for them to try to “earn
their place”. In doing this they become over reliant on the incentive-seeking dopaminergic system to give
them a sense of positive feelings and self-security. Indeed, people who become perfectionists, high striving
or feel they need to prove themselves will often only get temporary relief if they do succeed. The other thing
that strikes you when working with some perfectionist strivers, and those desperately wanting to prove
themselves, is that they will often talk about having this feeling of “being alone”—not really connected to
others, not feeling a sense of belonging. They may also struggle to feel happy and contented states of
intimacy and closeness to others. These are feelings that we discussed above in relationship to shame and it
is important to explore them because they may become a focus for therapy, linked to the problem in
activating the connectedness and soothing systems.

Self-focused vs. compassionate goals

Crocker and Canevello (2008) explored two types of self-image goals. One was compassionate, wanting to
help others. Concerns with altruistic goals, with wanting to help others, have been observed in very young
children (Warneken & Tomasello, 2009). The other goal was self-focused, wanting to achieve and create
good impressions on others. We see here similarities with value versus need achievers, growth versus
validation seekers, and the motives of getting along versus getting ahead. However, importantly, compassion
goals were associated with feelings of closeness, connectedness and social support and inversely related to
conflict, loneliness and being afraid and confused by feelings. In contrast, self-focused goals showed the
reverse relationships. Reed and Aquino (2003) suggested that caring, kindness and honesty attributes can
become important for a self-identity, which they call a moral identity. Wanting to be a kind and
compassionate person (i.e., harness the care-giving mentality for self and social role co-creation) contributes
to more benevolent behaviours and values—especially to outgroups. So, self-identity imagery goals, the kind
of self one wants to be, tries to be, and practises to be, have important effects on social behaviour and well-
being. This research should be kept in mind when we look at developing the compassionate self (Points 21–
29).

Doing and achieving

It is incorrect to see CFT as only concerned with one (soothing) system. The main focus of CFT is a
balancing of the three affect systems not just stimulating the soothing system. Sometimes it is important to
work on the drive and achieving system as well, by shifting threat-based achieving to value achieving.
“Doing” is very important in behavioural activation approaches to depression (Gilbert, 2009b; Martell,
Addis, & Jacobson, 2001). Nor is CFT the end of ambition. The Dalai Lama travels the world trying to
promote compassion; in ACT, committing one-self to the values and goals is key to personal growth and
development.

Compassionate behaviour can also involve providing people with things that are important for their
flourishing in life. For example, at Christmas time, providing a child with a long-wanted bicycle so that he or
she can be like his or her friends could be an act of compassion If we have a self-deprivation psychology
she can be like his or her friends could be an act of compassion. If we have a self-deprivation psychology
then learning how to enjoy and provide for ourselves can be important. Learning to take joy from our and
other people’s achievements is also important for balancing our emotions. However, as in other therapies,
there is a distinction between taking joy in having and “feeling a need and must have” (Dryden, 2009).