Dialogical Meetings in Social Networks

DIALOGICAL MEETINGS
IN SOCIAL NETWORKS
Other titles in the
Systemic Thinking and Practice Series
edited by David Campbell & Ros Draper
published and distributed by Karnac
Asen, E., Neil Dawson, N., & McHugh, B. Multiple Family Therapy: The Marlbor-
ough Model and Its Wider Applications
Baum, S., & Lynggaard, H. (Eds.) Intellectual Disabilities: A Systemic Approach
Bentovim, A. Trauma-Organized Systems. Systemic Understanding of Family
Violence: Physical and Sexual Abuse
Boscolo, L., & Bertrando, P. Systemic Therapy with Individuals
Burck, C., & Daniel, G. Gender and Family Therapy
Campbell, D., Draper, R., & Huffington, C. Second Thoughts on the Theory
and Practice of the Milan Approach to Family Therapy
Campbell, D., Draper, R., & Huffington, C. Teaching Systemic Thinking
Campbell, D., & Grønbæk, M. Taking Positions in the Organization
Campbell, D., & Mason, B. (Eds.) Perspectives on Supervision
Cecchin, G., Lane, G., & Ray, W. A. The Cybernetics of Prejudices in the Practice of
Psychotherapy
Cecchin, G., Lane, G., & Ray, W. A. Irreverence: A Strategy for Therapists’ Survival
Dallos, R. Interacting Stories: Narratives, Family Beliefs, and Therapy
Draper, R., Gower, M., & Huffington, C. Teaching Family Therapy
Farmer, C. Psychodrama and Systemic Therapy
Flaskas, C., Mason, B., & Perlesz, A. The Space Between: Experience, Context,
and Process in the Therapeutic Relationship
Flaskas, C., & Perlesz, A. (Eds.) The Therapeutic Relationship in Systemic Therapy
Fredman, G. Death Talk: Conversations with Children and Families
Hildebrand, J. Bridging the Gap: A Training Module in Personal and Professional
Development
Hoffman, L. Exchanging Voices: A Collaborative Approach to Family Therapy
Jones, E. Working with Adult Survivors of Child Sexual Abuse
Jones, E., & Asen, E. Systemic Couple Therapy and Depression
Krause, I.-B. Culture and System in Family Therapy
Mason, B., & Sawyerr, A. (Eds.) Exploring the Unsaid: Creativity, Risks,
and Dilemmas in Working Cross-Culturally
Robinson, M. Divorce as Family Transition: When Private Sorrow Becomes a Public
Matter
Smith, G. Systemic Approaches to Training in Child Protection
Wilson, J. Child-Focused Practice: A Collaborative Systemic Approach
Work with Organizations
Campbell, D. Learning Consultation: A Systemic Framework
Campbell, D. The Socially Constructed Organization
Campbell, D., Coldicott, T., & Kinsella, K. Systemic Work with Organizations:
A New Model for Managers and Change Agents
Campbell, D., Draper, R., & Huffington, C. A Systemic Approach to Consultation
Cooklin, A. Changing Organizations: Clinicians as Agents of Change
Haslebo, G., & Nielsen, K. S. Systems and Meaning: Consulting in Organizations
Huffington, C., & Brunning, H. (Eds.) Internal Consultancy in the Public Sector:
Case Studies
McCaughan, N., & Palmer, B. Systems Thinking for Harassed Managers
Oliver, C. Reflexive Inquiry: A Framework for Consultancy Practice
Credit Card orders, Tel: +44 (0) 20-8969-4454; Fax: +44 (0) 20-8969-5585
Email: shop@karnacbooks.com
DIALOGICAL MEETINGS
IN SOCIAL NETWORKS
Jaakko Seikkula & Tom Erik Arnkil
Foreword by
Lynn Hoffman
Systemic Thinking and Practice Series

Series Editors
David Campbell & Ros Draper
KARNAC
KARNAC
LONDON NEW YORK
CONTENTS
SERIES EDITORS ’ FOREWORD vii

ABOUT THE AUTHORS ix
FOREWORD
Lynn Hoffman xi
PREFACE xvii
Introduction:
on networks and dialogues 1
PART I
CHAPTER ONE
Dialogues at the boundaries between and within
professional and personal networks 13
CHAPTER TWO
Frustrating network meetings 33
v
vi CONTENTS
PART II
CHAPTER THREE
Open Dialogues as crisis intervention 51
CHAPTER FOUR
Anticipation Dialogues for lessening worries 65
CHAPTER FIVE
Alike but different 89
CHAPTER SIX
Healing elements in dialogues 104
PART III
CHAPTER SEVEN
Dialogue and the art of responding 131
CHAPTER EIGHT
Effectiveness of dialogical network meetings 149
CHAPTER NINE
Research and generalizing practices 167
Epilogue:
on power and empowerment 187
REFERENCES 195
INDEX 205
S E R I E S E D I TO R S ’ F O R E WO R D
W
e are proud to publish the English translation of this
groundbreaking book. The two Finnish authors, Jaakko
Seikkula and Tom Erik Arnkil, have been among a
group who have pioneered a new community-based approach to
adult mental illness and, particularly, the onset of psychosis. What
makes the approach unique is its grounding in a method called
“open dialogues”—dialogues that take place between the profes-
sionals and the patients and their families.
The book carefully explains the theoretical basis for dia-
logical communication, drawing heavily on the work of Mikhail
Bakhtin, but its great strength is the way these principles have
been transformed into a practice of community meetings. Because
the authors are drawing on years of experience as well as outcome
research studies, they present a framework for a model that is
proven, and they describe the techniques that make it work. For
example, there are helpful sections that spell out what questions
to ask a psychotic patient, how to respond to delusions, and how
to approach other professionals when the network is becoming
stuck in its task.
vii
viii S E R I E S E D I TO R S ’ F O R E WO R D
Although the book describes their model in detail, many

practitioners find that the authors’ ideas about dialogical com-
munication can be applied across a wide range of clinical settings,
such as family therapy, but are particularly useful when multi-
disciplinary or multi-agency teams meet to devise treatment for
patients. The book is a guide for facilitating conversations among
people from different points of view.
This Finnish team is widely recognized for the work they are
doing, as shown by their writings and presentations at confer-
ences, and this book gives them the space to explain their thinking
and practice in greater detail than ever before. It is a major step
in bringing their work to the attention of the English-speaking
community.
David Campbell
Ros Draper
London, May 2006
A B O U T T H E AU T H O R S
JAAKKO SEIKKULA is a clinical psychologist and family therapy

trainer. He is acting professor in the Department of Psychology
at the University of Jyväskylä and is an adjunct professor at the
University of Tromsso. He has been involved for more than twenty
years in developing, studying, and implementing the Open Dia-
logues approach for the most severe psychiatric and other crises.
Presently he is involved in several projects for developing social-
networks-based practices in many countries. In his works, psy-
chotic problems are primarily seen as answers to a crisis instead
of a stable condition. He is the author of more than a hundred
scientific articles and the author or co-author of twelve books.
TOM ERIK ARNKIL is a research professor at STAKES (National

Research and Development Centre for Welfare and Health, Hel-
sinki, Finland) and an associate professor of social policy at the
University of Helsinki. For two decades he and his team have stud-
ied “multi-problem”—or “multi-agency”—situations where multi-
ple helpers may get stuck, and they have sought to develop means
for enhancing cooperation between professionals and with the
clients and their personal networks. His studies have mainly been
ix
x ABOUT THE AUTHORS
on child- and family-related questions, where social work, therapy,

schools, and so on are involved. He and his team, together with
front-line professionals, have conducted a still-ongoing series of
projects since the mid-1980s. Besides family-related situations, the
set of methods called Anticipation Dialogues has been applied and
studied in other multiple-helper contexts, such as senior-citizen
care and work with the long-term unemployed. He has authored
and co-authored numerous articles and seventeen books.
F O R E WO R D
Lynn Hoffman
T
his book—which introduces two therapy innovations,
called “Open Dialogues” (OD: Seikkula et al., 1995) and
“Anticipation Dialogues” (AD: Arnkil & Eriksson, 1995)—
marks the discovery of a new mine offering two kinds of precious
ore. With regard to Open Dialogues, I had learned in the mid-1980s
that a number of professionals in the north of Finland had started
to use inclusive and ongoing “treatment meetings” as their main
intervention with first-time psychosis. At around the same time,
Tom Arnkil was developing a network method to deal with social
agency “muddles”—where the various professionals working with
children and families cannot see what the others are doing. In each
case, a dialogical conversation between the client’s network and
the professional network takes centre stage. The first approach
(OD) leans in the direction of establishing a more open and spon-
taneous exchange between family and staff, using Russian philolo-
gist Mikhail Bakhtin’s concept of “dialogism” (Holquist, 1981). The
other (AD) draws from the same general principles about dialogue
but adds ideas from social network theory in offering a set of
structured questions that literally moves all parties into a better
future.
xi
xii FOREWORD
Placing each method in its own context, there is a big difference

between the demands made by the respective settings that OD
and AD sprang up to deal with. The OD method was developed
while looking for a more effective way to address acute psychosis.
Instead of presenting people with plans based on a diagnosis and
set up by the hospital, the staff decided to open the meetings to
everyone involved from the beginning, and to make all decisions
together. Network meetings would take place on a daily basis for
as long as necessary and would be held in as familiar a place as
possible, usually the home. Most importantly, the enterprise itself
changed from a “monological” discourse, the aim of which was to
eradicate the symptom, to a “dialogical” one focused on finding a
common language for what was disturbing people. This practice
gave words to frightening ideas that, up to then, had been repre-
sented only by strange or threatening gestures. It was not unusual
for the psychotic behaviour to lessen dramatically during even one
session (Seikkula & Olson, 2003).
The OD format was clearly revolutionary in the context of a
psychiatric hospital. The open meetings, the preference for anxi-
olytics rather than neuroleptics in the choice of medications, and
the leaning away from hospitalization except as a last resort went
against standard practice but turned out to prevent chronicity. The
most striking artefact of the OD work has been a five-year study
that described the statistics on the persons treated at Keropudas
Hospital and in local psychiatric clinics in the north of Finland,
where Seikkula’s group introduced their innovations. Compared
to what one might call “treatment-as-usual” treatment in another
context, the Keropudas outcomes were amazing. In North Finland,
in a population that represented first-time psychotic breakdowns,
more than 80% were working, studying, or looking for a job at the
end of five years. In the comparison setting, 62% of patients ended
up on welfare. In the first group, only 17% were on neuroleptic
medication at the five-year point, whereas in the second group
75% were. On other measures, like hospitalization or recidivism,
Keropudas also came out ahead: twelve years into the inception
of the Keropudas study, Seikkula says, the incidence of new cases
of schizophrenia in the area that the hospital served dropped from
33 per 100,000 per year to 7 per 100,000 per year—an astonishing
figure by any measure.
FOREWORD xiii
The AD method, on the other hand, more nearly resembles the

early ecosystems work of E. H. Auerswald (Hoffman, 2002) and the
family-plus-wider-systems approach of Evan Imber-Black (1998).
The genesis of AD, which I had not known about before reading
an article that appeared recently in Family Process (Seikkula, Arnkil,
& Eriksson, 2003), was aided by the unplanned chaos of routine
social-agency work. Arnkil describes his experience in working
with families who would come in with their own little ecosystems
attached from a variety of disciplines. How to manoeuvre through
the labyrinth of social experts, each with its own agency directing
matters in the background, could become a huge problem.
This is an example of what Bakhtin (Holquist, 1981) calls “hid-
den dialogicality”, meaning unseen presences that influence what
is going on. The opportunities for covert conflict and blaming are
easy to imagine. One result of that predicament in our field was
the invention of methods to move the discourse from a focus on
deficiencies and problems to one that dealt with strengths and so-
lutions. Ideas like the Miracle Question (de Shazer, 1994)—asking
what would tell a person that the problem that brought them in
had disappeared—or Michael White’s (2002) emphasis on deeply
held core values are all ways to substitute an emphasis on what
is esteemed or desired for an emphasis on what is feared or de-
spised. Anticipation Dialogues fall into that category and extend
and expand it.
Arnkil, who is a social scientist, not a psychologist, as Seikkula
is, explains his method by saying that “sectored and specialized
professional systems are desperately in need of intermediaries”.
One form of help that Arnkil describes is a method called “Recall-
ing the future”. The group convened by the AD consultant would
include the child and her extended family, as well as agency per-
sons involved in her situation. The group is told that the nature
of the dialogue would be the family speaking with the consultant
while the others listen, and then vice versa, but that there would be
no exchanges or interruptions. The consultants—a pair, trained in
Arnkil’s programme—would start by telling the family to imagine
that it is a year from now and that the child’s situation has im-
proved. He would ask what they were most happy about. Then he
would ask: “What did you do that contributed to this good future,
and who helped you and how?” A last, very important question
xiv FOREWORD
was: “What were you worried about a year ago, and what lessened
your worries?”
The same set of questions would then be asked of the profes-
sionals, while the family listened in. The professionals’ views on
both helpful measures and worries would be written down, and a
discussion would follow about the emerging future plan and who
could commit to what. In doing this, the consultant would not
push for any particular outcome but limit himself to clarifying and
summarizing what people said.
Arnkil also draws on his future-questions method in consulta-
tions with groups of professionals by themselves. He argues that
this makes everyone equal in an ongoing dialogue in which “an
endless polyphony of subjectivities appears to pave the way to a
postmodern expertise with fewer fantasies of control and a high
tolerance of uncertainty” (Seikkula, Arnkil, & Eriksson, 2003, p.
198). In this sense, his inventions are certainly related to the aims
of OD. However, I would say that the very chaotic nature of the
setting compels a more controlling protocol. And the feedback
from the groups he has worked with is impressive. Family mem-
bers say that they are relieved to imagine a more hopeful future in
the presence of professionals who usually look for shortcomings
and for what is wrong. Equally important, the silo walls are tempo-
rarily melted, so that all present hear from the many stakeholders
involved. This can be incredibly enlightening for the professionals,
who had previously had no way of knowing whether they might
be working in tandem with each other or at cross-purposes.
At this point, I want to cite a distinction that is offered by Roger
Lowe (2005) in an article where he brings up the difference between
methods that are characterized by what he calls “structured ques-
tions”—as in the narrative- or solution-focused approaches—and
the kind of free-form interviewing that Harlene Anderson and
Harry Goolishian popularized in their not-knowing approach and
Tom Andersen furthered with his reflecting process. Drawing from
Lowe’s article, I would like to use the term “structured dialogue”
for the more interventive methods offered by Arnkil and “open-
ended dialogue” for the work that Seikkula has proposed. Another
difference is that Arnkil’s Anticipation Dialogue is used occasion-
ally, as a consultation, rather than as a direct treatment method like
FOREWORD xv
Open Dialogue, and for this reason it will take longer to evaluate,
although studies of its efficacy are being undertaken.
But that is a quibble in the face of the achievement of the last
three chapters of this book. In this section, we are offered one
of the most cogent critiques I have ever read of the randomized
experimental study that is the gold standard for evidence-based
research. In describing the comparison studies used by the Keropu-
das teams, the authors outline a “naturalistic” design for research,
as opposed to the single-variable design that experimental research
calls for. The outcomes of this naturalistic research, done in the
field rather than in a laboratory, offer a whole new framework for
evaluating psychosocial help. For instance, in Finland, guidelines
for best practice have been based on a meta-analysis derived from
surveys of evidence-based studies. This protocol maintains that
patients require neuroleptic medication from the start of treatment.
But the OD studies showed better results among patients who had
no neuroleptic medication or had it at a later point. It seems that
the treatment indications supported by the OD study were almost
the opposite of those supported by the Finnish Psychiatric Asso-
ciation’s Guidelines.
What I particularly appreciated in reading this work is that we
are introduced in both Open Dialogues and Anticipation Dialogues
to important elements of Bakhtin’s dialogical theory, together with
fresh ideas about networks talking with networks. Both approach-
es, in my view, owe an enormous debt to Tom Andersen’s reflecting
process and, seen on a larger scale, are part of Andersen’s vision of
a “Northern Network”—a project that is introducing open-ended
dialogue to acute teams in hospitals all across the rim of Northern
Europe (Andersen, 2005). On a smaller scale, this book feels to me
like a trumpet blast, not just in announcing the achievement of a
more contextualized approach to human dilemmas, but in offering
some solid proof of its efficacy. Having watched this evolution play
out over the course of forty years, I may be forgiven for saying that
I will now take some time out to rejoice.
P R E FAC E
W
e invite professionals in the psychosocial fields—thera-
pists, social workers, teachers, counsellors, and so on—to
network dialogues: that is, dialogues between a client’s
personal networks and the professional networks. Why would the
professional be interested? Because there are surprising and unex-
pected resources to be found through thinking together.
Both “networks” and “dialogues” are popular catchwords. We
know we are taking a chance. The word “network” is used to de-
scribe such a vast variety of phenomena—from railway networks,
social networks, neural networks, to the “network society”—that
the concept stands close to losing its meaning. In the psychosocial
professions, there are additional reasons to be guarded or sceptical
when hearing someone talk rapturously about networks and net-
working. In our country, Finland, there is hardly a professional in
the psychosocial field who has not attended “network meetings”.
Almost any get-together—with or without clients—is called a net-
work meeting. Many of these assemblies turn out to be fruitless
repetitions of long-experienced interaction patterns, and the more
complex the issue and the more worrying the situation, the more
likely that these “network meetings” prove to be overt or covert
xvii
xviii P R E FAC E
attempts to control others. There will be many who, unsurpris-

ingly, say “been there, seen that—not interested”. Some may recall
the faded enthusiasm for network therapy. In the 1970s the work
of Ross Speck and Carolyn Attneave inspired many in the psycho-
social professions. Towards the end of the decade the appetite wore
thin. In Scandinavia and Finland, the idea continued for a while to
gather momentum, until eventually it began to wane here too. The
initial work with social networks aroused great enthusiasm, but
soon the difficulties of adopting this new idea into the toolboxes of
psychosocial work emerged. Without a specific organizational base
within the multi-sectored professional system, the new approach
was effectively left homeless.
It is a further risk to introduce the concept of dialogues. First
of all, the concept “dialogue” is used to denote almost any form
of conversation between people. It has become a catchword in the
consultation literature of organizations, in management guide-
lines, political pamphlets, and the like. While there might be a
common ground of interpretation about dialogues, the prolifera-
tion has also lead to vagueness. Furthermore, with reference to the
possible audiences for this book, the idea of dialogues is not alto-
gether novel in psychotherapy or in psychosocial work in general.
It can be justly argued that dialoguing is at the core—if not being
the core—of all work aimed at empowerment.
On the one hand, the ideas we introduce follow on from the
long tradition of dialoguing. On the other hand, they also place
dialogues in a new context—in the no-man’s-lands of multiple
participants, and not the traditional one-to-one settings of psycho-
therapy or the family-team contexts of family therapy. A meeting
between a client’s personal networks and the associated profes-
sional networks in a worrying situation calls for a high tolerance
of uncertainty. In such a situation, dialoguing is of particular im-
portance and, at the same time, particularly at risk. It is tempting
to resort to an expert-centred professional means of controlling
situations, but this is to be avoided if thinking together is to be
fully brought into play.
The approaches we introduce have not appeared simply through
merging dialogue-oriented professional methods and multi-party
situations. Nor are they a modification of network therapy. We
have, of course, benefited and combined ideas from numerous
P R E FAC E xix
sources, inventing and trying out novel ways—and, above all, we

have tried to listen closely to how the context talks back to us. Our
work has called for a two-way flux between practice and theory:
we have modified our activity and analysed the experiences. In
truth, the book is a summary of some twenty years of research
and development.
The aim of this book is to demonstrate how—in our experi-
ence—the resources of social networks can be brought into good
use in alleviating psychosocial problems and to analyse the core
dimensions of such networking practice. We, the authors, came to
this work via different paths and were surprised to notice the simi-
larities in the dilemmas we each encountered in practice and how
alike were our conclusions about the essential elements of valid
practices. Not unnaturally, we started to consider analysing and re-
porting our experiences together. Since our experiences were of de-
veloping network-oriented work, it is obvious that there are more
than two people behind the ideas. We have been fortunate—each
of us in our own contexts—to be members of an innovative team
and part of a network of inspired practitioners and researchers.
Jaakko had a fortunate start as a psychologist in the small Kero-
pudas psychiatric hospital in Tornio, a town in Finnish Lapland.
Being a member of the staff from 1981 until 1998 gave a wealth of
experience in recognizing the resources that families had within
them and in their other relations, even in the most severe psychiat-
ric crises such as psychosis and schizophrenia. The staff developed
its practices through collaboration. No individual towered above
others, and each one could contribute in the search for new ap-
proaches. Around 1988 Jaakko started analysing systematically the
newly emerging collective practice and to write research reports
on this joint achievement. Therefore, on the pages of this book,
the entire staff of that psychiatric system are present. We want to
acknowledge all of them, forgetting no one.
Whereas Jaakko developed practices as a member of a team car-
rying out those practices, Tom approached teams from “outside”.
He had the opportunity to connect with a number of innovative
communities of practice, first with teams of social workers in wel-
fare agencies, then teams within adolescent psychiatry, next the
entire network of psychosocial professionals in two municipalities,
followed by projects with regional multi-professional networks.
xx P R E FAC E
The core that ran from project to project was to join together in
co-developing multi-professional approaches of empowerment-
oriented work to bring about good cooperation between families
and their personal networks. Therefore, Tom’s contributions also
include many voices, and we want to acknowledge these co-devel-
opers, each and every one of them. Tom would like to mention by
name his closest research partners, Esa Eriksson and Robert Arnkil
(who is also his brother).
Writing this book was made possible in 2003/2004, when Jaak-
ko took a leave of absence from his university position and entered
the employ of STAKES, where Tom also works. The great opportu-
nity of working together for eighteen months enabled this book to
be realized. We appreciate the opportunity given to us by STAKES
and wish especially to thank Division Manager Sirpa Taskinen
for her support and wise decisions. Last but not least, we want
to thank Mark Phillips, who did a wonderful job as our language
consultant at STAKES.
Jaakko Seikkula
Tom Erik Arnkil
Helsinki and Jyväskylä, May 2006
DIALOGICAL MEETINGS
IN SOCIAL NETWORKS
Introduction:
on networks and dialogues
P
eople live in social relations even if professionals are
approaching them individually. When the professional and
the client meet one-to-one, the client’s personal network is
already an audience to proceedings, as is the professional network.
Private and professional networks are dimensions of social net-
works. Individuals have, in varying ways, access to relationships
where they maintain their social identity, where they get spiritual
and material support, information, and new relationships.
When we ask the client about the persons close to her/him,
these persons join the conversation through the inner dialogues of
the client—and echoes of their voices are there even if we do not in-
quire about these people. Each question and comment adds voices
to such dialogues. Asking and commenting about the important
people in the client’s life is relations-oriented work, even if there
was no intention to orient towards networks. How we talk has an
effect on both the inner dialogues of the client and the dialogues
that follow between her/him and those close to her/him.
The client also brings into the conversation the professionals
she/he has been in contact with. When we comment on the client’s
words or formulate questions about his/her situation, we in fact
1
2 D I A L O G I CA L M E E T I N G S I N S O C I A L N E T WO R K S
also talk to the rest of the professional network. The client may, at
first, be a go-between in the sense that she/he compares what we
say to something said by others. The other professionals are also
present as voices in our inner dialogues, and they echo our experi-
ences with those professionals. Therefore, one is in a network of
relationships even when only two people are present. We form
those relationships all the time as the voices of others echo in our
conversations. In this book, we describe work where those people
are actually invited to the dialogues.
Professional work takes place in relation to what other help-
ers are doing or have done with the client. People have, in mod-
ern societies, connections to agencies, services, and institutions at
birth, in babyhood, as toddlers, at school age, as adolescents and
young adults, as providers, in middle age, and as senior citizens.
One simply cannot find a person who has appeared in the world
and lives totally outside a personal network of relationships or has
never been in contact with professional systems.
Professional networks are likewise in the picture through their
actions. Our professional measures are linked to the measures of
others. They may complement each other or they can make a poor
fit. Especially if the problem is not clear-cut, there may be a number
of professionals from various agencies working with the client or
the family. In psychosocial work, multi-helper situations are the
rule rather than the exception.
If one sees social networks as connections that afford access to
help and support, information, and new personal relationships, all
professionals in psychosocial work are networking already. In this
sense, it is impossible not to do network-based work. The question
is not whether to do it but how to do it.
Cross-boundary work in worrying situations is far from simple.
The networks may get “stuck” or the work may produce outcomes
that no single party intended. However, in spite of the complexity
of boundary-crossing work, very little attention is paid to it in sys-
tematic practice-development. Multilateral collaboration is, rather,
seen as something that occurs spontaneously. In comparison to the
variety of methods, approaches, and training for individuals and
team work, there is not much available in the way of preparing
professionals for multi-stakeholder situations.
INTRODUCTION: ON NETWORKS AND DIALOGUES 3
This book focuses on encounters between personal and profes-

sional networks in psychosocial work and discusses how such
encounters could be made dialogical. Networking has been a topic
in therapy and social work for at least three decades. Dialogical
approaches have been developed in several branches of individual
and group therapy, education, and counselling. However, dialogues
between networks have not been discussed to the same extent. At
least, they have not been analysed and reported so far.
When several actors meet to discuss and treat a disturbing situ-
ation, it is difficult for one person to control the state of affairs. In
a worrying situation, it is not easy to put up with uncertainty and
lack of control. The temptation for monological discourse—control
over the thoughts and actions of others—is particularly great in
such circumstances.
Mikhail Bakhtin (1981) points out that authoritative discourse
is finite and demands that we acknowledge it, that we make it our
own. Dialogue, in contrast, is open. Meanings are generated and
transformed from response to response. The more voices incorpo-
rated into a “polyphonic” dialogue, the richer the possibilities for
emergent understanding. Dialogue is a way of thinking together,
where understanding is formed between the participants, as some-
thing that exceeds the possibilities of a single person. To achieve
this, the parties need to turn towards responses, to listen and be
heard.
Superficially, the conditions for dialogical processes are not
the most favourable when a crisis occurs, worries run high, multi-
stakeholder networks get stuck, clients are dissatisfied, relatives
are concerned, and professionals blame each other. In our experi-
ence, however, precisely these situations are where dialogues are
called for and where they display their greatest strengths. In the
chapters to follow we shall discuss the preconditions for dialogical
encounters in network situations and trace the essential dimen-
sions of dialogism.
Networking in psychosocial work

means boundary-crossing
Networks have been a topic in psychotherapy and social work

since the 1970s. A variety of network-oriented practices have been
developed. “Networking” has also become a catchword in other
areas of psychosocial work, such as education, counselling, reha-
bilitation, and so forth. The network metaphor is used in a wide
variety of different fields, ranging from global business to informa-
tion technology, from transportation to neurons, and from personal
relationships to social capital. Manuel Castells (1996) says we live
in a “network society”.
According to Bruno Latour (1996), there are some simple prop-
erties common to all networks. Connectedness is the keyword.
Networks can transform from poorly connected to highly con-
nected and back. Andrew Barry (2001) points out that the network
concept is not used merely to represent something “out there”. The
concept provides a basis for re-imagined and refashioned reality.
Thinking in terms of networks affords planning activity in terms
of networks.
John Barnes (1954) may justly be called the inventor of the con-
cept social network. He studied a Norwegian village and realized
that he could follow the social connections that criss-cross social
classes, family boundaries, and so on, if he envisioned them as
fibres with knots making social networks. Ross Speck and Carolyn
Attneave (1973) were pioneers in networking in the field of psy-
chosocial work. They developed a set of methods called “network
therapy” for combining the resources in social networks.
Networks are not always the same. The private or personal
networks of citizens—that is, the family, the extended family, the
close-community, the work community, and so forth—can change.
John Barnes (1972) thought that his creation—the “social network”
concept—would fade away like previous fashionable concepts,
but to date that has not happened. The concept is exact enough
to direct the gaze to connections and vague enough to not define
those connections. The family is in transformation, and the exact
meaning of the concept is not unambiguous. The meaning of “fam-
ily” is not the same in all cultures, and it has changed over the
course of time. The significance of the extended family or the kin
or local communities to the individual is also changing, and so are

work-community relationships. In general, the sources of social
support and control are transforming. Individuals enjoy unprec-
edented freedoms in terms of traditional control—yet, at the same
time, they are more removed from traditional support than ever.
However, in the middle of this, the social network concept is alive
and well. The vagueness of the definition may mean it can survive
societal transition. At the same time, social networks are not today
quite what they were in the 1950s, in, for example, a Norwegian
fishing village. Even in the late- or postmodern process of indi-
vidualization—the changing relations between the individual and
the society—it makes sense to trace the fibres of relationships and
to find resources in those relationships.
One dimension of social networks that has been transformed
fundamentally is that of professional help. In the 1950s there was
nothing like the deep vertical expertise, the sectored psychosocial
professional system that exists in late-modern societies. After the
Second World War, a need emerged to provide health, education,
welfare, and other services on a large scale. Individually craft-
ed healthcare, education, and so on were replaced by a type of
mass-production organization and work with deepening vertical
specialization. A sectored “silo system” with boundaries to mark
lines of activities and professions emerged, each sector with their
management and budgeting arrangements, with far less horizontal
structures. Demarcations have also led to the need for boundary-
crossing or “flexible networking” (see Castells, 1996) and the need
to look at outcomes of the whole instead of just the outputs of each
tube in the silo.
The wider context for network dialogues is the encounter be-
tween the life-world of the citizens and the compartmentalized
professional system. The sectored organization tries to cope with
late-modern problems of individualization. How can one extend
over boundaries—both within the professional system and to-
wards the clients and their personal networks—in a way that
combines resources?
At its best, the multi-professional system comprises well-fit-
ting complementary parts. The professionals know who to contact
when they need supplementary expertise. At its worst, there is un-
certainty about responsibilities, endless negotiations, a toing and
froing, attempts to tell others what to do—while those seeking help

are always worse off. At its best, the system provides citizens with
expert help and advice. At its worst, the system does not listen to
the client, defines people in ways that are foreign to them, carries
out measures that do not fit the client’s life situation, or pulls the
client in opposite directions.
Boundary-crossing is necessary both within the professional
system and between professionals and non-professionals. These
dimensions merge. When the patient’s personal network is invited
to join the treatment, these people are not invited to be treated but
to provide help and support. Their voices are necessary for form-
ing mutual understanding about the confused nature of the matter.
It is difficult to make an absolute distinction between an expert
and a “layman” in such situations. It goes without saying that the
doctor is still a doctor, the psychologist a psychologist, and a social
worker a social worker. Their expertise does not evaporate because
of boundary-crossing. However, in dialogues the treatment does
not rest on them alone. In effect, the personal network—the “lay-
men”—are adding to their resources. A psychic crisis or other such
worrying situations do not touch the client alone; it also touches
the people close to them. In a mutual dialogue, it is possible to
form an understanding that no single party could reach or manage
alone. As a result, the client’s personal network becomes a resource
in the treatment and a co-creator of the process. When networks
meet, shared expertise can emerge.1 Such expertise exceeds the
capabilities of any single actor.
We have done research and development on network dialogues
for some two decades: Jaakko working within psychiatric care; and
Tom studying multi-professional work done with children, ado-
lescents, and families in social-work-related matters. In this book
we describe and analyse Open Dialogues—an approach developed
by Jaakko and his colleagues—and Anticipation Dialogues—an ap-
proach developed by Tom and his colleagues. We are aware that
there are several other dialogical approaches and that more are be-
ing developed all the time. We use our own experiences as material
for more general reflections on the dimensions of dialogism. Because
dialogism does not manifest itself “in general” or outside the ac-
tual activity, we want to describe carefully the practices that are the
groundwork for our conclusions. We assume that dialoguing has
a lot to offer to practices other than the ones we talk about— but
whether it does and how it might is for others to decide who work
in such contexts and know them from within.
The book presents methods and techniques, but it is not an in-
struction manual. We approach dialogism as a way of thinking and
acting that can be enriched with all kinds of methods that further
being listened to and thinking together. In the latter part of the
book, we discuss research—how to assess the effects of network
dialogues and how to utilize such feedback for further developing
the work.
Structure and content of the book

Part I
We begin, in the first of the three parts into which the book is
divided, by describing our surprises: the unexpected turns we
took in response to theoretical and practical impasses. We did not
arrive at dialogical practices according to some plan. Jaakko de-
veloped practices as a member of a team in a psychiatric hospital.
Meanwhile, Tom developed practices in collaboration with teams.
Practice led the way, most of the time, while theory often directed
the gaze.
In chapter 1, we describe briefly the paths that led to Open and
Anticipation Dialogues. Those paths are not straight. We believe
that showing the twists and turns explains the idea better than
giving a polished picture of the innovations as if they appeared
complete and ready to implement.
We discuss in chapter 2 the factors that may lead to network
meetings falling flat on their face. Our assumption is that, like us,
many have experienced the rivalries going on in multilateral meet-
ings—of who is competent to define the matter, say—or attempts
to define the problem as something common to all parties, as if
from a bird’s-eye view. Or similarly, many will have experienced
participants warmly welcoming cooperation while also trying to
ensure they don’t get involved, or experienced that the interactions
in a meeting oddly resemble the interaction patterns that the par-
ticipants encounter in client work. How do you avoid the pitfalls of
multi-stakeholder processes? That is a central topic in this book.
Part II
The guidelines for Open Dialogues, which are outlined in Part II,
were created through analysing the essential features in successful
network-centred psychiatric treatment. They were not drawn up in
advance as rules to be followed. Open Dialogues are both a way
of organizing treatment and a special way of discussing matters
when networks meet. The experiences summarized in the guide-
lines were gathered during many years of development work,
aided right from the beginning by research in Western Lapland in
Finland. Thus, chapter 3 gives an account of how network-centred
psychiatric treatment can be organized as a whole. In chapter 4 we
describe Anticipation Dialogues that have been developed, in the
first place, in the context of multi-professional work with children,
adolescents and families. They are a way of bringing the child’s/
adolescent’s/family’s everyday life into the centre and planning
cooperation with that as the hub. A special method of “recalling the
future” was developed for creating multi-voiced understanding in
worry-laden situations.
Chapter 5 presents a comparison of Open and Anticipation
Dialogues. We also take some distance from them for a while in
order to discuss more general dimensions of dialogism. However,
as our experiences are connected especially to these practices, we
use them as material in our reflections. In chapter 6, we look in the
direction of theory, discussing dialogism by weighing up first and
foremost the concepts of Mikhail Bakhtin. We try to find out what
it is about dialogues that heal and help.
Part II
The third and final part of the book deals with effectiveness studies
and possibilities for transferring or generalizing good practices.
In chapter 7, using Open Dialogues as the material, we ask
whether it is possible to distinguish between the dialogue pro-
cesses that yielded good results and those that yielded not such
good results. It seems that there are dialogues and then there are
dialogues, illustrating differences in Open Dialogues practices. We
explore whether research can help in detecting factors that enhance
the development of good dialogues.
In chapter 8, the treatment outcomes of Open Dialogues are

studied in the light of follow-up data on the patients’ psychosocial
coping. The results are baffling—namely, they are very different
from the practices and outcomes that are the usual basis of rec-
ommendations for psychiatric treatment, and, in many respects,
they are the opposite. The need for neuroleptic medication can
be minimized in work involving networks; the patients’ recovery
and social coping were better. The outcomes suggest that there
is an alternative to the traditional patient-controlling psychiatric
treatments.
Chapter 9 discusses what may have caused the differences
among the results, while also broadening the picture. We dis-
cuss the basic assumptions of evidence-based research. Behind
Valid Treatment recommendations are research designs that do
not incorporate the most essential features of dialogues. Designs
that simplify the active variables are on their way to becoming
the yardstick of scientific competence. Broader evidence-based re-
search is needed that also recognizes something other than mere
monological encounters, in which actors intervene on objects. At
the end of the chapter 10, we discuss the problem of transferring
or generalizing good practices. Good practices simply cannot be
duplicated. The difference of contexts and actors should always be
taken into account. Today there is more and more discussion about
the need for research that is better contextualized than laboratory
research and can therefore produce socially more valid knowledge.
The need to form cross-boundary arenas and learning spaces has
been emphasized in recent research-policy discussion, as well as
the need to further dialogues in those arenas. Our closing words,
in the Epilogue, reflect upon transformations in power relation-
ships.
The book has taken shape through a mutual process. Both of
us wrote chapters and sections in response to our discussions. We
then studied, discussed, and worked through the material thor-
oughly. The Introduction and chapter 1 consist of sections by both
authors. Chapter 2 is based on Tom’s manuscript, and chapter 3
on Jaakko’s. Tom wrote the original outline for chapter 4, while
chapter 5 has sections written by both authors. Chapters 6, 7, and
8 are mostly Jaakko’s work, whereas chapter 9 and the Epilogue
were written by Tom. Most important for us, however, is that we
have written a joint work that integrates each of our particular

points of view.
Note
1. It is emblematic that there is no name for the position between experts
and non-experts or between professionals and the uninitiated, other than
“client”, “patient”, “pupil”, and other labels of institutional roles. What should
private-network members be called when they help the professionals?
PA R T I
CHAPTER ONE
Dialogues at the boundaries

between and within professional
and personal networks
W
e have primarily sought practical solutions. Jaakko
worked right from the beginning “within” the practices
he was researching, being a member of a psychiatric
treatment team. For Jaakko, practical developments have most
often preceded theoretical analysis of the experiences. For Tom,
developing concepts has been central, albeit in close contact with
practice. He was not a member of the open-care teams he has
researched, which work with children, adolescents, and families.
However, network-dialogical practices were developed in close
cooperation with the grass-root practitioners.
Although we describe in detail the dialogical practices we have
been involved with, our aim is also to discuss the more general
dimensions of dialogism. After all, technical skill is not sufficient.
Attitude, outlook, and one’s way of thinking are equally if not
more important. Neither is networking simply a series of network
methods; it is, rather, an understanding of the significance of rela-
tionships for an individual—an orientation towards networks. Of
course, therapeutic tools are not unimportant. Some ways of dis-
cussion are more favourable than others in generating dialogues.
The guidelines for Open Dialogues and Anticipation Dialogues
13
took shape over a long period. The course was not straight, how-
ever. We have had to totally retrace our thoughts a couple of times.
Below, we describe those twists and turns in the development
path, to highlight how the central ideas were arrived at.
No common definitions of the problem, after all
Anticipation Dialogues were developed in complex multi-profes-

sional situations where the helping process seemed to be leading
nowhere, in spite of attempts by a number of helpers to do the
professionally right thing. Tom was studying encounters between
the compartmentalized professional system and the comprehen-
sive everyday life of the citizens. Anticipation Dialogues were de-
veloped in a series of projects in several municipalities in Finland
conducted by Tom’s team at the National Research and Develop-
ment Centre for Welfare and Health (STAKES). Tom’s point of de-
parture was in social sciences: social policy and sociology. Tom is
not a clinician. His colleague through the years was Esa Eriksson,
a psychologist and family therapist.
“Multi-problem situations” create client relationships with sev-
eral agencies. Multi-agency contexts are complex, and such com-
plexity certainly does not reflect on the family alone. The effective
combination of different agencies may get hindered even when the
work of the individual professionals is sound and good. In a way,
the outwardly well-organized professional system gets into disor-
der when it encounters phenomena that cannot be compartmental-
ized in a way that parallels how the expert system is divided. Such
disorders are not infrequent. Cases can revolve around repeated
attempts to try to control others and to make them do what is
seen as necessary. Anticipation Dialogues were developed to re-
vitalize such situations. Together with the front-line professionals,
Tom and Esa and their team tried to develop methods that would
be helpful in situations where the parties seemed to be repeating
unsuccessful patterns of activity. Professional work with children,
adolescents, and families afforded plenty of such material.
Tom and Esa conducted a series of projects together with front-
line professionals from the mid-1980s on. The first project was with
D I A L O G U E S AT T H E B O U N D A R I E S 15
three social welfare offices and an clinic for substance abusers.

The next partnership involved the personnel of two social welfare
offices and a psychiatric clinic for adolescents. The third develop-
ment community to become involved was extensive: all the profes-
sionals—from antenatal clinics to psychotherapy, from day care to
family counselling clinics, from schools to child protection—work-
ing with children, adolescents, and families in two municipalities.
The next community was somewhat larger again: the equivalent
professionals plus the police force in fourteen municipalities and
rural districts. Finally, the Anticipation Dialogues’ set of practices
are being used in a training programme being carried out in vari-
ous parts of the country.
At first, dialogism was not at the forefront. It became central,
however, through experience. In the beginning, the focus was on
the means for changing one’s own activity, instead of trying to
change others directly. The main method for this was through an-
ticipating the outcomes of one’s own acts and reflecting together upon
the actual outcomes—and trying to learn the lessons. The aim was
to find a means that, on the one hand, would not repeat recurrent
patterns but would, on the other, likewise not be so exceptional as
to break the contacts with the client. The idea was to find appropri-
ately different ways through anticipations.
Tom and Esa sought a new approach through varying their own
patterns of activity with the help of anticipation, facing intended
and unintended consequences, gaining surprises, and reflecting on
their own positions in the network of relationships. This develop-
ment was a merger of at least three strands of previous thought:
(1) In the work of analysts of first- and second-order change, such
as Paul Watzlawick (Watzlawick, Weakland, & Fisch, 1974), we
can see the idea that more of the same problematic attempt to solve a
problem is itself problematic. The problem is, in part, the problem-
atic problem-solving. Thus, the problem-solving efforts need to be
changed. (2) Tom Andersen’s work (1991) was also a source of in-
spiration: the idea of finding ways that are appropriately unusual for
the client. More of the “usual” does not bring about change, while
attempts that are too unusual can break off the contacts or make
the client feel defensive. Tom and Esa sought an “area” in between.
(3) The idea of using anticipation for probing what is appropriately
different or unusual came from the work of two Finnish psycholo-

gists, Johan Weckroth (1986) and Totte Vadén (1985).
In “multi-problem situations” there are several parties trying
to change the client’s or the family’s activity. The professional
gaze is fixed on the client from various directions, and each of
the specialized players tries to change the client according to their
professional frameworks. If the whole does not progress as wished,
the professionals will intensify their attempts and may also try to
change not only the client but also the activity the other profession-
als. A pattern of activity may emerge, a recurrent figuration, that
is not a single-handed “creation” of the client/family or the work
of the professionals alone but something in-between: a co-creation.
It does not belong to the “system-world” alone, nor only to the
“life-world”, and it is—or was—very hard to put a name to it. The
parties maintain a pattern of interaction together, even with many
players pursuing change instead of constancy. The cooperation
plan they are following may embody something quite different to
what is actually realized as a figuration of reciprocal actions.
Changing multilateral interaction to one’s liking by unilateral
commands or appeals is hard to achieve. However, what one does
and how is of significance. Therefore, instead of focusing only on
how the others should change, it is fruitful to look at one’s own
attempts in the pattern. Varying one’s own activity is not as hard
to achieve. The outcomes will very seldom be exactly what one
anticipated. Instead, one is likely to face a mix of intended and
unintended consequences. In order to learn what promotes change
through being different enough while also maintaining contact
by being familiar enough—in other words, is being appropriately
different—it is useful to reflect upon the outcomes in the light of
one’s anticipations. Controlling networks is not possible, but it is
possible to learn more about how others position you. In this vein,
one can become a little more aware of the anticipations one makes
regardless.1
Evan Imber-Black’s work (Imber-Black, 1988; Imber-Copper-
smith, 1985) inspired Tom and Esa to combine anticipations with
analysing multi-problem/multi-agency situations.
Tom and Esa tried, together with the teams in the projects, to
simplify and clarify the anticipation method. Finally, it material-
ized as three questions for the team members to consider when

trying to vary their own activity in multilateral networks. The
questions were:
1. What would happen if you did nothing?

2. What could you do to help that was appropriately different?
3. What would happen if you did that?
These questions were originally for team members within an agen-

cy. They would discuss each other’s anticipations about a case so as
to find a way where they would act in a way that was appropriate-
ly different—that is, was acceptable for the client/family but not a
“more-of-the-same” recurrent activity pattern. Therefore, trying to
anticipate “what happens if . . .” requires personal involvement, so
that not only a cognitive means of orientation (general professional
knowledge) but also emotional and moral means of “feeling out”
the particulars of the unique situation can be made use of.
In Tom’s and Esa’s later projects, the situations were no longer
approached from the direction of a particular agency or team. The
hub was a multi-agency case: a “multi-problem situation” that
pulled in professionals from various agencies around it. Tom and
Esa acted as dialogue facilitators for professionals’ network meet-
ings. The conveners of such sessions were professionals who felt
increasingly worried about where the cooperation—or the lack
of it—was leading. The professionals invited were those actually
working with the client/family in question.
The three above-mentioned questions were applied. The dis-
cussions were organized with the aim of leaving plenty of room for
mutual listening. The participants could only talk in turns. When
one had a turn, the others listened. The idea behind this was to try
to foster inner dialogues instead of commenting on each other’s ut-
terances. The idea of separating speaking and listening came from
Tom Andersen’s (1991) way of setting up reflecting teams. The an-
ticipation method began to evolve towards a deliberate dialogism.
Organizing the sessions in a way that would maximize possibilities
for generating rich inner dialogues became the aim.
Earlier, in the context of team discussions, it had seemed fruit-
ful to try to bring into focus the full spectrum of the team members’
anticipations. The new context was not a team plus its clients, but
a whole professional network around a family that had simulta-
neous contact with a number of professional helpers. It seemed
equally productive to bring to the fore the spectrum of anticipa-
tions. The sessions were organized in the no-man’s-land between
the agencies, with one involved agency as the “host/hostess”. The
same three-part question structure was applied. Each of the pro-
fessionals answered in turn what they thought would happen if
they did nothing, what could they do to be helpful, and what they
thought would happen if they did that. Talking and listening were
separated in the same way it was done in team reflections: each
professional could talk without others interrupting to comment.
At the end of the session there was a discussion on who will do
what with whom next. Bringing to light the dissimilarities between
each other’s anticipations was very powerful in enriching the par-
ticipants’ ideas. With such experiences, enhancing the polyphony of
voices became a central aim in developing the approach. It seemed
that it was exactly the multi-voicedness that helped to broaden
the narrowing patterns of thinking and interacting. No overall
definitions of the problem were made in the sessions. Instead, the
facilitators asked about the professionals’ own activity and the
anticipated outcomes.
The professionals seemed be curious to hear each other’s an-
ticipations and were interested in what each specialist thought
would be the consequences in other parts of the network if she/he
did this or that, or nothing. Each participant was equal regardless
of her/his status in the professional hierarchy in the sense that no
one, no matter what her/his speciality, could say with certainty
“what would happen if . . .”. Tom and Esa also observed that most
of them knew very little about the non-professional resources—the
private networks—of the clients and that the discussions raised cu-
riosity towards these possibilities. The only summaries at the end of
the discussion were expressed in activity-language—that is, a braid
of actions took shape: who does what with whom next.
Finally, when everything was “ready”, in the sense that the
anticipation procedure for the sessions was tested and stream-
lined, clients entered the sessions and everything had to be re-
appraised.
Everything has to be changed when clients are present
The professionals had been encouraged to bring along clients.

When, without warning, this actually happened in one particular
meeting, Tom and Esa were faced with the realization that the
situation was different with the client present. Thorough revisions
of previous thinking had to happen almost instantaneously. They
realized that the set of anticipation questions would not be ap-
propriate for an encounter between the client’s network and the
professional network, so they had to make revisions in the short
timeframe before the session began. A meeting between clients and
professionals is a direct intervention into the lives of the clients,
and the minimum requirement for such an encounter is that the
clients leave the session more empowered than they entered it.
The carefully prepared set of anticipation questions did not seem
appropriate for this. Even the opening question—what would hap-
pen if nothing was done?—could be offensive. In a case where
there was a risk of suicide or some other great danger, the question
would be absolutely inappropriate. The model had to be revised
thoroughly—in some twenty minutes. Tom and Esa discussed all
the resource-oriented work models they knew and weighed their
personal experience in ways that could arouse credible hopeful-
ness in those present. Making use of means they had developed for
in-organization consultation and supervision and combining ideas
from client work, they made a plan for interviewing the client, the
family, and the professionals from a future perspective. The set
of questions now called “Recalling the future” was outlined. The
anticipation structure was turned around: instead of starting from
one’s actions and anticipating what may happen, the starting point
was to be in desired outcomes, and actions were to be “derived”
from these.2
In the first-ever meeting having the client present, the mother
of a client family had arrived all by herself. Five or six professional
helpers of her family were also present. Two adjoining round
tables were organized, and Tom and Esa announced the follow-
ing rules. Talking and listening are separated; one is to refrain
from commenting, so that each and every person has room for
listening to the impressions generated in her/his inner dialogues;
everybody will be interviewed in turn; each will be asked to think

aloud. The client will be interviewed first, then the professionals.
Moreover, the questions will approach the present from the near
future, as if we were already there. This way a plan for cooperation
will be outlined, and it will be put together at the end. To help this,
everybody’s views will be recorded on a flip chart, for all to see.
Tom started interviewing the client at one table, while the pro-
fessionals listened from the other table. After a joining conversa-
tion, he asked the client the first of three questions: “A year has
passed and things are quite well in your family. How are they from
your point of view?” Tom helped the client with concrete questions
about her everyday life. The second question was: “Who helped
you to carry out these good developments?” At times he repeated
word-for-word what the client had said and inquired “did I hear
you correctly when I heard you say . . .”. He tried to help the client
to catch her thoughts and also to signal that she was listened to
attentively. Tom and the client were in eye contact; others were not
looked at, nor were they spoken to directly. The third question was:
“Were you worried about something ‘a year ago’, and what less-
ened your worries?” The question aimed at charting present wor-
ries from a perspective of relief. While Tom interviewed, Esa took
notes on the flip chart. The client’s words were used. The sentences
were shortened, but trying not to alter the idea. After the client had
had the chance to think aloud, Esa started to interview the profes-
sionals. The client listened. Tom took notes. The first question to
the professionals, each in turn, was: “As you heard, things are
quite well in the family now that a year has passed. What did you
do to support the good developments—and who helped you and
how?” The second question was: “Were you worried about some-
thing ‘a year ago’, and what lessened your worries?” The flip chart
gradually filled up with notes about the good year, its support, and
its worries, plus the matters that lessened them. Finally, no longer
now assuming a year had passed, a discussion took place on mak-
ing a plan of cooperation. The notes on the flip chart were used
as guidelines. Discussed were whether everyone felt they could
actually be involved as outlined, who does what with whom next,
who coordinates the cooperation, and whether it is necessary to
meet again and when. The flip charts were signed with the future
and present dates and handed to the client.
Tom and Esa were amazed how well the session went. The
plan was down-to-earth and concrete. The client left smiling hap-
pily, and the professionals radiated satisfaction. Dozens of similar
dialogues were then carried out, modifying the method. Instead of
single clients, families arrived, and many families brought people
from their personal networks—children, parents, grandparents,
significant friends, and so on. What made the sessions run in
such an interested atmosphere, one after another? The dilemmas
in question had been entrenched for quite a while in most cases.
Something a client once said gave a clue: Tom and Esa were near
the clients during a break, when the wife took a long draw on her
cigarette and said to her husband: “Isn’t it strange that we are
talking with those there [pointing in the direction of the professionals]
about how things are well for us. We’ve never discussed with them
about things being well!”
Joint planning turned out to be best treatment
The origins of Open Dialogues were cultivated in the early 1980s

in the small Keropudas psychiatric hospital in Tornio in Finland. In
the process towards an open system, the most important step—in
which families’ and professionals’ cooperation had a central role—
was taken in 1984. Instead of a division into closed treatment-
planning meetings by the staff and then family therapy sessions
according to the plan, open meetings were put into practice for
both planning and treatment. It was decided to have the patient
present from the very beginning of his/her treatment. This meant,
literally, a practice in which the staff did not plan in advance the
meeting or the proposals for the patient concerning treatment deci-
sions. The family was invited into the joint meetings in every case
without a specific indication for family therapy. This practice of
therapy meetings had been initiated in Turku University Hospital
(in Finland) a year before by Professor Yrjö Alanen (1997) and his
team.
The team wanted to build up a family-centred psychiatric hos-
pital. In the beginning, this was attempted within the traditional
psychiatric system, in which—after the patient had been admitted
to inpatient care—the team took charge of planning the treatment

and the different therapeutic methods. The idea was to first plan
the treatment carefully and then put the plan into practice and to
follow the treatment up. Previously, planning the treatment was
seen as a preparation, not as a part of treatment. The prepara-
tions would take place in staff meetings without the patient being
present.
However, these early efforts towards family-centred treatment
were not satisfactory. Although the staff members thought they
had more than enough family therapies to cope with, only 5–10%
of families of hospitalized patients were actually participating in
these. This was the outcome in a system in which the team first
analyses and defines the problems and plans the treatment. Family
therapy was seen as indicated only on a few occasions. The ideas
developed in Turku offered a solution to this problem. Instead of
first defining the problem and finding a reason to invite the family
to family therapy, it was realized that each hospitalization as such
already gave a good reason for inviting the family into discus-
sions.
The treatment-planning meetings were subsequently reorgan-
ized such that the patient was invited into the meeting from the
very beginning. This turned upside down the previous practice of
splitting planning and treatment into two separate processes. From
the very first open meeting, it was observed that after the team
members spoke openly of their ideas for the treatment and of what
they had heard of the patient, the patient behaved differently. All
this meant a radical step in the direction of seeing the planning and
implementation of treatment as parts of the same process, not as
separate units. On many occasions, the planning actually turned
out to be the best form of treatment.
The new practice opened the door for other surprises. The staff
members’ view of their professional identities changed. Jaakko
took a three-year systemic family therapy training from 1983 to
1986 and learned an approach that emphasized interventions for
influencing the family’s behaviour towards change. In the new
open practice, the families did not, however, act as expected. The
Keropudas team tried to apply systemic ideas in inpatient set-
tings and encountered confusing treatment experiences. Families
were no longer receivers of a family therapeutic intervention but

were actively affecting a mutual evolving between the team and
the family. Because the families were present—in every case, in
fact—the arrangement affected the entire treatment system. In
several dead-end treatment situations, the team found that their
change made it possible for the family to change. The conclusion
that change can first occur on the part of therapists was not an
obvious one. It was after several treatment impasses that the staff
members started to think that perhaps in this new, open system
the team’s re-evaluation of its action really could lead to a situa-
tion where the family can make use of this re-evaluation to initiate
their own change.
There are several documented examples of these new, open
treatment processes (Seikkula, 1991, 1994). The clash between two
different treatment cultures was often tragicomic. On the one hand,
there was the culture of a traditional treatment system, in which
therapists are in charge of interventions for changing the patient or
the family. On the other, there was the mutual and open interaction
focusing on dialogues. The following case exemplifies this.
Case illustration: “I will kill her . . .”

Matti was admitted into the ward as behaving psychotically
after a heavy drinking period. The danger of delirium was duly
noted. His wife, Liisa, announced in the very first meeting that
she was afraid of Matti being violent towards her. An agree-
ment was reached to admit Matti, and family meetings were
agreed. Rapidly, within a week, Matti became coherent and no
longer spoke of any psychotic experiences. His wife, however,
said repeatedly that she did not wish to have Matti home for a
weekend holiday since she was still afraid of his violence.
Staff member became sympathetic towards Matti, with nega-
tive feeling towards Liisa increasing. As she was persistent
in not being willing to have Matti home, the team decided
between themselves to have a meeting to introduce their plan
of Matti’s weekend holiday. Matti and Liisa participated in the
meeting, together with several staff members. After an initial
period of discussion on other themes, the doctor proposed to

Liisa that Matti should return home for the weekend. Immedi-
ately on hearing this, Matti stood up and started to swear and
threaten Liisa’s life, saying “I will kill her . . .”. Staff members
calmed him down, and discussions moved swiftly to other mat-
ters. After a while, the suggestion of the weekend holiday was
brought up again, and the same thing happened: Matti started
to swear and to threaten his wife’s life. After the same thing
happened for a third time, one of the female nurses started to
ask other staff members “what in the whole world are we do-
ing here? Every time we propose the holiday, Matti threatens
his wife’s life. How can we do something like this, increasing
the risk for violence?”
After her comments, team members started to discuss with

each other what had happened in the meetings and realized
that the plan they had prepared in advance was not a realistic
one. All of a sudden their antipathy towards Liisa disappeared,
and they saw how extremely difficult her family situation was.
The plan the team had tried to impose was based on their idea
that treating psychosis meant taking care of the symptoms
and then discharging the patient. However, the plan seemed
to clearly increase the risk of violence, and so the team had to
adapt their work to this fact. Gradually, Liisa’s voice was given
more weight in the treatment, and it affected the mutual proc-
ess. It became evident that their marriage was unhappy, and
after some months of family discussions the couple actually
decided to divorce.
Repeated experiences of this kind offered a conclusive insight: in

an open system, the treatment team have to adapt their behaviour
to the family’s way of life. This notion was the basis for more care-
ful analysis of the boundary between a hospital and a family. It
seemed that at the boundary between the family and the hospital
(and between other professionals involved in the specific treatment
process), a specific interactional system was being formed—a sys-
tem that was no longer controlled by the ward team alone. Refer-
ring to the fact that all this was taking place across the boundary
between the therapeutic team in the hospital and the family, it was
named the “system of boundary” (Seikkula & Sutela, 1990). Identi-

fying and naming the phenomenon was particularly informative.
It helped the staff members realize that the most important events
were those taking place at the boundary, not those occurring on
the ward. Of course, the dogged emphasis on ward events was
not a problem specific to the Keropudas Hospital, since the entire
psychiatric system aimed at analysing the patient’s symptoms,
finding the causes, and putting into practice treatment methods
that affected such causes in such a way as to remove the symp-
toms. The theories available at the time focused on the phenomena
taking place within the patient and advised strategies for changing
them.
In retrospect, the conclusions the team reached are quite easy
to describe, though in actuality they were difficult to arrive at. The
understanding of the therapists’ place in the joint process had to
be radically changed. The team’s experiences of the open approach
made it evident that it was impossible to control the therapeutic
processes and that the therapists should be adapting their behav-
iour to a joint and mutual co-evolving process. Changes could be
expected if the team were themselves to change their own activ-
ity. All this meant challenging the principles of systemic family
therapy.
From problems to dialogues
One clear lesson learned was the realization that therapists should
think more about their own position in each therapy process. After
previously focusing entirely on the patient and on the family’s
interaction, the staff members now found themselves repeatedly
focusing on their own behaviour.
In the mid-1980s, the team did not have access to theories of
dialogues and dialogism. Now, in retrospect, it can be pointed
out that the basic principle of dialogism was actually qualified
in the team’s conclusion: if several voices in the patient’s life are
present in the meeting, one participant alone cannot control the
entire interaction. Instead, the aim should be increasing joint under-
standing.
In search of concepts
When Tom started inquiring into “multi-problem situations” in

the mid-1980s, he was puzzled by the conceptualizing that he ob-
served. The sociological/sociopolitical concepts that he had avail-
able did not seem to embody the most interesting phenomena: the
co-created activity patterns between the “system world” and the
“life-world”.3 Organizational and institutional concepts seemed to
be distanced from encounters between people, as if social structures
existed independently, far removed from people and their activity.
On the other hand, concepts for analysing the details of interaction
and the microcosm of activity seemed far removed from struc-
tures.4 Analysing the encounter between the complexity of every-
day life and the compartmentalized expert psychosocial system
is difficult, particularly if it alternates between a macro- or micro-
approach.
Tom’s attention was caught by something Evan Imber-Black
had formulated within family therapy discourse. She pointed out
that instead of “multi-problem families”, we should talk about
“multi-agency families” (as Imber-Coppersmith, 1985; as Imber-
Black, 1988). If a family has problems that do not fall neatly into
the category of one single specialist, the family will have to see
a number of professionals and agencies. In a specialized system,
a client or a family with a spectrum of dilemmas will inevitably
become a multi-agency client or family. Multi-agency situations
can be complicated, and Imber-Black emphasized that this was
a problem in itself. Tom was thrilled by the notion. What Imber-
Black was talking about from the family therapy perspective also
made a lot of sense from a sociological perspective.5
Multi-agency complexity is not something that of necessity
affects the professional system, since the professional system does
not get into muddles in all multi-stakeholder cases. If the clients’
or families’ problems are clear-cut and are comprised of more
or less separable elements, the system will rapidly sort out the
responsibilities of the different professionals. If the problems are
multidimensional and fuzzy, then the process of dealing with the
division of labour will likely be so too. Complexity is not a func-
tion of the client-world alone, either, as the compartmentalization
of professions, disciplines, and agencies is not the making of ordi-

nary people.
Multi-agency muddles occur when the problems that the spe-
cialized system encounters do not fit the pre-specified grid of spe-
cialization. Everyday life is not sectored, even if bureaucracies often are.
Donald Schön (1983) says that professionals create “junk categories”
for phenomena they cannot categorize according to their rigorous
techniques. In this vein, the term “multi-problem client/family”
could be called a multi-junk category. All it expresses, though, is
that the professional system is confused. When the clients’ prob-
lems do not fit the categories of a sectored system, the system has
to negotiate its division of labour again and again. It has to be ne-
gotiated in each “multi-problem” case. Such boundary negotiations
do not always manifest themselves as negotiations. Often they
materialize as a toing and froing: referring clients without a clear
idea where they ought to go, or simply turning a blind eye, leaving
the problem to those who cannot opt out of responsibility.
The dilemmas occurring at the boundaries between the profes-
sional system and the life-world of the citizens and, correspond-
ingly, within the professional system could be seen as late-modern
dilemmas. In the earlier stages of modernity, when the fields of
psychosocial work were only just differentiating, one after another,
the branch was not specialized enough to produce such multi-
agency complexity. Today, these dilemmas are not infrequent or
exceptional. Boundary-crossing—in some form or another—is an
essential part of late-modern psychosocial work.6
On the one hand, there are the processes of individualization.
On the other hand, there are the processes of societal differentiation.
Individual–society relationships are experiencing a fundamental
transformation. According to the German sociologist Ulrich Beck
(1986), individualization is the core process of societal moderniza-
tion. The mediating structures—such as close communities, kin-
ship bonds, even family structures—are transforming and their
significance for the individual changing, with many societal inter-
mediaries even fading away to leave the individual facing society
more immediately than before.
Specialization increases the possibilities for tackling problems
through expertise. On the other hand, it also increases complexity,
as can be seen in the “multi-agency” cases. The one thing that the
late-modern psychosocial expert system cannot get on top of by
means of specialization is the very complexity created by speciali-
zation.
While individuals are breaking free from traditional bonds—
and paying the price in the form of a loss of traditional sources of
support—there is also a parallel, deep vertical specialization in the
professional system. Each professional compartment is empower-
ing citizens to live relatively independent of traditional bonds.
In-between these social upheavals are the encounters between the
clients and the professional system. This is where we find the fuzzy
multi-agency–client–family formations.
In the late-modern society, citizens with multiple or extensive
problems have to see a number of specialists and to try to cope
with and negotiate a complex multi-professional system. In turn,
professionals also have to coordinate and adapt to the complex
system-world of other experts and agencies involved with the
client, in addition to the life-world of the client. Within this multi-
stakeholder environment, boundaries have to be crossed. The concept
of boundary—and boundary systems—is, in our view, essential for
analysing and developing psychosocial work.
The idea that a system emerges through interaction across
boundaries turned out to be very useful for understanding how
the problems of the late-modern society were realized in concrete
situations. Jaakko had analysed his team’s interactions with the
families and developed a concept that could embody the inde-
pendent nature of such co-created activity patterns. He called the
co-created system the system of boundary. Tom got hold of the con-
cept by a stroke of luck, but its effect was to electrify his search for
concepts. At last there was something to bridge the analysis of the
“macro” developments of the differentiating expert system and the
individualizing individuals and the “micro” processes of interac-
tion. The processes met in the systems of boundary between both
the professional system and that created by the family. The sys-
tems of boundary were co-creations. Jaakko made the concept the
cornerstone of his doctoral dissertation in psychology (Seikkula,
1991), while Tom made it the basis of his doctoral dissertation in
social policy the following year (Arnkil, 1992). At that time, we did
not know each other personally.
In Jaakko’s conceptualization, boundary-crossing took place

at the interface between the hospital and the family. The hospital
was not a party as such. It was the psychiatric treatment team that
met the family. The very persons that were interacting in each
given situation from the family’s side and from the team had an
effect on what form the encounter took. The very interaction of
actual persons has to be analysed, not just the abstract encounter
of societal institutions.
Jaakko was interested in developing the team’s activity in help-
ing in family-related therapy and crisis work. According to his
analysis, the interaction with a family does not become curative
by the mere fact that a psychiatric treatment team takes part in it.
Only if the team can bring structure, flexibility, and a multifaceted
problem analysis into the boundary system can it help to make
the interaction curative. That this will happen can by no means be
taken for granted.
Jaakko and his colleagues had noticed that in several difficult
treatment situations the conduct of the treatment team would be-
gin to mirror the family’s mode of interaction. If the family was
tightly knit or entwined, the treatment team would in many cases
end up discussing almost endlessly to reach an understanding.
If the family’s interaction was rigid—as seemed to be the case in
the preceding example—the team, too, could repeat its preferred
attempt stubbornly, without yielding. On such occasions, change
could only start to take place when the team began to add flex-
ibility and to change its own activity.
As only those persons who took part in the discussions seemed
to have the possibility to affect the events of the boundary, it was
important to analyse the independent nature of such interaction. The
boundary system concept seemed to be able to catch something of
that independence. It was defined as the reciprocally developing
activity system of those who take part in the interaction. Such mu-
tual development can be called co-evolution. It is an essential idea
that only the participants taking part in it can define its meaning.
The meanings cannot be controlled from outside.
The family brings the problem and its activity culture to the sys-
tem of boundary. The treatment team can bring its treatment tasks,
its organization, its flexibility, and its versatility. The emphasis was
no longer on the general expertise of the team: the crux was the
team’s ability to adapt to each situation. If the family’s accustomed

mode of interaction “conquers” the system of boundary, recovery
may not occur. Instead, if the team can both adapt flexibly to the
family’s activity mode and bring in organization and versatil-
ity, then co-evolution may enhance recovery and mutual learning
(Seikkula, 1991).
The new emergent boundary system opens completely new
viewpoints to the client’s crisis situation. These cannot be seen
from one perspective alone, in the compartmentalized view. The
interaction at the boundary may afford access to unexpected re-
sources.
One can envision systems of boundary emerging not only
at the interface of hospitals and families but also in the mid-
dle grounds between, for example, social welfare offices, schools,
day-care centres, family counselling clinics, clinics for substance
abusers, and mental health clinics and families (Arnkil, 1991a,
1992). Tom was interested to know whether the professionals and
teams involved were actually organized, flexible, and versatile so that
they could bring such elements to the systems of boundary. When
a system of boundary emerges at the interface of the family’s
private network and the different agencies around it, one cannot
take for granted that the “presence” of the professionals promotes,
when put together, organization, flexibility, or versatility. On the
contrary, in the context of rigid bureaucracies, changing person-
nel, enormous caseloads, haste, and organization cultures that
encourage the staff to cope alone, then organization, flexibility,
and multifaceted problem discussion seem, if anything, to be en-
dangered species. The course of problematic interaction patterns
may, therefore, be bi- or multidirectional, not only from families
towards boundary systems but also from the professional system
towards the systems of boundary.7
When the sectored psychosocial system and the everyday life
of the clients meet, interaction patterns emerge that are not made
by, or at the control of, any single party alone. From a theoreti-
cal perspective, such a conceptualization enables bridging gaps
between micro and macro views. From a practical perspective, an
angle opens up that allows for an attempt to impact the co-evolu-
tion processes in the boundary systems through trying to make one’s
own activity more flexible, organized, and versatile.
As already said, we had not begun our cooperation at the time

of the above-mentioned conceptual and practical developments.
The Open Dialogues approach was developed within psychiatric
crisis work, where Jaakko was active. Anticipation Dialogues were
developed within child-, adolescent-, and family-related work,
Tom’s field of research. Thus the practices that we describe later are
different in many ways, even though they both deal with network
encounters and seek dialogism.
Notes
1. People anticipate the consequences of their actions constantly, but most-
ly unconsciously. The Russian psychologist P. J. Galperin (1969) points out that
the human psyche uses every means it has (cognitive, emotional, and moral)
to anticipate what happens (to me) next and that this subjective orientation is
essential for making sense of the world. Anticipation is mostly routine-like,
tacit. Only when something surprising happens—that is, when something very
different from the anticipation occurs—might one realize that one is actually
constantly “guessing” at what happens if I do or don’t do this or that.
2. Peggy Penn described “future questions” in the 1980s (Penn, 1985). Tom
and Esa learned about the article only long after their own formulations. At the
time, they were familiar with Steve de Shazer’s and Insoo Kim Berg’s “miracle
question” (e.g. de Shazer, 1994), which orients to future solutions instead of
past causes. However, instead of avoiding all problem-talk, Tom and Esa en-
gaged in analysing subjective problems of activity and worries as signs of them.
They also emphasized the active role of the subject. Tom had carried out, in
in-organization consultation together with his brother Robert, a series of ques-
tions where all “ifs” were left out of envisioning the future. Instead of asking
“if things were okay”, people were encouraged to envision “how things are for
you now that they are okay”. The present—the point of time of observation—was
“taken” to the future, and the path there was viewed as if in retrospect. In
addition, subjective action—own initiative—was emphasized by asking “what
did you do to get there?” And subjective worries were addressed—problems
that the persons felt they were facing—stressing the elements that lessened
those worries. Thus influences from resource-oriented therapies merged with
Vygotskyan activity theory.
3. Jürgen Habermas (1984) makes a distinction between the “life-world”,
where free communication is possible (in principle, in the ideal case) between
people in their everyday life, and the “system-world” of finance and
bureaucracies, where communication is instrumental and self-interested (for
profit or control) and therefore not free and equal.
4. Anthony Giddens (1979) wrote that it is essential for the social sciences
to return the active subject into sociology without lapsing once again into
subjectivism. He suggested that structures ought to be analysed as continually
structured by activity.
5. According to the German sociologist Niklas Luhmann (1989), differen-

tiation is what societal modernization is about. In the partitioned whole, each
specialized system and subsystem has its particular area of competence and
its specialized codes of communication. The differentiated society can respond
to a vast array of questions or problems, but the price is that it cannot respond
as a united front or in an uniform way. Each specialized subsystem responds
according to its structure and in its own particular way.
6. Although current developments are given different names by sociolo-
gists—postmodern by Lyotard (1984), post-traditional by Habermas (1984), re-
flexive or late-modern by Beck, Giddens, and Lash (1994), or liquid modern
by Bauman (2002)—they all seem to agree that something fundamental is oc-
curring through, for example, globalization, individualization, digitalization,
environmental consequences, and the full-employment crisis.
7. The process of “contagion” of interactional patterns may be called
different names—like transference and parallel processes in psychoanalytic
vocabulary, or isomorphic processes in the systemic camp—but the core idea
is the same: the practitioner/team may come to mirror and replicate the pat-
terns she/he is connected with, and this may reflect as far as supervision, case
conferences, or even administration (see Sachs & Shapiro, 1976; Schwartzman
& Kneifel, 1985; Stanton & Schwartz, 1954)
C H A P T E R T WO
Frustrating network meetings
I
t would be hard to find a professional—be it a doctor, a thera-
pist, a teacher, a social worker, a counsellor, a rehabilitation
professional, a mental health nurse, and so on—who has never
been in a network meeting. Multi-professional meetings and as-
semblies with clients, family members, and other parties are so
common that there are many who ask whether this is all worth-
while. Network meetings may wind up in frustrating shilly-shally-
ing, instead of being the ground for new ideas and solutions. There
are many who have just about had enough. However, boundary-
crossing is not likely to be on its way out; nor is the need to meet
and negotiate.
We would like to ask the reader:
• have you attended network meetings where, at some manifest

level, the discussion is about the client and his/her problem
and, at some other level, there is a competition going on over
who is competent to define the case;
• have you been in network meetings where the parties try to
define the common problem—the problem common for all—as
33
if from a bird’s-eye view, and you somehow can’t express your

own subjective view, nor can anyone else;
• have you attended network meetings where cooperation is
welcomed warmly while participants try to avoid committing
themselves to joint activity;
• have you been in network meetings where the professional help-
ers’ interaction curiously resembles the interaction the same
professionals encounter in their client relationships?
Four assumptions about the problematic

of multi-stakeholder meetings
Sometimes network meetings are identified with networking: co-

operation calls for assembling a number of people in the same
room. Network meetings are, however, a special case of network-
ing. The main part of networking runs well—often even better—
without meetings. Networking calls for taking into account that
neither the clients nor the professionals are actors with no con-
nectedness, and this does not necessitate meetings as such. It is
possible to be network-oriented even in one-to-one encounters
with a client. If one considers who else is involved in the matter
other than the client–worker dyad, one’s thoughts are gravitating
towards network relationships. The questions used and the tone of
the discussion are different when people are considered as being
in a network of social relationships, instead of merely being seen
as singular actors. If one actually makes a move towards involving
significant others, perhaps by contacting some of the other parties,
one is already many steps down the road of networking, even if
no multi-member meetings are arranged. Meetings are necessary
if there is a need to discuss and agree on matters multilaterally.
They are particularly necessary when parties need to meet face to
face in order to reach a joint understanding. Dialogue can generate
mutual understanding between the participants, as a co-creation
that cannot be reached by any party alone.
Network meetings are often organized without much thought
to what would be the wisest way of organizing them. Although
F R U S T R AT I N G N E T W O R K M E E T I N G S 35
multilateral cooperation is more complex than dyadic discussions,

it is often expected to run spontaneously. When several people
meet to discuss the kinds of problems psychosocial work involves,
there is at least as great a likelihood of talking past each other and
on over the top of each other’s words than to enter horizon-broad-
ening dialogues.
Network meetings are prone to tipping towards monological
instead of dialogical discussions—with the parties trying to define
how others should think and act. This, we think, is due to:
• meta-communication structures
• problem-definition procedures
• commitment-regulation
• isomorphic processes.
Actors define mutual relations
It is not exceptional to experience meetings where psychosocial

professionals argue over what the client’s situation actually is
about. Some may emphasize, say, medical causes, whereas others
emphasize social causes. The quest for a correct definition of the
“hyperactivity” problem could serve as an example. On the one
hand, there are convincing ADHD explanatory models that un-
derline the meaning of relationships and interaction. On the other
hand, there are equally powerful models that stress the significance
of hereditary elements—nowadays, the gene factor. The environ-
ment-or-genotype argument is readily available. Depending on the
definition, responsibilities are distributed in varying ways.
Along with seeking after the explanatory basis, the defining
parties position themselves as to who is the competent definer.
Language always depicts the speaker of the words and the “audi-
ence” of the utterances. Pierre Bourdieu (1993) argues that these
social functions may be more significant than the actual message.
Gregory Bateson (1972) points out meta-communication processes:
people communicate constantly—and not only verbally—and in
doing this, they also communicate mutual relationships.
When there are professionals present who are connected to the

matter from their own directions—according to their basic tasks—
they communicate not only the “matter itself” but also who they
are to be defining the matter. A physician may have a more or less
defined position within the health care system, the social worker
within the welfare system, and the teacher within education, but
what are their mutual relations when they meet in the no-man’s-
land between their agencies? Those relations have to be negotiated.
Anselm Strauss (1978) analyses the negotiated social order and
points out the active nature of mutual positioning both within and
especially between organizations. There is seldom an item “mutual
relations” on the agenda of a network meeting. Nevertheless, the
client case-matter will be there routinely, and mutual relation-
ships are, rather, discussed in the process of discussing the client-
matter.
The need to cooperate over sector boundaries is increasing.
Consequently, the forums for cross-boundary negotiations are mul-
tiplying. This means that, in particular, those situations where the
parties do not have defined mutual relations, or negotiated social
order, are becoming more common. When someone takes the floor
to discuss the situation, she/he inevitably communicates her/his
suggestions of her/his position as the definer of the situation. It
is done in tacit ways—stress and strain of voice, gestures, expres-
sions—rather than explicitly. Discussions run smoothly if the par-
ties are in agreement about the matter and the mutual relations.
The disturbances are not likely to be great if the parties disagree
about the matter but respect each other’s positioning as a definer
of the situation. A troublesome situation occurs if the parties mutu-
ally belittle both the definitions of the matter and their authority
as a definer. It is not necessary to express out loud the recognition
or the belittling of the other’s competence. The positioning will be
done by including or excluding the person’s view in some way or
another in the discussion.
Although the personal relations and characteristics are at issue,
the negotiated order cannot be reduced to them. The professionals
have different structural resources. Through the persons, the insti-
tutions are also interacting. When John the general practitioner and
Mary the social worker are interacting, so too are the health centre
and the social services centre. Defining mutual relations is not just
“chemistry”, but neither are the relations defined directly between

the institutions, over the heads of the very persons at issue. When
the client meetings discuss the client-matter, the social order of the
parties is negotiated at the same time.
Gender and attitudes towards gender are central in negotiated
social order. If we switch between “John” and “Mary” in the previ-
ous example, we get general practitioner Mary and social worker
John instead of doctor John and social worker Mary. In Finland we
hardly ever come across male social workers (outside substance-
abuse work, at least). But if we do, does it make a difference—and
does the gender of the general practitioner, psychologist, teacher,
or client make a difference? The actors make the difference. An
integral part of the negotiated social order is the negotiated gender
order. The gender order is hardly ever on the agenda in psycho-
social network meetings, but it is certainly always at hand. The
order will be negotiated continuously, as an important dimension
in defining who is competent to define the matter and the neces-
sary actions. And here, too, gestures, tone of voice, including or
excluding initiatives, as well as minor and major verbal means are
in use.
There are as many problems as there are actors
When professionals meet, the task is often to try to define the

problem at hand. A common definition of the problem is sought
so as to find the basis of coordination. However, there are no com-
mon problems to be defined. The problems of the various parties
can be connected and intertwined, even caused by each other,
but, ultimately, each actor has his/her own problems. Sometimes
quasi-clarity is achieved: the problem is defined in such a way that
it appears only as the problem of the client, not at all as a dilemma
for the professionals.
When, say, a doctor, a health nurse, a teacher, and a social
worker meet to discuss between themselves or with the family
what is the common problem in the son’s, mother’s, and father’s
situation, it may be lost that each actor has his/her own problems,
which are not interchangeable. The doctor’s problem in relation to
the family does not equate to the teacher’s problem—and for none
of the professionals is it the same as it is with the family members.
Even to the family members it is not exactly the same. Thus, strictly
speaking, there cannot be a “family’s problem”.
Let us assume that the father is dispirited and the other fam-
ily members are worried. The professional’s dilemmas in helping
the father and the other family members are not alike. They have
different basic tasks, different angles of approach, and different
contacts—even to different family members—and, therefore, very
different ways to be of help. Neither is the son’s problem the same
as the mother’s or the father’s problem. The problems of the dif-
ferent people interacting touch each other and in a way even cause
each other, but they are not congruent. The son may worry about
the father’s situation, and that may reverberate in his school at-
tendance and thereby manifest itself also to the teachers. But the
problem that this brings the teachers in their basic tasks is not the
same as the son’s problem. Neither are their problems exactly the
same as the father’s—or anyone else’s. The attempt to define the
problem as if it were something common to all dispels point-of-
viewness. Furthermore, it can be argued that only unique points
of view are open, since none of the persons share exactly the same
place in social relations.
Even when there are cases of a clearly diagnosable illness, the
problem is different for the various parties. The illness may affect
the family member’s life and lead to the need to see professionals.
Network relations may thus be formed. But the problem is differ-
ent for the patient and for the relative(s), as well as for the relevant
professionals. The possibility to define a common problem would
require a single point from which the situation would appear the
same for all parties. However, there are as many viewing points
and definitions as there are actors, and who would say whose is to
be accepted as the common or correct one?
A viewpoint always requires a point from which to view. The
French sociologist Pierre Bourdieu (1998) writes that each actor
occupies a point in social space. That point is the ground for the
gaze, the perspective that opens for that person, and the form and
content of that perspective is determined by the objective position
the person occupies. The point in the social space or the view that
opens from it are not voluntary. When several professionals meet,
they all have their own point of view—even in the literal sense,
as the point from which their perspective opens. This applies, of
course, to the clients and their personal network members, too—
that is, to everyone who is connected to the issue. Each subject has
his/her subjective point of view, which cannot be selected arbitrar-
ily. The points of view are not interchangeable, but each actor can
diversify his/her views by learning more about how the perspec-
tive opens up from the position of others. The others’ different
views offer a contrast, an observable difference that affords one
possibilities to learn more about one’s own point-of-viewness.1
Subjects may not have objective views—because views always
have a viewer—but the fact that subjects have views according to
their viewing points is objective. In hermeneutic philosophy objec-
tivity is understood as intersubjectivity.2
Point-of-viewness does not mean that any unexamined notion
could pass as the explanation. However, it does mean that even
the closest examination will not open the same perspective to
the actors. A bird’s-eye view is not available, but there is the op-
portunity to form richer views. Inspected facts are of help in this.
They afford points of reference. However, even inspected facts
cannot be adopted without a perspective towards them. They are
approached from a point of view. Even the most objective facts are
available for the subject within some activity, and they make sense
in that activity.
Actors regulate their stress load

Putting a name on a problem—for example, “ADHD”—does not
yet tell what the problem is. The condition has a name, and the
categorization leads to a distribution of responsibilities in the pro-
fessional system. Another categorization could lead to a different
distribution of responsibilities and tasks—problems, one could say.
When professionals negotiate their social order, sometimes through
competing as to who is the most competent definer of the problem,
they also “cause” problems for themselves and each other.
A recent study on Finnish rehabilitation cooperation showed
that a “common client” is likely to be a “nobody’s” client. Meet-
ings may end in as-if agreement that the client is the professional
network’s common client, but the follow-ups show that the client
fell through the net. The “shared” responsibility was not clearly
anyone’s (Kokko, 2003).
When multi-stakeholder situations are discussed, the issue is
also a binding force for each party with regard to the matter. If co-
operation is planned, each party has to weigh up its own commit-
ments. It is important to regulate commitments in psychosocial
work, where committing oneself means being exposed to not only
quantitative but also qualitative loading in the form of the mental
load one experiences in nursing, professional help, teaching, and
so forth. Therefore, one of the tensions in a network meeting is
between seeking solutions and regulating stress.
Psychosocial practitioners may be working in conditions where
increasing amounts of responsibilities have to be met with down-
sized personnel and budgets. When more or less worn-out profes-
sionals meet to discuss what ought to be done, there is not only a
competition of who will take on the problem to be seen, but also
attempts to control the case load.3
Multi-agency situations—where a number of professionals
work with the same client/family—call for getting in touch with
others, through phone calls, meetings, and so on. On the one hand,
there is more or less one’s “own” case load. On the other hand,
there are the clients who are also seeing other professionals. What
can one commit oneself to without being overwhelmed?
Network therapists have noticed that in network therapy ses-
sions the route to personal commitment goes via experiencing
hopelessness (Klefbeck, Bergerhed, Forsberg, Hultkranz-Jeppson,
& Marklund, 1988; Speck & Attneave, 1973). If one does not share
with the clients the feeling of how difficult the matters are, it is easy
to suggest to others what they should do. Such suggestions tend to
be exaggerated in relation to the circumstances.
In network therapy the process was called the “spiral process”
(see also Speck & Attneave, 1973). When the actual issue is taken
up in the session, after the joining phase, the views begin to po-
larize. Different and opposing points of view are expressed. The
network—the child’s family members, relatives, and other invited
personal network members, plus the professionals who are deal-
ing with the matter—begins to divide into “camps”. The process
is guided by a network therapist-team, and experienced therapists
lead the process to a head. This enables mobilization; the network

may be taken over by solution-seeking. After a while, the stage will
be reached where all the clever suggestions have been made and
still there are no solutions on the horizon. Hopelessness begins to
make itself felt. The matters appear as much more difficult than
expected. This feeling of getting stuck helps the participants to
share the emotional load of the persons who are in the middle of
the crisis: people start to feel personally what it is to be like in the
difficult situation—by feeling the feelings the persons concerned
are feeling. If the network does not give up—and this is what
the therapist-team will not let them do—there is a chance for a
breakthrough. Someone suggests something life-size but hopeful,
something she/he could do to alleviate the crisis. Another one
joins by adding her/his suggestion, and soon a third, and so on.
The network creates both routine and surprising solutions. Actors
like grandmothers and godfathers, whose resources have so far not
been in use, suggest how they could do their bit. Outlining a crea-
tive and plausible combination of resources lifts the network from
its distress and takes it to the last phases of the spiral: enthusiasm
and, finally, exhaustion at the end of the intensive process.
The process described above is a generalization of crisis-session
events. Crisis sessions, of course, have special tensions. But the
more customary meetings can also drift into a spiral process—
especially if the issues discussed polarize the group. Many pro-
fessionals have experiences of “innocent” meetings that become
hopeless: opinions split, arguments arise, suggestions are made
but none of them feel credibly viable—and sooner or later the par-
ticipants find reasons to leave the meeting “to attend to important
things elsewhere”.
As long as the participants suggest what others should do, they
may safely suggest even major changes without necessarily giving
credence to their initiatives themselves. In that sense, they remain
distanced from or outside the matter and the solutions. When one
falls personally into hopelessness and cannot leave the situation,
one becomes personally interested in factors that lead the way to
a brighter outcome. One’s personal hopelessness becomes a problem
that needs a solution. One ceases to be an outsider, since one is
not outside one’s hopelessness, the problem that has to be solved.
The solutions are often smaller in scale and more ordinary in
implementation than the suggestions one may have made as a non-

committed outsider. One may even come up with ideas of what
one could do oneself. If someone else suggests something personal
at such a juncture, something they could do themselves, one may
find ways to join in by adding ideas. This is how network solutions
emerge—through combinations of available resources.
To avoid committing oneself requires that one manages to
avoid sharing the emotional load of the distressed people. The profes-
sional system has a wide variety of means for this. Riitta-Liisa
Kokko (2003) analysed such means in encounters between clients
and multi-professional groups in rehabilitation settings in Fin-
land. She concluded that the interaction was often talking past
the other person: sometimes the professionals talk but do not act
accordingly, sometimes they act behind the clients’ backs without
talking about it, sometimes they are careful not to talk in order to
avoid obligations, sometimes they talk in a manner that shifts the
responsibilities to others, sometimes they talk and act together but
without the client.
A responsible practitioner commits her/himself to providing
high-quality services, and regulating one’s tasks and stress load is
an essential part of such responsibility. It is important to regulate
the case load in order to maintain the high level of work. Practi-
tioners at the grass-root level are in a difficult situation: unless the
management system prioritizes, the front-line workers have to do
it—and they have to do it case by case. Protecting the high level
of work through regulating case loads is, however, also paradoxi-
cal. If the professional manages to keep away from cooperation,
he/she may cause more future load for him/herself, which, in
turn, threatens the quality of the services. If matters are not dealt
with early enough, they may become a lot more difficult. In other
words, the client’s situation may be more and more difficult while
the professionals are busy regulating their commitments and shift-
ing the responsibilities from one to another.
Management by results in a sectored system encourages the
turning of a blind eye: “with luck” the problem manifests itself on
the other side of the sector boundary, in someone else’s in-tray. The
“silo system” stares at outputs at the other end of each silo and
overlooks the outcomes of the whole.
Delegating the control tasks is a common pattern in psychosocial

work.4 Help and support are often seen as positive tasks; however,
the professionals have to regulate their commitment to them, too.
Controlling, watching, and restricting are tasks that many would
prefer to leave for “control professionals”. Even in situations where
the professional feels that it would be necessary to restrict the
client’s dangerous or damaging activity, taking up the matter or
dealing with it is often evaded. This happens in particular in situ-
ations where the professional feels that her/his relationship to the
client is not “strong” enough to “bear” taking up control. The pro-
fessional may hope that the relation would grow stronger and that
someone else—for example, the child-protection worker—would
be active in controlling in the meantime.
However, as Barbara Cruikshank (1999) points out, there are no
social relations without elements of social control. Michel Foucault
(1980) emphasizes that practising power is not problematic as
such—but subjugating is. He points out that modern governance
prefers that citizens control themselves: the ideal is not external dis-
ciplining, but helping subjects to self-help. Professions with the prefix
“psycho-” (like psychologists, psychiatrists) or “socio-” (like social
workers) have their origin in restoring individuals’ social viability.
Surveillance and therapy gained ground over straightforward pun-
ishing. But how does one exercise control if the relationship with
the client is not strong? The temptation is to maintain the support
but delegate the control—that is, split the work.
All psychosocial professionals are under various forms of ob-
ligation to notify. The most common is the obligation to notify
child-welfare officials if the child is in danger. This means that all
the psychosocial professionals are controlling authorities. Since
there is no “outside” to control relationships, the crucial point is
how each professional carries out her/his control tasks. Helping
and supporting can similarly be either empowering or subjugat-
ing, and so too can controlling. Empowering help does not make
the receiver helpless; however, subjugating help makes the receiver
dependent. Empowering control enhances the other’s self-control;
however, subjugating control narrows down his/her autonomy.
Interaction patterns may be repeated
When professionals who are in contact with complex interaction

patterns through their client/patient relations meet to discuss cases,
it is not exceptional that the experienced interaction is carried over
to the relations between the professionals. A central dimension in
this process is identification (see Sachs & Shapiro, 1976; Stanton &
Schwarz, 1954), and it is not a professional flaw. However, it can
produce peculiar processes in network sessions, even deadlocks.
As we described earlier, Jaakko and his colleagues analysed
the systems of boundary that are in effect at the interface of the
hospital (team) and the family and the interaction processes that
may “conquer” the boundary system. The work of the team did
not manifest as treatment by virtue of the fact that the team was
named a treatment team. More important was whether it could
bring organization, flexibility, and multifaceted reflection into the
interaction. It might equally happen that the family’s crisis inter-
actions—the ways it deals with the crisis and the distress linked
to it—sweeps the team along. Tom applied the boundary system
concept for analysing the no-man’s-land situations where several
agencies and their professionals are connected through a common
case (in social-work-related situations, especially dealing with chil-
dren at risk). He became interested in from where the professionals
and teams in the various agencies drew their flexibility, organiza-
tion, and many-sidedness. This led to analysing patterns of man-
agement, resource allocation, and organizing work in connection
with the micro-interaction at the client interfaces. If organizations
and agencies are in a state of disorder, inflexibility, or one-sided
reflection, it is not evident that the professionals will bring order,
flexibility, and versatility into the interaction they take part in.
Central to this reasoning was the assumption about isomorphic
patterns—interaction patterns that become alike. The assumption
embodies the idea that interacting systems—such as a team and a fam-
ily—co-generate structure and become alike. Humberto Maturana and
Francesco Varela (1980), among others, argue that systems, if they
can connect, co-evolve and that both/all interacting parties change
in the process. Helen Schwartzman and Anita Kneifel (1985) ar-
gued that teams working with child-related matters replicate the
patterns their client families maintain: teams working with tightly

knit families tend to be quite tightly knit and somewhat distanced
from the rest of the professional scene, whereas teams working
with loosely connected families find it hard even to find shared
time for joint reflections—and these teams have huge case loads.
Tom and his colleague Esa Eriksson analysed the core pro-
cesses in isomorphic patterns and concluded that central to such
co-evolvement is the “contagion” of emotions (Arnkil & Eriksson,
1995). Persons have the ability to identify with each other—that
is, to feel each other’s emotions. Identification is essential for
mutual understanding: if one cannot feel how the other person
feels, one has only slight possibilities for understanding the other.
Understanding is not only a cognitive process: the other person’s
joy transmits, and so does horror; grief transmits, and so does
enthusiasm. Veikko Surakka showed in his study “Contagion and
Modulation of Human Emotions” (1999) that the transmission of
emotions is corporeal to a high degree; for example, people echo
each other’s facial expressions by diminutive micro-motions and
replicate the movements of speaker’s vocal cords. Unless they did
this, they would not understand the meaning of the expressions or
the words. Understanding through identifying is essential in all social
interaction. According to Surakka, people are able to “dispatch”
more than a hundred interaction signals per second. The phenom-
enal “receiver” that can read all that, even when there are several
parties in interaction, is human emotions.
Emotions are important explorers or scouts of one’s fields of
interaction —emotions “tell” what the situations and other people
are for the subject. As we described earlier, Galperin (1969) argues
that the psyche makes constant use of its cognitive, emotional, and
moral means of orientation to find out what will happen to the
subject in their potential fields of activity. Because it is possible to
identify with others’ feelings, they can be understood at least to
some extent. And when one understands the emotions of the other,
one can understand what direction the interaction may be taking.
In this way, the subject can at least faintly understand something
about what is going to happen to them.
When professionals of psychosocial work take part in net-
work meetings, they bring with them echoes of their client
relationships—of which they have more than just cognitive analy-

ses. In their client relationships, they constantly feel out—without
necessarily noticing—where the interaction is leading and what is
happening to them. They might have reached identifying under-
standing. They incorporate their identifications (sometimes unwit-
tingly) into their approaches to the discussions where their client
relationships are discussed. It is not strange, therefore, that the
interaction patterns the professionals have been in contact with
find their way into the network meetings, too.
Identifying with the clients and being entangled in the interac-
tion is essential for understanding. On the other hand, the interac-
tion patterns may become so alike that possibilities to be helpful
are lost. There is nothing problematic per se in identification or
repeating interaction patterns. Such activity becomes problematic,
however, if the professionals do not notice that they too are parties
in these humane processes, even as more or less passionate par-
ties in network meetings. It may happen that the opportunities to
enrich the points of view and to change the patterns of activity are di-
minished instead of broadened. There may be dominating parties
who try to make others understand what the matter is about and
what should be done about it. Or there may be parties that get no
response to their initiatives and are treated like children. Some may
join forces in order to increase the weight of their arguments, and
some others may whisper significantly and without giving much
value to others’ attempts to have a joint discussion.
Attempts to win others over to one’s own stance increase in
worrying situations, when one’s own subjective worries increase.
Network meetings that deal with considerable worries are particu-
larly prone to isomorphic processes. There are plenty of tensions in
customary multilateral meetings. Meetings that discuss emotion-
ally touching, worrying, professionally difficult situations without
panacea are particularly charged. Psychosocial work offers them in
abundance. Such situations are prone to monological attempts to
define for the others what they should think and do.
What feeds monologism in network meetings
We made four assumptions about factors behind frustrating net-

work meetings. The factors could be summarized as temptations
towards monologism:
1. Participants communicate relations and negotiate gender order.

When persons communicate, they position themselves in rela-
tion to others. When problems are defined, the definers also
define themselves as competent definers of the problem—and
the others as the audience for their definitions. Competition over
competencies may arise.
2. When multi-stakeholder assemblies seek ways of cooperating,
they often try to define a common problem as the basis for col-
laboration. However, there are no problems common to all. The
quest for common problems displaces the analysis of each participant’s
unique circumstances of acting upon the problem.
3. Participants explore individual commitments to joint activity,
and while it may be necessary to collaborate, it is also necessary
to protect one’s boundaries. The participants will both consider
their commitments and regulate their stress load.
4. The interaction patterns the professionals encounter through
their client work tend to recur also between professionals—even
when the clients are not present.
Notes
1. The anthropologist Renato Rosaldo (1989) argues that in the cultural
borderlands that form when cultures are in contact, one learns more about
one’s own culture than the foreign one—and that it is indeed impossible to get
to know the other thoroughly.
2. Tested knowledge, purified of subjective bias, is indispensable for mak-
ing accurate assessments. However, when a subject gets hold of such knowl-
edge, she/he cannot but form a subjective interpretation of the objective
source. The Russian psychologist A. N. Leont’ev (1978) points out that cultural
meanings (like the arguments in a book) are acquired as subjective senses. What
makes subjective sense and how is not determined by the cultural meaning as
such, but by life, the activity of the subject.
3. In Finland there were sharp cuts in the 1990s in practically all the psy-
chosocial branches, and they have not recovered yet. Most of the psychosocial
services are public, and this means that the agencies have to serve the whole
population and that the work is not a business for the professionals. Thus,
there is not competition for clientele.
4. Tom, together with Esa Eriksson, analysed such processes (Arnkil &
Eriksson, 1996). Their book is in Finnish, with an English summary.
PA R T II
CHAPTER THREE
Open Dialogues
as crisis intervention
W
e have described Jaakko’s first experiences in embedding
a network perspective into psychiatric practice. In the
very beginning, the boundaries and points of interaction
between the family and hospital team and across the professional
systems became relevant. The focus was on the collaboration with-
in the social network of the patient, not so much on the treatment
process itself. During 1988–91, the first large research project was
conducted (Keränen, 1992; Seikkula, 1991). It was the first window
into the content of interaction within a new, open system. The
dialogue ideas of Mikhail Bakhtin began to have meaning. As an
unexpected surprise, his ideas on literature and language research
seemed to fit our increasing understanding of the processes in
psychiatric treatment as well.
Jaakko, together with Jukka Aaltonen, analysed two years of
treatment processes of first-episode psychotic patients during
1985–94. In this qualitative study, some important elements of an
optimal treatment were highlighted. They compared the differ-
ences between patients admitted during 1985–89, when the system
was focused on hospital treatment, to patients admitted during
51
1990–94, when a new community psychiatric system had been

introduced.
During the later period, a system was followed in which the
same procedure is followed in all cases of psychiatric crisis, regard-
less of the specific diagnosis. If it is potentially a hospital-treatment
case, the crisis clinic in the hospital will arrange a treatment meet-
ing, either before the decision to admit in the case of voluntary
admissions or during the first day after compulsory admission.
At this meeting, a tailor-made team consisting of both outpatient
and inpatient staff is constituted. The team usually consists of two
or three staff members (e.g., a psychiatrist from the crisis clinic,
a psychologist from the patient’s local mental-health outpatient
clinic, and a nurse from the ward). The team then takes charge of
the entire treatment sequence, regardless of whether the patient is
at home or in the hospital and irrespective of how long the treat-
ment period is expected to last. In cases of other types of crisis,
where hospitalization is not considered, the regional mental-health
outpatient clinics take the responsibility by organizing a case-spe-
cific team, inviting members of the different facilities that are rel-
evant to the patient. For instance, in cases of multi-agency clients,
the team may consist of one nurse from the outpatient clinic, one
social worker from the social office, and one psychologist from the
child-guidance clinic. The principles of this organization have been
embedded in the entire state social and health-care network in the
province. In fact, the same concept is also applied in situations
other than psychiatric and social crises—for example, in organiz-
ing debriefing in different types of post-traumatic situations.
Several effectiveness and treatment-process evaluations of the
model had been made through action research (Aaltonen et al.,
1997; Haarakangas, 1997; Keränen, 1992; Seikkula, 1991, 1994;
Seikkula, Alakare, & Aaltonen, 2001b) By summarizing the ob-
servations contained in these studies, seven main principles were
summarized: (1) responding immediately; (2) including the social
network; (3) adapting flexibly to specific and varying needs; (4)
taking responsibility; (5) guaranteeing psychological continuity;
(6) tolerating uncertainty; and (7) dialogicity.
O P E N D I A L O G U E S A S C R I S I S I N T E RV E N T I O N 53
Practical guidelines
It is worth noting that the seven principles emerged out of research;

they were not principles planned in advance and then followed.
More general ideas on good treatment were added later. In the
following, we describe the principles as guidelines for treatment
focusing on dialogue. Although most of the studies have focused
on the treatment of psychotic problems, they are not diagnosis-
specific but describe an entire network-based treatment that is
especially practical in crisis situations.
Responding immediately
The best kind of start for treatment after a crisis is to act imme-
diately, and not, for instance, to wait for the psychotic patients
to become more coherent before arranging a family meeting. The
boundary for an immediate response has generally been accepted
as 24 hours. The units should arrange the first meeting within 24
hours of the first contact, whether made by the patient, a relative,
or a referral agency. In addition to this, a 24-hour crisis service
ought to be set up. One aim of the immediate response is to pre-
vent hospitalization in as many cases as possible. All, including the
psychotic patient, participate in the very first few meetings during
the most intense psychotic period.
A common observation seems to be that patients experience
reaching something that is unseen by the rest of their family.
Although a patient’s comments may sound incomprehensible in
the first meetings, after a while it can be seen that actually the
patient was speaking of some real incidents in his/her life. Often
these incidents include some terrifying elements and threat that
the patient has not been able or possible to express in spoken lan-
guage before the crisis. Psychotic experiences most often include
real incidents, and the patient is bringing forth themes for which
he/she has not previously had words. This is also the case in other
forms of difficult behaviour. In an extreme emotion, such as anger,
depression, or anxiety, the patient is speaking of themes that have
not previously been discussed. In this way, the main person in the
crisis—the patient—reaches something unreachable to others in

his/her surroundings. The aim in treatment becomes generating
forms for constructing words for those experiences that did not
have words or shared language.
During the first couple of days of a crisis, it seems possible to
speak of things that later are difficult to introduce. In the first days,
the hallucinations may be handled and reflected upon, but after
that they easily fade away, and the opportunity to deal with them
may not reappear until after some two or three months of individu-
al therapy. It is as if the window for these extreme experiences may
only stay open for the first few days. If the team manage to create
a safe-enough atmosphere by a rapid response and by listening
carefully to all the themes the clients speak of, then critical themes
may find a space in which they can be handled and the prognosis
improves. Later in this book (chapter 7), a case is presented that
indicates the danger if the team does not manage to respond to the
psychotic or confusing themes spoken of by the patients.
Including the social network

The patients, their families, and other key members of their social
network are always invited to the first meetings to mobilize sup-
port for the patient and the family. The other key members may
be representatives of other bodies, including State employment
agencies and State health-insurance agencies (whose task is to sup-
port vocational rehabilitation), fellow workers or the head of the
patient’s workplace, neighbours, or friends.
Social networks can be seen as relevant in defining the problem
itself. A problem becomes a problem after it has been defined as
one in the language of either those nearest the patient or by the
patient him/herself. In the most severe crises, the first notion of
a problem often emerges in the definition of those nearest the pa-
tients after they note that some behaviour no longer responds to
their expectations of them. For example, if a youngster in the fam-
ily is suspected of using drugs, the young person will seldom see
that as problem, but their parents can be terrified by the first signs
of possible drug abuse. Anderson and Goolishian (1988, 1992) said
that the one seeing the problem becomes a part of the problem-
defining system. From a network perspective, all these individuals

should be included in the process, because the problem becomes
dis-solved only if all that have defined it as a problem no longer
communicate it as such.
It is good to adopt a simple way of deciding who should be
invited to meetings. It can be done, for instance, by asking the
person who made the contact in the crisis the following questions:
(1) who knows of the situation and has been concerned; (2) who
would be of help and is also able to participate in the first meeting;
and (3) who would be the best person to invite them—the one who
contacted the services or the treatment team?
By doing it in this way, the participation of those nearest the
patients is suggested as part of an everyday conversation, which
decreases the possible suspicion towards the invitation. In this,
the one who has made contact with the services can decide who
they do not want to participate in the meetings. If the proposal for
a joint meeting is done in an official tone, by asking, for instance,
“Will you allow us to contact your family so as to invite them to
a meeting?” then problems have emerged in motivating both the
patients and those close to them, One element of deciding the
relevant participants is to find out whether the clients have either
contacted any other professionals in connection with the current
situation or have had previous contacts with helpers. All of these
actors should be invited, and the sooner the better. If it is not pos-
sible for the other professionals to attend the first meetings, a joint
meeting can be agreed for later.
The social relations of our clients can be included in many
forms. They can be present, or, if some of them cannot manage to
attend meetings, then the clients can be asked if they want others
who know of their situation and who could possibly help. Some
member of the network can be given a task of contacting them
after the meeting and relaying the absent person’s comments in
the next joint meeting. Those present can be asked, for instance,
“What would Uncle Matti have said if he was present in this con-
versation? What would your answer be? And what would he say
to that?”
The social network perspective is shifting the focus to the cli-
ent’s personal network, which may include all important people,
regardless, if they are family or not. The family is always relevant,
including the members of the extended family. Fellow employees

or schoolmates can be important in some circumstances, as can
friends and neighbours. The professional-network members can
discover new meanings in their contact with the client, with whom
they may have experienced difficult processes.
Adapting flexibly to specific and varying needs

The flexibility is guaranteed by means of adapting the treatment
response to the specific and changing needs of each case using
therapeutic methods best suited to each case. The treatment meet-
ings are, with the approval of the family, organized at the patient’s
home.
Each patient needs to be treated in a way that best suits their
specific language, way of living, possibilities for making use of
specific therapeutic methods, and the length of treatment time that
fits the actual problem. This is instead of applying a generic pro-
gramme without variation from case to case. The need is varying,
as well. During the first 10–12 days of a crisis, the need is quite
different compared to that of three weeks later. For instance, dur-
ing the most acute phase, it is advisable to have the possibility of
meeting each day, which will no longer be needed in more secure
situations later. In that later period, families will know how often
it is best for them to meet. These ideas follow the need-adapted
treatment first initiated by Yrjö Alanen (1997) and his team.
The meeting place should be jointly selected. In some situations
the best place is the patient’s home, if the family approves; in some
others, it might be an accident-and-emergency unit or a polyclinic,
if the family sees that as more suitable. Home meetings appear
to prevent unnecessary hospitalizations, since the family’s own
resources seem to be more available in a home setting (Keränen,
1992; Seikkula, 1991).
New ideas for psychosocial treatment of psychosis have been
developed lately. Main parts of new programmes follow an ill-
ness model, in which psychotic reactions are seen as signs of an
illness and of which it benefits the family to learn about so as to
avoid overstimulation and relapses. In these approaches, psych-
oeducational models are used. Families are informed about the
illness, and family members are trained in stress management for

family interactions. In most cases, this will involve a therapeutic
programme that is broadly followed in a similar way in each case.
Such programmes are relatively easy to follow scientifically, but
the problem remains of adapting them to individual needs. Fami-
lies can easily refuse to participate (Friis, Larsen, & Melle, 2003). To
avoid this, the need-adapted approach seems better at taking into
account the uniqueness of each treatment process. It seems to suit
the Nordic system, in which psychiatric units have total responsi-
bility for all clients in their catchment area.
Taking responsibility
Organizing a crisis service in a catchment area is difficult if all the
professionals involved are not committed to providing an immedi-
ate response. A good rule of thumb is to follow the principle that
whoever is contacted takes responsibility for organizing the first
meeting and inviting the team. The one contacting the professional
may be the patient her/himself, her/his family members, a refer-
ring practitioner, or other authorities such as, for instance, a school
nurse. Organizing a specific crisis intervention or acute team is
one possibility. All staff members will know where to contact if
they themselves have been contacted. This principle means that it
would no longer be possible to answer a request for help by saying
“this hasn’t got anything to do with us, please contact the other
clinic”. Instead, one can say, for instance, “it sounds to me that
alcohol abuse may be involved in your son’s problem. Would you
accept the idea that I invite someone from the alcohol abuse clinic
to join us in the meeting tomorrow?” In the meetings, the decisions
are then made as to who will best form the team that will take
care of the treatment. In multi-problem situations, the best team is
formed from professionals from different units—for instance, one
from social care, one from a psychiatric polyclinic, and one from
the hospital ward.
The team mobilized for the first meeting should take all the
responsibility needed for analysing the current problem and plan-
ning the future treatment. All that is needed for a good-enough
response is available in the room: there is no other authority
elsewhere that will know better what to do. This means that all
team members should take care of gathering the information they
need for the best possible decisions to go on. If the doctor is not
able to attend the meetings, she/he should be consulted from the
meeting by phone, and if there is a difference of opinion over the
decisions, a joint meeting is advisable to discuss the choices openly
in the presence of the family. This empowers family members in
participating more in the decision-making.
Guaranteeing psychological continuity

The team takes responsibility for the treatment for as long a time
as needed, in both the outpatient and inpatient setting. This is
the best way to guarantee the continuity arising in the very first
meeting, and it should form part of the entire process. Forming a
multi-agency team as such already increases the possibilities for
crossing boundaries of different treatment facilities and prevent-
ing drop-outs.
In the first meeting, it is impossible to know for how long the
treatment will continue. In some instances, one or two meetings is
enough; in others, however, intensive treatment for two years is
needed. Problems may occur if the crisis-intervention team meet
for three or five times and then, after that, refer the case to other
authorities. In these circumstances, even in the first meetings,
too much focus is on the acts taken and not on the process itself.
The representatives of the patient’s social network participate in
the treatment meetings for the entire treatment sequence, includ-
ing when other therapeutic methods are applied. The process of
an acute psychotic crisis can be expected to last for two to three
years (Jackson & Birchwood, 1996). In a study (Seikkula, Alakare,
Aaltonen, et al., 2003) (see also chapter 8), 65% of treatments were
discontinued at the end of the second year.
One part of psychological continuity is to integrate different
therapeutic methods into a joint treatment process so that meth-
ods do not compete with each other, but support each other. For
instance, if in a crisis meeting an idea is generated to have indi-
vidual psychotherapy for the patient, the psychological continuity
is easily guaranteed by having one of the team members act as
the individual psychotherapist. If this is not possible or advisable,

the psychotherapist could be invited to one or two joint meetings,
in which ideas are generated that can serve as the basis for an
individual-therapy process. The therapist should be invited every
now and then to joint meetings with the team and the family. Prob-
lems may have occurred if the individual psychotherapist has not
wanted to participate in the joint meetings. This has increased the
family’s suspicion towards the therapy, which has, in turn, been di-
rected to the joint process. This is particular important to consider
in the case of children’s and adolescent’s problems.
Tolerating uncertainty
The first task for professionals in a crisis is to increase the safety
of the situation, where no one yet knows the answers to the actual
problem. The aim then is to mobilize the psychological resources
of both the patient and those nearest to him/her so as to increase
the agency in their own life. This means creating stories about their
most extreme experiences. This is strengthened by building up a
sense of trust in the joint process. For instance, in psychotic crises,
for an adequate sense of security to be generated, this means meet-
ing every day at least for the first 10–12 days. After this, meetings
can be organized on a regular basis according to the wishes of the
family. Usually no detailed therapeutic contract is made in the cri-
sis phase, but, instead, at every meeting there is a discussion as to
whether and, if so, when the next meeting will take place. In this
way, premature conclusions and treatment decisions are avoided.
For instance, neuroleptic medication is not commenced in the first
meeting; instead, its advisability should be discussed at least three
meetings before implementation.
There are slight differences compared to other practices. In
some illness-orientated approaches, the uppermost task in the early
phase of treatment is defined as decreasing or removing the symp-
toms, and for this purpose, medication is prescribed; for psychotic
patients, this is neuroleptic medication. Medication can help, of
course, but the risk is that medication, by affecting symptoms, at
the same time removes psychological resources. Neuroleptic medi-
cation has a sedative effect that calms down psychological activity,
and thus it may be a hindrance to psychological work. The chal-

lenge is to create a process that, at the same time, increases safety
and encourages personal work. It is good to consider permanent
medication in at least two or three meetings before starting. This
conclusion is verified in the studies we describe in chapter 8.
Besides the practical advice of seeing that the family is not left
alone with its problems, increasing safety means generating a qual-
ity in the therapeutic conversation such that everyone comes to be
heard. In a crisis with loaded emotions, working as a team is one
prerequisite. One team member may start to listen more carefully
to what the son says when the son is saying that he does not have
any problems—it is his parents who need the treatment. The other
team member may become more interested in the family’s burden
in not being successful at stopping his drug abuse. Already in the
very first meeting, it is good to spare some time for reflective dis-
cussion among the team from these different or even contradictory
perspectives. If the team members can listen to each other, it may
increase the possibilities for the family members to listen to each
other, as well.
A situation in which professionals are in a hurry to get to the
next meeting and therefore propose a rapid decision is not the
best possible way of making use of the family members’ own
psychological resources. It is better to define the situation as open.
One way to put it into words might be: “We have now discussed
for about an hour but have not reached any firm understanding
of what this is all about and what the best way of going ahead is.
But we have discussed very important issues. Why not leave this
open and continue tomorrow?” After that, concrete steps should
be agreed on before tomorrow’s meeting to guarantee that family
members have a feeling that they know what they should do if
they need help.
Dialogicity
The focus is primarily on promoting dialogue and secondarily on
promoting change in the patient or in the family. Dialogue is seen
as a forum through which families and patients are able to acquire
more agency in their own lives by discussing the problems (Haar-

akangas, 1997; Holma & Aaltonen, 1997). A new understanding
presupposes a dialogical conversation. New meanings are gener-
ated as if in the area between those participating in the discussion
(Andersen, 1995; Bakhtin, 1984; Voloshinov, 1996). For a profes-
sional, this all means inviting new aspects for being an expert on
which clients can trust. Professionals have to become skilful in
dialogues through which their specific expert knowledge becomes
rooted in the context as well.
Treatment meeting as joint wondering
The main forum for dialogues is the treatment meeting. As said,

here the major participants in the problem together with the pa-
tient gather to discuss all the issues associated with the actual
problem. All management plans and decisions are also made with
everyone present. According to Alanen (1997), the treatment meet-
ing has three functions: (1) to gather information about the prob-
lem, (2) to build a treatment plan and make all decisions necessary
on the basis of the diagnosis made in the conversation, and (3) to
generate a psychotherapeutic dialogue. On the whole, the focus is
on strengthening the adult side of the patient and on normalizing
the situation instead of focusing on regressive behaviour (Alanen,
Lehtinen, Räkköläinen, & Aaltonen, 1991). The starting point for
treatment is the language of the family: how each family has, in
their own language, named the patient’s problem. Problems are
seen as a social construct reformulated in every conversation (Ba-
khtin, 1984; Gergen, 1994, 1999; Shotter, 1993a, 1993b, 1997). Each
person present speaks in his/her own voice(s), and, as Anderson
(1997) has noted, listening becomes more important than the man-
ner of interviewing. Team members can comment to each other on
what they hear as a reflective discussion while the family listens
(Andersen, 1995).
The meeting takes place in an open forum. All participants sit
in the same room, in a circle. The team members who have taken
the initiative for calling the meeting take charge of conducting the
dialogue. On some occasions, there is no prior planning regarding

who will take charge of the questioning, and thus all staff members
can participate in interviewing. On other occasions, the team can
decide in advance who will conduct the interview. This is the best
option when the treatment unit is accustomed to conducting fam-
ily meetings in a structured way. The first questions are as open-
ended as possible, to guarantee that family members and the rest
of the social network can begin to speak about the issues that are
most relevant at the moment. The team does not plan the themes
of the meeting in advance. From the very beginning, the task of the
interviewers (or interviewer) is to adapt their answers to whatever
the clients say. Most often, the team’s answer takes the form of a
further question, which means that subsequent questions from
team members are based on, and have to take into account, what
the client and family members have said.
Everyone present has the right to comment whenever she/he is
willing to do so. Comments should not interrupt an ongoing dia-
logue, and the speaker should adapt his/her words to the ongoing
theme of discussion. For the professionals present this means they
can comment either by inquiring further about the theme under
discussion or by commenting reflectively to the other professionals
about what they have started to think in response to what is said.
Most often, in those comments, new words are introduced to de-
scribe the client’s most difficult experiences. When the staff mem-
bers have to point to their obligations, it is advisable to focus on
these issues towards the end of the meeting, after family members
have spoken about what are the most compelling issues for them.
After deciding that the important issues for the meeting have been
addressed, the team member in charge suggests that the meeting
may be adjourned. It is important, however, to close the meeting
by referring to the client’s own words, by asking, for instance, “I
wonder if we could begin to close the meeting. Before doing so,
however, is there anything else we should discuss before we end?”
At the end of the meeting, it is beneficial to briefly summarize the
themes of the meeting, especially whether or not decisions have
been made, and, if so, what they were. The length of meetings can
vary, but usually an hour and a half is adequate.
How to increase the polyphony

and tolerance of uncertainty?
Facilitating dialogue among a multiplicity of voices is not, in the
main, achieved by training in specific interview techniques or
interventions. In fact, following some specific interview technique
may hinder dialogue, since one of the most basic ideas is that every
interlocutor adapts his/her words to what was previously said.
One can learn to support dialogue by focusing more on one’s own
way of speaking. From our background in the most severe psychi-
atric crises, we propose the following guidelines:
• Make sure that every participant in the meeting has a chance to

say something as early as possible. First, everyone is asked to
give her/his view of the present situation. Most often the very
first utterances are monological, since they have not yet been
shared with other participants. If team members show that
they are interested in listening to what is said and value what a
participant has said, then those present start to become curious
about each other’s utterances as well. As said, this is especially
important with conflictual opinions or if psychotic things are
said.
• Fit your very first comments to what the other has said. You can,
for instance, before going on, repeat word for word what was
said: “You said that it was terrible when father moved away?”
After this it is good to leave a short pause in speech to give
the client space to consider if this was really the thing she/he
wanted to say. If it is, the speaker can hear her/his words back
from another and consider what it actually means to hear those
words. This short sequence constructs a joint language area
between the team, family, patient, and others present. In this
shared language area, it becomes possible to understand issues
and treatment decisions in the same way.
• Do not interpret or “reality orientate” psychotic comments but,
instead, ask for more information about the patient’s experi-
ences. Discussion of the psychotic experiences can take the very
same form as any other comments. If the patient or family mem-
bers prefer that the experiences are discussed as symptoms of
illness, then the professional may also use this same language.
Furthermore, If the speaker is expecting an answer on whether

the experiences are real or hallucinations, one possible way to
answer might be that “I have not experienced something like
that. Could you tell more about them so I can understand more
of your experiences?”
• Reflect your observations and thoughts with other profession-
als. During this type of reflection, more space is created for
alternative descriptions and solutions. In this way, the family
role is emphasized, because now it is more a question of choices
made in life instead of treatment decisions known in advance.
If the family is speaking a concrete language, try to adapt your
words to be concrete as well.
If the discussion is stuck, you can create a surprising question,

such as with regard to differences in how the situation is perceived
by different family members. One can use circular questions—for
example, the three-way wondering about each other’s perspectives
can be elicited, for instance, by asking “after you got these prob-
lems, how did it affect your father and mother’s relationship?”
Surprising comments or questions may also include positive sides
to the family’s extreme life situation: “When you said for how long
you have been struggling with these difficulties, I started to think
of the enormous resources you have built up to handle all this.
Many people lack such resources”.
These should be seen as examples, not as concrete guidelines
for how to formulate questions and comments. Dialogue is always
created in the specific situation. This is explored further in chapters
6 and 7.
CHAPTER FOUR
Anticipation Dialogues
for lessening worries
I
n the previous chapter we described Open Dialogues, which
are well suited to psychiatric-crisis situations. In this chapter
we describe Anticipation Dialogues, which are appropriate for
less acute situations. Anticipation Dialogues are not a treatment
practice. They are a means for carrying out network meetings in a
manner that helps the participants find ways of coordinating their
actions. Empowerment is also an aim.
Anticipation Dialogues make use of facilitators who are out-
siders in relation to the matter under discussion.
Anticipation Dialogues are useful especially when:
• there are multiple parties involved in the matter

• it is unclear what each party is doing or even who are the parties
involved
• the actors are dissatisfied in what the others are doing
• worries grow and resources ought to be combined to lessen
them, but
• coordination is somehow lost.
65
Besides anxious situations of child-, adolescent-, family-related

matters, Anticipation Dialogues have been carried out in tangled
situations in senior-citizen care—for example, complicated nego-
tiations about discharge or institutionalization—and in work
with the long-term unemployed. Long-drawn-out and helping
situations that have become “stuck” are the terrain of Anticipa-
tion Dialogues. The “method family” of Anticipation Dialogues
also includes variations for negotiations outside immediate client
work.1 However, we shall not deal with them in this book. In the
following, the focus is on Anticipation Dialogues involving fami-
lies and their helpers.
The initiative to seek a network meeting

is a plea for help
The initiative for assembling a network meeting is made by some-

one who is worried about the well-being of a child or an adolescent.
A social practitioner, a health nurse, a doctor, a teacher, or some
other professional working with the child, the adolescent, or the
family may feel that their worry will not lessen unless there is im-
provement in the cooperation within the professional network and
with the family. Clients can also suggest a meeting. The dialogue
will be carried out through applying a method called recalling the
future, and is conducted by a facilitator pair. The pair will put a set
of questions to each participant to reflect by thinking aloud. Antici-
pation Dialogues (which apply the future-recalling method) target
the near future: the aim is to make a joint-action plan for lessening
the worries. However, there is also an immediate aim that has to
do with the dialogue situation itself. To be heard and listened to
and listening to others may have empowering effects in the “here-
and-now”. The cooperation plan is devised in such a way that the
process itself can give the people concerned hope and energy.
The first step is acknowledging the worry with the parent(s)/
caregiver(s) and suggesting the meeting. It is important to notice
that the professional practitioner is requesting help in lessening
her/his own worry. If the practitioner feels that the situation of the
child or adolescent does not improve even though the practitioner
A N T I C I P AT I O N D I A L O G U E S F O R L E S S E N I N G W O R R I E S 67
is doing all she/he can, the worry about the child/adolescent has
to do with the practitioner, too. The practitioner is concerned with
what is happening with regard to her/his possibility for helping and,
consequently, worried about what is happening to her/himself as
a helper. Professional help does not take place in a vacuum. One
may experience that the ability to be of help is ebbing away if one
does not get a positive response or support from others. Therefore,
there is often at least a trace of a network in the worry of the prac-
titioner—a network of people that the practitioner views as con-
nected to the matter. They may be family members, other people
in the private network, other professionals working with the child,
adolescent, or family, and so on. The practitioner may have tried to
be in contact with these parties, perhaps one at a time, but without
any positive effect on her/his worries.
Preparations for Anticipation Dialogues begin with a discus-
sion with the parent(s)/caregiver(s). If the practitioner suggests a
meeting by way of a practitioner’s plea for help to the caregivers, tinges
of blame can be avoided. The plea is sincere if the practitioner re-
ally feels she/he needs the caregivers’ help so as to lessen her/his
own worries. If one requests help for oneself, the tone is very differ-
ent from taking up the matter as a demand for others to change. The
tone and the message are very unalike in the following moves:
—“It worries me whether my attempts to help have really been

helpful enough to your child. I would need to be clearer about
the context of the work. Could we call the practitioners that
are connected with the matter and your family and perhaps
people you feel are close and supportive? We could plan how
to support the well-being of your child together. This would
lessen my worries.”
—“You and your child’s problems are so complicated that we
need to summon a group of experts to determine what the
problems are and what measures need to be taken. It would be
of help if your family were present.”
The initiator—the person who originally takes the initiative to

call the meeting—needs help in conducting the meeting. The initia-
tor calls the professionals and the caregivers summon the personal
network, but it is not advisable that the initiator has a conducting
role in the meeting itself. In relation to the meeting, the initiating

practitioner is the client (who “ordered” services”).
Outsiders—persons who are not involved in the case—are
needed to facilitate the dialogue. All the professionals that are
invited are involved, and so are the family and personal network
members. Therefore, the initiator asks for two facilitators. In Fin-
land, “facilitator pools” are now being organized.
The initiator is usually a practitioner worried about the coop-
eration and his/her own possibilities to be of help and about what
is happening to the child/adolescent. The initiative can also be
taken by a caregiver who wishes to have the overall picture clari-
fied. If the initiator is a practitioner, he/she naturally describes the
dialogue to the clients. They will, of course, need to have a general
outline of what will take place.2 The initiator asks the clients whom
they would like to invite from their personal networks, discussing
with them who could be helpful in planning a cooperation that
lessens the concerns. The initiator also discusses with the clients
which professionals from various agencies are connected to the
case and which of these they think should be invited to the dia-
logue. The group of people invited is, thus, for the clients to deter-
mine. Only those agreed upon with the clients will be invited.
Dialogue facilitators
The persons conducting Anticipation Dialogues are peers and col-

leagues, not experts “above” the rest. In Finland they are therapists,
social workers, youth practitioners, doctors, teachers, kindergar-
ten workers, health-nurses, midwives—that is to say, practition-
ers from all branches of psychosocial work. They have received
a two-year on-the-job training from Tom’s team at STAKES. They
facilitate the dialogues as pairs.
It seems to be fruitful for the Anticipation Dialogues that the
facilitator pair are sufficiently detached from the matter under
discussion. What is deemed sufficiently detached and whether
the facilitators are outside the matter enough is for the clients to
determine. The initiator and client together ensure that the avail-
able facilitators are not involved in the case in such a way that
the clients are uncomfortable with. If they are, another pair will
be suggested. Such an outsider position is necessary for clarity. If
the facilitators are, in fact, involved with the case but facilitate as
if they were not, they will be fulfilling two roles, and that does not
contribute to clarity. Experience shows that the outsider position is
more necessary the greater the concerns and the more intense the
network members’ attempts to change each other. In such a case,
there are good reasons to suppose that the ways of interacting that
were already in place will recur in the network meeting too. If the
facilitators are part of such patterns themselves, their capacity to
carry out a different way of interaction—the future recalling—is less
than for a pair that has not been involved in the case. When not-so-
significant worries are discussed, a sufficiently detached position
can be reached simply by agreeing that two people are “freed” to
conduct the dialogue for a while. Many planning sessions are, in
fact, like this.
Client feedback indicates that the facilitators’ outsider position
is valued highly. It is seen to increase feelings of security. In order
to preserve that position, the facilitators have to avoid two things
in particular. First, they must not acquire or hear too much advance
information about the family and the work done with the family.
It suffices that they know what the initiator’s worry is and what
sort of assemblage has been invited. Voluminous descriptions and
pieces of information tempt the facilitators to make interpretations,
but interpreting the situation is not the task of the facilitators in
Anticipation Dialogues. It is clear that they form interpretations
even from scant elements; they construct meanings as all people
inevitably and constantly do. But the more detailed the “case-his-
tory information” the facilitators are offered, the more difficult it
is for them to stick to asking non-suggestive questions. Second,
the facilitators have to refrain from giving advice or expressing
opinions and thus involving themselves in the handling of the
“case”. The fact that Anticipation Dialogues progress as interviews
makes such refraining easier. The task of the facilitators is to ensure
that the dialogue enables those present to be heard, that manifold
impressions are obtained, and that planning cooperation to lessen
worries develops as well as possible. That is it. The facilitator pair
are not in an expert position of handling the “case”, but are there
to foster dialogue.
William Isaacs (1999) opposes the use of facilitators for dia-

logues. He says that it is important that the work community
encounters its crisis—the bewilderment that fills the mind when a
person understands that his/her expectations will not be fulfilled
and that a group cannot fully control the outcomes. The facilitat-
ing that Isaacs criticizes is of a different kind to that carried out
by Anticipation Dialogues facilitators. Isaacs refers to facilitators
who offer analyses and visions for “smoothing” the discussion. In
Anticipation Dialogues, it is vital that the facilitators do not inter-
pret on behalf of the participants. Giving advice or summarizing
explanations is precisely what they do not do.
Facilitator training emphasizes ways of facilitating, talking, and
listening in multi-stakeholder impasses. Such situations are often
charged. Various parties may have been trying to resolve the di-
lemmas for some time already. The expectation is for others to
change, while there is perhaps disbelief that change is possible.
Open Dialogues, the approach described in the previous chapter,
is a practice developed for crisis situations. Anticipation Dialogues
are carried out not in acute crises, but in multi-agency situations
that have got into a rut. They are a more “chronic” than acute. The
parties may not be in a state of high alarm, but worries smoulder
and eat away the confidence of those involved.
When the dialogue is arranged on an agreed date, it begins with
various ways of joining (warm-up activities)—that is to say, of al-
leviating tension. Many of those present do not necessarily know
each other, and the meeting is called because of an accumulation of
worries. The initiator opens the occasion by thanking those present
for coming to help in lessening their worries. They describe his/her
worry briefly and hand over the conducting role to the facilitators,
who will organize the dialogue.
The pair of facilitators are, in a way, in a service role. They facili-
tate—that is to say, foster—dialogism. They are employed3 in order
to get the situation unstuck or to avoid it getting even more stuck.
Their service to the networks can be summarized as follows:
• In the first place, the facilitators organize the process. Thus they
lighten the load of the participants. In meetings of several peo-
ple it is not easy to make one’s voice heard—especially if one is
nervous about how matters will be dealt with. The facilitators
structure the dialogue into turns of talking and listening. Each

person will have the space to think aloud without others com-
menting—and, correspondingly, the others have a chance to feel
out their impressions without interruption.
• The facilitators help to plan the joint activity. They inform the
participants that at the end of the meeting, enough time will be
allowed for discussion on what will be done next and who will
do what with whom. The participants can enter into reflection
and listening in peace. The facilitators try to make sure that the
meeting will discuss and produce practical conclusions.
• The third kind of support is that the very thinking process is
facilitated. It is not easy to start thinking aloud in a big crowd—
especially if one is a non-professional and there are several ex-
perts present (some of whom may think critically of the family’s
activity). It is not necessarily easy for the family members to ex-
press thoughts about worrying situations between themselves.
Neither is it easy for practitioners who may feel that they are on
the defensive or are swimming against the tide of opinion. As
a whole, thinking aloud in public is a demanding task even if
there are no tensions. Thoughts are not there ready in the mind,
and words are not there waiting on the tongue. Rather, thoughts
form in the process of speaking, and thinking aloud gives one an
opportunity to get hold of one’s thoughts. Therefore, Anticipa-
tion Dialogues aim at creating a favourable situation for think-
ing aloud and getting to know one’s own thoughts. In addition
to that, everyone’s turn is shielded from comments and inter-
ruptions; the person thinking aloud is helped with questions
when necessary as part of the interviewing process. It is easier
to answer thought-evoking questions to one person than to start
thinking aloud without aid in the presence of a large audience.
The thinking process is also aided by the fact that everyone
hears each others’ spoken thoughts and thus gets impressions
to consider.
• The questions are open; they do not have right or wrong an-
swers. They are also quite surprising, for the participants are
requested to “recall” the positive future and the actions that
have led there. The answers to the questions are subjective, and
they await other’s subjective responses as complements. The
answers form a loosely woven pattern of the near future. An es-

sential service by the facilitators is to offer the future-recalling
format as a tool for thinking together.
• The fifth kind of support to the participants is, thus, that the
facilitators try to further new understanding between people:
the forming of shared understanding. When all hear each oth-
ers’ spoken thoughts, all will have those elements in their inner
dialogues. Though it is not likely that there are even two peo-
ple among the participants who understand the matter alike,
all have had an opportunity to form their understanding in a
shared situation and with the voices of the participants echo-
ing each other. The situation would be totally different if the
facilitators went and interviewed the same parties separately,
asking the family and private-network members and profes-
sional-network members the questions and hearing the answers
without the parties hearing each others’ spoken thoughts. The
facilitators would be likely to have a lot of voices in their inner
dialogues, but all others would lack precisely that most essential
aid to mutual understanding.
• The facilitators also further the opportunities for the participants
to talk about their present hopes and worries. There are very few
people who think precisely how their life will be in (say) a year
and which of one’s own and other people’s actions will take one
there. Both the aim and the means are more likely to be more or
less sketchy (with the exception, of course, of such areas of life
like work, which may sometimes have to be planned step by
step). However, most people probably have hopes and worries
about the near future. The facilitators help the participants to
think about these, as well as about the actions that would lessen
their worries. This is done by “taking” the present “into” the
near future as if one were already “there”. It is different look-
ing at matters of the present from a good situation in the future,
rather than viewing the future from a problematic present situa-
tion. The former gets its driving force from subjective outlining
of a desired situation, the latter from a focusing on undesired
phenomena. When matters are looked at from a relieved per-
spective, worries can be expressed more freely. Namely, if one
starts from present worries, all participants will have to consider
that the persons with whom the worries are connected are not
offended. If the starting point is a situation where the worries
have been lessened, “recalling” the path there is likely to have
less offending tones.
• One of the services of the facilitators is, therefore, that through
Anticipation Dialogues an opportunity is afforded to think about
a positive future as actually possible. This can have empowering
effects.
• The facilitators also help to bring the family’s (multi-voiced)
future into the centre. The focal point of professional measures
will be the family members’ positive everyday life.
• The facilitators try to aid the making of a concrete plan in which
the initiator (the worried person who initiated the meeting)
will not be left alone—nor anyone else, and especially not the
child.
• Although the perspective in Anticipation Dialogues is in the
future—that is, the meeting aims at making a worry-lessening
cooperation plan—there is also an immediate aim. The process
also targets the very moment of the dialogue. At their best,
Anticipation Dialogues help the participants to get a handle on
their hopes and worries, to feel heard, to hear their thoughts
echo in what others say, to gain more understanding of the
viewpoints of the others, to feel the participants share under-
standing, and to experience a mutual support spreading in the
network. That means also being emotionally touched.
The process of recalling the future
We briefly described the process of recalling the future in chapter

2. Now we shall represent it in some detail and make some sup-
plementary observations about the intentions underlying this ap-
proach.
The facilitators work as a pair. One interviews, the other keeps
a record (they may, of course, take turns). The records are made as
public notes—for example, on a flip chart. The notes are necessary
in order to help make a plan at the end where everyone can find
his/her own focal point(s).
The participants are organized around two adjoining tables—
family and friends around one, professionals around the other. It
is important that everyone can hear and see each other. The fam-
ily group consists of those persons the family members invited
from their personal networks. The professionals are those who are
personally connected through their work and whom the family
wanted to be present.
The interviewing starts with the family group. It is first agreed
which is the appropriate order of turns. Cultural and ethnic mat-
ters are taken into consideration. Within mainstream Finnish cul-
ture, it seems acceptable to start with the children instead of the
parents or to interview the mother before the father. In some
cultures, that would be inappropriate. Starting the process with
the family group is important because it is exactly the family’s
everyday life that can provide and should be the hub or the focal
point of support measures. If the professional measures do not fit
the family’s everyday life, even excellent measures are not neces-
sarily helpful. The picture of the family’s positive future has many
shades: each family member has his/her own view about it, from
his/her own unique viewpoint. Therefore, they are interviewed
separately—with all others listening, of course. The facilitators
agree with the family what is a suitable timeframe for viewing the
future. Often it is a year “ahead”.
Case illustration
In an Anticipation Dialogue with a Muslim family, the facilita-
tors had been prepared to start with the father, and this was, in-
deed, the way the family wanted it. When the facilitator asked
the father the first question through an interpreter—“a year has
passed and things are quite well in your family; what are you
particularly delighted about”—the father refused to answer.
The interpreter explained: a good Muslim cannot answer such
a question—the future is in the hands of God. The interpreter
advised: “Ask ‘God willing’ . . .”. The facilitator did that, and
the interview passed off excellently.
The thought of the present in the near future
The facilitator’s questions are presented here in a reduced form. In

actual dialogues, the questions are not a mechanical set, applied
independently of the situation. However, there is a certain core that
is linked with the facilitators’ facilitating task. They are in service,
supporting the formation of mutual understanding. The skeleton
model we present highlights this core.
The first question takes the viewpoint into the future. The core
of the question is: “A year has passed and the state of affairs in
your family is quite good. How is it from your point of view?
What are you particularly delighted about?” When the issue is a
worry attached to children/adolescents, the fixing point (of view-
ing) is especially in the alleviation of their situation. The future is
“entered” in order to find the actions and the support that could
make their situation easier, and then to make a plan from out of
them.4
Not just the future but a good future is envisaged in order to
let the alleviation of worries illuminate the present, instead of
the present worries overshadowing the thinking about the fu-
ture. Although the imagined present is set in the near future, the
participants are, in fact, thinking aloud about today’s hopes and
worries—about their future.
Each person is asked to think about matters from her/his own
point of view, because each person truly has only her/his point
of viewing. At the same time, such an address conveys a mes-
sage: your point of view is valuable. Every now and then the fa-
cilitator asks “did I hear you correctly when I heard you say. . . .”
The facilitator repeats the utterances verbatim, using the person’s
verbal expressions, without trying to interpret, although perhaps
condensing the sentences. This is to help the participants to think
about their thoughts. Hearing their own words echoed, they can
catch what they just thought aloud. The feedback (“did I hear cor-
rectly when . . .”) also communicates that the intention really is
to listen to the person—and to hear specifically what the person
says—without trying to interpret the meanings behind what is said.
As the facilitator interprets each individual in their turn, in close
eye-contact, putting the questions precisely to them, and listening
to them carefully, a “talk-to-me” situation is formed. It is as if two

people meet and others around them sort of fade away, although
the bodily presence of the others is felt and their presence echoes
in the thoughts of the person interviewed. This “one-to-one-ness”
aims at facilitating thinking aloud in a multi-actor situation.
Mikhail Bakhtin has said that each utterance is a response to
another utterance and the person who uttered it (see chapter 7).
That “other” may not necessarily be present. An utterance in this
moment may, for example, be a response in a conversation that
the person was engaged in in another circumstance yesterday. In
Anticipation Dialogues, the process of addressing is indirect. The
facilitator asks, and the other person, in turn, answers. Everyone
understands, however, that those present to whom the words
are meant as responses will hear the words. The utterances are
responses to something said during someone else’s turn and may
also be a continuation to a conversation that took place in another
context. Herein is one of the particular powers of Anticipation
Dialogues. On the one hand, the individual’s chances to talk are
facilitated by creating a one-to-one “talk-to-me” situation and by
putting questions that do not have right answers. Everyone’s own
subjective view is what is particularly interesting. On the other
hand, all this is created in a situation where everybody knows that
all will hear all the utterances. In this way, the presence of others
echoes in everyone’s talk.
Case illustration
In one Anticipation Dialogue, the mother listened to her son
with an amazed expression on her face. The interview had
started with the son. The single mother—and the professionals
present—heard the boy say that “now that a year has passed
and things are quite well, I’m a vocational school student”. The
mother’s astonishment only grew when the boy also told what
he had done to achieve this and with what support—from his
mother among others. When the facilitator asked “what were
you worried about ‘a year ago’ and what made your worries
lessen”, the boy replied that “a year ago” he had been worried
about the fact that his mother never listened to him but always
scolded him for doing nothing. “What made that worry lessen”,
the facilitator asked. “The fact that my mother for once heard
what I want.” “When was that”, the facilitator asked. “When
you asked me the questions ‘a year ago’” the boy replied.
If the person addressed “via” the facilitator is present, the situation

may be blocked if the person talking has to watch his/her words.
The Anticipation Dialogue questions are precisely about a good
future and the lessening of worries. The way of asking is designed
and refined with the intention of increasing the person’s chances
to talk about his/her worries. We shall come back to the worry
question a bit further on.
While the facilitator interviews, the co-facilitator takes public
notes about the good future on a flip chart or any means that al-
lows them to be seen by all. The co-facilitator tries to capture the
actual utterances used. When writing them down, the facilitators
ask every now and then whether the notes reflect the thoughts well
enough. Public notes are important not only for having a record of
the elements for a collective study at the end of the meeting, but
also for transparency and clarity. Everything takes place on view;
documents are not worded somewhere behind people’s backs.
The positive-future question can be asked in many ways, and
in the actual dialogue situations there are several specifying ques-
tions. The facilitator does not repeat the set of questions me-
chanically, but generates ways of asking so that the essence will be
realized. The essence is to help people to think and express their
hopes and worries, to be heard, and to get impressions of others’
thoughts and utterances in order to plan activity that increases
hope and lessens worries. Approaching the present from the future
is a means for this, not an end. People are helped to think about
(pleasing) matters of everyday life because the support of every-
day life is the rationale of the professional help. It is important
to talk in ordinary language—that is, to avoid professional lingo.
People are not interviewed as figures in institutional roles (pupils,
patients, clients) but as people living their lives—who may, from
time to time, do business with educational or medical institutions,
welfare agencies, or the like.
An important prerequisite for dialogism in Anticipation Dia-
logues is that the facilitators really listen very carefully to what
each person says and that everyone is interviewed as an individual.
The facilitators’ way of asking is their dialogism. Dialogism entails

responsiveness, directedness to replies, and reciprocity. Therefore,
the questions form as a response to what the participants say. A
skilful facilitator asks in every Anticipation Dialogue about the
positive future, actions, and support and about the worries and
how they are lessened, but he/she speaks in each dialogue specifi-
cally to the unique person—listening to that person’s expressions
and proceeding according to them. The fact that facilitators do not
need to—and indeed must not—make interpretative summaries
or explanatory conclusions of the utterances helps them to be
responsive. They can remain open to what each person expresses
and concentrate on their service task: to help in listening and fa-
cilitating the participant to feel listened to.
Each family member and the others from the family’s private
network are interviewed—adults and children—in this same man-
ner. It often seems that the family members are hearing each other’s
thoughts about a positive future as if for the first time. It may well
be that families do not make annual plans through discourse. But
it also seems that if there are discussions about the future and if
there are worries about it, those worries feed the need to talk rather
than listening. Especially when worries increase, listening tends
to be on the losing side. There is no commenting in Anticipation
Dialogues, so that each person has the peace to think aloud in turn
and to listen during other people’s turns. This is also a way of re-
specting each person’s subjective views on the positive future. The
pictures conjured up around the family-group table may have a lot
in common, but they may also differ a lot. A single picture is not
even aimed at. A multi-shaded picture about a positive future is
the point of reference for the next step, “recalling” in “retrospect”
the path of actions that lead to that positive future.5
Recalling what one did and what was helpful

The second main question maps personal actions and experienced
support. The imagined present is still in the near future, and the
family group is still interviewed.
The core of the question is: “What made that development possible?
What did you do, and who supported you and how?” These reflections,
too, are facilitated by feedback (“did I hear correctly when I heard

you saying . . .”), and brief public notes are kept. The question—
supplemented by specifying questions—aims at helping each per-
son to outline a subjective functional, active perspective and to take
responsibility in that. The scene for committing oneself is set, first
of all, by addressing subjective plausibility. If the participants do
not believe that the acts could actually lessen the worries, they are
not likely to carry them out. Therefore, the facilitators try to check
whether the person feels that what they said is possible and moti-
vating for them. It is important to keep personal, subjective acts in
view. If the talk is about activity in general (e.g., “we started to do
things together”) without unfolding the acts that form the activity,
the plan may well remain as good intentions. (A specifying ques-
tion could be: “What did you yourself do that helped you start
doing things together?”) Committing oneself is also backed by the
publicity of the thinking and planning. When the participants con-
sider aloud their own acts, the process is somewhat akin to making
promises. At the end of the meeting, the facilitators make sure that
the persons concerned believe that the acts can actually lessen the
worries and that they really intend to carry them out.
The “what did you do” part of the question points to the
person’s own activity. The “who supported you” part points to
others. The questions are formulated with the intention of encour-
aging each person to talk only about him/herself; if others are
talked about, they are mentioned within a perspective of support:
who helped me and how did they help. The idea is to prevent a
situation where the participants determine mainly what others
should think or do. This is done so that people would not have the
need to defend themselves against such definitions. Commenting
is not “aloud”, even in situations where the person thinking aloud
expresses clear misunderstandings—for example, of a profession-
al’s mandate and possibilities. Commenting not only interrupts
the flow of thoughts of the speaker, but also cuts off the commen-
tator’s own inner dialogue, which is where the speech of others
is integrated into his/her own system of views. The facilitators
who regulate the outer-dialogue setting aim thus at furthering the
polyphony of each person’s inner dialogues. There will be the oppor-
tunity to set straight possible misunderstandings when the person
who feels her/his position is misinterpreted has her/his turn in
the interview. She/he has the chance to “recollect” from her/his

point of view.
It is important that no one is “pushed into a corner”, in the
sense that one is not forced to adopt the same elements of other
people’s “recollections”. The narrative formed does not have to
be one formed in unison. Because of the format of “recalling”,
there is plenty of room for thinking. The facilitator offers such
room by framing his/her questions as “and how do you remember
this . . .”. The respondent has the opportunity to say, “I remember
it differently, namely like this . . .”. Indeed, the participants use the
opportunities that the recollecting format offers as their instrument. One
can observe in Anticipation Dialogues a progression from a some-
what bewildered start to a situation where participants adopt the
future structure as a communication tool. Dialogism would hardly
be possible if the utterances have to be locked into place as a thou-
sand-piece puzzle. Breathing space and polyphony are called for.
Worries and their lessening

The third question (also to be supplemented by auxiliary ques-
tions) maps today’s worries. These are approached as if in the
future already. The family members and others in the family group
are requested to “recall” “what were you worried about ‘a year ago’
and what lessened your worries”. The “year ago” phrase points back
to the very moment of the interview. Present worries are viewed
from an alleviated perspective. This adds to the freedom to express
anxieties without having to worry about the implications for the
social context. People regulate their social relationships and do
not generally want to make their situations worse. Experience
from Anticipation Dialogues indicates that expressing worries can
be facilitated in at least two ways. The first way has already been
mentioned: starting from a vision where the worries have already
dissolved or have lessened and by mapping the paths of actions as
if in retrospect. The second facilitating factor is the many-sidedness
of the network: there may be people present who have a better
chance than those who might be in vulnerable positions to talk
about their worries.
Case illustration
In one session, the father, mother, and children had already
outlined a good future and the actions that made it possible.
Worries had also been expressed. When the grannies—the fa-
ther’s mother and the mother’s mother—had their turn, they
both expressed that “a year ago” they were worried about the
father’s heavy drinking, and they also spoke about matters
that made those worries lessen. Alcohol abuse had so far been
almost out of the picture. Maybe the grandmothers were in a
position that allowed them to express such things safely.
Asking about worries and their lessening is, in a way, mapping the
minimum requirements: what, at the very least, must happen and
what, at least, has to be done. Typical auxiliary questions would be:
“What in particular made your worries lessen?” “What did you do
to lessen them?” With their worry-mapping questions, the facilita-
tors try to help the participants to express worrying matters on the
one hand, and to make a subjectively plausible plan of action on
the other. If necessary, they ask in detail: what specifically, when,
and by doing what were the worries lessened and what did the
person him/herself do in those important phases. This is to help
the participants to think about the alleviated situation as feasible
and to gather concrete material for putting together the plan.
Recalling the future differs a bit from usual communication. It
has features of fantasy. Using the future as an “instrument” allows
a lot of creativity. Often those present make use of rapid situational
humour. Although there is playfulness in recalling the future, it
is not play. It is a way of interviewing that allows a mapping of
worries felt in the present, the hopes for a better situation, and the
desires for mutual support—in a manner that the very dialogue
strengthens hopefulness. Also, the mapping of worries is impor-
tant in creating plausible hopefulness. If worries and matters that
lessen them are not charted, the subjective plausibility of the plans
may remain thin. This applies to the professionals, too, not just the
family and the private network.
Two questions to the professionals
After interviewing the family group, the professionals are inter-

viewed. The facilitators may take turns. Usually the facilitator
who made the public notes while the other interviewed the family
group will interview the professionals.
The practitioners are put two main questions, one at a time.
The principle is still that each of them “is” in the future and that
the processes that lead there will be recalled. Also, the rule that
everyone talks aloud and others hear is in effect, as well as the
rule that talking and listening are separated. The first question
goes directly to acts and support. The second is about worries and
their lessening. For the family group, the first question charted
the family’s positive future—the situation “now that a year has
passed”. The professionals are not asked about their view about
how the family’s situation is when it is good. The practitioners do
not live in that family. Instead, their job—each according to their
basic task— is to support the family members. It is exactly that that
they are asked about. The fixed point is the multi-voiced picture of
the near future that the family expressed.
The supportive actions may be linked to material things or con-
ditions, or they may be more of a spiritual kind. If decision-making
is necessary, it is not done in the Anticipation Dialogue session.
Instead, the practitioners are asked whose mandate the decisions
are and where will they be made. Decisions, too, are mapped from
a future perspective—for example, in the following vein: “a deci-
sion was made about that matter; who made it and what did you
do to promote the decision?”6
The first question to each professional is: “A year has passed. As
you heard, things are quite well in the family/for the child. What did you
do to support this positive development, and who was helpful to you.” In
this way, questions about the helper’s actions are supplemented by
questions on what support the helper got in his/her activity. The
practitioners may “recall” the help they got from both the family
and their private network, as well as from various professionals.
Each professional gets the opportunity to view things from his/her
own point of view and perspective of action. When the hub or the
focus of coordination is the family member’s positive everyday life
as she/he just described it, there are several points available where
the professionals can connect with the process. Opportunities are
available for connecting professional actions with the everyday
life of the family, instead of determining the measures from pro-
fessional routines. A whole is outlined where the resources of the
family members, their private network, and the professionals can
combine to relieve the situation of the child or the adolescent.
The second question (which is also supplemented by auxiliary
questions) is “what were you worried about ‘a year ago’ and what
lessened your worries”. That is to say, the professionals also get the
opportunity to express their present worries by putting the future
perspective to good use. Because they can look at their present
worries as relieved, from the perspective of decreased worries,
they do not have to be emphatically worried about the conse-
quences of their utterances. In other words, they need not fear that
they offend the clients by expressing their worries, as much as they
would have to in a situation where problems are the focus.
Also, the practitioners’ thought experiments are facilitated by
brief feedback like “did I hear correctly when I heard you say-
ing. . . .”. The professionals are also listened to intensively and in
a one-to-one contact. A situation is created where the person think-
ing aloud has the opportunity to reply to one person while her/his
utterances are responses to the utterances of others. Responsive-
ness does not necessarily require immediate replies. As a matter of
fact, delaying the response creates sufficient room for multi-voiced
inner dialogues. As with interviewing the family, the professionals’
reflections are recorded in the public notes, for all to see.
Thus, the facilitators also try to facilitate the professionals’
thinking. Like the family members, the professionals do not have
a clear vision of what is what and what is a good way to continue.
And they—like the family members—may have lots of thoughts
about what the problem is and what ought to be done. Future-
recalling is not embarked on by defining the problem, nor are
views collected on how others think the “problem bearers” should
change their activity. Problems are not charted at all as problem
definitions—and certainly not as a single common problem. Instead,
there is talk about worries and their lessening. Worries are subjec-
tive. They tell about one’s experienced opportunities to act in a
perceived network of potential resources. Therefore, not only the

worries are subjective, but also the problems of what to do. When
people express their worries, they are talking about themselves in
connection to others. When they talk about the help they get and
the acts that could lessen their anxieties, they are likewise placing
themselves in relationships.
Future-recalling opens up the space that dialogism requires.
Everyone has the opportunity to “remember” matters from their
own point of view. Participants do not have to recall the things
exactly as someone else already has. If that were true, their own
possibility to speak would be reduced to making supplements. It
is essential in dialogism that new meanings are created through utter-
ances that respond to each other. In monological conversations, utter-
ances are closed: others can only admit or deny. In future-recalling
the facilitators try to make sure that each has the opportunity to
think his/her thoughts against the sounding-board of the impres-
sions he/she gets from other people’s utterances. Quite often the
participants’ utterances start to echo each other, so that the sketch-
es outlined in the reflections begin to form a pattern.
Case illustration
One particular mother once made clever use of the recalling for-
mat. When she recollected what sort of help she got from each
professional in supporting her son, she said that “the teacher
was particularly helpful by not calling me on the phone so often
any more, since he heard in the dialogue ‘a year ago’ how well
things are going and how many helpers our family actually
had”. When the teacher sometime later got his turn at the pro-
fessionals’ table, he recalled his acts of support and the help he
got: “When I heard ‘a year ago’ how well things were progress-
ing for the boy and how well the family was supported, I didn’t
call so often any more because I did not have to worry. I was at
the family’s disposal. Sometimes the mother called me and told
me what was happening. It was very helpful for me.”
In that session the teacher linked his recollections (in other words,
his future plans) quite closely to what the mother expressed. He
had, however, the opportunity to “remember differently”.
There is one thing that is fixed before the professionals are

interviewed. The picture of the positive future acquired from the
family group is the point of reference to which the professionals’
reflections are connected. The professionals are not asked to delin-
eate how the family lives are when things change for the better.
However, the professionals can recall in so many different ways
how they supported the family and who was helpful to them. They
are not tied to what the family members “recalled” about profes-
sional help. If that were the case, the dialogue would stiffen into a
briefing. Each professional also has space in relation to what other
professionals say. They may continue the narrative if they want
to, or they can choose to do otherwise. For them, too, the recalling
format offers a tool for thinking together.
Another possibility for space seems to open up from the op-
portunity the professionals have to recall what sort of support they
got from the family and their network not just from other profes-
sionals. In the passage above, the teacher recalled that the mother
called him every now and then to tell him how things were. The
mother heard this. Perhaps the teacher made a suggestion, indi-
rectly, through answering the facilitator’s question. Recollections
of help from others are also requests for support. This kind of com-
munication differs from putting in claims or dictating terms.
The requests made in Anticipation Dialogues are indirect in at
least two ways. First, the recollection format affords indirectness.
To say “as far as I remember he helped me like this . . .” is different
from saying “he should help me like this. . . .” In the recall format
the addressed person has dialogical space. Second, the utterances
are expressed as answers to the facilitator’s questions. The facilita-
tor interviews in a one-to-one mode within a multilateral setting.
He or she makes sure that the interviewee does not turn attention
to someone else and start to put forward what this person ought
to do or think. Thus, the facilitator is also an intermediary, there to
safeguard indirectness and space. Of course, this is the job of the
facilitators in interviewing the family group as well as the profes-
sionals.
The outline of the plan

and the support for coordination
The session is drawing to a close. The facilitators are present be-

cause an initiator invited them—because they were worried. It is,
therefore, their task to make sure that a plan is made at the end—a
plan where the resources of the parties form a good combination.
The future approach is not applied in the discussion in the last
part of the session. The facilitators take some care that the par-
ticipants “return” from the imagined future. Innovative techniques
can be applied for agreeing that the present is actually not in the
future but where it was left before the exercise—the atmosphere
usually is very creative at that stage. The facilitators clarify that the
task is now to use the elements gathered from recalling the future
so as to make a plan for cooperation. The public notes serve as
material for the discussion.
The facilitators once again make sure that the notes sufficiently
represent the participants’ ideas. In the closing discussion, the
facilitators help the participants to put together an activity plan
from the expressed and recorded acts, the mutual support, and
the worry-lessening elements. No “ideological” summary is aimed
at. The building blocks of the whole are actions. The core is to
find out who does what with whom next. This possibility to go on is
decisive, because without the first step to carry out the plan, there
are hardly the following ones. The personal acts and hoped-for
support make a pattern whose core elements can be gathered from
the replies to the question of what in particular lessened your worries.
The (recorded) answers to the anxieties question serve as the base
elements of the plan.
It is important to ensure that the people in distress are not left
alone. At the closing stage of Anticipation Dialogues, this takes
shape as a guideline for the facilitators: make sure that the support
plan does not leave the child, adolescent, or family alone—or the
initiator either, because leaving the initiator (the party that initiated
the meeting) alone in the professional network may leave the cli-
ents without the support just outlined.
Right at the end of the session, the facilitators hand the organ-
izing function back to the initiator. They mention that their part is
ending and that summarizing is naturally the task of the person

who originally wanted to summon the networks to dialogue. The
facilitators request the initiator to agree with the participants what
takes place next, when the follow-up meeting will be arranged,
and so on.7 The initiator is also asked to agree with the partici-
pants what will be done with the notes, whether a summary is
to be made, who will get it, and so forth. (Usually the clients and
the core network receive a summary with the plan.) It is also good
practice to check what decisions have to be made and where they
will be made.
After summarizing the closing discussion, the initiator thanks
the participants for the help he/she has received for lessening
his/her worries.8
Notes
1. Variations of Anticipation Dialogues are used for joint-action planning in
various context, such as residential areas, schools, launching projects and units
of work, or municipal medium-term planning. Versions have also been used
for supervision and consultation. A significant application area is evaluation.
2. In the municipalities where such network activity is being developed,
leaflets and other information are available in libraries, agencies, and so on.
The activity has a coordinator who takes care of the facilitator pool, informa-
tion dissemination, follow-up, and so forth.
3. In Finland, they are provided free within the public psychosocial-serv-
ice system. The intersecting mutual services balance out the efforts. Among
the facilitators there are professionals from most psychosocial branches. The
sectors provide on a give-and-take basis. They give some professionals the
opportunity to act as facilitators alongside their jobs and to make use of the
services of the cross-sectional facilitator pool.
4. The “recalled” period can, of course, be shorter or longer than a year.
In the warm-up discussion the facilitators ask the clients what could, in their
view, be a period in which changes could actually begin to show. In other
words, the time span for future recalling is negotiated at the beginning. More
than three years is too far away, though: variables become very complex and
discussion loses concrete reference points.
5. Situations where the family members have mutually incompatible future
visions are challenging. Custody battles or quarrels about deinstitutionaliza-
tions are examples. Although Anticipation Dialogues have been tried out in
situations where rather diverging views have been expressed, and have been
thought-experimented to bring them into dialogue, fiercely divisive situations

are not the terrain for Anticipation Dialogues.
6. To date, the experience is that the plans formed for professional help in
the sessions do not call for exceptional resources. On the contrary, quite typical
supportive actions are reorganized into a better coordinated whole. Starting
the interview with the family is decisive: it is exactly the positive everyday
life that the professional measures should connect with. When there is a fo-
cus on coordination, professional measures gain their sense from the positive
everyday life.
7. Follow-up meetings aim at supporting the continuity of the coordination
found and at keeping up the dialogue.
8. The facilitators ask the participants to fill in a form for collecting par-
ticipant feedback. They are also requested to give written consent for a later
follow-up.
CHAPTER FIVE
Alike but different
I
n the two preceding chapters we described two different ap-
proaches for working with social networks. In this chapter
we discuss some basic elements of dialogicity. We believe that
many forms of dialogicity are needed in psychosocial work, and
therefore it is necessary to go beyond the approaches just de-
scribed. However, since dialogicity does not exist outside dia-
logues in practice, we make use of our experiences in Open and
Anticipation Dialogues as a point of departure for a more general
discussion. We also compare the common and distinguishing fea-
tures between them.
In Open Dialogues, each meeting creates its own structure
based on the context and circumstances. In Anticipation Dialogues,
the structure for the meeting is planned in advance. In Open
Dialogues, each person can participate as an “insider” in the treat-
ment process. In Anticipation Dialogues, facilitators are used.
89
Common factors
What are the common factors in these two approaches? They are
both based on a case-specific group of people.1 Both the professional
people and the private network of the client include precisely those
who are connected to the actual situation. It may be, for example,
a psychiatric crisis or concern for a child’s welfare. In psychosocial
work, there are practices where the team of professionals remains
unchanged while the cases they handle vary. There are even prac-
tices where the professionals represent their “home unit”—a hos-
pital ward or a social office, for instance—and participate in the
meetings representing their unit’s generalized point of view, not as
themselves as individuals. These may include different types of re-
sponsibility-teams or rehabilitation-cooperation teams in a munici-
pality or catchment area. These teams handle each case regardless
of whether any of the participating professionals have a personal
relation to the client whose matter will be under scrutiny.
Both Open and Anticipation Dialogues differ from such prac-
tices. It is very important in both that the individuals participating
in the meeting are connected with the actual case—either as a part
of the client’s everyday life or through professionals tasks. They
cannot be randomly replaced by other people, because building
up mutual understanding presupposes the participation of those
very individuals that already are personally connected to the case.
Understanding emerges between individuals, not between institu-
tions. Both Open and Anticipation Dialogues are based on ideas of
a multiplicity of languages (the heteroglot principle) and social lan-
guages. Precisely those individuals who are connected to the actual
issue and the people around them participate in the actual con-
versations. Each case-specific team member has personal contacts
to the clients and thus creates a social language for this situation,
and therefore the problem of the new community will be different
compared to any other conversation. Anticipation Dialogues are
organized in situations in which helping efforts may have contin-
ued for a while already. The individuals in the meetings may have
had a lot of personal contact with each other over the course of
time. In Open Dialogues—especially in a new crisis—it is possible
that the people present are meeting for the first time. Anticipation
ALIKE BUT DIFFERENT 91
Dialogues are meant to be single interventions with restricted fol-

low-up, whereas Open Dialogues often include several sessions in
succession. Thus, generating a joint language takes a different form
in each practice. The idea of situation-specific social language is
central to both approaches.
In both approaches, the composition of the actors goes across
boundaries. Both private-network and professional-network mem-
bers participate, and practitioners most often come from different
units and professions. In the meetings, the aim is to find fruitful
ways to go on and collaborate. Boundary-crossing like this creates
a new type of expertise: a co-created understanding that presupposes
laymen and professionals coming together. To be viable, the exper-
tise actually requires the clients and their personal networks to be
actively involved, not just the professionals with their specialized
knowledge. In both practices, a polyphonic world view is essential.
The aim is not unanimous understanding of the problem as the
basis for a plan of action. The point of departure is, on the con-
trary, that each person has his/her own specific view of the matter.
It is important to try to understand each point of view. The new
understanding is created at the boundary of the parties, as no one
person’s point of view takes precedence as the single and correct
definition.
The approaches also have in common the aspiration to embed
the helping efforts in the client’s everyday life, as near as possible to
his/her everyday circumstances and relationships, rather than basing
them on institutional communities or relations alone. Everyday
language is preferred in conversations, which means avoiding
professional jargon. A common factor in both approaches is that
the focus is on the next step for going on instead of aiming at firm
plans either for the treatment and helping processes or for life
in general. Although the “recalling-the-future” conversations in
Anticipation Dialogues makes use of a one- to two-year horizon,
the goal is to find out “who is going to do what with whom next”.
In Open Dialogues, it is decisive that a perspective for new joint
understanding opens up as the next step. These steps should not
be hurried, because they have to follow the specific rhythm of each
process. Although treatment or rehabilitation plans may give a
perspective for the treatment system, life itself is taking the natural
course within the systems of boundary that are created by those

participating in the meetings. This illustrates both the heteroglot
principle and dialogical nature of language. Firm, far-reaching,
and monological plans are not aimed at; instead, the next step for
each person is held in perspective. At their best, treatment plans
become parts of a dialogical process in the sense that they are
discussed in each conversation and are actually opened up every
time the participants in the dialogue meet with each other. In every
conversation, new meanings will emerge, although the same issues
are also handled as in the previous sessions.
In both approaches, the observable external changes in activities
are rather minimal and non-dramatic. After participating in an An-
ticipation Dialogue meeting, the professionals return to their units
to take actions that are already familiar. Open Dialogues do not
replace any other traditional treatment method. What does take
place in both practices, however, is that the activity gains new
meaning. The actions themselves are mostly the same, but the ac-
tivity of which they are composed will have a new meaning for the
parties after the joint experience. Dialogues allow each participant to
find her/his connection with the wider process. Finding one’s own place
in the process is enhanced by seeing how others place themselves
in the process. This makes it easier for each to ask for help from
the others and to understand the significance of his/her activity
in relation to others. This is not merely a cognitive process but is,
instead, an embodied emotional experience. It is not only “see-
ing” or “understanding”, but also becoming touched as a human
being. The new understanding is generated in a shared emotional
experience, which means that people become connected with each
other in a new, active way. Orientating to the response becomes,
in a way, a part of the inner orientation of each interlocutor in
the conversation, and thus thinking becomes a joint process. Al-
though the perspective is subjective—respecting each participant’s
point of view—the process aims at a multi-agency and polyphonic
understanding. An individual perspective does not exclude joint
experience. Everyone participating in the discussion carries within
their inner orientation the voices of other participants in the con-
versation.
Different factors
Open and Anticipation Dialogues are very different from each

other as well. For the most part, Open Dialogues have been applied
in psychiatry, where dialogues are conducted by a team committed
to the actual crisis. The same team has the possibility to follow and
guide the process as intensively and for as long a period as needed.
Anticipation Dialogues, on the other hand, are mostly applied as a
one-time intervention in impasse situations of multi-agency confu-
sions—with a follow-up session. Although both approaches em-
phasize the tolerance of uncertainty through polyphony, they offer
different possibilities for this. Open Dialogues continue with the
same group of people if needed, whereas the group that takes part
in Anticipation Dialogues disperses, returning to the no-man’s-
land between the various agencies of social, health, and so forth.
The follow-up meeting is designed to support the cooperation of
the parties after the Anticipation Dialogue meeting.
Open Dialogues do not presuppose an outside team for con-
ducting the meetings, since one basic assumption is that every
one in the meeting should have their own experience of sitting
together with others in this particular interactional field. Anticipa-
tion Dialogues presuppose outside facilitators. Of course these,
too, will become involved in a joint experience, but they do not
continue meeting with the people involved. After their service in
facilitating dialogue, they leave and join the process again only in
the follow-up session some six months later. In Open Dialogues,
uncertainty is tolerated by, for instance, continuing the search for
a new understanding instead of aiming at an exact definition of
the problem in the early phase of the treatment. The content of
the psychic crisis is welcomed into the meeting. For example,
the specific interactional styles of the family may be repeated in
the joint meetings, where they are lived through. Many types of
turbulence may be encountered in the conversation. Anticipation
Dialogues try to control such turbulence. The idea is to avoid re-
peating in the joint meeting the dysfunctional ways of conduct that
were the reasons for the impasse and for calling the joint meeting
in the first place. It is exactly because of such interaction proc-
esses within and between the networks that new coordination is
necessary. Uncertainty is tolerated by creating plenty of space for

listening to the many voices in the session. In order to safeguard
space for multi-voicedness, the facilitators are somewhat resolute.
Talking and listening are secured by separating them from each
other. Each person listens silently to the other’s talk and waits for
his/her own turn. Turns are not taken freely as in the flow of con-
versation; rather, each one is speaking by answering the questions
from the interviewer.
For the one who takes charge of conducting a dialogue, the
question of being an insider or outsider in the process is second-
ary. Both practices aim at generating dialogue, which presupposes
the same type of actions by the facilitators regardless of the spe-
cific form of the meetings. To this end, you have to formulate your
questions in an understandable everyday language; you have to
ask about concrete acts and incidents; you have to proceed slowly
to allow time for the formulating of answers and searching for
the right words; and you have to be sensitive to the client’s emo-
tional experiences and embodied messages as responses to your
questions. You will become moved yourself, as well, when the
participants speak of sad things. The interviewer has to join the
shared language area. This is essential for creating dialogical con-
versation. Although the facilitators have not met with the clients
nor been in contact with the problem at issue, in the meetings
knowledge from within relationships (Shotter, 1993b) is created.
Facilitators have three basic questions for the family group and
two for the professional group, but a mechanical interview, insen-
sitive to the previous utterances in the session, would inhibit the
dialogue.
Both approaches are different from interviews, where inter-
viewers have some type of diagnostic map in mind into which
they try to place the answers in order to analyse the problem. In
these types of interviews, there is a danger that the interviewers
emphasize questions that fit their hypothesis. In both Anticipation
and Open Dialogues, unconditional acceptance of the answers is
a prerequisite. The aim is not to gather information for a correct
interpretation of the client’s problem. There are no wrong answers
for the questions in Anticipations Dialogues, and the answers do
not have any diagnostic weight. No distinction is made between
“healthy” or “sick” utterances in Open Dialogues, either; they all
are voices in constructing a new joint understanding. Both prac-

tices differ from the more traditional social network interventions
in which the aim is to conduct a network spiral process within one
session (Speck & Attneave, 1973). In spiral-process interventions a
neutral team is invited to take charge of the meeting. The team’s
task is to emphasize a spiral group-dynamic process, and the dis-
cussion themes are subordinated to such a process. In Anticipation
Dialogues the themes are primary.
Basic elements of dialogicity
“Dialogue” (dia’logos in Greek) has many definitions. One of the

most basic literary dialogues was written by Plato, usually seen
as the father of dialogue in Western culture (Nightingale, 2000). In
his works, Plato presents the main characters in dialogue. Plato re-
fused to write a treatise, because he saw that it is impossible to give
a rational description of understanding generated in dialogues.
Plato, for one, saw the self as a social construction (Nightingale,
2000). In his early texts, Socrates in particular was described as
one who helps interlocutors to create the truth in an on-going
dialogue; it was not his task to find the answers (Bakhtin, 1984).
The blossoming period of the ancient Greek sciences (philosophy,
medicine) and arts (poetry, sculpture) took place in the same classi-
cal era when assemblies of citizens assumed the form of dialogues.
Perhaps the idea of open meetings with social networks includes a
return of some of the elements of the ideals of the ancient Greeks.
David Bohm (1997) and William Isaacs (1999) define dialogue
as a flow of meanings: In the word, the first part, dia means either
“through” or “via” and second logos, either “word” or “meaning”.
Bohm and Isaacs make a distinction between conversation and
dialogue. They see dialogue fulfilling deeper needs than merely
acceptance. In a conversation the aim is to find out an agreement
between two or more disagreeing parties, but in a dialogue the aim
is to gain a new understanding as a foundation for later thinking
and actions. Dialogue aims not only at an agreement, but at creat-
ing a new context for new deals and creating a foundation that
could assist in coordinating both values and joint actions.
For Paolo Freire (1970) dialogue is essential in the empower-

ment of the oppressed, in the process where they can become
subjects instead of objects. Dialogues are central in the “pedagogy
of the oppressed”, where students and teachers are simultane-
ously teachers and students. Dialogues require and create faith in
people.
Ivana Markova (1990) points out that it is important to notice
the difference between dialogue and dialogism. Dialogue refers to
communication between people. Dialogism is a specific epistemol-
ogy, an outlook. The German theologian and philosopher Martin
Buber (1987) has emphasized the fundamental difference between
the relationships of people to people (I–Thou relationships) and
those of people to objects (I–It relationships). In the former, there
is a possibility for mutuality, openness, and dialogue. I–It relation-
ships are necessary, but if the other is instrumentalized, dialogue
cannot be reached. Dialogical relationships require two separate
persons who meet each other as comprehensive beings. You are
another I. For Buber, the essential does not exist in you or I, but
in-between.
The Lithuanian philosopher Emanuel Levinas, who lived and
worked in France, deliberated on encountering the Other (Levinas,
1985). According to him, the Other is always outside the experi-
ence of the I. Differing from Buber, he saw that the relationship
between the I and the Other is not symmetrical. Complete mutual
understanding is not possible. The I can never have final knowl-
edge about the Other. The asymmetry of the I–Other relationship is
fundamental, according to Levinas. The Other is always more than
the I can grasp. Because no one else can be in the place of the I, the
I has the inalienable responsibility of the Other. That responsibility
means responding to the Other. In every reaction—even a bodily
reaction or deliberate act—the I is responding. Such responding
is, according to Levinas, the subject’s first and most fundamental
relation to another person.
Following Buber’s deliberations, dialogicity requires that the
Other is seen as another I and not as It. Following Levinas’s
contemplation, the other I can never be understood or explained
totally. The otherness of the Other is beyond similarity. But the I
is responsible for the Other. Both Buber and Levinas analyse rela-
tionships, not singular or isolated individuals or consciousnesses.
There are relationships between I’s, multi-subject networks. In

them, others are potential means for my ends but also for other I’s,
like me and more, others that I can never understand completely. I
am in responsive relationships to them and am therefore responsi-
ble for my actions. Responsiveness on my part and on the part of
the others can generate dialogical relationships.
Responsiveness is central also in Mikhail Bakhtin’s analysis on
dialogues. “For the word (and, consequently, for a human being)
there is nothing more terrible than a lack of response” (Bakhtin,
1986, p. 127). He pointed out that “(b)eing heard as such is already
a dialogical relation” (p. 127). Bakhtin analysed social languages
and thereby also the social character of consciousnesses. The unit
of analysis is not a separate unit in the language nor a separate in-
dividual but individuals and thoughts in dialogue. “The idea is not
a subjective individual–psychological formation . . . in a person’s
head; no, the idea is inter-individual and inter-subjective—the
realm of its existence is not individual consciousness but dialogic
communication between consciousness” (Bakhtin, 1984). Also, Ba-
khtin considered dialogue as corporeal: bodily dialogue between
subjects in space and time.
Below we discuss dialogicity mainly in the light of Bakhtin’s
theory. We distance ourselves from our material—Open and
Anticipation Dialogues—for a while. After the discussion, we
return to the theme before closing the chapter.
Dialogicity, polyphony, and social languages
Although the term “dialogue” is used for describing all sorts of

discussions, dialogism is more of an epistemological stance.2 It is
an attitude towards life in which a dialogical relation is the start-
ing point (Bakhtin, 1986). It is not merely exchanging utterances
between interlocutors by, for instance, asking and responding. In
it, the guiding centre for our behaviour is created between people.
Isaacs (1999) describes this as the art of thinking together. There is
no longer a single subject who does the thinking; the thinking sub-
ject is all the participants in the dialogue. In this sense, dialogicity
contradicts monologism, in which the guiding centre for behaviour
is located within an individual. The speaker is referring to inner

thoughts in giving meanings for things, and thus the truth of each
utterance is defined according to the personal map of the speaker
(Crowley, 2001). In dialogicity the speaker is relating to the social
field around him/her in the way that he/she is constantly adapt-
ing to the utterances of the others present and to the social (and
geographical) context, and leaving space for answering words in
formulating the utterance. Answering, on the other hand, does
not close the theme or give a final answer or solution, but, on the
contrary, opens up ever-widening perspectives to the theme under
discussion.
In dialogicity, the focus is on the borderline, in the area between
people, in which they meet with each other. As Valentin Voloshi-
nov noted (1996), “the location of the organizing and formative
centre is not within [that is, not in the material of inner signs], but
outside” (p. 85). In a dialogical relation the word becomes shared
between speaker and interlocutor. The word belongs to both the
speaker and the interlocutor. As Voloshinov notes further, “the
word is on the border zone between himself and his addressee—
still—it does in part belong to him” (p. 86). The speaker owns part
of the word, but half of it belongs to the addressee, and thus it is
always built jointly in the specific discussion. To take advantage of
this, the speaker has to consider the listener(s) throughout. While
speaking, the speaker has to read the body language, such as posi-
tions, tears in eyes. The speaker has to listen both to the content
and the tones of the answers; she/he has to take into account the
circumstances such as the presence of other people, if the room is
noisy, and other such conditions. A countless number of embodied
emotional factors are constructing the shared dialogue. The speaker
is the subject for her/his speech only in a physical sense; she/he is
the one producing the words in her/his vocal cords. But the social
context determines the structure of the speech situation already as
an inner orientation (Voloshinov, 1996). The entire context becomes
part of the inner structure of the speaker. In her/his utterances the
speaker is constructing the words according to her/his embodied
emotional situation. And this is affected by surroundings: whether
it is warm enough or too warm; how the furniture is located; does
one have to raise the voice to be heard; are there one or more in-
terlocutors sharing the listening. These types of countless details
become inner parts of the speaker, and they change with every
answer. In treatment or help situations, all this highlights the im-
portance of how we construct the situation for the dialogue. If we
create pleasant physical surroundings and if we start to speak in
a way that considers all the people present, we can encourage the
network members towards a more active stand in the conversation,
which again forms the direction of the conversation in the meeting.
Individuals participating actively in the conversation more read-
ily become part of each one’s inner structure rather than sitting as
passive onlookers in the room.
The social reality is polyphonic:3 it speaks in many voices. In
every social situation a variety of different voices are present. The
term “voice” refers both to the speaking subject and to the con-
sciousness. It is a situation-specific incident, in which the speaker’s
message is not ready-made in his/her mind and then submitted
to the receiver. Instead, it will be constructed in the area between
the interlocutors. As Wertsch (1991) points out, in each conversa-
tion at least two voices are present, and thus using the word in
the plural (“voices”) is relevant. We are living in a multiplicity of
voices that will be turned on and playing simultaneously accord-
ing to what, where, how, and with whom we are talking. Social reality
is always polyphonic. The term “social roles” has also been used
for describing the changes of our tasks according to the changes
in the social context. However, it is worth noticing the essential
difference between these two concepts. The concept “social role”
is a firmer description of the individual’s tasks and obligations in
various social realities. In the polyphonic reality as Bakhtin and Vo-
loshinov delineate it, no such firm social structures exist that could
be moved from one place to another without taking into account
the actual actors. In the polyphonic reality, each issue receives a
new meaning in a new conversation, in which a new language
for the things under scrutiny is constructed. Each person’s social
meaning and social identity is created in the actual conversation
instead of thinking that they would stay the same from one social
situation to another.
All the voices present participate in constructing new meanings
in the actual conversation. Part of these voices—a minor part—will
be said aloud, part of them are living as qualities of embodied
actions and as qualities of the social and physical context, and
another part as inner experiences. In the polyphonic reality, no

selections can be made as to whose voices are the right ones and
whose are the wrong ones: they are all contributing to the new
understanding where all the voices play an important role. They
are equally valuable. In monological reasoning, the voices do have
a hierarchy in the sense that, for instance, the opinion of the chief
psychiatrist is more important in defining the diagnosis. In a poly-
phonic dialogue, the professional hierarchy becomes secondary:
understanding of the problem situation becomes richer the more
voices there are participating in constructing new meanings. Ba-
khtin (1984) found the idea of polyphonic life in the novels of the
Russian classic writer Fjodor Dostoyevsky. In these novels it was
not possible to define one hero (the main character). The entire
reality seems to be generated in the dialogues between the charac-
ters, and the new reality cannot be defined in advance. Everything
that is said is true and constructs new understanding. When the
reality becomes generated in the dialogue between the characters,
the author no longer has the possibility to define in advance the
meaning of the actions of the characters. The author has to stay in
dialogue with her/his characters. Bakhtin calls this a polyphonic
novel (Bakhtin, 1984). Subsequently, the idea has been used in
describing the basic attitude of dialogicity as an epistemological
stance. All our knowledge in dialogicity is constructed in situation-
specific conversations, in which all the things discussed receive
new meanings regardless of how many times the same group of
people has gathered to discuss the same issues.
This illustrates the heteroglot principle of language and the
idea of social languages (Bakhtin, 1984). Discourses happen in
specific social contexts. A conversation in a different social con-
text generates new meanings even if the same type of discussion
with the same participants took place in some other context. The
idea is very useable in help-discourses and treatment situations.
Each conversation creates its own social language, which is not a
firm coding system but is realized differently compared to previ-
ous conversations. Cultural differences become essential. Female
language is different from male language, even as each female’s
language is different; the language of fathers is different from
children’s language; the language of the psychologist differs from
the language of the social worker; the family guidance clinic’s lan-
guage differs from the language of psychiatric outpatient clinics.

The list could be continued endlessly, and the categories further
broken down. The languages are not, however, positioned as stable
coding systems in each place; they do not stay overnight in the psy-
chiatric outpatient clinic. They only exist if the actors start to talk
of their tasks in each helping process. And, as the list indicated,
we live in many languages simultaneously. When we participate
in a network meeting, we are present both as a psychologist and
as a family therapist, but also as a father if issues that refer to chil-
dren’s problems are discussed and as a man who has experienced
sadness after a loss of some dear person if death is discussed. The
different voices are “switched on” according to each specific topic.
Professionals always participate in dialogues as both bearers of
some professional titles and as living human beings.
This is half of the heteroglot principle. Each conversation cre-
ates its own language that did not exist last week when the same
social network sat together and that will not exist in the next meet-
ing next week with the same social network. All this is regardless
of whether the same issues are handled or not. For the participants
in a therapy meeting, all this is both very challenging and reliev-
ing. It is challenging because it reminds us that each conversation
is unique in creating possibilities for new understanding, and thus
we could not afford to act carelessly with this possibility. But at the
same time, it is relieving, since every conversation is a seed for new
possibilities and perspectives in cases where, perhaps, we have
become tired after meeting several times without any remarkable
improvement. The potential power resides in the idea of social net-
work. As we said earlier, the parts of the social network each have
their own idea of the problem—actually their own problem—that
cannot be the same as that of any other party of the actual case.
There always exist as many problems as there are parties in the
dialogue. Every new discussion has the possibility for new ideas
since all the persons have meanwhile participated in other conver-
sations in which their view of matters has altered. The task would
be very different if we aimed at a joint definition of the problem
in each situation.
The last theoretical notion to be made at this point is to high-
light the active nature of understanding. In constructing new un-
derstanding, the words of reply become the most essential part. A
dialogue constructs a new shared reality. The aim is to understand

more about what the other person said and, at the same time, to be-
come more aware of what oneself thinks about what is discussed.
In a discussion, there never exists the first speaker. Everything
said, even the opening utterance, is a reply to what has happened
earlier (Bakhtin, 1986).
The fact is that when the listener perceives and understands the
meaning (the language meaning) of speech, he simultaneously
takes an active, responsive attitude toward it. [p. 68]
And the speaker himself is oriented precisely toward such an
active responsive understanding. . . . Moreover, any speaker is
himself a respondent to a greater or lesser degree. He is not,
after all, the first speaker, the one who disturbs the eternal
silence of the universe. [p. 69]
At the moment the interlocutor perceives the meaning of what is

said, he takes an active, responsive position to it. All understand-
ing is striving for a response in which the new understanding is
crystallized. In the conversation, replying becomes more important
than asking questions. The speaker is throughout in a mutual posi-
tion with the interlocutors, and at the moment they start to reply,
the speaker becomes an active responder, who, with his own re-
sponse, affects the one who talks.4 In dialogue this chain is endless
and open since new meanings are generated throughout. In a con-
versation with clients we can affect the situation by our responses.
We have an essential impact on what type of space is left for the
clients and how their responses can affect the joint forthcoming
process of dialogue.
Response is important in any type of conversation, even in a
monological one. A monological utterance cannot live if it does not
have a responding listener. Response belongs to the basic qualities
of language. A monological utterance awaits a reply that either
approves or rejects it, and after the reply the circle is closed, be-
cause only one definition can be correct. In monological dialogues
the speakers often have to defend what they are saying and thus
take a defensive position. This type of conversation often has a
tendency to deal with power relations by, for instance, defining
who has the power in deciding the criteria for the correct answer.
The significance of the response in constructing new understand-
ing is different in a dialogue. There is a shift to an area of thinking

together (Isaacs, 1999). A dialogical utterance does not wait for an
approving or rejecting answer, but, rather, for a reply that opens up
new perspectives to what is said. The form and timing of the reply
often becomes secondary. The response may appear as an immedi-
ate comment or it may appear as delayed speech after the speaker
has had plenty of time for thinking the reply through.
In chapter 2 we put forward hypotheses of factors that may lead
network meetings into an impasse. The elements could now, after
the discussion above, be seen as factors leading the conversation in
a monological direction, where the aims for defining and changing
the other participants of the conversation take the lead.
Notes
1. By “case” we mean the actual matter that brings together the networks
for a joint process. Thus, the case is not an individual.
2. In the following we use the concept “dialogicity” to describe dialogical
interactions.
3. In music, polyphony is created by multiple voices that are independent
of each other, in both melody and rhythm.
4. Both Bakhtin and Levinas write about the significance of the face of the
other. According to Bakhtin (1986), we never see ourselves completely. Instead,
we see ourselves via the face of the other. From them we see how our utter-
ance is received and to what effect. In this way, speakers “know” themselves
and shape their utterances. According to Levinas (1985) the otherness of the
Other—their strangeness—is especially manifested in the face of the Other. By
the “face”, Levinas means the whole corporeal being of the other. One cannot
know the other. The face appears in the sphere of the subject, but the other-
ness they express does not show as such. The I can, facing the face, step out
of its reserve and grasp that it exists in relationships, but it cannot have final
knowledge about the Other.
CHAPTER SIX
Healing elements in dialogues
A
fter changing the approach from searching for explana-
tions of client’s behaviour and pathologies to a joint shar-
ing with those nearest the patient, new and deeply human
elements of the therapeutic work can be incorporated. The crisis
threatens the most basic life conditions, especially when we meet
such extreme problems as psychosis, domestic violence, or trau-
matic experiences after a loss in the family. The good praxis for
helping our clients does not take place simply by removing the
symptoms, but by meeting living persons. The entire interaction
system becomes our focus, not just specific therapeutic techniques.
In this chapter, our aim is to discover what elements in these meet-
ings form the basis of the change experience for our clients such
that their symptoms may start to ease off and behaviour starts to
change.
104
HEALING ELEMENTS IN DIALOGUES 105
From handling things to a shared experience
Sitting in meetings handling extreme life situations, we often

become involved in deeply emotional and comprehensive expe-
riences. If family members have had difficulties speaking of the
issues connected to the crisis or concerns they have, an optimal
dialogue in the meeting may create the possibility to jointly live
through the hopelessness of the situation for the first time and
thus create the feeling of togetherness. This can happen even in a
situation where the emotional loading is heavy because of inter-
nal conflicts. The feeling of togetherness can lead to an emotional
reaction in the meeting, so people may weep or show some other
embodied signs that they have been moved. The emotions of the
network members’ touch the professionals as well, and they share
in the emotions. The meeting is asymmetric in the sense that the
professionals have more of a chance to control the process, but,
nonetheless, the professionals may be moved themselves by the
heavy emotions. This is an important aspect of dialogue. We are
present as an entire human and living person and not only as a
neutral professional applying some method. The emotional expe-
rience of the professionals may become so intense that it is quite
easy to understand those writers (e.g., Patterson, 1988) who write
of dialogue that is not so much a conversation method as an act
of love.
After an optimal dialogue, it is often impossible to define what
each of us did to aid the positive outcome and what happened.
What is remembered is the shared experience, and this is difficult
to form into an exact rational description. Bakhtin (1981) sees the
dialogue emerge as a communication between consciousnesses
rather than an internal processing on the part of individuals. The
basic criterion for joining this communication between conscious-
nesses is if the speaker was heard and accepted. Becoming heard
presupposes having an answer. As Bakhtin said, for a human being
nothing is so terrible than remaining without a response.
The aim of joint sharing in the dialogue through listening to
each other becomes the basis for the treatment and consultation.
In open dialogue, professionals meet without pre-planning and
openly, whereas clients are present from the very beginning. This
may be a new experience in the sense that now all is transparent,

and, at the same time, a form of real team work. For Isaacs (1999), a
dialogue is thinking together. Professionals do not succeed in their
search for a new, shared understanding if they do not likewise aim
at understanding each other in the present conversation. In many
ways, their focus changes from looking for solutions to the prob-
lem to emphasizing the form of the conversation. For our clients
the content of the dialogue is the most important thing: they want
to find a way ahead in their lives after this actual crisis. For the
professionals, the content is important, of course, but the thing that
makes the difference is how the dialogue is generated. By aiming
at generating dialogue, professionals can assist a process in which
clients have more possibilities for creating new words for their ex-
periences, which currently present themselves as symptoms. Even
if the content is important, it is relative, since dialogue can actually
be generated from whatever theme is important for the clients. In
the conversation, content and form are inseparable. For instance,
in the discussion of anxious themes, the form of the conversation
becomes very different compared to the opening phase of the meet-
ing, where practical issues are usually handled.
Joint sharing becomes essential in Anticipation Dialogues as
well. A specific structure is planned for listening and talking in the
dialogues, but generating joint language means a mutuality that
does not emerge simply by each answering according to his/her
turn. Although the perspective of a positive future and the actions
needed to achieve it are thematized, the dialogue itself also be-
comes something real. The atmosphere can be very intensive. It can
include becoming emotional when each person has the possibility
to talk about his/her concerns and the possibility for a more posi-
tive perspective and especially of perhaps being heard for the first
time. The content is not pre-planned, but is created from the issues
the family members want to speak of at that moment. In this, it is
important that each can speak both of his/her own themes and to
listen to what others say. Each one has the freedom for “remember-
ing” whatever he/she want; no one person’s remembering is taken
as the basis for others to comment on. Polyphony is aimed at not
finding one dominant voice.
Case example:
In one meeting, the father spoke towards the end of it. He
was on leave from prison so as to participate in the dialogue
together with his wife. Their family life was in a muddle, and
many worried professionals were involved. After the very last
round in the dialogue—after the questions to the professionals
of “what did you do to support the family?” and “if you were
worried about something, what helped to lessen your con-
cern?”—the father said: “I did not think that we have so much
support. I never have thought of you as helpers”.
The guiding centre in-between
One way to understand dialogue is to distinguish it from mon-

ologue. Bråten (1988) describes monologue as “passivating the
other”. Interpersonally, monologue involves silencing the other by
domination or by control of the available means of explanation.
At times, dominant members of the network may impose their
single-minded view of the situation onto the others. More com-
monly, there are several competing views struggling to dominate
the situation. Although some individual dialogical utterances may
emerge, these do not become the main form of conversation. No
one is truly responding or listening to the others, as each clings
doggedly to their own understanding. The conversation persists
primarily in the monological domain, which in such situations
is maladaptive, as the network members’ understandings of the
situation have failed to resolve the situation, and no new ideas
can emerge if everyone is stuck in monological mode. Distressed
network members are caught in a dilemma: to find their way out
of their situation they must shift into dialogue, but dialogue by its
nature is unpredictable and therefore particularly threatening for
people struggling with trauma (Kamya & Trimble, 2002).
Monological conversation emphasizes mostly the patient’s be-
haviour and diagnosis. An example would be the verbal exchange
between a patient and a physician to rule out a heart attack. The
physician is guided in questioning of the patient by a well-estab-

lished internal map of the pattern of symptoms of a heart attack and
a clear set of instructions for action if the diagnosis is confirmed.
The patient’s responses to the physician are under the control
of this monological discourse. In traumatic situations, discourse
tends towards monologue among members of a network affected
by the extreme situation. In such situations, monological conversa-
tion aiming at rapid and right diagnosis is of vital importance. In
situations that are based on the client’s emotional experiences, the
entire meaning is better constructed in dialogical conversation. The
guiding centre of behaviour is created in the area between the par-
ticipants in the actual dialogue. Clients take the initiative for the
issues to be discussed. Professionals aim at responding in such a
way that all the participants can express their concerns. Each utter-
ance is adapted to what was previously said and the way towards
new experiences is opened. In the shared guiding of the process,
clients take charge in leading into the themes that they can handle
and proceed for as long as is possible for them at that moment.
When aiming at a joint understanding—in which our clients can
re-incorporate the mastery of their lives—it is not enough to have
a rational explanation about the problems and their causes. For the
new understanding to come about on how to go on in their lives, it
is necessary that it is embodied in comprehensive experience.
Bakhtin (1984, 1986) understood dialogue as the condition for
the emergence of ideas. It is in the particularities of exchanges be-
tween persons in the moment that meaning develops, not within
either party’s head alone, but, rather, in the interpersonal space
between them. In “borrowing” words already richly endowed with
the meanings they carry from their history of prior use, participants
in dialogue will craft meanings for those words that are unique to
the particular occasion of their use. An utterance derives its mean-
ing as much from the listener as the speaker; for words to have
meaning, they require response. This dependence on response for
meaning contributes to what Bakhtin calls the “unfinalizability”
of dialogue (Holquist, 1981). Meaning is constantly generated and
transformed by the intrinsically unpredictable process of response,
response to response, followed by further response, in a process
that may be interrupted but can never be concluded. The more
voices incorporated into a “polyphonic” (Bakhtin, 1984) dialogue,
the richer the possibilities for emergent understanding. Thus, team

members strive to draw out the voices of every participant in the
room. For each theme under discussion, every individual responds
to a multiplicity of voices, internally and in relation to others in
the room. All these voices are in dialogue with each other. Thus
the aim is not to find one description or explanation. Dialogue is a
mutual act, and focusing on dialogue as a form of psychotherapy
changes the position of the therapists, who no longer act as in-
terventionists, but as participants in a mutual process of uttering
and responding. Instead of seeing the family or the individual as
objects, they become parties in subject–subject relations (Bakhtin,
1984).
Creating new shared language
In dialogue, producing words in our bodies is the basis for the

new joint meanings. Speaking is an action in which the speaker
creates possibilities for her/his own self to understand what the
things she/he said mean for her/him. A Russian psychologist Lev
Vygotsky (1934) suggests a theory of the social origin of language.
Originally, the internal speech in which we can form thoughts
was outer, taking place in the social interactions between the par-
ent and the infant. For the infant, language first emerges in the
meaning ecology created by her parent. In the phase of egocentric
speech approximately between 3 and 7 years of age, the child starts
to incorporate this meaning-creating system into his/her inner
psychological tools. By speaking aloud, the child starts to guide
his/her own behaviour and so behaviour originally emerging in
the social becomes a part of the internal. Internal speech—mainly
thinking—becomes possible. But the social origin of speech stays
throughout our lives as one aspect of language, the inner speech
as well as the social speech. Interlocutors in a dialogue continue
in their inner speech the dialogue started in the social exchange
of ideas. And in the outer social speech, the speaker can speak to
him/herself and continue it in his/her inner dialogue, again. In
Open and Anticipation Dialogues, the importance of listening and
thinking in peace the words that others said is not only a situation
in which someone has inner speech and the other social speech.
The person speaking is, at that very moment, speaking within
his/her inner dialogue as well.
The mediator between monologue and dialogue can become
the egocentric speech, speech to our selves, presupposing that the
words uttered become heard. In a crisis meeting the speaker is
concretely forming words in his/her vocal cords of his/her most
difficult experiences. This often is an anxious process, since dif-
ficult emotionally loaded things are included that have not been
spoken previously. In this very process producing the first words,
the speaker can hear to him/her self, what that words he/she cre-
ated means for him/her. Tom Andersen (1995) describes this by
saying that speech is at the same time informative and formative.
The speaker informs the interlocutors of his/her experiences, but,
at the same time, forms his/her experiences and becomes more
aware of them. If the interlocutors hear what he/she is saying,
the things spoken are landing in between all the participants in
the dialogue, and through this the process can become a healing
experience. The speaker is at the same time dealing with the issue
in the social reality and in his/her inner dialogue. The speaker is
acting linguistically outside and increasing his/her inner agency.
A sign of this type of healing experience can be the moment when
the speaker, while speaking, becomes touched by the words he/
she has said. The task for the interlocutors becomes to accept what
the speaker said entirely and not to give extra meaning to it by, for
instance, interpreting it. Giving an interpretation can give rise to
the need for the speaker to defend him/herself, thus preventing
the new process.
Vygotsky’s idea of the “zone of proximal development” pro-
vides a frame of reference for understanding how the actions of
team members support the flow of emotion in treatment meet-
ings. The zone of proximal development is the metaphorical space
between the student who strives to learn new skills just beyond
the limits of her/his current ability and the teacher who, already
having mastered those skills, draws the student forth, offering the
teacher’s skills as a scaffolding to support development of the stu-
dent’s skills. The process is, however, not a one-sided act directed
from the more skilful person to the student, but a mutual coopera-
tion, in which the one in charge must constantly adapt his/her
activity to the learner (Bruner, 1985). In the case of Open Dialogues,

it appears that the experienced mastery of strong emotions of the
team members in meetings provides a secure framework in which
network members discover their abilities to sustain conversations
about the most difficult of experiences. Although moved by the
emotions in the room, team members are still not as fully embed-
ded as are network members. Not having participated in the past
events that have shaped the current crisis, they are less vulnerable
to being overwhelmed emotionally. They do not share the intensity
of the network members’ bodily involvement in the feelings in the
room. Their experiences with other crises in other networks have
shown them that the current crisis can be survived. The particular
experience of the team members is embodied in their presence
in the room, as they radiate calm confidence and compassionate
engagement. Demonstrating with their embodied presence that it
is possible to talk-through extremely difficult experiences, they af-
ford feelings of safety that make it possible for network members
to venture forth from their monological impasse.
In Anticipation Dialogues, the questions of the interviewer ini-
tiate a searching for one’s own thoughts. The meeting is not of a
crisis, and thus no one exact theme for the meetings exists. Themes
for discussions are reviewed by asking about each positive future.
The future hopes and concerns and ideas of one’s own actions
and of support are not that clear in each person’s mind that, when
allowed, they are simply spoken. The interviewer assists in scruti-
nizing themes by asking for thinking aloud. Thus both the person
speaking and the others listening are encouraged to also find their
own thoughts in the inner dialogue. The interviewer often repeats,
“did I hear right when I heard you saying that . . .”. Emphasis is
on hearing and precise listening, not on interpretation of what we
hear. In Open Dialogues, a shared language is aimed at, and thus
one can say: “I did not quite follow what you said. I have not expe-
rienced something like that—Could you tell me more about it?”
Symptoms exist in comprehensive, embodied emotions. The
new language is generated in comprehensive, embodied experi-
ences as well, and not in rational explanations. Professionals aim at
understanding both the circumstances and the emotions concern-
ing the crisis. This takes place both by giving words to experiences
and by living through the actual emotions that have not been aired
through words. For instance, a big part of the experiences of a psy-

chotic patient exists in the embodied emotions of the participants
in the meeting, because psychotic experiences in the beginning can
be impossible to understand. In the process of shared understand-
ing, the family’s capability for creating coherent stories of what
has happened increases. This means possibilities for incorporating
traumatic events into one’s life history and scrutinizing and study-
ing one’s own reaction and emotions.
Creating a new community
In developing shared experiences and joint language, a new com-

munity is generated either through continuing meetings with the
same people several times or through continuing dialogues after
one meeting. A safe-enough community is not self-evident: the
professionals may not have seen the client they are meeting to
plan rehabilitation (Kokko, 2003). In these meetings, professionals
may act as if the understanding is not generated in the specific
context, but out of a general knowledge of how this and that type
of problem should be handled, as if an expert who is sufficiently
trained can do exact diagnosis and select the right methods with-
out personal contact.
In dialogue this is not possible, since new understanding arises
between participants in the actual conversation. Interlocutors have
to be present and available to each other since the new understand-
ing cannot be generated by some trick of the professional. Being
present means hearing, and not only listening. An expert has to
adapt to the client’s language and not only focus on her/his own
specific professional language. One has to expose oneself to being
moved by what is discussed.
In Open Dialogues the social network is gathered or disbanded
through a joint decision. The treatment community is based on
guaranteeing continuity and psychological security and is differ-
ent compared to traditional psychiatric care. It is not an artificial
community in the ward, but a network of real living persons who
are important in the patient’s life. The community lives though the
crisis for as long as needed in a way that is possible for the family.
In the community, new treatment methods can be included based

on specifically defined needs, but the meetings continue with the
core network for the entire treatment process, even after the worst
crisis has eased off. The deeper the crisis that can be lived through,
the more versatile a process it seems to predict. The community is
not, however, a stable structure for controlling the entire life. It is
more an experience of how the people participating in the dialogue
become important in finding the answer for how to go on.
In Anticipation Dialogues, this experience is different, because
a succession of meetings is seldom organized. After the dialogue,
each person returns to the agency he/she works in, having had
the opportunity in the meeting to deal with the difficult problem
with other participants. The active social network becomes smaller,
but connections most often become tighter as if a core network has
been generated. Many professionals and members of the private
network can conclude that there is no need to be so worried about
the situation, and they can inform others that they are available
if needed. Other may conclude the opposite by noting that they
need to be in deeper collaboration. A good outcome of a meeting
could be that even though no joint meetings are organized, the
“understanding network” does not disappear. It can leave its signs
by echoing in the minds of the participants in their next meetings
with other networks and by decreasing the threshold for contacts
among the participants in this actual dialogue. Most often a follow-
up session is organized with the most involved professionals.
The everyday language of everyday life
The professional vocabulary includes as few possibilities for inter-

pretations as possible. Professional terms cannot, however, avoid
the fact they need to be connected to specific discussion in order
to define their meaning in an actual situation. Meanings flow into
them in dialogue.
In network meetings, professionals have much more power to
define what it is all about. In their words echoes the authority of
science and expertise. In generating the shared language area, it is
important to carefully listen to what the words mean for clients.
This can include following their utterances literally by repeating

what they said. The speaker has the possibility of hearing through
another’s utterance what it means for her/him. Bakhtin (1984) sees
dialogue as a practical conversation that is more important than
theoretical language. In theoretical language—as used in medicine
or in social work—we can choose what word we use, but in practi-
cal dialogue we are at the mercy of the context. Our challenge is
throughout to adapt to the actual conversation, to its words and
form.
The following example illustrates the difference between practi-
cal and theoretical language:
Case illustration: Two types of schizophrenia

Lars was a severely psychotic young boy. He would sit in a cor-
ner of the ward and have no contact with anyone. After three
months without any noticeable improvement in his condition,
the therapeutic team decided to have a joint meeting to discuss
the serious situation. The team invited all the professionals
involved in his treatment, from both outpatient and inpatient
care, and his family. At some point in the conversation, Lars’s
older sister said, “The last two weeks have been hard on the
family.” When asked what was making things so hard, neither
she nor any other family member answered. After a while,
Lars’s brother replied that, “after hearing what the doctor said,
it was tough”. He was asked what the doctor had said, and for
a second time the conversation on this subject dried up. After
a while, the sister, for a third time, took up the same issue by
saying that “it has been a tough period for the family after
hearing the doctor’s words”. She was asked to tell the group
more about the situation and the doctor’s diagnosis. She said
that the doctor had given his opinion about what was wrong
with their brother, and his diagnosis was hard to bear. She was
encouraged to repeat what the doctor had said. After a moment
of silence, she answered in a soft voice, “the doctor said that
our brother has schizophrenia”. Upon hearing this word, all the
family members started to weep.
The team responded to this incident by sitting silently, thus
making space for the emotional moment, after which the family
members were asked to say what schizophrenia meant to each

of them. They started to tell, at first hesitantly and then more
and more straightforwardly, how their father’s mother was
diagnosed as having schizophrenia and that she had been hos-
pitalized for 35 years. The family had tried to have the woman
live with them, but this always failed because she had strong
delusions that they would either poison her or make use of her
property in some other way. This history was traumatic for all
the family members, and they never talked about it. It was a
history without words.
The doctor who made the diagnosis was asked to describe the
things that led him to view the problem as schizophrenia. He
did so and said that he wanted to start Lars on the best possible
treatment. He did not think that Lars should stay in the hospital
for the rest of his life. At this point, a new type of conversation
emerged: one between the doctor and the family members. This
helped everyone to see the seriousness of the situation. In the
same conversation, it became possible to talk in a new way of
their experiences with the father’s mother (whom they began to
speak of as “grandmother”) and to supply words for a narrative
that previously had had none.
In this example, the doctor perhaps aimed to find the rule—the

right diagnosis, “schizophrenia”—governing the boy’s behaviour.
This one and correct diagnosis meant an end to the doctor’s inter-
est of this research problem. But in the joint meeting, this defi-
nition triggered an avalanche of new meanings, which opened
up in the shared conversation and prompted new understanding
between the discussants. In the meaning-network constructed be-
tween these individuals, the diagnosis of schizophrenia of course
had its place, since it formed the theme of conversation. The talk,
however, no longer focused on the meaning of schizophrenia to the
inner psychological or biological structure of the patient, but on
the actual conversation then and there, on what “schizophrenia”
meant to every participant. This led to a polyphonic deliberation
of each person’s own experiences of schizophrenia and of matters
related to the grandmother and to Lars’s future. The originally
one-voiced, monological words started to receive multi-voiced,
dialogical aspects.
In Anticipation Dialogues, the dialogue is started by first ask-

ing the family members to tell of their positive future. It is the
everyday life of family members that should be supported. If the
interview started with the professionals, the problem would be
focused too much on the specific terminology possessed by each
professional. There is also the danger that such a discussion would
repeat the fragmented view that existed before the meeting took
place. The interviewers do not ask the professionals what they see
as a good future for the family. The professionals do not live the
family’s everyday life and it is not their task to define how the
family members should see their future. Professionals can support
everyday life and lessen distress.
Tolerating uncertainty—
increasing psychological resources
In Anticipation Dialogues, facilitators regulate the social space.

They have a specific structure to follow in the meeting. This is
meant to increase tolerating the uncertainty of whether the actual
meeting is leading to no better outcome than the previous ones that
often have ended in frustration. Another factor of uncertainty is the
question of what each person dear to the family members is to say.
Facilitators contribute to the tolerance by describing in the begin-
ning the frames for the meeting. They say that each is interviewed
as if from the future and after that they will return to the present
to plan the next steps to be taken. They have to guarantee that the
dialogue is respectful, polyphonic, listening, and that the speaker
is not interrupted. They have to remember to end the meetings
with specific decisions on the future steps as well.
The process of remembering the future creates an exploration
of the tentative possibilities for thinking about it. This thinking is
done in the presence of the social network, and thus it has a rel-
evant contribution to the actual meeting, not only for future plans.
The dialogue itself generates a way of thinking of the future that
empowers the participants in claiming more agency in their lives.
If they become heard, if they are affected by the thoughts of others,
if they give shape to their own thoughts and experience by par-

ticipating in thinking together, they will experience the situation
as empowering.
There are some specific conditions for Anticipation Dialogues
to enable increasing the tolerance of uncertainty. This is taking
place under the following circumstance:
1. If participants have a feeling of being handled in an equal way.

In the opening phase, they are often so excited that they do not
fully follow the orientation of the facilitators. Through seeing
how the interview is going on and continuing from one person
to another, they have the possibility of recognizing the equal
interest in listening to every speaker and seeing that everyone
is asked the same questions and no blaming is occurring. In
follow-up interviews many clients have wondered “when does
the blaming begin again?” and had been surprised when it had
not happened in the meeting(s).
2. If themes are allowed to become freely saturated. Facilitators
have to tolerate no single discussion topic being selected for
the main theme for the meeting; instead, if the speakers start to
respond to each others’ concerns, then subjective themes start,
step by step, to become saturated as central themes.
3. If a trustworthy plan that includes concrete acts is discussed.
Facilitators inform everyone at the outset that at the end of the
meeting there is time reserved for the concrete planning of co-
operation. They increase safety by showing through their own
actions that the structure for the meeting is both flexible and
dependable. Plans for the future should be generated in a way
that the participants in the meeting see the theme as realistic
from their perspective. It is especially important to focus on the
last questions of “what concerns did you have a year ago, and
what made it possible for your worries to be become less?”
In Open Dialogues, the activity of constructing a new shared

language—incorporating the words that network members bring
to the meetings and the new words that emerge from dialogue
among team and network members—affords a healing alternative
to the language of symptoms or of difficult behaviour. The team

helps cultivate a conversational culture that respects each voice
and strives to hear all voices. Essential team actions towards this
purpose include the following:
1. Asking for information in a manner that makes telling the

stories as easy as possible and as de-stressing as possible. This
includes using everyday language, pursuing details, and invit-
ing comments on people’s responses, thus generating a multi-
voiced picture of an incident.
2. Listening intently and compassionately as each speaker takes
a turn and making space for every utterance, including those
made in psychotic speech. By so doing, the psychotic comments
become a part of the ordinary conversation, and in a way they
are “normalized”. This does not mean disparaging the dif-
ficulty of the experience, but actually the opposite: the team
members try to catch as accurately as possible the severity of
the problem as it is experienced by the family members. Often
this means showing appreciation for the extreme life situations
that engender psychotic ideas and feelings of hopelessness.
3. Reflective dialogue is conducted among team members, com-
menting not only on the network members’ utterances, but on
each other’s utterances about the network members’ utterances.
One of the main elements is guaranteeing space for both talking
and listening. In Open Dialogues, the reflective conversation is
done by shifting the position from interviewing to commenting
with other professionals. In this it is important to emphasize
looking at the one with whom we speak and not the clients.
Clients have more space for their inner thoughts about what is
said. By tolerating this uncertainty, network members discover
in their sharing of the situation the psychological resources to
bring to the question of how to move on.
After team members have entered the conversation by adapting

their utterances to those of the patient and his/her nearest relations,
the network members may in time come to adapt their own words
to those of the team. It helps one to understand more when one ex-
periences the other as understanding oneself. If one discovers that
one is heard, it may become possible to begin to hear and become

curious about the experiences and opinions of others. Together,
team and network members build up an area of joint language, in
which they can come to the same understanding of the words used
in the situation. This joint language, emerging in the area between
the participants in the dialogue, expresses their shared experience
of the incidents and the emotions embedded in them.
The shared experience unites the actual embodied emotion,
the psychological description of it, and the social meaning of the
described things. These cannot be separated, but they emerge si-
multaneously, in the “once occurring event of being”, as Bakhtin
(1984) formulated it. The main task of the team is guaranteeing
the possibility for this comprehensive moment in the dialogue.
While speaking, it is good to proceed slowly in order to guarantee
that there is space for the emotions to stay connected to the issues
described. If team members try to move the conversation forward
too quickly at such moments, there is a risk that it will shift to a
solely rational level. The most difficult and traumatic memories are
stored in nonverbal, bodily memory (van der Kolk, 1996). Creating
words for these emotions is a fundamentally important activity.
For the words to be found, the feelings have to be endured. Em-
ploying the power of human relationships to hold powerful emo-
tions, network members are encouraged to sustain intense painful
emotions of sadness, helplessness, and hopelessness. A dialogical
process is a necessary condition for making this possible. To sup-
port the dialogical process, team members attend to how feelings
are expressed by the many voices of the body: tears in the eye,
constriction in the throat, changes in posture and facial expression.
Team members are sensitive to how the body may be so emotion-
ally strained while speaking of extremely difficult issues as to
inhibit speaking further, and they will respond compassionately
to draw words forth at such moments. The experiences that had
been stored in the body’s memory as symptoms are “vaporized”
into words, as Vygotsky (1934) described it.
The team’s task is to realize that the emphasis throughout is on
what is taking place at the moment, here and now. The expertise
needed focuses on how to relax into the present interaction and
how to be present for the clients, who are speaking of their most
extreme experiences. In doing this, the team, of course, is working

pedagogically, as well. Through their actions, they inform about
their way of handling difficult and emotionally laden issues. Their
earlier experiences of many other crises are, in a way, incorporated
in their embodied actions. They increase the safety for handling
threatening issues in a detailed way.
It is important to remember that all the members of the network
are struggling with the emotional intensity of the incidents and
experiences that comprise the crisis, albeit from different positions.
Network members may have acted to bring the crisis on, or lived
through the effects of the crisis, or both. The hallucinations of a pa-
tient having psychotic problems may incorporate traumatic events
in metaphoric form. Although the symptoms’ allusion to the trau-
matic events may thereby be inaccessible to network members,
they themselves may have been affected by those same events,
and their own embodied emotional reactions are stimulated. The
emotional weight from these collective interactions and amplifica-
tions of emotional states make the network meeting very different
from a dialogue between two individuals. The intensity seldom
manifests as a huge explosion or catharsis. It emerges most often
as small surprises that open up new directions for the dialogue.
By its nature, the emotional exchange occurs in the immediate
moment, and the experience cannot be moved as such to another
time or place. The outcome of the meeting is experienced more in
the embodied, comprehensive experiences of the participants than
in any explanations offered for problems or decisions made at the
end of the meeting. This may be unusual for professionals used to
working in a more structural way.
As multiple voices join in the sharing of the situation, new pos-
sibilities emerge. These possibilities seldom emerge as a single un-
ambiguous response to the question of how to move on. Different
network members live in very different, even contradictory situa-
tions and thus have very different ideas about the problem. Con-
sider a crisis surrounding a mother, father, and son, in which the
son, suspected of drug abuse, becomes nearly psychotic. The father
may be concerned primarily with the family’s reputation among
his co-workers, the mother about her son’s health, and the young
man may protest angrily that he does not need any treatment, that
his parents are crazy and should seek treatment themselves.
Change is co-evolution
Generating a joint language area is the condition for joint deci-

sions. When others understand, it helps the person to understand,
presupposing that everyone shares the core of the experience. In
this sharing is created an experience of the themes under discus-
sion that is sufficiently similar. And vice versa: only after reaching
an area of joint language can we become part of decisions in which
we understand the words in a sufficiently similar manner. If the
team members start to adapt their words to clients’ words, clients
may have more possibilities for adapting their utterances to each
other’s utterances.
Our experience of living through many crises in many net-
work meetings is embodied in our presence in the room; the
team can thus radiate confidence and compassionate engagement.
Vygotsky’s idea of a “zone of proximal development” is helpful
in understanding how this team stance facilitates the meeting
process. In borrowing this concept from developmental psychol-
ogy, the processes could be illustrated in the following way. The
crisis consisting of emotional experiences can be seen as a block to
development, where a future perspective is lacking, which leads
to a situation where the family cannot access their own natural
psychological resources. The possibility for a dialogue has disap-
peared for a while. In joining with professionals, a new commu-
nity is created. Professionals come into this new community with
their organization, flexibility, and multiplicity of choices. All this
emerges as devoted listening to the family members, as a flexible
handling of different alternative perspectives, as tolerating the
uncertainty. In the joint system, their organization can start to be-
come the basis for the interaction, and the family, being a part of
it, can again become an agent in dealing with their own difficult
issues. The perspective for the future becomes possible. We have
previously referred to this as co-evolution (Arnkil, 1991a, 1991b;
Seikkula, 1991), a mutual development, in which all the partici-
pants themselves change as well as the interaction between them.
When the family—and the patient as a part of the family—first act
in the system of boundary, they can thereafter repeat the mastery
of their own actions without the presence of the treatment system.
All this is, of course, only a very rigid description of a process that,
in real life, does not follow a specific order. Things may happen
very rapidly, as if with a single leap from one position to another,
where people may have more control in dealing with their prob-
lems. On some other occasions the progress can be through “two
steps forward, one back” throughout.
In Anticipation Dialogues, family members are interviewed
first, and professionals are not asked of their views about the fami-
ly’s positive future. Everyone is encouraged to speak only from
his/her individual point of view. For each an individual zone for
proximal development is opened that is a possibility for learning
about his/her own action. This is also the case with the facilitators.
If they do not learn something new in the dialogue, the dialogue
most probably has not been an optimal one. Remembering a posi-
tive future can be done in a mechanical way, and nothing new is
learned, but at those meetings it is rarely an indication of a dialogi-
cal meeting-point.
The whole body, not only talking heads
The experiences of the patient and those nearest to the patient are
entangled: the patient is not an isolated actor in the social context.
Part of the crisis may be directly caused by these people present;
part of it exists as a shared experience with these people. For in-
stance, in the psychotic utterances the origin of the experiences
become fuzzy, and the traumatic events may not be seen clearly
because the patient is speaking not clearly of the things he/she
has lived through.
All this forms an exciting tension at the very beginning, and
this “fusion loading” is verified in the meeting, in that most often
the one behaving in a psychotic way discontinues this way of
behaving during the meeting. He/she is heard, and psychotic-
speak is no longer needed in the context of the meeting. When
therapists are working as a team, they have to become responsive
to and thus connected with each other’s language in the presence
of the context of the patient’s social network. The patient, too, is
speaking in the presence of those who most probably are the core
persons in the difficult experiences in his/her life. Together with

the therapists, they form a new, mutually responsive community
for not only living through the crisis, but also to construct a new,
joint language for the as-yet unspoken experiences.
The same type of emotional loading exists in Anticipation Dia-
logues as well. Although meetings in this situation are not in
response to an intense crisis phase, the concerns about a child, an
adolescent, or an adult have their origins in the field of social rela-
tions, and the concerns are directed towards everyone’s action in
this field. There may be strong expectations for someone to change
his/her behaviour, and many can be deeply frustrated that things
are not improving. The main goal is increasing the coordination of
actions after the meetings, but for that goal to be realized, everyone
present in the meeting should be heard and have the possibility to
hear the others. Concerns will diminish in the actual dialogue.
Case illustration: A violent, caring father

Jaakko was involved in one consultation in which the com-
prehensive quality of the meetings was illustrated. We met
for consultation at the request of the family therapist with a
21-year-old woman, her husband, her mother, and the family
therapist. The therapist wanted to avoid hospitalization in an
extremely difficult situation. Maija, the young woman, had be-
come more and more afraid that she was about to be killed. She
was so afraid that she cried from fear through most of each day.
Fearing also that someone would kill her mother, she would go
to meet with her for a couple of hours every night to make sure
that all was well with her. She told us that she would stay in bed
every day until her husband came home from work.
As the mother offered her description, she asked if their family
history with Maija’s father could be contributing to the extrem-
ity of the situation. She said that her former husband used to
be very violent, and that before the divorce, when Maija was 9
years old, Maija had seen him attacking her several times. As
the consultant asked for more details about the father, all gave
a very negative description of how bad a man he had been for
the family. There had been no contact since the divorce; their
only information about him since then was through letters and
occasional accounts from relatives. Asked for more detail about
the family incidents before the divorce, the mother became up-
set and said that there was no use starting to speak about old
things. Yet she asked if it would be of any help to talk about
them. The consultant asked, would it be possible for her to dis-
cuss the old things, if she knew that it would help her daugh-
ter? She was uncertain but promised to think it over.
Maija’s husband added that he knew that the father had been
a psychiatric patient, and that they had heard that he was
currently in the hospital. Towards the end of the meeting, the
question was raised as to what if the father had been present
in the meeting; what would he say to help in Maija’s situation?
This question made Maija and her husband very upset. Maija
started to cry, and said that her father would have killed them
all. Her husband said that this kind of question would ruin the
help that Maija needed, and it should not be asked. Surpris-
ingly, the mother started to cry while answering the question,
saying that the father had been very worried and had tried to
help. All of a sudden, in the middle of a very negative descrip-
tion of the father, a new voice describing him was presented by
the mother. She said that the father was not all bad, that they
had had good moments in their marriage and in the family
as well, and that during those moments he had been a caring
father for the family.
In this case, the network members were encouraged to present
their worries about a difficult situation. The team accepted
their words without hesitation, and without interpretation.
They did not, for example, suggest that the father’s violence
could be behind the daughter’s psychotic symptoms. We were
very interested in descriptions of the order of daily activity,
and of everyday life incidents. Family members could speak
in an emotionally involved way of the themes under discus-
sion—for instance, Maija cried while speaking of her father.
In many ways, the father’s presence in their stories was as a
significant, unspoken subject. It was relatively easy to access
more of his presence in an as-if form by asking, “If he were
present, what would he be saying?” This stimulated the inner

dialogues of the network members, as they imagined engaging
with the father as a speaking presence in the conversation. The
mother was suddenly moved by this, and began to remember
the good periods in their family life. A new voice emerged, a
voice of the family and of the father that had been impossible
to hear before the meeting. If the father had only remained as
the violent man, who seemed to be the one responsible for the
problems in their life, no dialogue with the other things he
represented would have been possible. The opportunity for a
more polyphonic picture of the family and the father would not
have opened up. Of course, more meetings would be necessary
to expand on this.
This case illustrates well the comprehensive quality of the meeting

experience. People’s utterances included expression of the strong
emotions associated with their content. New meanings emerged
in the context of new, emotionally shared and embodied experi-
ences in the very meaningful social context of the patient’s nearest
personal relationships. The voice of the father as a caring person
was new and surprising, and speaking of it made a new opening
possible towards a more polyphonic description of their present
life and of their history.
Dialogue is experience. It may lead to living through something
that was not possible before. Experiences are experiences of living
persons; they are embodied, and there is not always a need for
formulating them into words. In dialogue the new experiences
often emerge as bodily relaxing emotions, and it might be that just
this moment of relaxing in the dialogues about the most extreme
concerns in one’s life is the turning point towards healing and
problem solving. Symptoms can be seen as ways of living that are
settled down in our bodies, and they are difficult to formulate into
spoken narratives because of their most anxious content.
Through joint sharing, a new language may emerge. In this
new language, stories can be told that were not previously possible
because of either the anxiety aroused or the traumatic content of
the experiences. Words arrive in the place of the symptoms, pro-
vided that it is a dialogical meeting. The body can start to function
towards dissolving the symptoms after a relaxing of the tension in

the meeting.
Conclusions as simple instructions
Out of what has been described in this chapter, some concrete

conclusions can be drawn for a worker in the meeting aiming at
dialogue.
1. Generating safety and minimizing anxiety. Everyone has to feel

him/herself as important in the conversation. The team and
other professionals should guarantee safety and increase pre-
dictability in the meeting. The latter is not in the sense that the
themes for discussion should be defined in advance or that
the conclusions on the problem and the decisions of treatment
should be rapidly concluded: predictability means that clients
can rely on the professionals being available to them and that
the members of the private social network become an option.
Clients should be able to trust that the treatment is planned
together and that they are not left alone in a difficult situation.
Increasing the predictability helps to generate the perspective
towards the future so that it becomes possible to handle the is-
sues related to the actual crisis or concerns. This predictability is
different compared, for example, with a traditional psychiatric
ward, in which the predictability is guaranteed by an exact
daily programme. This type of predictability increases safety,
but, in addition to it, possibilities for a psychological process
can be increased: the dialogue in a treatment meeting is one
such positive process.
2. Opening discussion and demonstrating a genuine interest in what
each person is saying. Do not give the message that someone said
something wrong. The professional vocabulary easily creates
situation in which the professionals may sneer at the client’s
comments if he/she does not know the exact code or meaning
of some utterance. The big challenge for the professionals is
how they manage to follow the client’s language, instead of
presupposing that the client learns their language. For instance,
“reality-orientating” the psychotic comment of a patient by

saying that “what you just described is a psychotic experience
that is not true” seldom includes a genuine interest in what the
patient said.
3. Generating dialogical utterances. Because of the strong anxiety in
the crisis, dialogue is reached almost without any effort (if we
do not manage to hinder it). Demonstrating emotions should
not be inhibited; it is not dangerous. Emotions bind the said
with the embodied experience, and thus the “intensity” be-
comes more effective. Making this possible often presupposes
that professionals have a calm way of discussing, give sufficient
time for searching for words for the reply and forming the
utterances, and empathize with the answers through their en-
couraging comments. It is important to note that our respond-
ing occurs much earlier than before we form the first words
for our answer. The way we sit, the way we gaze, whether we
are relaxed or not, whether we take into account all the people
present—these are all acts that tell our clients if we really are
present or not.
4. Responding always to what is said. This does not mean giving
an explanation or interpretation, but noticing what is said and
giving perhaps new nuances or perspective to it (“yes, and
in addition to that . . .”). You can begin your response by, for
instance, repeating word for word what the previous speaker
has said. All this does not mean interrupting every utterance
as often as possible to give a response, but, rather, finding the
natural rhythm of the conversation and adapting your com-
ments to that rhythm.
5. Speaking for your own sake and in I-form. Through the ques-
tions, the team members can demonstrate a habit that each
one is speaking of his/her own views and emotions and not
of another’s. There is often a need, at least at the beginning of
a conversation, to emphasize what is the speaker’s own sub-
jective view. Thus, everyone is participating in generating the
dialogue in his/her own voice.
All these instructions are relative. More important than follow-

ing exact rules is to be present. Being present is shown by many
aspects in each specific meeting. Dialogue emerges in the once

occurring event of being (Bakhtin, 1984), and thus each meeting is
a unique occasion creating new and fresh meaning for the things
spoken. The meanings we receive in the previous conversation
become a part of the current one, but, at the same time, old themes
receive new senses as well.
PA R T III
CHAPTER SEVEN
Dialogue
and the art of responding
S
ymptoms are replaced by words, as we said in the previous
chapter. Our aim in the present chapter is to analyse the
possible ways to realize this process. Are there ways of dis-
cussing that better enable dialogue or ways that are not so good
at generating dialogue? Are there ways that actually inhibit new
words emerging? Our goal is to define factors enabling or inhibit-
ing dialogue.
In the previous chapter we outlined some elements of dialogue
that can generate change. Going further in analysing the healing
elements of dialogue presupposes specific analyses to be used.
Some studies have already been conducted that analyse the differ-
ences between productive or poor dialogues (Haarakangas, 1997),
and our aim is to further that discussion.
Generating dialogue in psychotic crisis
The therapeutic aim in Open Dialogues is to develop a common

verbal language for the experiences that would otherwise remain
131
embodied within the person’s psychotic speech and private, inner

voices and hallucinatory signs (Holma, 1999; Seikkula, Alakare, &
Aaltonen, 2001b).
The Bakhtinian idea of dialogue and its adaptation to the psy-
chotic situation derive from a tradition that sees language and
communication as primarily constitutive of social reality. The crisis
becomes an opportunity to generate new stories, in which the ex-
periences emerging in the form of symptoms are clothed in words.
Open Dialogues thus translate Bakhtin’s concept of dialogism into
a co-evolving process of listening and understanding (Hoffman,
2002). Actually, listening becomes more important than the process
of interviewing. For this reason, the first questions in a treatment
meeting are as open as possible to give maximum opportunity for
the family members and the rest of the social network to be able
to speak about whatever issues are most relevant to them at that
moment. The team does not decide the themes in advance.
To generate dialogue from the very beginning, one of the tasks
of the interviewer(s) is to “answer” what the patient or others
have said. However, the answers usually take the form of further
questions that are based on a previous utterance of the patient.
Every spoken statement, or utterance, requires a reply. The fitting-
together of utterance and reply makes the dialogue “dialogical”,
rather than “monological”, which would be a speaker without a
contributing listener (Voloshinov, 1996).
Psychotic comments as a response to the ongoing dialogue
In Open Dialogues, the discussion starts with everyone present in

the same room with as little preplanning as possible by the team.
Compared to a systemic family therapy, the team is not aiming to
find the rules of family behaviour and to provide some interven-
tion for changing the family interaction. Therefore, the need to
plan a strategy for the meeting does not arise. When the aim is
generating dialogue so as to create new words for the experiences
lived through the hallucinations, it is important to ensure that the
participants have a common history within the dialogue, both the
team and the private social network members.
DIALOGUE AND THE ART OF RESPONDING 133
In dialogue, each participant becomes involved in his/her own

language. Each one participating in the dialogue starts to give
words to the most acute experiences. For a patient referred to treat-
ment, hallucinations are most often a frightening and incomprehen-
sible phenomenon. Such phenomena are expressed in situations
that touch on themes in some way connected with the psychotic
experiences. One aspect of hallucinations seems to be that earlier
extreme or traumatic experiences are included in them. Patients
often speak of real, pre-existing occasions in their lives, although
they may not be understandable to everyone present. In fact, in a
way hallucinations may increase the possibilities for some relevant
themes in their past experiences to be given words, although in a
metaphorical way. Aiming at a dialogue, they should not be “real-
ity-oriented” by saying “these experiences belong to your psycho-
sis and are part of your illness”. In aiming at generating dialogue, a
better way could be to open up interest in the psychotic utterances
by pausing other conversational themes and asking, for instance,
“wait a moment—what did you say? I did not follow. How would
it be possible for you to control your neighbours thoughts, I have
not done that. Could you tell me more about it, please? When did
it start? Does it happen all the time or only in the mornings or
nights? “And so on. Others present can be asked how they un-
derstood what the patients said. By so doing, the team creates an
atmosphere or an attitude that every type of utterance is valued
and possible and even psychotic ways of expressing oneself can
become a part of the ways of expressing within the conversation.
It becomes possible to talk about those very frightening and odd
experiences and thus for the voice to become one voice among
many other voices in the conversation. In this conversation, both
the patient and the family participate in a process of constructing
new narratives of restitution and reparation (Stern, Doolan, Sta-
ples, Szmukler, & Eisler, 1999; Trimble, 2000).
The points at which the patient opens up with an incomprehen-
sible utterance that the therapist may think includes a psychotic
experience are especially important for the future dialogue. More
psychotic utterances can be expected in the beginning of the first
meeting than will emerge towards the end of the meetings. This
is an understandable reaction to a new and exciting situation, in
which no one in the family knows how to behave—at least in the
first psychotic crisis. If it happens in the middle of the conversa-

tion, most probably themes are being handled that are connected
to the experiences that live in the psychotic speech. At that point
in time, it is advisable to break away from other ongoing conversa-
tions and thus create space for psychotic voices to become a part of
any other voices. The following case example illustrates a situation
in which the behaviour changes into a psychotic one. This seemed
to happen after the team, through its questions, opened up new
themes in the dialogue.
Case example: The rapists are coming

Anita herself contacted the psychiatric polyclinic on her moth-
er’s advice. At the beginning of the first meeting, she told of
her fear of becoming mad. She first spoke of her fear in a very
coherent way and said that over a couple of months she had
lost her memory. Although she remembered older things, she
was uncertain of the recent events in her life. She also said
that she suspected she had been involved in fights and even of
hitting someone, but that she could not accurately recall this.
She herself said that perhaps she is paranoid. The following
sequences1 describe the very first utterances in the first meeting.
Anita is represented as A, and T represents the therapists. Anita
and three therapists are present in the meeting.
T1: Where should we start?
A: I can’t really remember anything of my whole life
T1: Has it been like this for a long time now, that you can’t
remember anything?
A: Well, I don’t know if it has been that way for two months. I
do remember whether I’ve been in contact with someone. . . .
But then when I leave my place, I don’t even know if I was
there, and where I might suddenly pop up is here. . . .
T2: With whom are you living?
A: I’ve been living by myself, but now I’ve gone to my par-
ents. . . .
T1: And for how long have you been living by yourself?
A: Hmm . . . for three, four years. Three years.
In her answer to the first question, Anita opened up the core

theme of her experience. The team constructed a number of
questions as answers to Anita’s reply. From the very beginning
the conversation was very informative, so that in a short time
relevant information to Anita’s life situation was gained. She
herself had a suspicion that she might have severe problems,
but she could describe her situation well. Although she spoke
of odd experiences, she was not psychotic in her speech. A
change started to emerge after she started to tell more about
her family members.
T1: Whose idea was it that you came here?
A: Well . . . Mother’s.
T2: And what was Mother worried about?
A: I don’t know if I’ve been talking with her. I can’t really
remember anything. I have a feeling that I have even hit
someone, but I can’t even remember.
T2: Has someone said that to you?
A: No. But I am paranoid and lost my memory. You think that
something has happened.
T1: What about your father. Is he worried about some specific
issue?
A: I don’t know, but yesterday night when we were watching
TV he went to bed and in the morning he left to work.
T1: And what was the situation when you returned home?
A: Well . . . I was afraid of others, I was quarrelling with these
kind of guys, umm . . . afraid of them and you see . . . they
had keys made to my place and then . . . they came in and
raped me and did all these things.
T1: In May?
A: I was living in my apartment. You see, someone who came
into my flat, had they been blackmailing or something . . . ?
And forced to steal a key. And they made a copy of it and
could come in whenever they wanted. I don’t know if that
happened when I was asleep . . . and they gave me some pills
and I got mixed up and started to . . . I don’t know. Or if you
take some drug without knowing it and then when you get
into her flat, they wait until you are asleep and after that
come in with their own keys . . .
In the beginning Anita described her odd experiences in a way
that she could see those as her thoughts. The team could un-
derstand the difficult situation. Although the experiences she
described may have included psychotic experiences, she herself
was not psychotic until the coherence of her story started to dis-
appear after the team member’s questions about her parent’s
concerns. Especially in speaking about her father, her anxiety
increased. She told how a gang from the street could come in
to her apartment and commit sexual violence against her. Her
story changed into a more frightening one, so that she no longer
described her fear of having lost her memory, but, instead she
was terrified and telling of what was for her a true situation.
All this seemed to happen after the team asked about her
father’s concern about her. In this meetings all this seemed
incomprehensible, but later in the treatment process severe
marital problems between the father and mother became evi-
dent. Her father had a drinking problem, and her mother was
depressed. The patient also had a belief that people who disap-
pear are dead, and this might also have affected her description
of her father’s concerns since he had disappeared off to work in
the morning. In a way, the psychotic behaviour was “caused”
by the team’s question about her parents and especially about
her father. The team cannot, of course, avoid this type question,
because they cannot know in advance what the unspoken expe-
riences are that may be connected to the psychotic behaviour.
But what the team should do is to respond by encouraging all
present to tell more about what the patient has said has hap-
pened. The psychotic story becomes one voice among others in
the dialogue, and the task of the therapist is to understand this
story, as well.
In a dialogical therapy, not too much information is needed in

advance; all that is relevant is present in the therapy room. There
is no need to define the rules of behaviour or the function of the
psychotic symptoms, as in systemic family therapy, only, rather, to
be present by responding in the dialogue. Dialogue becomes an
aim as itself.
The team creates space for the new joint language
In Open Dialogues, the most relevant social relations of the patient

participate in the meeting, and thus the new understanding that
starts to emerge is, from the very beginning, a socially shared phe-
nomenon. The nearest family members have their concerns, and
in generating dialogue they are heard and can receive support for
themselves. Most often, in the very first meeting, the patient may
come out of the psychotic behaviour (Alanen, 1997), and this sense
of a new social community with the most important individuals in
the patient’s life may be one aspect in making it possible. It is the
team’s task to create a space in which the most extreme experiences
of the family members can be shared. To do that, the challenge is
for the team to be present in their responses.
Reflective dialogue between the professionals may be one way
to answer. In some phase of the meeting, the therapists often want
to discuss with each other their observations of what they have
thought about what the family members have previously said. This
reflective conversation takes places between all those professionals
who have participated in the dialogue, without any specific reflect-
ing team. After the reflective conversation, family members are
asked if they have some comments on the reflections. The meeting
is ended by summarizing what has been discussed and what has
been agreed. In a crisis, it is advisable not to aim too rapidly for
either firm conclusions or firm decisions, but to structure it as an
open the situation.
The team discussion has another important aspect, as well,
because all treatment decisions are made transparently while eve-
ryone is present. Decisions on hospitalization, discharge, medica-
tion, or choice of different treatment methods are all themes to be
openly processed, where the clients can both see and participate
and thus affect their content. It is advisable to bring forth several
alternatives, to avoid the idea that only one way of proceeding
exists. For instance, in a decision for compulsory treatment, it is
good to hear the sceptical voices that are trying alternative ways
to progress. All this empowers the family to take charge in their
own life.
In the discussion, a paradoxical situation is generated. We em-
phasize that the family has the initiative in introducing the themes
and ways of discussing them and that the team’s task is to adapt
its behaviour to this. However, at the same time, we emphasize
the importance for the team to take charge of the process and to
be guided by it through, for instance, reflective conversation. The
team has more possibilities for continuing the process. Bakhtin
(1984) sees different positions and asymmetry as a precondition
for dialogue. It is the team who structure the time for reflection, in
general. This is the team’s way of applying expertise, in the meet-
ing, towards therapeutic ends. It may be useful to call it dialogi-
cal expertise, and we analyse this more closely at the end of the
chapter.
Team responds on the family’s language area

In what follows, two specific analyses of dialogues from the Open
Dialogues approach are considered. Kauko Haarakangas (1997)
analysed how the family and the team are connected, how the
polyphony and dialogicality are verified in the meeting, how new
meanings are generated, and what is the significance of reflective
dialogue. What seemed to be most important for connectedness
was the specific language area for conversation between the team
and the family. If the family members spoke in a very concrete,
indicative way, the situation was especially demanding for the
team. The words in such dialogue are always being used to refer
to some factually existing thing or matter. This is different from
symbolic meanings in language, where the words are being used
in a symbolic sense—that is, they refer to other words rather than
to an existing thing or matter (Haarakangas, 1997; Seikkula, 1991;
Vygotsky, 1972; Wertsch, 1985). Conversation may, for instance,
concentrate on specific treatment decisions or treatment planning.
In discussing this type of dialogue theme, the team had to learn
the family’s specific way of being in the conversation, and, at the
same time, to open the conversation up into symbolic meanings
in a way not too strange for the family members. A skilful team
manages to respond in a way that includes new voices in their
comments, which might continue with the exact word or phrase
of the clients. In the following sequence (from Haarakangas, 1997,
p. 74 ) involving the therapist (T), the wife (W), and Sirpa (S), the
family’s 11-year-old daughter, one of the team members interviews

the family by connecting to the family’s language area:
W: . . . that Erkki [her husband, who is the patient] became ex-

tremely lively and fresh, simply. No illnesses. No vomiting,
no diarrhoea. Not even any pain in the head . . . You could
see it, in a way, in his face.
T: In what part of his face?
W: Around his eyes. I have many times looked at it afterwards,
as well. I could see that he was better
T: And how was that different compared to how it is now,
here?
W: A lot more tired and . . . I see the situation.
T: Are there others apart from you who see it in the face.
W: No.
E: Irma sees it.
T: Irma, who is she?
W: She is one of our family friends
.......
T: How would she say if she was present here?
.......
T: What about you Sirpa, do you see that your father is tired? . . .
When you look at your father’s face, do you see the differ-
ence if he is tired or well?
S: He is tired.
T: Tired. Hmm. What about when he looks like well. How does
your father’s face look like then?
S: Happy.
T: And where do you notice that? The mouth or eyes . . . ?
S: Eyes.
The sequence describes a family session with a family in which

the father had severe invalidating somatic symptoms combined
with depressive mood. The therapist answered the wife’s speaking
by continuing with indicative language. The family seemed to be
very sensitive to each other’s body signs. The concern emerged by
looking at the other’s face, and the therapist followed this by ask-
ing, in more detail, what part of the face they looked at. And they
gave more and more detailed descriptions of it.
Problems in generating dialogue would emerge if the team ap-
plied circular questioning too rigidly. Other problems might have
occurred in the case of a therapist with so much experience of
being in dialogues. On some occasions, their preconceptions were
apparent in their answers to what the family said, and they did
not actually hear the words of the family members. Autio (2003)
noted that problems may appear in the sense that the team have
more power than the family to affect the form of the dialogue. Thus
the team may at some points use indoctrination to introduce new
themes for discussion that are more suited to the team’s aim. This
could happen if a decision had to be made on whether the patient
should be hospitalized or not. If the team were insufficiently sen-
sitive, they might not hear the family’s wish for hospitalization,
preferring their own desire to see outpatient treatment.
In the responses, people were heard—or not
In analysing the first meetings in psychotic crisis, Jaakko conducted

a comparison between good- and poor-outcome patients (Seikkula,
2002). The objects for analysis were transcripts of two or three of
the first- or initial-phase treatment meetings. The transcripts were
based on video-recordings of the meetings. The study’s author,
who was also responsible for developing the categories for the
sequential analysis, made the transcripts.
A specific analysis of the topical sequences was conducted by
combining the practice of the treatment meeting with both the
dialogue theories and the research methods developed to analyse
the dialogical process. Sequence analysis involves taking topical
episodes in the narrative as the main object for analysis (Linell,
1998). Topical episodes were defined in retrospect, after the entire
dialogue generated in a treatment meeting had been divided into
sequences. In each sequence, three variables were identified:
• Dominance. The first variable was labelled as dominance of inter-

action, and it includes quantitative, semantic, and interactional
dominance. Quantitative dominance simply refers to who is

speaking most within a sequence. Semantic or topical domi-
nance refers to who is introducing new content words. This
individual contributes the most content to the socially shared
world of discourse. Interactional dominance refers to control
over communicative actions, initiatives, and responses. This
individual has more control of other parties’ actions than the in-
terlocutors (Linell, 1998; Linell, Gustavsson, & Juvonen, 1988).
• Indicative versus symbolic meaning. This distinction indicates
whether the words used in the dialogue are being used to al-
ways refer to some factually existing thing or matter (indicative
language) or if the words are being used in a symbolic sense—
that is, they refer to other words rather than to an existing thing
or matter (Haarakangas, 1997; Seikkula, 1991; Vygotsky, 1972;
Wertsch, 1985).
• Monological versus dialogical dialogue. This distinguishes the qual-
ity of dialogue. Monological dialogue refers to utterances that
convey the speaker’s own thoughts and ideas without being
adapted to the interlocutors. One utterance rejects another one.
In dialogical dialogue, utterances are constructed to answer pre-
vious utterances and also to wait for an answer from utterances
that follow. New understanding is constructed between the in-
terlocutors (Bakhtin, 1984; Luckman, 1990; Seikkula, 1995).
The procedure is described in detail elsewhere (Seikkula, 2002).
Symbolic language in good outcomes

A comparison of notes on sequences for the three categories
(dominance, indicative vs. symbolic, monological vs. dialogical)
shows some differences between good and poor outcomes. In
the good-outcome group, sequences had a tendency to become
longer. It appeared that when a dialogical dialogue was reached,
the themes persisted for longer compared with monological con-
versation. In interactional dominance, in the good-outcome group,
clients (including both the patient and the family members) domi-
nated in over half (55–57%) of the sequences, as against only in
10–35% in the poor-outcome cases. The patient and the family
had more possibilities for control over initiatives and responses
in new topics of discussion. This could be interpreted as a sign

of greater possibilities to become agents in relation to the story of
their lives that was being narrated in the crisis meeting. In con-
trast, in the poor-outcome cases, this was not present during the
first meetings. In terms of semantic dominance, in all the good-
outcome cases the family showed dominance in the majority of
the sequences (70%), whereas there was more variation in poor-
outcome cases (40%–70%). Concerning quantitative dominance,
no difference emerged between good- and poor-outcome patients.
A striking difference emerged in the language area. In poor-out-
come cases, the conversation took place in symbolic language on
only a few occasions (0–20% of sequences), whereas this was the
rule in good-outcome cases (38–75%). When a family had the pos-
sibility to become involved in a symbolic-language area, dialogue
persisted longer around a specific theme, whereas in the case of
indicative language, the team often asked questions one after an-
other and the conversation itself remained in a question–response
form. In monological versus dialogical dialogue, poor-outcome
cases showed greater variation. Dialogical dialogue could occur as
well (10–50%), but it was not the rule, as it was in the three cases
with a good outcome (60%–65%).
Dialogue of violence
To demonstrate the difference between the dialogues in good- and
poor-outcome cases, two conversation extracts now follow. They
occurred in the treatment meetings of two patients, and in both
cases there was a situation of violence within the family that was
discussed.
Good-outcome case: Reflective dialogue on violence

After the first treatment meeting at home, a father and a son
had a serious quarrel concerning the son (M, the patient) who had
not been taking care of his studies and was talking about vivid
hallucinations and producing peculiar philosophical theories. The
father did not like this, and when he started to speak about them,
M began to talk of his difficult experiences during his childhood.

This led to a big quarrel, which ended in the father and son wres-
tling. This theme was taken up towards the end of the second
meeting, in a conversation between M, his mother (Mo), a female
team member (TF), and a male team member (TM):
M: It was wrestling.
TF: But did you mean it quite seriously?
M: It was like pitting oneself against . . .
TM: Which of you was wrestling?
M: I think that he got furious.
TM: Which of you was wrestling?
TF: Which of you took hold of the other?
M: Well, I took hold of his neck.
M: It was a hard headlock although I have not practiced any
bodybuilding. He has been doing this for several years. I
have not done any sports at all, so I got a little bit fright-
ened.
TM [turning towards his team-mates]: Well, it is quite, it is a kind
of outrage when you realize that your own child has . . .
TF: Yes, but, on the other hand, a father can be proud that his
own son is so strong that he himself was . . .
TM: Yes, but when it was directed towards himself, it cannot
be . . .
TF: Yes, but he could still be proud that M is an adult man and
that he is able to win in wrestling and . . .
TF: I was thinking about what M was speaking of, about
whether the dream is a whole one. But he is disputing with
his parents.
TF: Haven’t you heard of this kind of thing at the usual age of
puberty? At the age of puberty you are disputing everything
and you may have a fight to see who is going to win.
TM: So this would be a kind of . . .
TF: . . . late puberty.
TM: Quite a spurt, or is it the form which is different?

TF: The form is different. And, of course, it can be quite mad-
dening, I was thinking, when the father is a maths teacher
and it is a quite difficult situation.
M: Yes, and it is like the last straw when I make that two is
equal to one.
TF: Yes, it must’ve been the last straw.
Team members seemed to be surprised when the family mem-
bers started to speak of wrestling. But they listened and took it
seriously. After some specifying questions and comments, they
turned towards each other and started in a reflective conversa-
tion to wonder about this astonishing and frightening situa-
tion. In their reflections, they recognized the seriousness of the
father–son conflict, but they also used normalizing comments.
They started to question whether the wrestling could be seen
as a teenage revolt and as a father’s response to a teenager’s
revolt. The team commented on what they heard and they did
so in a dialogical conversation with each other.
In this sequence, M had the topical dominance, the team mem-

bers the interactional and quantitative dominance, the meanings
were created in a symbolic language, and it was a dialogical dia-
logue. Overall in this specific case, in the two treatment meetings
analysed, in 57% of the sequences the clients had the interactional
dominance and in 69% the semantic dominance. Symbolic mean-
ing-construction took place in 75% of the sequences, and a dialogi-
cal dialogue was reached in 65% of the sequences.
Poor-outcome case: The team does not answer

The patient (P) had arrived at the hospital where the first treat-
ment meeting was to be held. In this meeting, it appeared that P
had been violent towards his mother. This occasion is described in
the following sequence (T1 refers to a female therapist and T2 to
a male therapist).
T1: I thought that it happened during the last two weeks, not
before.
T2: Was it a threat or even worse?

T1: Hitting, I thought that P hit his mother.
T2: Was P drunk or did he have a hangover?
P: No, I was sober.
T2: Sober.
T1: I understood that P had tried to ask his mother some-
thing?
P: Well, it was last weekend; the police came to us. She was
drunk. When she didn’t say anything and started to make
coffee in the middle of the night, and I asked . . . I went out
and came into the kitchen, and she turned round and said
that it wasn’t allowed to speak about it. Then I slapped her.
She ran out into the corridor and started screaming. I said
that there is no need to scream, that why can’t she say. . . .
And then I calmed down. At that point I got the feeling. . . .
And the police came and the ambulance. But in some way
I have a feeling, that it is, of course, it is not allowed to hit
anyone. But there are, however, situations . . .
T1: Was that the point when you went into primary care?
P: Yes, it happened just before that
T2: Why did she not say that the police came?
P: What?
T2: Why did she not say that police had been at your place the
previous night?
P: It wasn’t the previous night, it was last weekend. I was
thinking, all the time I am thinking those strange things and
I knew that they were not true. But when you think about
them for a while, after that you have the feeling that things
like that can really happen. It is too much. . . . You are only
thinking of all kinds of futile things.
T2: And it all started last weekend, this situation?
T1: Yes.
This discussion about a situation involving violence took quite a

different form compared to the one in the good-outcome case. When
the patient was describing the situation in confused language, un-
able to use an unambiguous description, he ended by saying, “it

is not allowed to hit anyone.” He had the beginning of an inner
dialogue to deal with what he had done. However, the team did
not respond to this, but, instead, continued by questioning him
about how he contacted the health care system. This was not an
isolated example, given that in the next utterance, when the patient
continued his self-reflection on his “strange things” (meaning hal-
lucinations), the team did not help him to construct more words
for this specific experience he was speaking about. In this short
sequence, there were two utterances that were not answered, and
consequently no dialogue emerged.
The team members had both topical and interactional domi-
nance, whereas P had the quantitative dominance. Meanings were
created in indicative language, and it was monological dialogue.
Overall, in the three treatment meetings analysed for this case, in
25% of the sequences the clients had the interactional dominance
and in 60% the semantic dominance. Symbolic meaning-construc-
tion took place in 10% of the sequences, and a dialogical dialogue
was reached in 15% of the sequences.
“There is nothing as terrible as

being without a response”
In dialogues that had good treatment outcomes, clients often

seemed to take the initiative both for the content and especially
for the way of speaking. This meant that throughout the meeting
they were able to affect the team’s way of responding to their ini-
tiatives. One theme could be dealt with for a quite a long time in
the meeting. In poor treatment outcomes, on the other hand, the
team’s responses did not always fit the clients’ initiatives, and no
dialogical exchange of ideas was achieved. This may have been
affected partly by the family’s use of concrete and indicative lan-
guage, which is more difficult to respond to compared with the
symbolic use of words. The discussion easily turns into a question-
and-answer format—in effect, simply collecting information—with
the procession from one topic to the other proceeding according to
the wishes of the team: responding is not for the sake of generating
joint meanings.
The second case gave an example, however, of how the pa-
tient himself in his indicative language may take the initiative in
proceeding onto emotionally loaded issues. The specific challenge
for the team would be staying alert and responding to these faint
openings, because these openings might be the only possibility for
touching upon the more difficult themes. If they are not responded
to in the present dialogue, they will perhaps not emerge later as
the subject of conversation; they will be lost to the patient as an
opportunity to explore the experience from several alternative
perspectives. In this case, it is illustrative to note that the entire
two-year treatment period often included discussion of whether
the patient had enough insight into his problems. In fact, as we can
see in the analysis, it was he himself who first spoke of his violent
act and psychotic thoughts. The team were the ones who did not
connect with his utterances.
We noted that people participating in therapy meetings start to
speak of their most important experiences, such as hallucinations,
as a response to the themes. The team may open these themes up
through their questions. Being sensitive to these particular mo-
ments in meetings may increase the opportunities to give words—
in a new, joint language—to those experiences that have not been
clothed in words before. It seems to be important to avoid “real-
ity-orientating” the client, but to accept the psychotic utterances as
voices among all the others of the ongoing dialogue.
With regard to the team members’ responding comments, it
seems to be relevant to formulate the utterances along the same
lines as the family members’ language. If they speak of concrete
issues, team members’ comments should follow that line. The
indicative language is the starting point for opening up the possi-
bilities for more symbolic meaning-construction, using exactly the
words and phrases familiar to family members. This presupposes
that there is time given for formulating the words and actually
focusing on the words themselves instead of primarily focusing
on the theme discussed. On the other hand, focusing on words
cannot take place without including the discussion topic. Impor-
tant single points of time emerge when the patients speaks about
incomprehensible or unexpected issues, such as of hallucinations.

The best advice for taking into account what the patient said and
actually responding to it is, perhaps, to pause any other ongoing
dialogue and ask more about the theme the patient has opened up.
If that is not done, incomprehensible issues may be not acknowl-
edged and thus the patient cannot begin to give words to these
experiences or be heard.
Note
1. All the dialogues in this paper are translations from Finnish.
CHAPTER EIGHT
Effectiveness
of dialogical network meetings
O
ur aim in this chapter is to explore the types of outcomes
that dialogical network meetings produce compared to
established helping systems. To this end, the focus is
on studies on Open Dialogues in first-episode psychotic crisis.
Effectiveness studies on Anticipation Dialogues do not yet exist,
although immediate feedback gathered from the clients in the
meetings is available and is very positive.
Training and research

as a part of developing new practice
New practices always emerge in some specific contexts, not as a

replication of some generalized knowledge. The psychiatric system
in which the Open Dialogues approach was generated has func-
tioned since the 1980s in Tornio in Finland. Taking the very first
steps in the direction of a new treatment led to the conclusion that
basic professional training did not promote sufficient possibilities
for working in family-centred practice. For that reason, since 1989
149
a three-year specialist-level training programme in family therapy

has been conducted jointly with the Department of Psychology
at the University of Jyväskylä. It is a multi-professional training:
every member of the therapeutic staff can enter for free training
that legitimizes them as a psychotherapist according to Finnish
law. The idea of having this official criterion was to guarantee the
quality of treatment given by staff taking care of severe crises.
Another early conclusion was arrived at after several confus-
ing experiences in treatment. In a way, Jaakko and the team were
“forced” to systematically analyse the treatment processes and
their outcomes. No ready-made models or theories were available
for the new open family-centred approach, so the only possibil-
ity remaining was to analyse our own experiences in the specific
context. Since 1998, research programmes have been conducted
together with the University of Jyväskylä and Keropudas Hospital.
Most of them have been realized as social-action research, in which
the researchers are part of the system. Thus the information comes
directly to the practitioners from the daily practice and in the train-
ing programmes. In this chapter, as stated above, we refer to stud-
ies that consist of treatment of first-episode psychotic patients.
First-episode psychosis and Open Dialogues
The effectiveness of Open Dialogues was explored in the context

of the Finnish national multi-centre API (Integrated Treatment of
Acute Psychosis) project, which ran from 1 April 1992, through to
31 December 1993, with follow-up at two and five years from the
beginning of treatment, under the direction of STAKES and in con-
junction with the Universities of Jyväskylä and Turku (Lehtinen et
al., 1996; Lehtinen, Aaltonen, Koffert, Räkköläinen, & Syvälahti,
2000). Keropudas Hospital was one of the six research centres. All
first-episode cases of non-affective psychosis (DSM–III–R, Ameri-
can Psychiatric Association, 1987) were included. After 31 Decem-
ber 1993, it was decided to continue the project on a local level
in Western Lapland. The continuation period, named the Open
Dialogues Approach in Acute Psychosis (ODAP), ran from 1 Janu-
ary 1994 through 31 March 1997. The ODAP period can be seen as
E F F E C T I V E N E S S O F D I A L O G I CA L N E T WO R K M E E T I N G S 151
one of fully adopting the treatment principles that we presented

in chapter 3, whereas the API period can be seen as a pre-phase
for that.
The aim of the API project was to analyse how the principles of
the need-adapted approach can be applied in different treatment
cultures. One of the specific aims was to study the importance of
neuroleptic medication in treatment of psychotic episodes. Three
research centres—Keropudas Hospital among them—sought to
avoid the use of neuroleptic medication during the early stage of
treatment. The results of these three centres were compared with
three others where the use of medication in their treatment was as
customary. A specific procedure for deciding whether or not to use
neuroleptic medication was planned. During the first three weeks,
benzodiazepine was used in the event of need for medication, and,
after this, if there was no progress in the psychotic symptoms or
in the social behaviour of the patient, neuroleptic medication was
considered. The aim was that of finding ways for integrating the
medication as a part of a psychosocial treatment as any other treat-
ment methods that are used according to a specified need.
Samples and methods

Information on the samples as well as the concrete research de-
sign is given elsewhere (Seikkula, Alakare, Aaltonen, et al., 2003;
Seikkula et al., 2006). As an orientation, the information that fol-
lows is relevant.
At the two-year follow-up evaluation, complete data were
available for 34 API patients and 46 ODAP patients; at the five-year
follow-up, for 33 API and 42 ODAP patients. Disregarding some
minor attrition mostly due to people having moved to other parts
of Finland for work reasons, the material constitutes all psychotic
patients in Western Lapland during both periods.
The main sources of information were:
1. premorbid variables such as psychological and employment

status at the outset, and duration of untreated psychosis (DUP:
defined as the time between first psychotic symptoms and the
start of psychosocial intervention);
2. process variables: registered number of hospital days, number

of family meetings, and registration of the use of neuroleptic
medication and individual psychotherapy;’
3. outcome variables: registered number of relapses (defined
as making a new contact for treatment after terminating the
original treatment or an intensification of existing treatment
because of new psychotic or other symptoms), employment
status, and the ratings of the mental state of the patients by
the Brief Psychiatric Rating Scale (BPRS), by the Global As-
sessment of Function Scale (GAF), and by a five-category sub-
scale of the Strauss-Carpenter Rating Scale (0 = no symptoms;
1 = mild symptoms almost all the time or moderate occasion-
ally; 2 = moderate symptoms for some time; 3 = prominent
symptoms for some time or moderate symptoms all the time;
4 = continuous prominent symptoms: Opjordsmoen, 1991;
Strauss & Carpenter, 1972).
The ratings were jointly assessed by the researchers and the thera-
pists after each meeting to guarantee that the categorical and, in
some scales, pathologizing language would not disturb dialogues.
In Keropudas Hospital, the researchers were Jaakko together with
Birgitta Alakare. All the above-mentioned ratings were done at the
baseline and at two- and five-year follow-up. During the first treat-
ment meetings, the family were interviewed about the duration of
psychotic and prodromal symptoms before the first contact.
With the idea of producing knowledge in a specific context,
follow-up interviews were planned as learning forums for both
the professionals and the clients. They were conducted in the
presence of both the case-specific treatment team and the family.
At this interview the main theme was the realization of the seven
Open Dialogues principles (responding immediately, including the
social network, flexibility, responsibility, psychological continuity,
tolerance of uncertainty, and dialogicity). First, the patient and
the family were interviewed, after which the team commented
on what the family had said and reported their own experiences,
and, in the final phase, the family members gave their comments
on what they felt was important. By doing so, the team received
immediate feedback on their work, hearing how the family mem-
bers had experienced the treatment. In daily clinical practice, such
immediate feedback is extremely important. Usually, in studies,

the feedback information is received after several years when the
results are published.
The effectiveness was analysed in three different studies: (1) In
the first analysis, which used the entire material from the five-year
period (N = 78), a comparison was done between good and poor
outcome at the two-year follow-up (Seikkula et al., 2001b). (2) In
the second analysis, only schizophrenia patients were selected.
The comparison of the outcomes was conducted both historically
in Keropudas Hospital, between API (n = 22) and ODAP periods
(n = 23), and in another psychiatric unit—representing treatment
as usual—for the comparison group (n = 14) (Seikkula, Alakare,
Aaltonen, et al., 2003). (3) The third analysis was done by conduct-
ing a historical comparison between the API (n = 33) and ODAP
(n = 42) groups to see if the results from the API period persisted,
as well as to see if further changes or improvements were forth-
coming when the approach was applied in all cases. The main
results are discussed below.
Treatment starts earlier

In psychotic crisis, the duration of the untreated period (DUP)
before the beginning of the treatment has recently been focused
on. The longer the patient has psychotic experiences before the
treatment begins, the worse the prognosis. In traditional treatment,
where patients are hospitalized because of psychosis, patients have
been psychotic approximately two to three years before the first
treatment contact (Loebel et al., 1992). In Tornio during the new
Open Dialogues approach, the DUP declined to 3 months. The
longest duration was 13 months. The same type of decline has been
noted in several other active psychosocial programmes for early
intervention (Linszen, Lenior, De Haan, Dingemans, & Gersons,
1998, 1998; McGorry, Edwards, Mihalopoulos, Harrigan, & Jack-
son, 1996; Yung et al., 1998).
The early beginning of treatment may indicate that the thresh-
old to treatment has declined. People in crisis are making contact
instead of just staying at home and hoping for the best. The new
system has helped professionals in other systems to make contact
as well. All this means that psychotic problems are not present for
as long and that neither the patient him/herself nor the family
members become used to the odd behaviour and thus do not have
time to develop a strategy to deal with the odd experiences. This
conclusion is supported by the observations in Tornio that the in-
cidence of schizophrenia has essentially declined (Aaltonen et al.,
1997) and that the institutionalization in hospitals of chronically ill
patients due to schizophrenia has stopped (Tuori, 1994).
Treatment processes in social networks

instead of hospitalizations
In comparing the treatment processes of schizophrenia patients
between Open Dialogues and treatment as usual, the following
differences were noted at the two-year follow-up:
• In the comparison group the patients were hospitalized signifi-

cantly more often (approximately 117 days, compared to 14 days
in the ODAP group and 36 days in the API group).
• All the patients in the comparison group used neuroleptic medi-
cation, compared to one-third in both API and ODAP groups.
• Fewer family-treatment meetings were organized in the com-
parison group (approximately 9 compared to 26 in the API
group). The variation was large in each group: in the API group,
from 6 to 55; in the ODAP group, from 0 to 99; and in the com-
parison group, from 0 to 23.
The treatment-as-usual seemed to emphasize the controlling as-

pects of treatment, such as hospitalization and using the neurolep-
tic medication. Family members were invited to the discussion in
most cases, but family meetings were not focused as much as in
Open Dialogues. Individual psychotherapy was used as much in
each group—in about half of the treatments—and thus the inte-
gration of different therapeutic methods is taking place in both
traditional as well as Open Dialogues treatment.
In comparing the treatment processes within the Open Dia-
logues systems in Tornio, the following differences were noted at
the five-year follow-up between the API and ODAP groups:
. During the ODAP period patients had significantly fewer in-

patient days. During the first two years, API patients were hos-
pitalized for 26 days compared to 9 days in the ODAP period.
. Family-treatment meetings happened less often during the
ODAP period, especially after two years of treatment. API pa-
tients had approximately 11 meetings compared to 4 in the
ODAP group. During the entire five-year period, API patients
had approximately 37 family meetings compared to 25 with
ODAP patients.
No differences were noted in the use of neuroleptic medication or

of individual psychotherapy. During the entire five-year period,
29% of patients used neuroleptics, but it was most often discon-
tinued, since at the follow-up 17% had neuroleptic medication.
Dropout from treatment occurred more often during the API pe-
riod (18%) than during the ODAP period (5%).
During the ODAP period there were fewer treatment occasions.
We can hypothesize that the staff have more training and experi-
ence of Open Dialogues in the later years; they can tolerate more
uncertainty, and thus fewer treatment efforts are needed for the
same outcome. The treatment meetings may become more effec-
tive in the sense that they are needed less during the treatment
periods.
Symptoms ease off

and a return to work becomes possible
When comparing the outcomes, Open Dialogues patients seemed
to have better recovery from their crises in schizophrenia. The fol-
lowing differences emerged at the two-year follow-up:
• At least one relapse occurred in 71% of comparison group pa-

tients, compared to 31% in the API group and 24% in the ODAP
group.
• Comparison-group patients had significantly more psychotic
symptoms remaining compared to the ODAP group. Some
50% of comparison-group patients had at least mild symptoms
sometimes, compared to 17% of ODAP patients.
• The employments status was better with ODAP patients, of

whom only 19% were living on a disability pension compared
to 57% of the comparison-group patients.
At the five-year follow-up, no significant differences between the

API and ODAP occurred. At least one relapse occurred with 39% of
API patients and 29% of ODAP patients. Regardless of the decline
in relapse rates, it was not significant. Recovery from psychosis
occurred equally in both groups. After five years, 82% of ODAP
patients had no psychotic symptoms remaining. A difference was
noted in that the API patient seemed to recover somewhat slower.
At the two-year follow-up, ODAP patients had fewer symptoms,
but at five years no differences existed.
Employment status was better than in any other reported re-
search. Some 86% of the ODAP patients had returned to their
studies or work or to active job-seeking. ODAP-group patients
had some better ratings in all variables compared to API patients.
Because they had no worse ratings, it can be concluded that their
treatment outcomes were reached with fewer treatment efforts.
Treatment is not always successful and should be developed

The third analysis consisted of a comparison between good and
poor outcomes in Open Dialogues. To serve the study aims, two
groups were formed: (1) a poor-outcome group (N = 17), consisting
of those patients whose source of living was a disability pension
(n = 13) or who had residual moderate or more severe psychotic
symptoms (rated as 2 or 3 on the Strauss–Carpenter scale) (n = 4);
and (2) a good-outcome group (N = 61), consisting of those patients
who were working, studying, or job-seeking, with no more than
mild residual psychotic symptoms (n = 7). Thus the distinction was
based on functional criteria with practical implications for the pa-
tient’s quality of life. Compared to other studies, there were fewer
poor-outcome patients (Gupta, Andreasen, Arndt, & Flaum, 1997),
though problems still emerged in some treatment processes.
Already at the beginning of treatment, several differences were
evident between the good- and poor-outcome patients. Significant
differences emerged in employment status and in the quality of the
social network before the psychotic crisis. If the patients had been
living passively without searching for a job, they were more likely
to have a poor outcome after two years. There were, however,
many exceptions to this. Most of the patients with a poor outcome
at the two-year follow-up did not have problems in their occu-
pational history before the crisis. Three of the patients who were
living passively at the onset of the crisis had a better occupational
situation at the two-year follow-up.
At the onset, the symptoms ratings (BPRS, GAF) did not dif-
fer between the patients in the two groups. Significant differences
emerged in the duration of both psychotic (DUP) and prodromal
symptoms prior to treatment. The majority of patients with a poor
outcome had been diagnosed as either schizophreniform or schizo-
affective psychosis or schizophrenia (88%). On the other hand, it is
worth noting that the majority of schizophrenia patients were in
the good-outcome group at follow-up, which means that neither
diagnosis can be seen as the predetermining factor for treatment.
The poor-outcome group were more likely to have had a poor
social network either at the onset, with no change during the treat-
ment period, or to show further impoverishment during the two-
year period. There were, however, exceptions to this general trend:
3 of the poor-outcome patients did show a positive development
in their social network, and, vice-versa, 11 patients had a good
outcome regardless of the impoverishment of their social network.
After the two-year follow-up, however, the quality of the social
network seems to have been connected with treatment processes:
two-thirds of those with a poor outcome at the two-year follow-up
and with improvement in the quality of their social network had
made considerable progress by the five-year follow-up point.
Poor-outcome patients were hospitalized for a longer period
and were more likely to have used neuroleptic medication. This is
a remarkable result, as 57 patients (73%) had not used neuroleptic
medication at all, and of those patients only 8 (14%) were defined
as poor-outcome cases. Of these 8 patients who were not taking
neuroleptics, 4 had declined such medication and in 4 cases it had
not been considered necessary because these patients did not have
severe psychotic symptoms. In the treatment, medication was only
used in the most severe cases for those who really needed it. Evi-
dently, patients with a poor treatment outcome would have been
more likely to receive medication. Those patients might likewise

be expected to have difficulties at follow-up. What is interesting
to see is that avoiding the use of neuroleptics did not cause worse
outcomes, although that is so strongly claimed in some treatment
recommendations.
Practices behind outcomes

Statistical data is always given as group means or as other group
variables. They do not give much information about unique indi-
vidual treatment processes. In what follows, two examples are giv-
en, one with good outcome and the one with a poor outcome. We
can see that when statistical analyses approximate the outcomes,
the case description gives more precise and versatile information.
Case 1:
Good outcome and an intensive treatment process
Lisa’s twin brother had been brought into primary care during
the weekend following a suicide attempt. On Monday morning
his general practitioner contacted a psychologist at the local
mental-health outpatient clinic. He organized a team consist-
ing, in addition to himself, of a nurse from the outpatient
clinic and a doctor and a psychologist from the hospital’s crisis
clinic. The team made a home visit the same day. Present at the
first meeting were Lisa, her mother, father, twin brother, and a
younger brother. Lisa started to talk about her own personal
philosophical theories and about her delusions of seeing peo-
ple with the head of a bull. The team tolerated this unexpected
story and started to talk with Lisa and the rest of the family.
They were somewhat surprised, as they were under the im-
pression that they had come for Lisa’s twin brother, but her
mother told them that at home they were worried about both
siblings. It emerged that both were, in fact, severely psychotic.
Lisa had returned home one year earlier, and both twins had
isolated themselves during the preceding four-month period.
Lisa had had psychotic symptoms for twenty-five months. She

was evaluated at the outset as having a poor social network.
Treatment meetings were organized daily at the beginning, and
during the first two months, nine meetings were organized
altogether. The meetings developed an increasingly psycho-
therapeutic quality, and many episodes of reflective discussion
between the team members emerged. In the sixth and seventh
meetings, Lisa expressed anger and hatred towards her father
owing to her perceptions of his behaviour during their child-
hood. It seemed that for the first time it had become possible
for her to construct words to describe her difficult experiences
with her father.
At the beginning of the process, the general practitioner pre-
scribed neuroleptics, which Lisa tried to take on five occasions
but eventually decided to discontinue because, as she said in
the two-year follow-up interview, “the whole world became
dark and immobile, I did not have any thoughts left”. After
two months, individual psychotherapy was started with a psy-
chologist who was a member of the crisis team. This was done
with a joint understanding after being proposed by the team.
The team members took the view that after the acute crisis was
over, more systematic psychotherapy would be needed. Dur-
ing this phase, Lisa still occasionally had prominent psychotic
experiences. After six months, Lisa decided to stop the psycho-
therapy, moved away from home, and could not be contacted.
In the two-year follow-up interview, she said that she had
started to study philosophy and that she had no remaining
psychotic symptoms. She had decided that she herself had to
find a way out of her problems, and that was why she discon-
tinued the psychotherapy and moved away. She also said that
she had realized that it was not good for her to live near her
family, since this easily led to quarrels. It was better for her to
live without too many contacts with her family. She had had
psychotic delusions for about half a year after discontinuing the
psychotherapy, but since then the symptoms had not recurred.
In the five-year follow-up interview, she said that she had taken
a break from her studies. She had been working as a full-time
cleaner for three years time and was married, and she talked
about their plans to have a baby.
The seven principles had been realized to an adequate extent.
The first meeting had taken place immediately on the same day
after the contact with the general practitioner, and the closest
social network was involved from the outset, although the mo-
tivation of the parents subsequently declined. The same team
guaranteed psychological continuity throughout the process,
including the shift from family sessions to individual psycho-
therapy. The process allowed for tolerance of the uncertain peri-
ods of hallucinatory talk and the hatred Lisa expressed towards
her father. During the many critical phases of the process, the
dialogue was generated and maintained. This was already evi-
dent at the first meeting, in which the team did not focus on
diagnosing the psychosis, but, rather, emphasized generating
a dialogue in which the family could present the problem in
their own language.
Case 2:
Poor outcome, with prominent psychotic symptoms
in the follow-up
Matti returned home after a three-year period of employment
and reported that people were listening in to his telephone con-
versations and were trying to poison him. His mother brought
him to the hospital’s crisis clinic and asked for help. At this
point, however, Matti was very coherent, saying that he had
himself found the solution to his problems, and he rejected the
proposed home visit the next day. He was evaluated as hav-
ing a normal social network, and his psychotic symptoms had
continued for two months.
One month later, a general practitioner in the local health cen-
tre contacted the crisis clinic reporting that Matti had severe
psychotic problems. An intense process of discussing the crisis
with the whole family was started, but during this process
Matti ceased to have almost any communication with the out-
side world. Five months later he began to refuse to eat and,

after developing an infection, was hospitalized for a month. A
course of antidepressant medication was started, and, mainly
on account of his family’s wishes, he was discharged and re-
turned home, where the treatment meetings continued. He was
hospitalized for 34 days. After one year of treatment, neurolep-
tic medication was prescribed. Another six months later, while
Matti’s sister was now in crisis, Matti suddenly improved. He
started two occupational-training courses one after the other
and decided of his own volition to discontinue medication. Af-
ter this, he “slipped” into speechlessness and limited commu-
nication. It was during this phase that the two-year follow-up
interview was conducted. Treatment meetings were organized
every second week throughout the follow-up period. A reha-
bilitation programme was also planned for him. This included,
for instance, an occupational-training course for three months,
which he took.
Matti’s treatment was at all times very difficult and no joint un-
derstanding of the problem emerged. Considerable problems
existed already at the outset because of the one-month’s delay
after the first treatment meeting at the crisis clinic. Medication
seemed to help at one point during treatment, but not perma-
nently. The main treatment always took the form of an intensive
effort to achieve psychotherapeutic understanding, but without
success. On the whole, problems emerged in meeting many
of the Open Dialogues principles. Both immediate help and
psychological continuity became problematic when, after the
first meeting, the team did not succeed in motivating the fam-
ily in favour of treatment. Perhaps the team did not notice the
severity of the situation and did not hear the worry expressed
by Matti’s mother clearly enough at the very first meeting. Flex-
ibility was not realized, in the sense that disagreement often
occurred between the team and the family, which meant that
abrupt changes in Matti’s condition forced actions to be taken
rather than allowing them to emerge from a joint process. The
team tolerated uncertainty and shared the disappointment of
the family because no improvement in Matti’s situation was
achieved, but a dialogical process was not realized. This was

partly because of the concrete language used by the family. The
team did not find a way to construct a process that would have
suited the language of the family.
***
Some preliminary hypotheses can be drawn after comparing the
differences between good and poor outcome:
• At the onset of treatment, those patients with a long duration of

untreated psychosis and living passively without searching for
a job should be especially emphasized.
• Schizophrenia patients are at greater risk for poorer outcomes
compared to milder forms of psychosis.
• Mobilizing the social network of the patient seems to increase
possibilities for recovering from a severe crisis. This includes
both being active in inviting those important persons in the
patient’s private social network and connecting a broad spec-
trum of relevant professionals. For instance, in the most severe
situations, inviting the employment rehabilitation authorities
already during the crisis phase seems to help in returning to job
or studies. Another element seems to be organizing meetings at
the workplace if problems have emerged there.
• Postponing the start of neuroleptic medication in an active psy-
chosocial treatment seems to increase the use of the patient’s
own psychological resources and those nearest her/him. Neu-
roleptics should be used if needed as a part of active pro-
grammes. In cases where the entire treatment can be conducted
without neuroleptics and the patient recovers from psychosis,
the prognosis seems to improve.
• Hospitalization seemed to increase the risk of a poor outcome.
Hospitalizations are needed, however, in any practice every
now and then. During the inpatient treatment, specific focus
should be placed on the patient’s psychological activity and on
having active social ties to ensure connections remain with the
social reality outside the hospital.
New practice can be generated
Some interesting conclusions can be drawn from the treatment of

first-episode psychotic patients in an Open Dialogues approach
and about the factors affecting the outcome of treatment.
Before commenting on the results, the shortcomings of the data
and the limitations of the study design should be noted. Psycho-
sis, and in particular schizophrenia, is a rare problem, and in a
small catchment area the selection of cases is susceptible to the ef-
fects of many unforeseen factors. This risk especially concerns the
comparison group, which was selected over a 21-month period.
Thus the patients in the comparison group could have had more
severe symptoms than would normally be expected in that health
district. In small samples, a single patient being moved between
categories may affect the statistical significance. The influence of
chance events influencing sample selection from this small popu-
lation was reduced both by conducting a historical comparison
and by having a comparison group from conventional treatment.
On the other hand, selecting from a small catchment area is also
an advantage in the sense that it is possible to control for those
patients who were excluded for some reason, since the treatment
is the only kind available for psychiatric patients.
Another problem may emerge with regard to bias, since the
researchers performing the ratings were involved in developing
the Open Dialogues approach. There is a risk of applying bet-
ter ratings during the ODAP period, as it gives more favourable
results for Open Dialogues. However, in the data presented, both
the objective data on the use of treatment and employment status
as well as the ratings of psychological status and symptoms seem
to be in line with each other and thus verify the reliability of the
results. Because the information from the objective registrations of
treatment and the ratings made by researchers were in accord, no
visible signs of bias were noted.
In spite of these design limitations, some preliminary conclu-
sions can be drawn. First, we can see rather good results compared
to the traditional treatment of psychosis. Svedberg, Mesterton, and
Cullberg (2001) analysed the five-year outcome of first-episode
psychotic patients in Stockholm in Sweden. Table 8.1 summarizes
Table 8.1. Comparison of the differences in five-year outcomes

with first-episode, non-affective psychotic patients (DSM-III-R)
in Western Lapland and Stockholm
ODAP
Western Lapland, Stockholm,*
1992–97 1991–92
Diagnosis (N = 72) (N = 71)
schizophrenia 59% 54%

other non-affective psychosis 41% 46%
mean age (years)
female 26.5 30
male 27.5 29
hospitalization (days/mean) 31 110
neuroleptic used 29% 93%
ongoing 17% 75%
GAF at follow-up 66 55
disability pension or sick leave 19% 62%
*Data from Svedberg, Mesterton, & Cullberg (2001).
the differences between that study and our study in Tornio. In

Stockholm, 54% were diagnosed with schizophrenia, about the
same as in Tornio. In Stockholm, the mean age seemed to be higher
(30 years, compared to 27 years in Tornio). This might indicate
that in Stockholm the duration of untreated psychosis (DUP) was
longer. Compared to the present study, the DUP has been reported
to be much longer in a traditional setting, compared with either
API or ODAP, having an average of between one and three years
(Larsen, Johannesen, & Opjordsmoen, 1998; Kalla et al., 2002). The
mean period of hospitalization was 110 days with the Stockholm
patients compared to 31 days in Tornio. Neuroleptic medication
was used in 93% of cases in the Stockholm group compared to 29%.
As an outcome, 62% of the patients treated in Stockholm were liv-
ing on a disability pension, compared to 19% in Tornio.
In the Linszen, Dingemans, and Lenior (2001) study, after an ef-
fective psychosocial programme the patients were referred to other
treatment agencies and the positive results did not persist. Some
75% had at least one relapse and their social-function level was
poor. In the present study, most of these problems were avoided.
Of ODAP-group patients, only 29% had a relapse during the five-

year period, and 86% had returned to an active social life in the
sense that they were working, studying, or actively job-seeking.
The differences may be seen as an illustration of how the
network-oriented treatment with immediate crisis response may
enhance the treatment of first-episode psychotic patients. The
treatment processes become different in the sense that the need
for controlling treatment methods can be decreased. Hospitaliza-
tion can be decreased as well as the need for long-term neuroleptic
medication. Most patients can manage by using anxiolytics to de-
crease anxiety or to aid sleep during the most intense crisis phase.
Anxiolytics can be prescribed in urgent need and the medication
can be stopped rapidly, and thus a needless addiction to long-
term medication can prevented. This can mean empowering the
patients and—on the other hand—decreasing the power of experts.
Decreasing medication often increases use of own psychological
resources. The same conclusions was drawn by Johnstone and
colleagues (Johnstone, Crow, Johnson, & Macmillan, 1986; John-
stone, Macmillan, Frith, Benn, & Crow, 1990) when they noted
that patients using a placebo had better employment status after
two years.
Compared with other psychosocial programmes, some differ-
ences can be seen. While most of the programmes focus on in-
creasing the coping skills of the patient and the family, in Open
Dialogues a concrete cooperation is taking place in addition to this
by mobilizing the relevant parties in the patient’s future, such as
employers, fellow employees, or state employment agencies for
as long a time as seems adequate. In this way a new community
around the patients is created (see chapter 6). In this community,
more actors are involved and the field is larger compared to pro-
grammes that only intervene with the patient and the family. All
this seems to have a positive effect on the social management and
coping skills.
In the qualitative analysis of two cases, it was noted that the
statistical differences found did not, as such, predict the outcome.
The different variables have a co-evolutionary effect in clinical
practice. Thus, an adequate process of treatment can be started in
spite of data that point towards unfavourable outcomes, provided
that a constructive dialogue between the team and the patients and
their social network can be started (see Case 1). If the possibility of
starting a dialogical process is minimal, the treatment may lead to
a poor outcome, even when this is not predicted by the premorbid
social and psychological adjustment (Case 2). The main problem
of Open Dialogues seemed to be organizing treatment in cases
where the family had few possibilities for developing a linguistic
description of their life.
The results shown in this chapter are rather confusing. In Open
Dialogues, the evidence-based practice recommendation has not
been the basis for the development but, rather, the information
produced in the specific context in Tornio. Our own practice has
been analysed and the information obtained used to develop the
work further, focusing particularly on the weaknesses in what has
been done. Some results seem to be, on the surface, even contrary
to general recommendations: neuroleptic medication was not pre-
scribed at the onset for all patients, not even for schizophrenia
patients. Families were not invited to family psychoeducation,
patients were not mainly hospitalized, and the psychotic symp-
toms were not treated with, for example, heavy medication before
other treatment methods. And in spite of this, some parts of the
outcomes seem to be better compared to treatment-as-usual and
even to psychoeducational programmes.
All this opens up a new and versatile challenge, which is the
focus of the final chapters of this book.
CHAPTER NINE
Research
and generalizing practices
T
he outcomes of Open Dialogues described in the previous
chapter differ from the results reported in experimental
psychiatric research. Experimental settings, in turn, are the
staple of so-called evidence-based research. There is a danger that
the most important elements of psychosocial help are lost in the
control-study settings that have become the benchmark for compe-
tent outcome analysis. Designs that study interventions by some-
one on someone are valued above all others, dismissing mutuality,
responsiveness, and dialogue and thereby reinforcing monological
practices. However, evidence is necessary, and hence it is vital
to develop research settings that recognize responsiveness. One-
way intervention studies go hand in hand with top-down practice
governance. In the rational-planning model, good practices are in-
novated, evaluated, and implemented. We believe that the model
neglects the diversity of contexts, actors, and interests—and poli-
tics. We shall discuss a contextualized approach to good practice
generalization.
In the treatment of schizophrenia, evidence-based studies are
gathered together to form treatment recommendations that are
based on meta-analysis of randomized trials. These have been
167
done in many countries, starting with the PORT report in the

United States (Lehman et al., 2003). Following the same guidelines,
the Finnish Psychiatric Association has also published Guidelines
for the Valid Treatment of Schizophrenia. The recommendation
on medication can be pointed out as an example of the differ-
ences between typical Open Dialogues outcomes and those of the
evidence-based results. The Guidelines recommend neuroleptic
medication right from the beginning of treatment. The Open Dia-
logues ODAP studies showed better results among those patients
who had no neuroleptic medication or who had it at a later stage
in treatment.
In the Guidelines, starting neuroleptic medication is justified
by stating that it prevents relapses. The outcomes were different in
this respect, too: those patients in the ODAP research who were not
medicated from the beginning of the treatment had fewer relapses.
Thus, the outcomes are practically the opposite of the Guidelines.
In addition, the patients who were not medicated from the begin-
ning recovered better from their psychotic symptoms and could
return to working life in larger numbers than those who received
neuroleptic medication from the very beginning.
Should we conclude that the ODAP results are invalid or, vice
versa, that the research behind the Guideline recommendations is
not valid? It is, of course, possible that both see only what their
evaluative instruments allow; thus, they may both be valid but are
both reporting different things.
Utilizing research-based knowledge is essential in developing
practices. It is necessary to analyse methods and practices system-
atically, and such feedback should be at the disposal of practition-
ers. As a matter of fact, it would benefit both research and practice
if the local actors could take part in the research from the very first
stages. In the following, we discuss research that could expressly
foster dialogical- and network-oriented practices. As will become
apparent, such research itself has to be dialogical and emerges
within network-relations.
We shall discuss three interlinking themes:
• The first theme reflects upon the implications that design has for
outcomes. We argue that in studying dialogical encounters and
their outcomes, it is necessary to apply naturalistic designs—re-
RESEARCH AND GENERALIZING PRACTICES 169
search on genuine network contexts—instead of mere labora-

tory-like settings with single variables.
• The second theme is the generalizability and transferability of
practices. We argue that developing dialogical practices calls
for research of local contexts. Enhancing local skill for variation
and contextual flexibility is called for instead of mere attempts
to implement good practices by duplicating them.
• The third theme is the production of socially robust or durable
research-based knowledge. This calls for arenas where various
disciplines, research orientations, and approaches as well as the
various parties of research—researchers, “objects”, clients, part-
ners, and so on—can meet to cross boundaries. In such learning
spaces, perspectives can be formed that exceed the possibilities
of any single party.
Evidence-based research
and the increasing one-dimensionality of designs
Scientific knowledge has increased dramatically in psychosocial

fields and is increasingly being evaluated. Scientific evidence for
the validity of practices is required. Because the effective factors in
human activity are multidimensional, research designs need to be
defined and reduced, in order to be able to make visible explana-
tory differences. Within medicine and social work, specific libraries
of evidence-based research—the Cochrane Library for psychiatry
and the Campbell Library for social work—have been founded for
such important quality-control purposes. These libraries collect
reports of studies that are based on experiments and randomized
trials. Through meta-analyses of such reports, guidelines on valid
practice are constructed.
Evidence-based guidelines are used for controlling and steering
practice-development. With the rapid increase of research knowl-
edge, it is difficult to control it in treatment practices. The sys-
tematization of research-based knowledge is an important service
to practitioners, management, and steering administrators. Verifi-
ably successful treatment or other practice is in the interests of
the patient and the client. Therefore, research should analyse as

diversely as possible both the problems to be treated and the
treatment practices. Surprisingly, however, the evidence-based-
practice libraries accept only knowledge produced by one research
tradition alone—thus making knowledge produced by different
approaches secondary.
Almost all the research reports published in journals of psy-
chiatry are based on experimental settings. There are increasing
demands to also publish evidence of cause–effect relations in social
work. Constructing an experimental design in psychiatry requires
therapy models that allow the separating-out of single effective
factors. This, in turn, leads to the publication of only those stud-
ies where the problem to be treated is defined unequivocally as,
for example, a biological condition. The Cochrane Library ac-
cepts only reports of randomized experimental studies, although
the spectrum of disciplines and research methods has widened
enormously. There is a tension: on the one hand, the possibilities
to study psychic and social crises and their treatment diversely
have increased; on the other hand, the principles of publishing
have narrowed to such an extent that only one type of research is
emphasized.
When the aim is to find universal causal relations, local factors
are not of interest. Thus, for example, the above-mentioned Finn-
ish guidelines do not refer to the long Finnish research tradition
in schizophrenia treatment, but almost entirely to international
studies. A special Finnish guideline would therefore not have been
necessary, since the knowledge is taken to be universal. The di-
verging outcomes of the ODAP research—no medication right at
the beginning of the treatment, but other measures and with bet-
ter results—do not follow from studying the same phenomenon
with different methods. Rather, they are based on the fact that a
randomized control study does not capture the dimensions that are
the most central in dialogical practices. This does not necessarily
result from ignoring or not wanting to include those dimensions
in the setting. They would be interesting, for example, for making
comparisons. The reason they are not included in the rigorous
control-study settings is that encounters between networks, the
emergence of joint meanings and language, and reciprocal causa-
tions do not yield readily to experimental settings with reduced
variables. There is a danger that the most important factors of

treatment disappear because of the setting.
In order to produce outcomes that can be generalized, the re-
search has to be carefully designed to produce knowledge about
factors that explain problems or practices. In order for research re-
sults to be compared and the universally applicable outcomes thus
winnowed out, the studies have to be carried out with the same
methods and in the same way. The problem here is that studies
can be compared, but not the actual practices, which are far more
complex. When research outcomes are summarized into treatment
recommendations, they are, after all, not guides of practices, but of
virtual realities created in the research settings. In the experimental
settings, the explanatory factors have to be defined individually in
a controlled way. In practical treatment situations outside experi-
mental settings, such a control of variables is not possible.
Another significant problem in relation to developing practices
derives from the fact that treatment success is assessed on the
basis of group comparisons. Within a certain basic population the
patients are selected for the experiment and the control group(s)
by casting lots. The experiment-group patients get the studied
treatment, and the control-group patients receive the conventional
treatment. The groups are compared. This is done by contrasting
group means. However, mean values do not tell anything about
individual cases. Treatment guidelines can be based on studies
where less than a third of the research-group patients are outside
the variation of the control group. This means that 70% of both
research- and control-group patients get the same values. Never-
theless, treatment guidelines suggest the use of the same method
for all patients.
Studying single effective factors rarely corresponds with situa-
tions of treating severe crises where there are myriads of effective
factors and where the outcome of the treatment is constructed in
the totality of the treatment process. To be sure, countless factors
other than the treatment in question can have an affect on the
outcomes. Human life is not affected by only one or two factors
at a time. There are countless unknown factors. The reduction
problem has also been heeded among the experimental-setting
researchers. Kissling and Leucht (2001) have stated that the gap
between experimental-setting results and treatment practices is
far too wide. They recommend emphasis on naturalistic follow-up

studies—research done in actual treatment situations.
From a quest for explanatory models

to descriptive studies
Research should correspond to living practice. Thus research would

be central to developing practices. This calls for settings that can
take into account the ensemble of the treatment and the needs
of each patient. The methodological aim of the ODAP research
described above was to carry out follow-up studies in naturalistic
contexts, in actual practice. The treatment of each patient is real-
ized individually within the described treatment model, and the
outcomes are compared both with the outcomes of a unit carrying
out traditional care and as a historical comparison with the earlier
stages of the treatment model. Such a setting can clarify the suc-
cess of a whole treatment model in a given geographical catchment
area, in which the psychiatric system has a total responsibility for
the entire population. The aim in such designs is to construct not
an explanatory model but a descriptive study.
Conventionally, descriptive study is regarded as inferior to
explanatory research, because descriptive study does not yield
generalizable comparisons of effective variables. The French soci-
ologist Bruno Latour (1988, pp. 159, 160) argues that valuing ex-
planations higher than descriptions is connected with attempts to
control contexts at a distance. He discusses so-called weak and strong
explanations. In a strong explanation, a minimum amount of ele-
ments (explanans) provide the explanation of a maximum amount
of elements (explanada). Correlations are weaker explanations than
showing causal relations, and descriptions are even weaker. If the
practitioner or the team is in the very context of the explanations,
then weaker explanations are sufficient, in that they do not help
to reduce the complexity to a few facts. A strong explanation be-
comes necessary when someone wishes to act at a distance. If one
attempts to control several contexts at a distance, strong explana-
tions become especially important. According to Latour, centres of
calculation are formed for acting at a distance. Information travels

between them and the contexts to be controlled.
Valid treatment or valid practice guidelines are a means to
attempt to control contexts at a distance: practices in different
contexts ought to change in accordance with the guidelines. Para-
doxically, controlling contexts at a distance requires that all con-
text-specific data is carefully distilled from the research results. If
research yields knowledge valid only to a particular context, it is
not transferable. Thus, according to the controlling-at-a-distance
approach, research has to be carried out in settings where the lo-
cal particularity cannot have an effect on the outcomes. The idea
is that the more context-free—that is, universal—the knowledge
is, the more transferable it is. These are problems that one has to
face if one wishes to control contexts at a distance, but the scene is
different if one acts within those contexts. The problem of transfer-
ability is not the same. If one acts in the contexts one tries to under-
stand, one does not need explanations that are assumed to be universal
or transferable as such.
The research model and the development model go hand in
hand. The counterpart of experimental-setting research in the
sphere of development is implementation. Implementation aims
at duplicating, repeating the studied practice. The idea is that re-
search provides information that is taken to the contexts to be con-
trolled, where, as a consequence, the verifiably successful practice
will become general. This vision dismisses politics at the various
stages of the process. Implementation is not a direct passage from
experimental settings to changing practices, but involves politics
at every step.
First, the fact that experimental research is valued higher at a
cost to other approaches is not a matter internal to science, but a
consequence of political choices. The need for universal explana-
tions and experimental settings is necessitated especially by the
wish to control treatment and other practices at a distance. Deci-
sion-making power has been delegated to regional/local bodies,
but at the same time there is a striving to steer practices. The latter
is pursued through—preferably universally applicable—guide-
lines. Universal applicability is aspired through showing causal
relations, and they, in turn, are sought in experimental settings.
A different kind of steering paradigm would necessitate different

research paradigms.
Second, implementing practices is also political. Introducing
practices calls for negotiations, choices, prioritizations, decisions,
variations, and adapting to local conditions, which is not merely
repeating an activity in the form it was studied. Good practices are
not like articles that can be transferred from one place to another. In or-
der to be sustainable, a good practice needs not only good work in
the actual treatment situation but also good management, viable
relations between agencies/practitioners referring and receiving
clients, good local networking in the civil society, and so forth. The
route from scientific evidence to introducing practices is a course
through landscapes of conflicting interests and multiple actors.
The word “implementation” barely describes the political process
through which an activity is made a general practice.
Positioning control studies as the benchmark for effectiveness
studies is particularly problematic for the development of dialogi-
cal practices. The matter is not as problematic when elaborating
interventionist practices, although they are not movable articles
either. Mutuality/reciprocity are essential in dialogical practices. Cli-
ents are not just objects targeted with measures. When settings
are reduced to single out the effects of measurable variables, the
research reveals lineal rather than reciprocal causations. Guidelines
built on such research are prone to reinforce monological prac-
tices—interventions on objects—instead of dialogical encounters:
someone does something to change others with the method x, but
does not change her/himself in the process, or if it does, it is an
unessential point to the research.
Governance culture creates the call for control studies
The rise of control studies to a benchmark-position for effectiveness

studies seems to connect with the process of delegating decision-
making power in the post-expansion phase of the welfare state.
The central government used norms and subsidies—the carrot-
and-stick approach—to create the basic structures of the welfare
state, the institutions of education, health, welfare, and so on, so
as to be somewhat concordant throughout the country. The model

of steering encouraged by the European Union is much looser
nowadays. Skeleton laws and guidelines have replaced conclusive
norms and administrative rules. Framework budgets have taken
the place of “earmarked” allocations. Yet there is a need for the
central authority to have an impact on local practices—and this has
been necessitated by discoveries of inequalities in relation to public
services: the once integrated welfare state diversifies, the quality of
the services the citizens get varying according to the locality they
happen to live in. The attempt to steer practices by recommending
verified good practices is gaining momentum in the EU as an effort
to solve the contradiction between delegated power and control-
ling contexts. Large developmental programmes are constructed as
steering instruments, and an essential content in them is recom-
mending and prioritizing good evidence-based practices.
Good-practice recommendations lack the steering capacity that
the conclusive norms and regulated allocation of resources once
had. Recommendation-based policies require an “authority” in
order to be carried through, which they now seek in science. When
authorities are steering expert systems like welfare, health and edu-
cation, and their trained personnel, the recommendations have to
be convincing. Now that the command hierarchies are transform-
ing, a prestige that is convincing cannot be guaranteed merely by
a high administrative position. A recommendation is just a recom-
mendation to experts in the field, even if it descends from the top
of the administrative hierarchy. Therefore, the administrative sys-
tem borrows authority from the science system. Basing on meta-
analyses of research reports, the steering system demonstrates with
the prestige of science that this or that recommendation is not just an
ideological preference of the prevailing government, but a general,
universally applicable explanation of the valid means of having an effect
on the phenomenon. By sending science-authorized information to
contexts A, B, and C, they can attempt to control without vio-
lating the principles of decision-power delegation. Control-study
research fits this configuration almost ideally. Its determination in
purifying all contextuality from its settings and results makes it
look like an activity totally removed from politics. Thus, govern-
ance, on the one hand, presents itself as removed from politics
and only recommending scientifically verified practices—where
“neutral” science takes care of selection and prioritization. On the

other hand, a financially and politically prioritized control study
provides “neutral” evidence-based results for guideline construc-
tion. The route from research and evaluation to implementation,
the passage through landscapes of conflicting interests, seems to-
tally apolitical.
Developing psychosocial work calls for scientific evidence. Evi-
dence-based research is not problematic as such. Problems arise for
developing dialogical practices, however, if:
• effectiveness studies recognize only interventions on objects

• guidelines steer practices towards interventionism
• administration builds only on research that recognizes only in-
terventions
It is necessary to develop broad evidence-based research. Among

other things, this calls for discussion across paradigm boundaries,
not just discussion within camps. Crossing boundaries calls for
meeting places, arenas of encounter, and dialogues between par-
ties.
Agoras and contextuality
Local actors and their interests cannot be bypassed when trans-

porting evidence-based good practices to a local context. There-
fore, research that aims at developing practices should take part
in local processes. If one acts close to instead of at a distance, the
issue of the universality of explanations changes. Instead of study-
ing simplified causal relations, multifaceted settings are called for.
Developing professional practices requires research that can feed
local learning processes. If one wishes to transfer network dialogue
practices from one context to others, there are no shortcuts. In the
new contexts, local negotiations, local networking, and local learning
processes are required. Results from previous undertakings have an
important role in supporting such processes. Insights and crystal-
lizations do accumulate—but the end results of the learning pro-
cesses in one context cannot be transplanted.
Helga Nowotny, Peter Scott, and Michael Gibbons write in

their book Re-thinking Science: Knowledge and the Public in an Age
of Uncertainty (2002) that the striving for valid knowledge through
purifying it from its contexts produces less valid knowledge than
strongly contextualized research. They analyse science–society re-
lations against a wide body of material from a variety of fields and
contexts—from technical to social sciences. Their observations are
not directly from psychosocial activities, but they certainly inspire
reflection upon the challenges of research in this branch, too.
Nowotny et al. (2002) argue that
[r]eliable knowledge, although it will remain a solid and in-
dispensable criterion to strive for, will be tested not in the ab-
stract, but in the very concrete and local circumstances. . . . The
reliability of scientific knowledge needs to be complemented
and strengthened by becoming also socially robust. Hence,
context-sensitivity must be heightened and its awareness must
be spread. . . . One way to make science more context-sensitive
is to bring in people. [p. 117]
The authors call science weakly contextualized if its communication

patterns are determined largely by institutions. In other words,
“people” are aggregated, and their wishes and desires are, in a
sense, represented by institutions. The settings are purified to-
wards the ideal of context-free circumstances in order to find
universal cause–effect relationships. By contrast, strong contextu-
alization occurs when researchers have the opportunity, and are
willing, to respond to signals received from society. Strong con-
textualization embraces stronger interaction with “outsiders”, in-
creased uncertainty and variation—and selective retention through
success—and people entering the process as actors whose needs,
wishes, and desires are listened to, responded to, and anticipated.
The authors mention medical research that closely involves pa-
tients as a candidate for strongly contextualized science.
The opposite case—weakly contextualized research—is read-
ily available in studies in the psychosocial field. An illustrative
example is offered by psychosis-symptom-relapse research and
the meta-analysis of it. The usual reference in studies assessing
the number of relapses is the Gilbert, Harris, McAdams, and Jeste
(1995) article, which presents an analysis of 66 follow-up studies,
all of which were based on an experimental setting. The patients
were randomized to a neuroleptic-medication group and a placebo

group after one year of neuroleptic treatment. In two-thirds of
the experiments, the neuroleptic medication was abruptly discon-
tinued. During the following year, the number of relapses was
registered, and it was found out that the patients who contin-
ued to receive neuroleptic medication experienced fewer relapses
(15–25%) than those who were switched to placebos (55%). On the
basis of this meta-analysis, a treatment guideline is constructed
that states that neuroleptic treatment should be started right at the
beginning of treatment because it prevents relapses better than
treatment without medication. However, the study in question
does not even describe the situation it offers recommendations on.
To be exact, the situation in the experiments was one where all the
patients first received medication and for some it was randomly
broken off. The study aimed at maximal general applicability, but
lost its validity. It no longer describes what it meant to describe.
According to Nowotny et al. (2002, pp. 144, 159) the process of
contextualization is to be found mainly in the middle range between
weak and strong. Research fields and actual research practices are
heterogeneous, and contextualization cannot be done by forcing
it into a single mould or ideal type. In contextualization in the
middle range, “transaction spaces” between groups, disciplines,
research fields, and so on become central. Thus, “agoras” become
central, as does dialogue between the parties. Changes also take
place with regard to what the researchers see as their legitimate
concern. So far the perspective has been in the context of applica-
tion: what are the results applicable for. In contextualization in
the middle range, the perspective moves beyond, to the context
of implications: what are the implications for people, what are
the anticipated consequences and impacts that research activities
continue to generate.
People enter research in many guises—in societal experiments,
as objects, as partners, as parties in dialogue. Spaces for cross-
ing boundaries are necessary, as well as interaction as part of the
process of shaping the research. Nowotny et al. (2002) write that
contextualization “depends on a permanent dialogue between sci-
entists and diverse ‘others’ in society. It is multi-layered. Explicit
messages are communicated, interpreted and re-interpreted, as
well as implicit or yet-to-be articulated preferences, needs and

desires” (p. 134).
Nowotny et al. do not suggest that it is time to desert the
laboratories and other reduced experimental settings. In fact, a
great number of their case examples of strong- or middle-range
contextualization are from experimental settings. What they do
emphasize is the importance of stronger contextualization than
institution-determined settings for producing valid outcomes and for
anticipating implications for people. The authors maintain that the
more strongly contextualized a scientific field or research domain is, the
more socially robust is the knowledge it is likely to produce. Such robust-
ness is not abstract: it can only be judged in the specific conditions
it is used. Furthermore, distancing science from societal influence
and purifying its contexts does not increase possibilities for ro-
bust knowledge; on the contrary, robustness is produced when
research has been infiltrated and improved by social knowledge.
Such knowledge has a strongly empirical dimension; it is subject
to frequent testing, feedback, and improvement because it is open-
ended.
Nowotny et al. see evidence-based research as a residue: “The
climax of high modernity with its unshakeable belief in planning
(in society) and predictability (in science) is long past, even if the
popularity of ‘evidence-based’ research demonstrates the stubborn
survival of the residues of this belief” (2002, p. 5). Furthermore,
outlining science in the age of uncertainty, the authors see this
stubborn residue as transient epistemology: “Gone too is the be-
lief in simple cause–effect relationships often embodying implicit
assumptions about their underlying linearity; in their place an
acknowledgement that many—perhaps most—relationships are
non-linear and subject to ever changing patterns of unpredictabil-
ity.” Nonetheless, however transient the position of linearity-based
research may be in the transition from “high modern” to later mo-
dernity, in many respects it is still on the ascendancy at present, not
fading away. From the point of view of diversifying psychosocial
approaches this is alarming. Study on simplified causal relations in
special circumstances is put on a pedestal, not for epistemological
reasons, but for political and financial purposes. The perspective
of controlling contexts at a distance and leaving prioritization to
indubitable science is far more inviting for administration and

financing than the “swamp” of contextuality, social inflirtation,
mutual learning, and dialogues.1 At present there is a move in the
West to cut financing from approaches that cannot demonstrate
evidence—and what passes as evidence in meta-analyses is mate-
rial from randomized linear intervention studies.
It is urgent to develop broader settings for evidence-based re-
search. The obvious challenge is to develop effectiveness stud-
ies that also recognize non-linear approaches, those of mutuality
and responsiveness. Evidence-based research should come into the
open, from institution-dominated settings to agoras and dialogues.
It is important not to equate the search for scientific evidence and
careful meta-analyses only with control studies that drastically
reduce the phenomena under study and see only one-way causa-
tions.
Agoras of learning and generalizing practices
The differences in attempts to generalize practices can be illustrat-

ed with two examples. In many countries the ideas of multi-system
treatment (MST), a method developed in the US, have been put
into practice by conducting randomized studies. In what follows,
we describe one such study conducted in Norway (see Ogden &
Halliday-Boykins, 2004). This huge undertaking and the follow-up
are particularly interesting in relation to the themes discussed in
this book, since the attempt was to introduce a network approach,
not an intervention solely relying on an effect variable like medica-
tion. The goals of the process and the study were:
• to prevent placement outside the home of adolescents with seri-

ous behavioural problems
• to assess the extent to which MST can produce outcomes that
are superior to the child welfare services already provided to
youthful offenders in Norway
• to determine the degree to which outcomes obtained in the US
would be replicated in Norway for youths with serious behav-
ioural problems.
In MST, professionals work in close interaction with the adoles-

cents, their families, and other networks.2 The aim is ambitious
and valuable. There certainly is no oversupply for effective means
for such situations. However, MST promises verified effectiveness
precisely with that part of the population. One has to be trained
and licensed in order to use the method, so as to ensure that the
MST that is actually practiced—and followed up—is what it ought
to be. Practitioners receive both training and supervision in the US,
and they sign an agreement that they will not change the methods.
The follow-up carefully designed, based on randomization and a
control setting. Some of the adolescents get MST, whereas the con-
trol groups do not. Adolescents were allocated to MST groups or
control groups by lottery, thus ensuring that the unknown factors
were probably distributed evenly across the whole research popu-
lation. The outcomes showed better results with MST than if MST
was not provided. The difference was not great, but it was statisti-
cally significant. The clearest difference was in the growth of the
adolescents’ social skills (see Ogden & Halliday-Boykins, 2004)
The undertaking was an experiment on whether it is possi-
ble to replicate in Norway the activity verifiably effective in the
US. Therefore, one can ask whether the method was carried out
exactly the same way throughout the country. Hardly! There are
good reasons for assuming that the practitioners’ ability to use the
method generatively was an essential prerequisite for its viability.
It is unlikely that the trainers or the supervisors could prepare the
practitioners for all the concrete situations and all the particulars
they were about to face. More probably the practitioners were able
to be flexible according to the “requirements of the context”—cre-
ating variations that each interaction with the adolescents and
their networks required.3 Did they—uniformly—put into practice
only things belonging to the method, thus perhaps eliminating
“being themselves”? It is not likely that they succeeded even if
they tried.
The research aimed at gathering generally applicable knowl-
edge, striving after effective factors. However, what ultimately
caused the observed outcome was not revealed. After all, methods
do not help or cure anyone as such. Psychological methods—and
other interaction-based means—exist as their users’ activity. Learn-
ing a new method always has an effect on its users as a whole.
They may become more interested in their clients, they may gen-
erate new kinds of relations with their clients, they may observe
themselves and their activity differently, and they may start to
talk differently. Perhaps they seek and find more room to move
in their relations with the management, perhaps they act differ-
ently in their partner relationships. All these factors—and many
others—can have a holistic positive effect on how difficult situa-
tions are encountered.
Developing Open Dialogues and Anticipation Dialogues took
place in processes that differ from that of the MST projects. The
development of Open Dialogues proceeded as a regional process.
Four supporting and interlinking activities are:
1. Changing the treatment process by following simple rules of thumb

from the beginning. These straightforward rules are such that all
the practitioners can follow their application from day to day.
In Keropudas Hospital, there were two rules from the begin-
ning: (a) The patient must always be present when his/her
matters are discussed, and the relatives must be present when
decisions that affect their lives are made. (b) Discussions should
take place as far as possible with the team, and one-to-one ap-
pointments with the doctor, the psychologist, or others should
only take place for justified reasons.
2. Training the staff in the new activity, emphasizing dialogism as a
way of thinking instead of a “technique” or a restricted method. The
aim was to support both acquiring the new practice and gen-
erativity, creative variation. The entire staff and partners of the
psychiatric department got a three-year specialist-level family-
therapy training, with the main emphasis on Open Dialogue
ideas.
3. Research on both dialogues and the effectiveness of the new practice
(as described in chapters 7 and 8, respectively).
4. Decision-making and administrative support for dialogical practices.
Follow-up studies provide the basis for conclusions on the
central elements of a well-functioning treatment, and these, in
turn, have a direct impact on the practices.
These four activities—treatment, training, research, and decision-

making—have been brought to the interaction in a way where
they nourish each other. New experiences gained in treatment

situations afford the base for empirical data and points of refer-
ence for training. Research provides analyses and syntheses of the
material, ways of conceptualizing experiences, and subject matter
for training. Training strives at increasing the actors’ possibilities
to generate creative dialogical practices. Administration provides
continuity for the activity structure. To make the above described
possible, it has been necessary to create various arenas for encoun-
ters between the parties and such ways of interacting where no
single view demands an uncompromising authority, but, instead
new ways of meaning and continuing are sought.
Anticipation Dialogues have been generalized by means of
similar structures, but with two main differences. Up until the
2000s, the projects were time-limited and were carried out in vari-
ous parts of the country, in contrast to the continuous regional de-
velopment of Open Dialogues in Western Lapland. Largely due to
this, longitudinal studies are only now beginning to be established.
While both Anticipation Dialogues and Open Dialogues take place
in networks, Open Dialogues have a home base: the psychiatric cri-
sis team. Because the coordination of Anticipation Dialogues has to
be settled in the no-man’s-lands between welfare offices, hospitals,
educational institutions, therapy agencies, manpower offices, and
so on, new cross-sectoral supervision structures have to be created.
The “silo system” of sectored services has very few horizontal
management structures. Likewise, training has also been essential
in developing Open Dialogues. In each municipality/region taking
part in the process, a large training programme has been carried
out. A special task has been the training of network facilitators.
It has been important to bring the following elements into inter-
action:
• practices for sharing experience and developing activity

• training that enhances flexible variation
• research practices emphasizing client feedback
• cross-sectoral management structures.
The Japanese organization theorists Ikujiro Nonaka, Norio Konno,

and Ryoko Toyama (2001) have analysed the arenas of learning,
reflection, and dialogues in the landscape of innovative corpora-

tions. Their observations are connected with “knowledge-man-
agement” practices in enterprises. According to Nonaka et al.,
the challenge is not managing knowledge as if it were something
existing as such, nor knowledge creation as if there were no knowl-
edge yet, but, rather, bringing different kinds of knowledge into
dialogue in order to create new understanding and to exceed
separate perspectives. The personnel of enterprises have a great
deal of so-called tacit knowledge that does not necessarily articu-
late as initiatives and suggestions. Nonaka et al. have noticed that
progressive enterprises have succeeded in creating for the essen-
tial parties—from the shop floor to top management—arenas for
analysing the activity and creating innovations. In Japanese there
is a word for such spaces of coming together: the “ba”, a concept
especially in the vocabulary of the philosopher Kitaro Nishida. Ba
means both the geographical space and presence and also the men-
tal state generated in coming together face-to-face. According to
Nonaka et al., the various stages where tacit knowledge connects
with more explicit knowledge and develops into analyses guiding
the enterprise activity each require a special kind of ba. In each it
is essential that a meeting space is created where a perspective
exceeding the possibilities of the separate actors can be reached.
The contextualizations of the middle range that Nowotny et al.
outlined and the bas that Nonaka et al. describe are, in our view,
conceptually rather close. They have in common the pursuit to
cross boundaries, form new arenas on the boundaries, and gener-
ate between the parties dialogues that exceed the possibilities of
separate actors to understand what they are embedded in and
how to go on.
Notes
1. Donald A. Schön opens his Educating the Reflective Practitioner (1987)
with a powerful metaphor of the swamp: “In the varied topography of pro-
fessional practice, there is a high, hard ground overlooking a swamp. On the
high ground manageable problems lend themselves to solution through the
application of research-based theory and technique. In the swampy lowland,
messy, confusing problems defy technical solution. The irony of the situation
is that the problems of the high ground tend to be relatively unimportant to
individuals or society at large, however great their technical interest may be,
while in the swamp lie the problems of greatest human concern” (p. 3)
2. Teams of three or four therapists plus a supervisor were at the clients’
disposal 24 hours a day, seven days a week. There were six such teams in
Norway. Each therapist was responsible for three to six families.
3. Had they not, the outcomes could hardly have been positive. At least
this can be inferred in the light of the American Psychological Association’s
expert report of 2002: the therapists who followed their guidelines strictly
produced worse therapy outcomes than those who were flexible according to
needs and situations (see Norcross, 2002).
Epilogue:
on power and empowerment
N
etwork dialogues are, on the one hand, a new path to
follow. On the other hand, they are streams in a broader
flow of psychosocial work that aims at promoting the
clients’ autonomy and independence. The common direction can
be described as empowerment: helping clients to help themselves. In
the approaches we have described, the clients personal networks,
not only professional expertise, are essential in achieving this.
Network approaches transform interaction at the boundaries of the
private and the public domain and thus challenge the traditional
expert system, in which the professionals have the higher position
in defining how the helping situation is constructed. Do network
dialogues challenge the power positions as well? In this final sec-
tion we discuss power relationships in the new practice.
Power is ubiquitous, but not only domination. In our view,
power is an element of human relationships, not good or bad
in itself. Instead, its meaning is created in each meeting place.
Bakhtin (1986) wrote about asymmetric relations as a condition
for dialogue. Power relationships both enable dialogues and are
transformed by them. As Michel Foucault emphasizes, power
187
relations are everywhere,1 and it is necessary to be aware of their

force: “I hold to the fact that every human relationship is a power
relationship to a certain extent. We move about in a world of end-
less strategic relationships. Not every power relationship is bad
in itself but includes a danger. . . . A power relationship in itself is
not either good or bad but can be dangerous, so it is necessary to
consider on every level how it could channel its force in the best
possible way” (Foucault, 1988, p. 96). In this vein, it is important
to carefully channel the force created by making close connections
between the professional and private networks and generating
dialogues among them.
The dialogical practices we have discussed in this book are both
promising and threatening at the same time. On the one hand, they
promise good results in situations where clients face great risk of
exclusion and protracted distress. The practices have been devel-
oped precisely in difficult situations, not in the easiest enclaves of
psychosocial work. They may be threatening too, however, since
positive results can be redeemed only by changing significantly
some of the basic elements of the prevailing paradigm of profes-
sional help. Among others, the following factors characterize the
new paradigm emerging in the dialogical practices:
1. The professional helper asks for help from the client and other
“laypersons” in lessening their own worries.
2. The professional helper varies his/her own activity instead of
aiming at changing others according to some ends.
3. Development is seen as co-evolvement, where everyone—in-
cluding the professionals—changes.
4. The clients’ personal networks are seen as a resource, not as a
source or bearer of problems.
5. Common problem definitions are not pursued in coopera-
tion; the professionals become interested in how the situation
presents itself from each participant’s point of view.
6. Planning treatment or helping is not separated from the treat-
ment or helping process, and the processes are not planned
among professionals without clients being present.
7. Listening becomes more important that giving advice.
EPILOGUE: ON POWER AND EMPOWERMENT 189
8. Ways of thinking, attitude, and encountering become more

important than methods.
9. Crossing boundaries in the professional system becomes more
important than drawing boundaries.
All this means that our clients have a greater say in how they want
their problems to be dealt with. In the dialogues between the cli-
ents’ personal networks and the professional networks, a certain
kind of shared expertise emerges. This is a qualitatively different
process compared to the traditional idea of expertise, which aims
at controlling phenomena through special professional knowledge.
Professionals do not alone hold the keys to the solutions—nor does
anyone else. In this way, the laypeople also participate in creating
expertise. This may, of course, be disconcerting from the point
of view of the traditional expert paradigm. However, the profes-
sional knowledge of the doctor, social worker, therapist, teacher,
psychologist, or other professional does not evaporate, nor is it
seized by the network. The new challenge is to make professional
knowledge and experience useful for creating joint understanding
and collaborative solutions.
While the position of the professionals and the clients changes
when networks meet, the positions do not become symmetrical.
It is the clients or those near them that seek help—or sometimes
clients are coerced into contacts with agencies, as in severe cases
of child protection. The professionals remain wage-labourers or
entrepreneurs; helping is part of their job description or business.
The professionals are on home turf when problems, symptoms,
deviance, definitions, diagnoses, or normalcy are discussed, even
when the discussion takes place in the client’s home. The profes-
sionals have—each according to his/her specific branch and posi-
tion—the possibility to facilitate or hinder the client’s access to
material and non-material resources in the service system. Power
relationships do not leave the therapy room or joint-action negotia-
tions when dialogicity enters. However, power does not only mean
subjugation. Barbara Cruikshank (1999, p. 41) points out that em-
powerment is a form of power that promotes rather than represses
subjectivity, of power that produces and relies upon active subjects
rather than subjugation.
Expertise, mediating, and power

Pursuing a joint expertise alters the way the power relationships
are practised. In discussing expertise in governing the poor, Peter
Miller and Nicolas Rose (1990) point out the important mediat-
ing role of expertise. According to them, expertise is significant in
liberal democratic modes of government in establishing a distance
between formal organs of political power and the type of practi-
cal interventions on the objects of government deemed necessary.
The modulation of events in the “private” spaces of family life
and personal conduct may be called for or be required by political
forces, but it cannot be directly undertaken by the public powers
since that would violate the necessary relation with the domain
of “civil society” that liberal mentalities of government call into
being. Expertise thus plays a crucial mediating role between the
formal organs of political rule and the objects of its policies and
programmes (Miller & Rose, 1990).
Although the network dialogues we have described can scarce-
ly be called interventions in the sense of unilateral, monological
actions upon others, they are, nevertheless, interventions within
the broad context of governance: they are parts in the broad con-
text of attempts to shape conduct in the private domains of social
life through actions by the public domain—namely, the psycho-
social professional system. According to Michel Foucault (1980),
governance includes any programme, discourse, or strategy that
attempts to alter or shape the actions of others or oneself. Net-
work approaches allow the professionals to draw very close to
the clients and their personal networks. Thereby, in a way, they
“normalize” close ties between the given networks as well as the
public discourse of the clients’ private matters. The clients’ ways
of guarding and negotiating the private–public boundaries are
facing challenge.
Dialogicity and an emerging joint expertise transform the con-
figuration between the professionals and non-professionals and
bring about marked changes in the possibilities of a single party
to control the situation. However, the pursuit to normalize devi-
ance remains—be it curing mental illnesses, curbing dysfunctional
socialization, or the like. In multi-stakeholder dialogues, the stance
of not-knowing becomes both more important and more viable
than in one-to-one situations. As the aim is to generate shared

language and mutual forms of understanding, the professionals
have the challenge to use their professional knowledge and ex-
perience accordingly. Professionals (like others) cannot but have
presuppositions, and it is these presuppositions that make creating
subjective sense possible. However, the professionals’ presupposi-
tions—such as emerging diagnostic ideas and consequent treat-
ment schemes—can create “noise” that hinders listening to the
others and impedes dialogue-generating. Thus, it is most essential
that the professionals are prepared to change their views instead
of imposing them. Multilateral situations do not lend themselves
easily to unilateral control. Therefore, they afford both the need
and ample opportunities for experiencing and joining polyphony.
This can be seen as one of the feasible options for dealing with the
crises of professional expertise in social and health care. According
to Risto Eräsaari (2002), science, professions, and institutions—the
tripod of expertise—tremble simultaneously.
Critique of power from within power

According to Scott Lash (2002), critics of power tend to resort to
“transcendentals” in analysing power relationships. A domain free
of subjugation is outlined, which, in turn, is contrasted with the
analysed system.2 He calls for critique without dualisms (p. 9),
with its pure ideal on the one hand and the imperfect reality on
the other. The challenge is critique from within.
In analysing the details relating to Open Dialogues, Petteri
Autio (2003) pointed out four types of discourse of exercising
power: (1) In the expert–patient discourse, the team members de-
fined themselves as practitioners and, in relation to this, the clients
defined themselves as the patient and thereby as the one with a
lesser means of influence. (2) In the treatment-model discourse,
the practitioners defined their position in accordance with the
Open Dialogues principles, which enabled them to, for example,
avoid discussing diagnoses, which was actually against the will
of a family member of the patient. (3) In normalcy discourse,
the general societal attitudes about normality came into the fore.
(4) Democracy discourse followed the prevailing Western patterns

of placing high value on individuality.
Research on practices of power within network dialogue ap-
proaches is essential for learning. There are lessons to be learned
for the professionals—including the facilitators—that pursue dia-
logues. Professionals and facilitators need to become increasingly
aware of the forces they channel within the power relationships,
so as to better enhance the independence and autonomy of the
clients.
Power relationships are essential to Open and Anticipation Dia-
logues; they are not obstacles. Without such relationships the par-
ties would probably never meet in the combinations we described.
The Open Dialogues approach is made operational through a policy
in which networks are summoned immediately in crises and where
the other six guidelines (including the social network, flexibility
and mobility, responsibility, psychological continuity, tolerating
uncertainty, and dialogicity), along with immediate help and the
network approach, are followed.
Thus, the fact that a given psychic-crisis situation in Western
Lapland is dealt with in the Open Dialogues way is by no means a
matter of chance, relying on whether an anxious relative or the pro-
fessional in the given agency feels Open Dialogues would be the
path to follow. On the contrary, private and professional networks
are summoned without a pre-assessment of whether the symptoms
and other data indicate a network dialogue approach. As a matter
of fact, departing from the traditional family therapy route seems
to be a crucial point in making possible the healing dialogue-
communities. In the traditional approach, invitations for family
members to join the treatment are decided on the basis of defini-
tions of the problem and indications. Where the Open Dialogues
approach is systematically practised, decision-making, training,
and research support a policy to invite the private network and to
enjoin the professionals that have been contacted to stay on, not
as an alternative, but as the basis for cooperation across the spec-
trum of treatments. This, in turn, has been central for generating
the exceptional treatment outcomes. However, the positive results
are achieved as a consequence of exercising power—carrying out
a policy—not without or outside power relationships.
The perspective for Anticipation Dialogues is similar: to de-

velop a policy, where with multiple professionals around them,
clients—regardless of whether these are children, adolescents, par-
ents, senior citizens, long-term unemployed, or disabled—have the
right for dialogical means of generating and maintaining cooperat-
ing joint action, with the client’s positive everyday life as the point
of reference.
Dialogicity by real people,

not by elements in a strategy
Open and Anticipation Dialogues are practiced by people. They are
not mechanisms in a power machine. The dialogicity in them—as
far as it actually emerges—depends on the voluntary “submission”
of the participants to pursue reciprocity and responsiveness. No
macro-policy can produce dialogism. Policies can enable profes-
sionals, but they cannot operate past or without micro-relationships.
One can only imagine what the outcomes would be if the profes-
sionals were obliged or forced to carry out Open or Anticipation
Dialogues. The unintended outcomes would most likely outweigh
the policymakers’ intentions. If dialogicity is to be pursued, there
are few viable alternatives to promoting a will to dialogues among
the professionals. Only if they find a personal, subjective perspec-
tive in dialogicity will the professionals reach towards responsive-
ness. Thus, dialogicity needs to be personally persuasive for the
practitioner who is seeking ways to solve professional problems.
Moreover, even if the professionals become interested in crossing
boundaries via dialogues, dialogicity will not emerge without the
clients, and their significant persons, also seeing dialogues as a
promising perspective from their point of view. In short, although
network dialogues take place within power relationships, they
cannot be “put into operation” through domination.
Network dialogues open up promising perspectives in allevi-
ating distress and preventing exclusion. It is necessary to analyse
and discuss transparently the power relationships that enable the
practices and that the practices themselves transform.
Notes
1. “Between every point of a social body, between a man and a woman, be-
tween members of a family, between a master and his pupil, between everyone
who knows and every one who does not, there exist relations of power . . .”
(Foucault, 1980, p. 187)
2. Lash uses Jürgen Habermas’ notion of the ideal speech situation as an
example. Habermas (1984) described the conditions necessary for reciprocal
and egalitarian communication. According to him, such communication can-
not take place in the “system world” of money and power. The relationships
between the expert system of the social state and the client lead to strategic,
instrumental communication aiming at profit or power (with money relation-
ships aiming at profit). An ideal speech situation, free of domination, could,
according to him, occur in the “life world” of the citizens, but only if the condi-
tions of justifying, argumentation, questioning, and making logical statements
are met—and only in the ideal case. Lash calls for an analysis of actual situations
without “transcendentals” as the comparison. Hans-Georg Gadamer (1982)
pointed out that instead of stating rules and preconditions for a dialogue on the
basis of abstract ideas that are imposed onto actual situations, dialogues ought
to be analysed as situated, generated by those and only those in dialogue.
REFERENCES
Aaltonen, J., Seikkula, J., Alakare, B., Haarakangas, K., Keränen, J., &
Sutela, M. (1997). Western Lapland project: A comprehensive fam-
ily- and network-centered community psychiatric project. ISPS:
Abstracts and Lectures, 12–16 October.
Alanen, Y. (1997). Schizophrenia: Its Origins and Need-Adapted Treatment.
London: Karnac.
Alanen, Y. O, Lehtinen, K., Räkköläinen, V., & Aaltonen, J. (1991).
Need-adapted treatment of new schizophrenic patients: Experi-
ences and results of the Turku Project. Acta Psychiatrica Scandi-
navica, 83: 363–372.
American Psychiatric Association (1987). Diagnostic and Statistical Man-
ual of Mental Disorders (4th ed.). Washington, DC.
Andersen, T. (1987). The reflecting team: Dialogues and meta-dia-
logues in clinical work. Family Process, 26: 415–428.
Andersen, T. (1991). The Reflecting Team: Dialogues and Dialogues about
the Dialogues. New York: W. W. Norton.
Andersen, T. (1995). Reflecting processes: Acts of informing and form-
ing. In S. Friedman (Ed.), The Reflective Team in Action: Collaborative
Practice in Family Therapy (pp. 11–35). New York: Guilford Press.
Andersen, T. (2005). The network context of network therapy. In:
195
196 REFERENCES
A. Lightburn & P. Sessions (Eds.), Handbook of Community-Based

Clinical Practice. New York: Oxford University Press.
Anderson, H. (1997). Conversation, Language, and Possibilities. New
York: Basic Books.
Anderson, H., & Goolishian, H. (1988). Human systems as linguistic
systems: Preliminary and evolving ideas about the implications for
clinical theory. Family Process, 27: 371–393.
Anderson, H., & Goolishian, H. (1992). The client is the expert. A not-
knowing approach to therapy. In S. McNamee & K. Gergen (Eds.),
Therapy as Social Construction (pp. 25–39). London: Sage.
Arnkil, E. (1991a). Social work and the systems of boundary. Sugges-
tions for conceptual work. In: E. Arnkil, Keitä muita tässä on mukana.
Viisi artikkelia verkostoista (pp. 97–120). Sosiaali- ja terveyshallitus.
Raportteja 23. Helsinki: VAPK-kustannus.
Arnkil, E. (1991b). Peilejä. Hypoteeseja sosiaalityön ristiriidoista ja kehitys-
vyöhykkeestä [Hypotheses on the contradictions and developmental
zone of social work; English summary]. Sosiaali- ja terveyshallituk-
sen tutkimuksia 5. Helsinki: VAPK-kustannus.
Arnkil, E. (1992). Sosiaalityön rajasysteemit ja kehitysvyöhyke [The sys-
tems of boundary and the developmental zone of social work;
English summary]. Jyväskylä Studies in Education, Psychology
and Social Research 85. Jyväskylä: University of Jyväskylä.
Arnkil, T. E., & Eriksson, E. (1995). Mukaan meneminen ja toisin toimimi-
nen. Nuorosopoliklinikka verkostoissaan [Becoming alike and acting
differently; English abstract and summary]. Tutkimuksia 51. Saa-
rijärvi: STAKES.
Arnkil, T. E., & Eriksson, E. (1996). Kenelle jää kontrollin Musta Pekka
-kortti? Sosiaalitoimisto verkostoissaan [Who is going to have the
Old maid card of the control game? Social welfare office in its net-
works; English abstract and summary]. Tutkimuksia 63. Jyväskylä:
STAKES.
Autio, P. (2003). Indoktrinaatio avoimen dialogin hoitomallissa. [Indoctri-
nation in the open dialogues treatment approach]. Pro gradu tutk-
ielma, Joensuun yliopisto, Psykologian laitos.
Bakhtin, M. (1981). Dialogic Imagination. Austin, TX: Texas University
Press.
Bakhtin, M. (1984). Problems of Dostojevskij’s Poetics. Theory and History
of Literature, Vol. 8. Manchester: Manchester University Press.
Bakhtin, M. (1986). Speech Genres and Other Late Essays. Austin, TX:
University of Texas Press.
REFERENCES 197
Bakhtin, M. (1990). The Dialogic Imagination, ed. M. Holquist, trans. C.

Emerson & M. Holquist. Austin, TX: University of Texas Press.
Barnes, J. (1954). Class and committees in a Norwegian island parish.
Human Relations, 7: 39–58.
Barnes, J. (1972). Social Networks. Reading, MA: Addison-Wesley.
Barry, A. (2001). Political Machines: Governing a Technological Society.
London & New York: Athlone Press.
Bateson, G. (1972). Steps to an Ecology of Mind. New York: Ballantine
Books.
Bauman, Z. (2002). Liquid Modernity. Cambridge: Polity Press.
Beck, U. (1986). Risikogesellschaft. Auf dem Weg in eine andere Moderne.
Frankfurt/M: Suhrkamp.
Beck, U., Giddens, A., & Lash, S. (1994). Reflexive Modernization: Politics,
Tradition and Aesthetics in the Modern Social Order. London: Polity
Press.
Bohm, D. (1997). On Dialogue, ed. L. Vichol. London: Routledge.
Bourdieu, P. (1993). Questions of Sociology. London: Sage.
Bourdieu, P. (1998). Practical Reason: On the Theory of Action. Cambridge:
Polity Press; Stanford, CA: Stanford University Press.
Bråten, S. (1988). Between dialogical mind and monological reason:
Postulating the virtual other. In: M. Campanella (Ed.), Between Ra-
tionality and Cognition (pp. 205–235). Turin: Albert Meynier.
Bruner, J. (1985). Vygotsky: A historical and conceptual perspective. In:
J. Wertsch (Ed.), Culture, Communication and Cognition: Vygotskian
Perspective (pp. 21–34). New York: Cambridge University Press.
Buber, M. (1987). I and Thou (2nd edition), trans. R. G. Smith. New York:
Scribners/Macmillan.
Castells, M. (1996). The Rise of the Network Society—The Information Age:
Economy, Society and Culture, Vol. 1. Malden, MA: Blackwell.
Crowley, T. (2001). Bakhtin and the history of language. In: K. Hirsch-
kop, & D. Shepherd (Eds.), Bakhtin and Cultural Theory (2nd edition,
pp. 177-200). Manchester: Manchester University Press.
Cruikshank, B. (1999). The Will to Empower: Democratic Citizens and
Other Subjects. Ithaca & London: Cornell University Press.
de Shazer, St. (1994). Words Were Originally Magic. New York: W. W.
Norton.
Eräsaari, R. (2003). Open-context expertise. In: A. Bamme, G. Getz-
inger, & B. Wieser (Eds.), Yearbook 2003 of the Institute for Advanced
Studies on Science, Technology and Society (pp. 31—66). Munich:
Profil Verlag.
198 REFERENCES
Foucault, M. (1980). Power/Knowledge: Selected Interviews and Other

Writings 1972–1977, ed. L. C. Gordon. Sussex: Harvester Press.
Foucault, M. (1988). Politics, Philosophy, Culture: Interviews and Other
Writings, 1977–1984, ed. L. D. Kritzman. London: Routledge.
Freire, P. (1970). Pedagogy of the Oppressed. New York: Herder &
Herder.
Friis, S., Larsen, T. K., & Melle, I. (2003). Terapi ved psykoser. Tidsskrift-
et for Norsk Lægeforening, 123: 1393.
Gadamer, H.-G. (1982). Reason in the Age of Science. Cambridge, MA:
MIT Press.
Galperin, P. Y. (1969). The development of mental acts. In: M. Cole & J.
Maltzman (Eds.), A Handbook of Contemporary Soviet Psychology (pp.
249–273). New York: Basic Books.
Gergen, K. (1994). Realities and Relationships: SOUNDINGS in Social
Construction. Cambridge, MA: Harvard University Press.
Gergen, K. (1999). An Invitation to Social Constructionism. London:
Sage.
Gilbert, P., Harris, J., McAdams, L. A., & Jeste, P. (1995). Neurolep-
tic withdrawal in schizophrenic patients: A review of literature.
Archives of General Psychiatry, 52, 173—188.
Gupta, S., Andreasen, N., Arndt, S., & Flaum, M. (1997). The Iowa longi-
tudinal study of recent-onset schizophrenia: One-year follow-up of
first-episode patients. Schizophrenia Research, 23 (1): 1–13.
Haarakangas, K. (1997). Hoitokokouksen äänet [The voices in treatment
meeting. A dialogical analysis of the treatment meeting conversa-
tions in family-centred psychiatric treatment process in regard to
the team activity. English Summary]. Jyväskylä Studies in Education,
Psychology and Social Research, 130.
Habermas, J. (1984). The Theory of Communicative Action, Vol. 1: Research
and Relationship of Society. Boston, MA: Beacon Press.
Hoffman, L. (2002). Family Therapy: An Intimate History. New York:
W. W. Norton.
Holma, J. (1999). The search for a narrative. Investigating acute psy-
chosis and the need-adapted treatment model from the narrative
viewpoint. Jyväskylä Studies in Education, Psychology and Social Re-
search, 150.
Holma, J., & Aaltonen, J. (1997). The sense of agency and the search for
a narrative in acute psychosis. Contemporary Family Therapy, 19 (4):
463–477.
Holquist, M. (Ed.) (1981). The Dialogic Imagination: Four Essays by M. M.
REFERENCES 199
Bakhtin, trans. C. Emerson & M. Holquist. Austin, TX: University

of Texas Press.
Imber-Black, E. (1988). Families and Larger Systems: A Family Therapist’s
Guide Through the Labyrinth. New York/London: Guilford Press.
Imber-Coppersmith, E. (1985). Families and multiple helpers: A sys-
temic perspective. In: D. Campbell & R. Draper (Eds.), Applications
of Systemic Family Therapy. New York: Grune & Stratton.
Isaacs, W. (1999). Dialogue and the Art of Thinking Together: A Pioneering
Approach to Communication in Business and in Life. New York: Cur-
rency/Doubleday.
Jackson, C., & Birchwood, M. (1996). Early intervention in psychosis:
Opportunities for secondary prevention. British Journal of Clinical
Psychology, 35: 487–502.
Johnstone, E., Crow, T., Johnson, A., & Macmillan, J. (1986). The North-
wick Park Study of first episode schizophrenia: I. Presentation of
the illness and problems relating to admission. British Journal of
Psychiatry, 148: 115–120.
Johnstone, E., Macmillan, F., Frith, C., Benn, D., & Crow, D. (1990).
Further investigation of the predictors of outcome following first
schizophrenic episode. British Journal of Psychiatry, 157: 182–189.
Kalla, O., Aaltonen, J., Wahlström, J., Lehtinen, V., Cabeza, I. G., &
Gonzales de Chaver, M. (2002). Duration on untreated psychosis
and its correlates in first episode psychosis in Finland and Spain.
Acta Psychiatrica Scandinavia, 106: 265–275.
Kamya, H., & Trimble, D. (2002). Response to injury: Toward ethical
construction of the other. Journal of Systemic Therapies, 21: 19–29.
Kennedy, M., & Bourne, J. (1996). Oxford Dictionary of Music (4th edi-
tion). Oxford: Oxford University Press.
Keränen, J. (1992). The choice between outpatient and inpatient treat-
ment in a family centred psychiatric treatment system [English
summary]. Jyväskylä Studies in Education, Psychology and Social Re-
search, 93: 124–129.
Kissling, W., & Leucht, S. (2001). Results of treatment of schizophrenia:
Is the glass half full or half empty. International Clinical Psychophar-
macology, 14 (Suppl. 3): S11–S14.
Klefbeck, J., Bergerhed, E., Forsberg, G., Hultkranz-Jeppson, A., &
Marklund, K. (1988). Natverksarbete i multiproblemfamiljer [Net-
working in multiproblem families]. Botkyrka: Botkyrka kommun;
Frankfurt: Suhrkamp Verlag.
Kokko, R.-L. (2003). Asiakas kuntoutuksen yhteistyöryhmässä. Institution-
200 REFERENCES
aalisen kokemisen jännitteitä [The client in a rehibilitation coopera-

tion group: Tensions of institutional encounter; English abstract].
Tutkimuksia 72. Helsinki: Kuntoutussäätiö.
Larsen, T. K., Johannesen, J., & Opjordsmoen, S. (1998). First-episode
schizophrenia with long duration of untreated psychosis. British
Journal of Psychiatry., 172 (Suppl. 33): 45–52.
Lash, S. (2002). Critique of Information. London/Thousand Oaks, CA/
Delhi: Sage.
Latour, B. (1988). Politics of explanation: An alternative. In: S. Woolgar
(Ed), Knowledge and Reflexivity. New Frontiers in Sociology. Bristol:
Sage.
Latour, B. (1996). On actor-network theory: A few clarifications. Soziale
Welt, 47 (4): 369–381.
Lehman, A., Kreyenbuhl, J., Buchanan, R., Dickerson, F., Dixon, L.,
Goldberg, R., et al. (2003). The Schizophrenia Patient Outcome Re-
search Team (PORT): Updated treatment recommendations, 2003.
Schizophrenia Bulletin, 30 (2), 193–217.
Lehtinen, V., Aaltonen, J., Koffert, T., Räkköläinen, V., & Syvälahti, E.
(2000). Two-year outcome in first-episode psychosis treated ac-
cording to an integrated model. Is immediate neuroleptisation
always needed? European Psychiatry, 15: 312–320.
Lehtinen, V., Aaltonen, J. Koffert, T., Räkköläinen, V., Syvälahti, E.,
& Vuorio, K. (1996). Integrated treatment model for first-contact
patients with a schizophrenia-type psychosis: The Finnish API
project. Nordic Journal of Psychiatry. 50: 281–287.
Leont’ev, A. N. (1978). Activity, Consciousness, and Personality. Engle-
wood Cliffs, NJ: Prentice-Hall.
Levinas, E. (1985). Ethics and Infinity: Conversations with Philippe Nemo,
trans. R. Cohen. Pittsburgh, PA: Duquesne University Press.
Linell, P. (1998). Approaching Dialogue: Talk, Interaction and Contexts in
Dialogical Perspectives. Amsterdam: John Benjamins.
Linell, P., Gustavsson, L., & Juvonen, P. (1988). Interactional dominance
in dyadic communication: A presentation of initiative-response
analysis. Linguistics, 26: 415–442.
Linszen, D., Dingemans, P., & Lenior, M. (2001). Early intervention
and a five-year follow up in young adults with a short duration of
untreated psychosis: Ethical implications. Schizophrenia Research,
51 (1): 55–61
Linszen, D., Lenior, M., De Haan, L., Dingemans, P., & Gersons, B.
(1998). Early intervention, untreated psychosis and the course of
REFERENCES 201
early schizophrenia. British Journal of Psychiatry. 172 (Suppl. 33):

84–89.
Loebel, A., Lieberman, J., Alvir, J., Mayerhoffer, D., Geisler, S., &
Szymanski, S. (1992). Duration of psychosis and outcome in first-
episode schizophrenia. American Journal of Psychiatry, 149: 1183–
1188.
Lowe, R. (2005). Structured methods and striking moments. Family
Process, 44: 65–75.
Luckman, T. (1990). Social communication, dialogue and conversation.
In: I. Markova & K. Foppa (Eds.), The Dynamics of Dialogue (pp.
45–61). London: Harvester.
Luhmann, N. (1989). Ecological Communication. Cambridge, MA: Polity
Press.
Lyotard, J.-F. (1984). The Postmodern Condition: A Report on Knowledge,
Theory, and History of Literature, Vol. 10. Minneapolis, MN: Univer-
sity of Minnesota Press.
Markova, I. (1990). Introduction. In: I. Markova & K. Foppa (Eds.), The
Dynamics of Dialogue (pp. 1–22). London: Harvester.
Maturana, H., & Varela, F. (1980). Autopoiesis and Cognition. Dordrecht:
Reidel.
McGorry, P., Edwards, J., Mihalopoulos, C., Harrigan, S., & Jackson, H.
(1996). EPPIC: An evolving system of early detection and optimal
management. Schizophrenia Bulletin, 22: 305–325.
Miller, P., & Rose, N. (1990). Governing economic life. Economy and
Society, 19 (1): 1–31.
Nightingale, A. (2000). Genres in Dialogue. Plato and the Construct of Phi-
losophy. Cambridge: Cambridge University Press.
Nonaka, I., Konno, N., & Toyama, R. (2001). Emergency of “Ba”: A
conceptual framework for the continuous and self-transcending
process of knowledge creation. In: I. Nonaka & T. Nishiguchi
(Eds.), Knowledge Emergency: Social, Technical, and Evolutionary Di-
mensions of Knowledge Creation (pp. 13–29). Oxford: Oxford Univer-
sity Press.
Norcross, J. C. (Ed.) (2002). Psychotherapy Relationships That Work: Ther-
apist Contribution and Responsiveness to Patients. New York: Oxford
University Press.
Nowotny, H., Scott, P., & Gibbons, M. (2002). Re-Thinking Science:
Knowledge and the Public in an Age of Uncertainty. Malden, MA:
Blackwell Publishers.
Ogden, T., & Halliday-Boykins, C. (2004). Multisystemic treatment of
202 REFERENCES
antisocial adolescents in Norway: Replication of clinical outcomes

outside of the U.S. Child and Adolescent Mental Health, 9: 77–83.
Opjordsmoen, S. (1991). Long-term clinical outcome of schizophrenia
with special reference to gender difference. Acta Psychiatrica Scan-
dinavia, 83: 307–313.
Patterson, D. (1988). Essays on Bakhtin and His Contemporaries. Lexing-
ton, KY: University Press of Kentucky.
Penn, P. (1985). Feed-forward: Future questions, future maps. Family
Process, 24: 299–310.
Rosaldo, R. (1989). Culture & Truth: The Remaking of Social Analysis. Bos-
ton, MA: Beacon Press
Sachs, D., & Shapiro, S. (1976). On parallel processes in therapy and
teaching. Psychoanalytic Quarterly, 45 (3): 394–415.
Schön, D. (1983). The Reflective Practitioner: How Professionals Think in
Action. London: Temple Smith.
Schön, D. (1987). Educating the Reflective Practitioner. San Francisco, CA:
Jossey-Bass.
Schwartzman, H., & Kneifel, A. (1985). How the child care system rep-
licates family patterns. In: J. Schwartzman (Ed.), Families and Other
Systems (pp. 87–107). New York: Guilford Press.
Seikkula, J. (1991). Perheen ja sairaalan rajasysteemi potilaan sosiaalisessa
verkostossa [The systems of boundary between the family and hos-
pital; English summary]. Jyväskylä Studies in Education, Psychology
and Social Research, 80.
Seikkula, J. (1994). Sosiaaliset verkostot: Ammattiauttajan voimavara krii-
seissä [Social networks: Professional helper’s resource in crises].
Helsinki: Kirjayhtymä.
Seikkula, J. (1995). From monologue to dialogue in consultation with
larger systems. Human Systems, 6: 21–42.
Seikkula, J. (2002). Open dialogues with good and poor outcomes for
psychotic crises: Examples from families with violence. Journal of
Marital and Family Therapy, 28: 263–274.
Seikkula, J., Aaltonen, J., Alakare, B., Haarakangas, K., Keranen, J., &
Satela, M. (1995). Treating psychosis by means of open dialogue. In
S. Friedman (Ed.), The Reflecting Team in Action. New York: Guilford
Press.
Seikkula, J., Alakare, B., & Aaltonen, J. (2001a). Open dialogue in psy-
chosis I: An introduction and case illustration. Journal of Construc-
tivist Psychology, 14: 247–265.
Seikkula, J., Alakare, B., & Aaltonen, J. (2001b). Open dialogue in psy-
REFERENCES 203
chosis II: A comparison of good and poor outcome cases. Journal of

Constructivist Psychology, 14: 267–284.
Seikkula, J., Alakare, B., Aaltonen, J., Haarakangas, K., Keränen, J., &
Lehtinen, K. (2006). Five years’ experiences of first-episode non-
affective psychosis in open dialogue approach: Treatment prin-
ciples, follow-up outcomes and two case analyses. Psychotherapy
Research, 16 (2): 214–228.
Seikkula, J., Alakare, B., Aaltonen, J., Holma, J., Rasinkangas, A.,
& Lehtinen, V. (2003). Open Dialogue approach: Treatment prin-
ciples and preliminary results of a two-year follow-up on first
episode schizophrenia. Ethical Human Sciences and Services, 5 (3):
163–182.
Seikkula, J., Arnkil, T. E., & Eriksson, E. (2003). Postmodern society and
social networks: Open and anticipation dialogues in network meet-
ings. Family Process, 42 (2): 185–203.
Seikkula, J., & Olson, M. (2003). The open dialogue approach to acute
psychosis. Family Process, 42: 403-418.
Seikkula, J., & Sutela, M. (1990). Coevolution of the family and the
hospital: The system of boundary. Journal of Strategic and Systemic
Therapies, 9: 34–42.
Shotter, J. (1993a). Conversational Realities: Constructing Life through Lan-
guage. London: Sage.
Shotter, J. (1993b). Cultural Politics of Everyday Life: Social Construction-
ism, Rhetoric, and Knowing of the Third Kind. Milton Keynes: Open
University Press & University of Toronto Press.
Shotter, J. (1997). Dialogical realities: The ordinary, the everyday, and
other strange new worlds. Journal for the Theory of Social Behavior,
27: 101–113.
Speck, R., & Attneave, C. (1973). Family Networks. New York: Pantheon
Books.
Stanton, A., & Schwartz, M. (1954). The Mental Hospital. New York:
Basic Books.
Stern, S., Doolan, M., Staples, E., Szmukler, G., & Eisler, I. (1999). Dis-
ruption and reconstruction: Narrative insights into the experience
of family members caring for a relative diagnosed with serious
mental illness. Family Process, 38: 353–369.
Strauss, A. (1978). Negotiations: Varities, Contexts, Processes and Social
Order. San Francisco, CA: Jossey-Bass.
Strauss, J., & Carpenter, W. (1972). The prediction of outcome in schizo-
phrenia. Archives of General Psychiatry, 27: 739–746.
204 REFERENCES
Surakka, V. (1999). Contagion and modulation of human emotions.

Acta Universitatis Tamperensis, 627.
Svedberg, B., Mesterton, A., & Cullberg, J. (2001). First-episode non-af-
fective psychosis in a total urban population: A 5-year follow-up.
Social Psychiatry, 36: 332–337.
Trimble, D. (2000). Emotion and voice in network therapy. Netletter, 7
(1): 11–16.
Tuori, T. (1994). Skitsofrenian hoito kannattaa. Raportti skitsofrenian, tut-
kimuksen, hoidon ja kuntoutuksen valtakunnallisen kehittämisohjelman
10-vuotisarvioinnista [Treatment of schizophrenia is effective]. Rap-
ortteja 143. Helsinki: STAKES.
Vadén, T. (1985). Psykologisen asiakassuhteen jäsentäminen [Analysing
the psychological client-relationship]. Ylioppilaiden terveydenhoi-
tosäätiön tutkimuksia ja selvityksiä 20. Helsinki: YTHS.
Van der Kolk, B., & Fisler, R. (1995). Dissociation and the fragmentary
nature of traumatic memories: Overview and exploratory study.
Journal of Traumatic Stress, 8: 505–525.
Voloshinov, V. (1996). Marxism and the Philosophy of Language (6th edi-
tion). London: Harvard University Press.
Vygotsky, L. (1934). Thought and Language (9th printing). Cambridge,
MA: MIT Press, 1972.
Watzlawick, P., Weakland, J., & Fisch, R. (1974). Change: Principles of
Problem Formation and Problem Resolution. New York/London: W.
W. Norton.
Weckroth, J. (1986). Tapahtuman ennakointi psykologin yksilöntutkimuksen
menetelmänä [Anticipating events as the psychologist’s method of
individual assessment]. Oulun yliopiston työtieteen laitoksen tut-
kimuksia 6. Oulu: Oulun yliopisto.
Wertsch, J. (1985). Vygotsky and Social Formation of Mind. Cambridge,
MA: Harvard University Press.
Wertsch, J. (1991). Voices of the Mind: A Sociocultural Approach to Medi-
ated Action. London: Harvester/Wheatsheaf.
White, M. (1997). Narratives of Therapists’ Lives. Adelaide: Dulwich Cen-
tre Publications.
White, M. (2002). Addressing personal failure. International Journal of
Narrative Therapy and Community Work, 3: 33–76.
Yung, A., Phillips, L., McGorry, P., Hallgren, M., McFarlane, C., Jack-
son, H., Francey, S., & Pattor, G. (1998). Can we predict the onset of
first episode psychosis in a high-risk group? International Clinical
Psychopharmacology, 13 (Suppl. 1): S23–S30.
INDEX
Aaltonen, J., 51–52, 58, 61, 132, and joint sharing, 106
150–151, 153–154 lessening worries, 65–88
activity: and Open Dialogues, similarities
language, 18 and differences, 89–103
theory, 31 and power relationships, 192–193
agoras: role of facilitators in, 116
and contextuality, 176–180 use of, 65
of learning, and generalizing anticipation method/questions,
practices, 180–184 16–17, 19
Alakare, B., 52, 58, 132, 151–153 anxiolytics, 165
Alanen, Y., 21, 56, 61, 137 API (Integrated Treatment of Acute
American Psychiatric Association, Psychosis), 150–156, 164
150 Finnish national project, 150
American Psychological Arndt, S., 156
Association, 185 Attneave, C., 4, 40, 95
Andersen, T., 15, 17, 61, 110 Autio, P., 140, 191
Anderson, H., 54, 61 B
Andreasen, N., 156 Bakhtin, M., 8, 51, 76, 95–97, 99–103,
Anticipation Dialogue(s), 6–8, 109, 105, 108–109, 114, 132, 141
111, 113, 122–123, 149 authoritative discourse, 3
development of, 13–15, 31, 182– dialogue:
183 preconditiond for, 138, 187
and increasing tolerance of unfinalizability of, 108
uncertainty, 116–117 and emergence of ideas. 108
205
206 INDEX
Bakhtin, M. (continued): consciousness(es):

event of being, 119, 128 communication between,
polyphonic reality, 99–100 dialogue as, 105
social construct, problems social character of, 97
reformulated as, 61 individual, 97
theory of, 97 “contagion” of emotions, 45
Barnes, J., 4 contextuality, 175–176
Barry, A., 4 and agoras, 176–180
Bateson, G., 35 contextualization, 177–179
Bauman, Z., 32 conversation vs dialogue, 95
Beck, U., 27, 32 coping skills, 165
Benn, D., 165 crisis:
Berg, I. K., 31 intervention, 51, 57
Bergerhed, E., 40 adapting flexibly, 52, 56–57
Birchwood, M., 58 dialogicity, 52, 60–61
bodily memory, 119 immediate response, 53–54
Bohm, D., 95 including client’s/patient’s
boundary: social network, 52, 54–56
-crossing, 2, 4–6, 27, 29, 33, 91 Open Dialogues as, 51–64
negotiations, 27 practical guidelines, 53
system, 29–30, 44 psychological continuity in, 52,
concept, 29, 44 58–59
Bourdieu, P., 35, 38 taking responsibility, 52, 57–58
BPRS (Brief Psychiatric Rating tolerating uncertainty, 52,
Scale), 152, 157 59–60
Bråten, S., 107 session(s), 41
Brief Psychiatric Rating Scale cross-boundary negotiations, 36
(BPRS), 152, 157 Crow, T., 165
Bruner, J., 111 Crowley, T., 98
Buber, M., 96 Cruikshank, B., 43, 189
C Cullberg, J., 163–164
Campbell Library, 169 D
Castells, M., 4–5 De Haan, L., 153
catharsis, 120 democracy discourse, 192
change: de Shazer, St., 31
as co-evolution, 121 developmental programmes, 175
first-order, 15 developmental psychology, 121
second-order, 15 deviance, normalization of, 190
Cochrane Library, 169–170 dialogical communication, 97
co-evolution, 29–30 dialogical conversation, 61, 94, 108,
change as, 121–122 144
communication, reciprocal and dialogicality, 138
egalitarian, 194 dialogical relationship(s), 97
community, new, creating, 112–113 dialogicity, 3, 52, 60, 89, 97–103, 152,
concrete language, 64, 162 189, 190–193
connectedness, 4, 34, 138 basic elements of, 95–97
INDEX 207
by real people, 193 democracy, 192

dialogism, 3, 6–8, 13, 15, 17, 31, 70, expert–patient, 191
77–80, 84, 97, 132, 182, 193 family therapy, 26
basic principle of, 25 monological, 3, 108
vs dialogue, 96 normalcy, 191
dialogue(s) (passim): treatment-model, 191
as act of love, 105 dominance:
Anticipation: see Anticipation interactional, 140–141, 144, 146
Dialogue(s) quantitative, 141–142, 144, 146
and art of responding, 131–148 semantic, 142, 144, 146
asymmetric relations as condition or topical, 141
for, 187 topical, 141, 144, 146
as communication between domination, 107, 187, 193–194
consciousnesses, 105 Doolan, M., 133
as condition for emergence of Dostoyevsky, F., 100
ideas, 108 drug abuse, 54, 60, 120
content of, 106 duration of untreated psychosis
vs dialogism, 96 (DUP), 151, 153, 157, 164
different positions and
asymmetry as precondition Edwards, J., 153
for, 138 Eisler, I., 133
facilitating/facilitators, 63–64, emotions, “contagion” of, 45
68–73 empowering/empowerment, 28, 43,
need for detachment, 68 65–66, 73, 117, 165
healing, 192 and power, 187–194
vs monologue, 107–109 Eräsaari, R., 191
monological, 141 Eriksson, E., 14, 45, 48
vs dialogical, 141 evidence-based research, 9, 167,
multi-stakeholder, 190 169–170, 176, 179–180
mutual, 6 expertise, 190–191
Open: see Open Dialogue(s) expert–patient discourse, 191
“polyphonic”, 3 explanatory research, 172
productive vs poor, 131 F
in psychotic crisis, 131–132 facilitator(s), 17–18, 94, 116–117, 122,
reflective, 118, 137–138 183, 192
as thinking together, 3, 7, 72, 85, in Anticipation Dialogues, 65–89,
97, 103, 106, 117 93
“unfinalizability” of, 108 family(ies):
of violence, 142–146 multi-agency, 26
good-outcome case, 142–143 multi-problem, 26
poor-outcome case, 144–146 therapy, 21–22, 26, 132, 136, 150,
differentiation, 27, 32 192
digitalization, 32 discourse, 26
Dingemans, P., 153, 164 systemic, 22, 25, 132, 136
discourse (passim): Finnish national multi-centre API
authoritative, 3 project, 150
208 INDEX
Finnish Psychiatric Association, 168 hallucination(s), 54, 64, 120, 132–133,

first-order change, 15 142, 146–148
Fisch, R., 15 hallucinatory signs, 132
Flaum, M., 156 handling things vs shared
flexibility, 29–30, 44, 56, 121, 152, 192 experience, 105–107
flexible adaptation, 52, 56–57 case example, 107
follow-up interview(s)/meeting(s), Harrigan, S., 153
87, 93, 117, 152, 159, 161 Harris, J., 177
Forsberg, G., 40 healing experience, dialogue as, 110,
Foucault, M., 43, 187–188, 190, 194 192
Freire, P., 96 hermeneutic philosophy, 39
Friis, S., 57 heteroglot principle of language, 90,
Frith, C., 165 92, 100–101
“future questions”, 31, 66, 72, 84 Hoffman, L., 132
see also “recalling the future” Holma, J., 61, 132
Holquist, M., 108
Gadamer, H.-G., 194 hopelessness, 40–41, 105, 118–119
GAF (Global Assessment of hospitalization vs treatment
Function Scale), 152, 157, processes in social networks,
164 154–155
Galperin, P. Y., 31, 45 Hultkranz-Jeppson, A., 40
gender, 37, 47 I
attitudes towards, 37 identification(s), 44–46
order, 47 understanding through, 45
negotiated, 37 Imber-Black, E., 16, 26
Gergen, K., 61 Imber-Coppersmith, E., 16, 26
Gersons, B., 153 impasse situations, 23, 93
Gibbons, M., 177 implementation of research model,
Giddens, A., 31–32 173–174
Gilbert, P., 177 individualization, 5, 27, 32
Global Assessment of Function Scale inner speech/dialogue(s), 1–2, 17,
(GAF), 152, 157, 164 19, 72, 79, 83, 109–111, 125,
globalization, 32 146
Goolishian, H., 54 institutionalization, 66, 154
governance, 43, 167, 175, 190 Integrated Treatment of Acute
governance culture, 174–176 Psychosis (API), 150–156,
Greeks, ancient, 95 164
Guidelines for the Valid Treatment Finnish national project, 150
of Schizophrenia, 168 interacting systems, 44
guiding centre, 97, 107–109 interactional dominance, 140–141,
Gupta, S., 156 144, 146
Gustavsson, L., 141 interaction pattern(s), 7, 30, 44–47
H interviews, 69, 73, 77, 85, 94, 139
Haarakangas, K., 52, 61, 131, 138, 141 follow-up, 117, 152
Habermas, J., 31–32, 194 I–Other relationship, asymmetry
Halliday-Boykins, C., 180–181 of, 96
INDEX 209
Isaacs, W., 70, 95, 97, 103, 106 symbolic, 141–142, 144
isomorphic patterns, 44, 45 in good outcomes, 141–142
isomorphic processes, 32, 35, 46 meanings in, 138
J Lapland, Western, 8, 150–151, 164,
Jackson, C., 58 183, 192
Jackson, H., 153 Larsen, T. K., 57, 164
Jeste, P., 177 Lash, S., 32, 191, 194
Johannesen, J., 164 late-modern dilemmas, 27
Johnson, A., 165 late-modern theories, 5, 27–28, 32
Johnstone, E., 165 Latour, B., 4, 172
Juvonen, P., 141 Lehman, A., 168
Jyväskylä, University of, Finland, Lehtinen, K., 61, 150
150 Lenior, M., 153, 164
Leont’ev, A. N., 47
Kalla, O., 164 Leucht, S., 171
Kamya, H., 107 Levinas, E., 96, 103
Keränen, J., 51–52, 56 Linell, P., 140–141
Keropudas Psychiatric Hospital, Linszen, D., 153, 164
Tornio, Finland, 21–22, 25, liquid modern theories, 32
149–154, 164, 166, 182 Loebel, A., 153
Kissling, W., 171 Luckman, T., 141
Klefbeck, J., 40 Luhmann, N., 32
Kneifel, A., 32, 44 Lyotard, J.-F., 32
knowledge, valid, 9, 177
Koffert, T., 150 Macmillan, F., 165
Kokko, R.-L., 40, 42, 112 Marklund, K., 40
Konno, N., 183 Markova, I., 96
L Maturana, H., 44
language(s): McAdams, L. A., 177
area, family’s, team’s response on, McGorry, P., 153
138–140 meaning, indicative versus
concrete, 64, 162 symbolic, 141
everyday, 94, 113–116, 118 mediating, 190–191
case illustration, 114–115 meeting(s):
heteroglot principle of, 100 comprehensive quality of:
indicative, 139, 141–142, 146–147 case illustration, 123–125
joint, 63, 91, 106, 112, 119, 121, 123, instructions for those conducting,
147 126–128
generating, 106 multilateral, 7, 46
new, space for, 137–140 Melle, I., 57
multiplicity of, 90 memory, bodily, 119
shared, 54, 63, 94, 109, 111, 113, Mesterton, A., 163–164
117, 191 meta-communication processes, 35
creating, 109–112 Mihalopoulos, C., 153
social, 90–91, 97–103 Miller, P., 190
origin of, 109 “miracle question”, 31
210 INDEX
monological discourse, 3, 108 multilateral, 17

monologism, 47, 97 multi-stakeholder, 3
monologue, 107–108, 110 personal, 1–6, 21, 39–40, 55, 67–68,
as “passivating the other”, 107 74, 91, 187, 189–190
MST (multi-system treatment), as resource, 188
180–182 dialogues at boundaries
multi-agency: between and within, 13–32
client, 26 private, family’s, 30, 78
families, 26 professional, 1–3, 18–19, 39, 66, 86,
team, 58 189, 192
multifaceted problem analysis, 29 dialogues at boundaries
multifaceted reflection, 44 between and within, 13–32
multilateral networks, 17 properties common to, 4
multiplicity: social: see social network(s)
of languages, 90 society, 4
of voices, 63, 99, 109 therapy, 4, 40
multi-problem: networking, 2–5, 13, 34, 174
families, 26 neuroleptics/neuroleptic
situation(s), 16–17, 26, 57 medication, 9, 59, 151–168,
multi-professional system, 5, 28 178
multi-stakeholder: in treatment of psychotic
dialogues/meetings, 34–35, 47, episodes, 151
190 Nightingale, A., 95
networks, 3 non-affective psychosis, 150, 164
processes, 7 Nonaka, I., 183–184
multi-system treatment (MST), Norcross, J. C., 185
180–182 normalcy discourse, 191
multi-voicedness, 18, 94 Nowotny, H., 177–179, 184
mutual dialogue, 6 O
ODAP, see Open Dialogues
narrative(s) (passim): Approach in Acute
new, 133 Psychosis
of restitution and reparation, open-care teams, 13
133 Open Dialogue(s) (passim):
National Research and and Anticipation Dialogue(s):
Development Centre similarities and differences,
for Welfare and Health 89–103
(STAKES), 14, 68, 150 as crisis intervention, 51–64
network(s) (passim): and first-episode psychosis,
client’s, 19 150–162
dialogues between, 3 good vs poor outcomes, 156–
meetings, 7, 17, 65–66, 101, 103, 158
113, 120–121, 149 principles, 152, 161, 191–192
dialogical, 149–166 treatment, 52–61, 151, 160–161,
frustrating, 33–48 191
training and research, 149–150 outcomes, 155–156, 168
INDEX 211
Open Dialogues Approach in Acute 160, 162–164

Psychosis (ODAP), 150–156, first-episode, and Open
163–165, 168, 170, 172 Dialogues, 150–162
Opjordsmoen, S., 152, 164 non-affective, 150, 164
outcome(s), 163 psychosocial treatment of, 56
practices behind, 158–162 recovery from, 156
case studies, 158–162 schizo-affective, 157
schizophreniform, 157
parallel processes, 32 untreated, 151, 162, 164
paranoia, 134–135 duration of (DUP), 151, 153,
Patterson, D., 105 157, 164
“pedagogy of the oppressed”, 96 psychosocial coping, 9
Penn, P., 31 psychosocial work, networking in,
personal networks: see network(s), boundary-crossing, 4–7
personal psychotherapeutic dialogue, 61
Plato, 95 psychotic comments as response to
polyphonic novel, 100 ongoing dialogue, 132–136
polyphonic understanding, 92 psychotic delusions, 159
polyphony/polyphonic dialogue, psychotic episode(s)/experience(s)/
18, 91–93, 97–103, 106, 108, crisis, 23, 58, 63, 112, 127,
115–116, 125, 138, 191 133–134, 136, 140, 149, 153,
and emergence of understanding, 157, 159
3 acute, 58
of inner dialogue, 79–80 case example, 134–136
and professional hierarchy, 100 generating dialogue in, 131–132
techniques for increasing, 63–64 immediate response to, 52–54,
and tolerance of uncertaintty, 152–154
63–64, 93 and social network, 157
PORT report, United States, 168 psychotic patient(s), 51, 53, 59, 112,
postmodern theories, 5, 32 150–151, 163–165
post-traditional theories, 32 first-episode, 150, 163, 165
power: psychotic speech, 118, 132, 134
critique of from within, 191–193
four types of discourse of quantitative dominance, 142, 144,
exercising, 191 146
practitioners, worries of, 67
present, the, in near future, 75–78 Räkköläinen, V., 61, 150
problem analysis, multifaceted, 29 “reality-orientating”, 127, 133, 147
problem-defining system, 54 inadvisability of, 133
professional networks: “recalling the future”, 8, 19, 66,
see network(s), professional 73–81, 86, 91, 116
psychiatric practice, network case illustration: Muslim family,
perspective in, 51 74
psychological continuity, 52, 58–59, present in near future, case
152, 160–161, 192 illustration, 76–77
psychosis, 24, 104, 133, 150–151, 153, see also “future questions”
212 INDEX
reflecting team(s), 17, 137 creating, 109–112

reflection, multifaceted, 44 Shotter, J., 61, 94
reflective dialogue/conversation, “silo system”, 5, 42, 183
60–61, 118, 137–138, 144, social-action research, 150
159 social control, 43
reflexive theories, 32 social language(s), 90–91, 97–103
reparation, 133 social network(s), 1–2, 62, 89, 95,
research, 7 101, 112, 116, 126, 132, 157,
in dialogical network meeting(s), 159–160, 166
149–150 client’s/patient’s, 51, 58, 122, 162
evidence-based, 9, 167, 169–170, collaboration within, 51
176, 179–180 inclusion of, 52, 54–56, 152, 192
explanatory, 172 concept, 4–5
and generalizing practices, perspective, 55
167–194 treatment processes in, vs
linearity-based, 179 hospitalizations, 154–155
project, 51 social order, negotiated, 36–37
psychiatric, experimental, 167 social speech, 109, 110
social-action, 150 Socrates, 95
resource-oriented therapies, 31 Speck, R., 4, 40, 95
responding immediately, 52 speech:
responsibility, 40, 52, 57–58, 79, 90, inner, 109, 110
96, 152, 192 psychotic, 118, 132, 134
shared, 42 social, 109–110
restitution, 133 speechlessness, 161
Rosaldo, R., 47 “spiral process”, 40–41, 95
Rose, N., 190 Stanton, A., 32, 44
Sachs, D., 32, 44 Staples, E., 133
schizo-affective psychosis, 157 Stern, S., 133
schizophrenia, 114–115, 153–155, Strauss, A., 36
157, 162–164, 166–167, 170 Strauss, J., 152
two types of, case illustration, Strauss–Carpenter Rating Scale, 152,
114–115 156
schizophreniform psychosis, 157 subjugation, 43, 189, 191
Schön, D., 27, 184 suicide, 158
Schwartzman, H., 32, 44 risk of, 19
Schwartz, M., 32, 44 Surakka, V., 45
Scott, P., 177, 191 Svedberg, B., 163–164
second-order change, 15 system(s) of boundary, 25, 28–30, 44,
semantic dominance, 142, 144, 146 92, 121
Shapiro, S., 32, 44 Syvälahti, E., 150
shared experience vs handling Szmukler, G., 133
things, 105–107 T
case example, 107 therapeutic conversation, 60
shared language, 54, 63, 94, 109, 111, thinking together, dialogue as, 3, 7,
113, 117, 191 72, 85, 97, 103, 106, 117
INDEX 213
tolerance of uncertainty, 63, 93–94, and strong contextualization,

117, 152 177
increasing, 63
topical dominance, 141, 144 Vadén, T., 16
Toyama, R., 183 Valid Treatment, 9
training for dialogical network Van der Kolk, B., 119
meeting(s), 149–150 Varela, F., 44
transferability of knowledge, 169, violence:
173 dialogue of, 142–146
transference, 32 good-outcome case, 142–143
trauma, 104, 133 poor-outcome case, 144–146
treatment: domestic, 104
meeting, 52, 126, 132, 140, 142, voices, multiplicity of, 63, 99, 103,
144, 161 109, 120
functions of, 61 Voloshinov, V., 61, 98–99, 132
as joint wondering, 61–64 Vygotsky, L., 109–110, 119, 121, 138,
model discourse, 191 141
outcome, 163 Vygotskyan activity theory, 31
Trimble, D., 107, 133
Tuori, T., 154 Watzlawick, P., 15
Turku University Hospital, Finland, Weakland, J., 15
21–22, 150 Weckroth, J., 16
Wertsch, J., 99, 138, 141
uncertainty, tolerating, 3, 5, 52, 93– Western Lapland, Finland, 8,
94, 121, 152, 155, 161, 179, 150–151, 164, 183, 192
192
increasing, 63–64, 116–120 Yung, A., 153
principle of, 59–60
and psychological resources, “zone of proximal development”,
116–120 110, 121

Dialogical Meetings in Social Networks

Uploaded by

Copyright:

Available Formats

You might also like

Dialogical Meetings in Social Networks

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dialogical Meetings in Social Networks

Uploaded by

Copyright:

Available Formats

DIALOGICAL MEETINGS

Jaakko Seikkula & Tom Erik Arnkil

Systemic Thinking and Practice Series

SERIES EDITORS ’ FOREWORD vii

Although the book describes their model in detail, many

JAAKKO SEIKKULA is a clinical psychologist and family therapy

TOM ERIK ARNKIL is a research professor at STAKES (National

on child- and family-related questions, where social work, therapy,

Placing each method in its own context, there is a big difference

The AD method, on the other hand, more nearly resembles the

attempts to control others. There will be many who, unsurpris-

sources, inventing and trying out novel ways—and, above all, we

This book focuses on encounters between personal and profes-

Networking in psychosocial work

Networks have been a topic in psychotherapy and social work

or local communities to the individual is also changing, and so are

froing, attempts to tell others what to do—while those seeking help

Structure and content of the book

In chapter 8, the treatment outcomes of Open Dialogues are

have written a joint work that integrates each of our particular

Dialogues at the boundaries

No common deﬁnitions of the problem, after all

Anticipation Dialogues were developed in complex multi-profes-

three social welfare ofﬁces and an clinic for substance abusers.

different or unusual came from the work of two Finnish psycholo-

ized as three questions for the team members to consider when

1. What would happen if you did nothing?

These questions were originally for team members within an agen-

Everything has to be changed when clients are present

The professionals had been encouraged to bring along clients.

everybody will be interviewed in turn; each will be asked to think

Joint planning turned out to be best treatment

The origins of Open Dialogues were cultivated in the early 1980s

to inpatient care—the team took charge of planning the treatment

were no longer receivers of a family therapeutic intervention but

Case illustration: “I will kill her . . .”

period of discussion on other themes, the doctor proposed to

After her comments, team members started to discuss with

Repeated experiences of this kind offered a conclusive insight: in

named the “system of boundary” (Seikkula & Sutela, 1990). Identi-

From problems to dialogues

When Tom started inquiring into “multi-problem situations” in

of professions, disciplines, and agencies is not the making of ordi-

In Jaakko’s conceptualization, boundary-crossing took place

team’s ability to adapt to each situation. If the family’s accustomed

As already said, we had not begun our cooperation at the time

5. According to the German sociologist Niklas Luhmann (1989), differen-

Frustrating network meetings

• have you attended network meetings where, at some manifest

if from a bird’s-eye view, and you somehow can’t express your

Four assumptions about the problematic

Sometimes network meetings are identiﬁed with networking: co-

multilateral cooperation is more complex than dyadic discussions,

Actors deﬁne mutual relations

It is not exceptional to experience meetings where psychosocial

When there are professionals present who are connected to the

“chemistry”, but neither are the relations deﬁned directly between

There are as many problems as there are actors

When professionals meet, the task is often to try to deﬁne the