SUMMER 2003 VOLUME 42  NUMBER 2
Postmodern Society and Social Networks: Open
and Anticipation Dialogues in Network Meetings
JAAKKO SEIKKULA, Ph.D.
TOM ERIK ARNKIL, Ph.D.
ESA ERIKSSON, M.A.§
Network therapy flourished in the U.S.
during the 1970s, but has since dwindled
there and begun to find new applications
in Europe, especially in the Nordic countries.
State social and healthcare systems,
in developing deep vertical expertise,
seems to build up a need for complementary
horizontal expertise. The latest theories
of sociology are used to analyze the
need for networking, with the focus on language
and dialogue as specific form. Two
approaches developed in crisis service for
psychotic patients (Open Dialogue) and in
consultation for stuck cases in social care
(Anticipation Dialogues), are dealt with.
What becomes essential seems no longer to
be the therapeutic method itself but the
ability to see the polyphonic nature of clients'
reality. In this respect, language­and
dialogue as a specific form of being in
language­as the focus of treatment, makes
the practical forms of different approaches
secondary.
Fam Proc 42:185­203, 2003
The goal of the present article is to
analyze the emergence of dialogism
in psychosocial work and discuss its consequences
for expertise. The authors also
intend to promote integration of psychiatric
and social care systems that are based
on network therapy and to develop an
 Professor, University of Tromso and University
of Jyva¨skyla
 Research Professor, Stakes,1
Helsinki
§ Development Manager, Stakes, Helsinki
Correspondence concerning this article should be
addressed to Jaakko Seikkula, University of Jyva¨skyla¨,
Department of Psychology, P.O. Box 35, Jyva¨skyla¨,
Finland. Email: seikkula@psyka.jyu.fi. Fax:
358-14-26 02 841, Tel: 358-14-26 02 842.
1
Stakes National Research and Development
Center for Welfare and Health
185
Family Process, Vol. 42, No. 2, 2003  FPI, Inc.
integrative view of the nature of dialogues
in different settings. The authors come from
two traditions: developing a psychiatric
crisis intervention system especially for
psychotic patients (author JS) and developing
network orientation in social care
and psychosocial work across the boundaries
of multiagency systems (authors
TEA and EE). In both contexts, methods
based on dialogues appear to be very
promising. Applying recent sociological
theories in combination with new theories
of psychotherapy, the article analyzes developments
at the patient/client-professional
interface in a wider sociopolitical
context. Such developments challenge implementation
of first modern or top-down
expertise. Dialogism is viewed as mainly
professional expertise called for by post-
modern2
development.
Professions performing various types of
tasks of psychosocial work, in the private
and public sectors, are elements of societal
modernization. Tasks of social support
and control traditionally covered by
communities and personal networks have
one after the other been taken over by
experts (Hirsch, 1985). He refers to this as
socialization of support and control. Specialization
has been seen as central to
societal modernization (Luhmann, 1989).
Systems such as law, economy, science,
and politics differentiate and form subsystems.
According to Foucault (1977), all
professions with the prefix psycho- or socio-
(psychologists and psychiatrists, sociologists
and social workers, etc.) are
based on developing a normalizing gaze
that detects deviations. Normality is the
area between extremes. Normalizing
power is productive: new objects for
study, new knowledge, and new professions
emerge, forming a system of professions
(Abbott, 1988). According to Bourdieu
(1994), newcomer professions try to
conquer fields by demonstrating that the
controllers of these fields are outdated.
Core development in societal modernization
is individualization (Beck, 1992).
Traditional bonds (kin, clan, community)
tear apart as industrial and postindustrial
society emerges. The feasibility of
living a more or less detached life is aided
by the modern state. Within this process,
the normalizing gaze of psychosocial professionals
detects obstacles to normal individuation.
The attempt to foster healthy
individuation in an enmeshed or too
loosely knit family can serve as an example.
If healthy individuation is threatened,
the task of the expert system, e.g.,
family therapy, is to modify family interaction
in order to fulfill its task of producing
independent individuals in society.
Paradoxically, the expert system, while
refining competent ways of helping people,
creates both intended and unintended consequences
in problem solving (Giddens,
1979). In the most serious cases of psychosocial
help, individuals and families are
surrounded by a multitude of experts. Multiproblem
families become multiagency
families (Imber-Black, 1988). Multiprofessional
muddle are late-modern phenomena.
Growing awareness of these unintended
consequences has led to attempts at finding
new ways of helping. These can be defined
as a search for second expertise in the second
or late phase of societal modernization
(Giddens, 1990).
Export and import of network therapies
Although compartmentalization of care
was not, at least in Finland, the consequence
of a detailed master plan, the sectored
mode of producing healthcare and
2
Although sociologists call current developments
by different names, i.e., Postmodern by Lyotard,
1979; Post-traditional by Habermas, 1984; Reflexive
or late-modern by Beck, Giddens, & Lash, 1994; they
all seem to agree that something fundamental is
occurring, e.g., through globalization, individualization,
digitalization, environmental consequences,
and full employment crisis. Here the term postmodern
has been chosen to represent all these ideas.
186 / FAMILY PROCESS
other services did not emerge at random.
The rationalization of the system world
(Habermas, 1984) was, in fact, an intensive
phase in modernization, challenged
only recently. First, the production of consumer
goods, and even later on, healthcare
systems, including essential elements
of psychosocial work, were turbulent
areas of reshaping production along
cost-effective lines. Crafted means of
healthcare were replaced by large-scale
production of health services in huge hospitals
with streamlined specialization--
creating so-called health factories. Gramsci
(1971) called it Fordism. A silo view of
planning, management, and production
has only recently been challenged by
ideas of flexible networking (Castells,
2000). The basic assumption that division
into clear-cut portions can safeguard the
totality has proven to be a gross simplification
in dealing with complexity, such as
social, biological, and even technical sys-
tems.
In attempts to restructure professional
psychosocial help, networking became influential
in the therapy world (Trimble,
Kliman, Villapiano, & Beckett, 1984).
Networks and networking are also catchwords
in postmodern society. Network
therapy as such was not a postmodern
phenomenon, but interesting ways of
working with serious crises have been
embedded in it. These included in their
origins ideas of multiple realities and
polyphonic life, which are enhanced
through generating dialogue. As they
landed in the Old World, they started to
change, revealing some interesting postmodern
features, especially dialogism as
a central element.
Network therapy originated in the U.S.
in the mid-1960s, especially through the
work of Carolyn Attneave and Ross Speck
(Speck & Attneave, 1973), with implications
for both social care and adult and
child psychiatry. Originally, network
therapy meant mobilizing full-scale network
meetings around crisis situations
(Trimble et al., 1984) and undergoing profound
emotional experiences in these
meetings. Network therapy flourished
during the 1970s when therapists developed
a wider repertoire for collaborating
in building, supporting, and restoring caring
personal communities (Trimble, 1996-
1997). It was oriented toward analyzing
the structure of the invisible web of ties in
which individual and family life are embedded
and toward mobilizing networks
of people relevant to meetings in crisis
situations. Anthropological and other social
science literature provided tools and
sources of inspiration. After this period,
the practice of network therapy began to
decrease in the U.S. (Attneave, 1990).
One problem was the orientation toward
mobilizing full-scale social networks,
which could easily involve meetings of
30-50 people.
Economic reasons along with rigid models
for organizing social and healthcare
hindered the development of network
therapy. Some network therapists became
critical of the method, particularly
concerning the strong emotions generated
during and after the meetings (Garrison,
1981). Therapists experienced difficulty
in learning to take charge of network
meetings. The helpful ideas of system theory
in analyzing and mobilizing social
networks became available only after the
early 1980s, as Attneave (1990) noted, referring
especially to the concept of psychosocial
network created by Pattison and
his team (Pattison & Pattison, 1981). The
early network therapists were great enthusiasts
with abundant personal charisma
that was difficult or impossible for
novices to replicate. During the 1980s and
1990s, some family therapists worked
with multiprofessional systems (ImberBlack,
1988). Although traditional social
network interventions have decreased,
the central ideas of network bonds around
clients have been used and developed, as
SEIKKULA, ARNKIL, & ERIKSSON / 187
Fam. Proc., Vol. 42, Summer, 2003
Sluzki (2000) illustrates in his fine analyses
of therapy with elderly patients.
Many of the language-orientated approaches,
such as collaborative language
therapy (Anderson, 1997; Anderson &
Goolishian, 1988), different forms of narrative
(Smith & Nylund, 1997) and solution-focused
therapies (deShazer, 1991),
also include the nearest social networks
in the therapy processes.
Network therapy crossed over to Europe,
especially to countries where comprehensive
social and healthcare systems
already existed (Trimble, 1996-1997). The
basic idea that various clusters of people
are important factors in a clienťs life gave
an opportunity to integrate the work of
different authorities in the same crisis situation.
Involved professionals were invited
to participate in joint meetings. Professional
work began to take place collaboratively
instead of each expert
working separately with the client according
to the principles of his/her profession
in his/her consultation room. Multiple resources
could be mobilized in tackling the
crisis. Such development occurred especially
in the Nordic countries, starting
with work involving immigrant families
and their social problems in the suburbs
of Stockholm, Sweden (Klefbeck, Bergerhed,
Forsberg, et al., 1988).
In the Nordic countries, a natural presupposition
exists for the development of
network ideas. The welfare state has, in
recent decades, been very active in supporting
individualization. In contrast to
Southern Europe, where the larger family
still has a marked emphasis on social support
and control, policies in the Nordic
countries have held a relatively independent
nuclear family as the normal case. In
so doing, they have also produced this
family type. Shaping the family was an
unintended consequence of a mixture of
markets and policies. In Finland, which
underwent a severe labor shortage in the
1960s, mothers with small children were
recruited into the labor force. A doubleincome
nuclear family emerged. At
present, the majority of mothers with
small children work full-time outside the
home. Along with housing, taxation, and
other allowances and policies that support
the family, there are professional
support systems for children (e.g., daycare,
education, etc.), for the elderly (senior
citizen care), and for other groups
(the disabled, etc.). Lehtonen (1986) argues
that the modern double-income nuclear
family could not remain functional
without societal scaffolding. The family
has a multitude of systems, agencies, and
professions supporting it, and psychosocial
support if needed. Expert systems are
replacing many traditional sources of support
and control. The Nordic countries are
at the cutting edge of the process that
Beck (1992) calls the second phase of mod-
ernization.
We see three linked causes for the interest
toward network therapy in the Nordic
countries. First, as the takeover of
support and control proceeds, psychosocial
expert systems are, in many cases,
running out of problem-solving resources.
Interestingly, many professionals turn toward
clients' personal networks. Second,
the psychosocial expert system is getting
into complex self-imposed muddles, where
compartmentalizing problems to fit streamlined
specialization seems to be part of the
problem. Finally, the Nordic psychosocial
systems are accessible; they are mainly
public and free of cost or very affordable.
Expert systems at turning point?
Finnish psychosocial systems must
work with all cases, including the most
difficult. This increases the professionals'
interest in methods that promise ways to
get out of complex situations. They seek
cooperation over boundaries. Inviting
people from the clienťs personal network
to participate in problem solving meant a
change in the paradigm where problem
188 / FAMILY PROCESS
solving was seen as something done by
the expert for the individual. The parallel
development of societally-aided individuality
and specialized expertise created a
client-expert dyad expected to be relatively
omnipotent. However, additional
resources are often called for.
The routine method is to refer the client.
In most cases this is unproblematic.
Nevertheless, as Imber-Black (1988) points
out, there is an inherent trap. If clients
are referred with no idea of supportable
resources, with the goal of getting rid of
the hopeless case, they tend to become
seen as multiproblem clients or families.
Typically, they will have knocked at almost
every door in the professional system.
Network approaches have attracted
the attention of precisely those professionals
who were at the end of the referral
chains, with no one to pass the problem on
to. Since the public service system cannot
be selective, the number of professionals
finding themselves stalled or helpless
with complex situations is quite high in
the Nordic countries.
In our experience, professionals have
been more ready to accept the idea of networking
among themselves rather than
inviting people from the clients' personal
networks. The latter course poses a
greater challenge to the conventional expertise.
However, the urgent need for additional
problem-solving resources encourages
experts to take the step. When
the professionals became brave enough to
invite laymen, they were delighted to find
new resources for problem solving in crisis
situations. This meant, of course, a
considerable challenge for the concept of
expertise. It was no longer the professional
alone who held the keys to solutions.
New competencies were required
from professionals: instead of prescriptive
expertise, mobilization skills were called
for. The first modern society produced the
sectored service systems and specialized
expertise where the expert knew better.
In a network setting, the professional is
faced with a great deal of unpredictability.
In fact, unforeseen solutions by combinations
of people working together are
the goal. Top-down expertise begins to
give way to networking competencies.
In Beck, Giddens, and Lash's reasoning
(1994), high tolerance of uncertainty is
central to the postmodern expertise. In
the conventional expertise, professionals
were expected to have­and perhaps
thought they had­a high level of control of
consequences. In our view, the postmodern
expertise cannot be a characteristic of
an isolated professional. Rather, it is embedded
in a network of professionals and
other stakeholders. In complex, bewildering
situations­surrounding multi-problem
clients or families for example­expecting
ways out through streamlining
specialization even further appears far
less promising than joining forces, reframing
the picture, and tolerating uncer-
tainty.
All of this has also meant changing the
traditional idea of network therapy. Handling
these types of multiagency quandaries
has led us from seeing our position as
applying social network intervention ("doing
network therapy") to focusing on the
language created in the network meetings.
It is no longer a question of having a
network therapy method, but rather a
network-centered approach in most serious
psychiatric and social problems. The
clienťs network is not an object for interventions,
but rather an irreplaceable resource
for proceeding. In the beginning,
we were disposed to see network therapy
as one method of expertise. At the moment,
we are emphasizing the dialogue
between the participants in the meeting
and, of course, the inner dialogue of each
participant. The two approaches described
in what follows, open dialogue
(OD) and anticipation dialogue (AD), deal
with the uncertainty inherent in most serious
situations in psychiatry and social
SEIKKULA, ARNKIL, & ERIKSSON / 189
Fam. Proc., Vol. 42, Summer, 2003
care. Both approaches aim for a dialogue.
The pragmatics, however, are different.
OD has mostly been applied in severe psychiatric
crisis as a form for organizing
both the treatment system and the dialogues
themselves in the meetings. AD in
its origins is a form of consultation in
complex processes in social care. Giving a
general description of both is an attempt
to illustrate how, on the surface, very different
types of approaches may include
the central elements of postmodern exper-
tise.
OPEN DIALOGUES IN PSYCHOTIC
PROBLEMS
The roots of the OD approach lie in
Finnish western Lapland, in a small province
with 72,000 inhabitants. In the local
psychiatric hospital, where one of the authors
(JS) worked as a psychologist, family-
and network-oriented treatment was
the goal. In 1984, the traditional manner
of admitting patients was challenged. The
team started to organize open meetings,
referred to as treatment meetings, to analyze
the problem and prepare the treatment
plan after a patient was admitted to
the ward. The patient participated from
the outset. Staff members stopped having
their own separate gatherings. At the
same time, instead of inviting families
into family therapy after the team had
defined the problem, the team started to
invite families whenever a family member
was hospitalized. Gradually, it became
evident that this change in working
style caused a remarkable shift in the position
of the family and the patient. Families
were no longer objects for staffplanned
treatment; instead they became
active participants in joint processes. In
many impasse situations encountered in
the treatment, the team noticed that the
only way forward was to change the
team's own activity in the actual situation
(Seikkula, Aaltonen, Alakare, et al.,
1995). The team began to rethink the
structural paradigm on the principle that
it is the team's task to intervene, which in
turn effects change in the family (Boscolo
& Bertrando, 1993; Selvini-Palazzoli,
Boscolo, Cecchin, & Prata, 1978).
Opening doors for families to participate
in analyzing the problem, preparing
a treatment plan, and participating in
treatment meetings throughout the entire
treatment sequence were the first
steps in seeing all the problems as problems
in the actual social situation of the
patient. In such situations, many other
aspects and other parties in the social
network of the patient also proved to be
important. In family therapy based on the
structural paradigm, the nuclear family is
the basic unit, since symptoms are seen
as functions of the family system, be it the
nuclear family or the extended family (Kemenoff,
Jachimczyk, & Fussner, 1999). In
the new approach, it became natural to
invite all the important participants in
the patient s social network in order to
increase coping resources and to open up
new constructive perspectives.
In building up the family- and networkcentered
psychiatric system, the next step
was to realize the importance of holding
the first treatment meetings as soon as
possible after the crisis had occurred.
This led to a rapid decrease in the need to
hospitalize (Kera¨nen, 1992; Seikkula,
1991, 1994). It became necessary to organize
a mobile crisis intervention team in
each psychiatric outpatient clinic in the
province. Currently, all staff members
can be called upon to participate in these
teams according to the particular need.
Regardless of the specific diagnosis, if
there is a crisis situation, the same procedure
is followed in all cases. If it is a
question of possible hospital treatment,
the crisis clinic in the hospital will arrange
the first meeting, either before the
decision to admit for voluntary admissions,
or during the first day after admission
for involuntary patients. At such a
190 / FAMILY PROCESS
meeting, a tailor-made team consisting of
both outpatient and inpatient staff, is
constituted. The team usually consists of
2 or 3 staff members (e.g., a psychiatrist
from the crisis clinic, a psychologist from
the mental health outpatient clinic for the
area where the patient is living, and a
nurse from the ward). The team 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 treatment period is expected
to be.
In the treatment meeting, all the important
members of the social network,
together with the patient, gather to discuss
all the issues associated with the
actual problem. All management plans
and decisions are also made with everyone
present. On the whole, the focus is on
strengthening the adult sides of the patient
instead of regressive behavior
(Alanen, 1997). The task of the dialogue is
to construct a new language for the difficult
experiences of the patient and those
nearest him/her--experiences that do not
yet have words. In analyzing this OD approach,
Gergen & McNamee (2000) noted
that this could be seen as transformative
dialogue instead of disordering discourse.
Although OD is not a diagnosis-specific
approach for psychotic problems, treatment
of the psychotic crisis best illustrates
the central elements. In organizing
open meetings, our understanding of the
nature of psychosis began to change. In
the next sequence, an illustration of psychosis
from the social constructionist and
dialogical point of view is given.
Open dialogue and psychosis
Psychosis can be seen as one way to
deal with experiences so terrifying that
they cannot be expressed other than
through the language of hallucinations
and delusions. For example, most female
psychotic patients have experienced physical
or sexual abuse either as a child or as
an adult (Goodman, Rosenberg, Mueser,
& Drake, 1997). In clinical situations,
these traumatic experiences are often
present in the hallucinations or delusions
about which the patients are speaking
(Karon, 1999).
Case: Breaking windows
A female patient had been hospitalized
for more than 2 weeks, and a treatment
meeting was organized to prepare for her
discharge. Her husband, son, doctor, ward
team, and a two-person team from the
psychiatric outpatient clinic participated
in this meeting. The patient was asked to
describe what happened when she was
admitted to the hospital. She answered by
describing how one afternoon she was at
home with her son who had suddenly
asked if there was someone in the garden.
She was frightened, believing someone
was there, although she could not see
anyone. She was convinced it was the
man with whom she had been living for a
2-year period, 16 years ago. The following
day, when her husband returned home
from his work tour and drove into the
yard, she started to fear that he was under
the influence of drugs and was going
to kill her. She locked all the doors so that
her husband could not come in. He grew
irritated and started to yell while on the
front steps. She became terrified and in
the end, broke two large living room windows
by throwing chairs through them.
After this attack she was hospitalized.
The team became interested in her
former husband and asked her to tell
them about her relationship with him.
She said that it was difficult for her to
speak about it, never having done so before.
The man, she said, was a narcotics
addict who, when under the influence,
would always assault and beat her
heavily. She used to stay home long
enough for her bruises to disappear to be
sure no one knew she was a victim of her
husbanďs violence. After 2 years she
SEIKKULA, ARNKIL, & ERIKSSON / 191
Fam. Proc., Vol. 42, Summer, 2003
managed to divorce him. They had not
met since. She told that one night, 5 years
ago, while she was alone at home, the
telephone rang. She answered and found
that the man was calling to ask how her
life was. She became terrified, began to
tremble, and almost ran out of words. After
the conversation, she remained terrified
for a long time and had her first psychotic
break down two months later.
Since this was the first time that she
had been verbally able to express these
terrifying memories, the team began to
ask about concrete descriptions of how the
attacks had happened. For example, they
asked whether her husband had hit her
with his fist or with an open hand. The
intention was to have plenty of words
available for constructing a story of the
traumatic memory. In a stress situation,
difficult and terrifying experiences in
one's life may be actualized and can be
relived (Penn, 1998; van der Kolk, &
Fisher, 1995). The person can begin to
search for a way to express these experiences
in the form of a metaphor. As in the
case described, where the patient had a
delusion that her husband was under the
influence of drugs and was coming to kill
her, this was something that was not true
at the moment, but had actually happened
in a previous relationship.
To have an open dialogue with no preplanned
themes or forms for conversations
appears to be important in making
it possible to construct a new language for
describing and reflecting on difficult
events. Whatever the background for psychotic
speech, it is important in the starting
phase of treatment to take it seriously
and not in any way challenge the patienťs
sense of reality during the crisis situation.
Instead, the therapisťs questions
may be as follows: "I do not understand.
How could it be possible for you to control
the thoughts of other people? I have not
experienced that. Could you tell me more
about it?" The questions to other network
members in the meetings could be as follows:
"What do the rest of you think about
this? How do you understand what M is
saying?" In this way, the task is to afford
a variety of voices for the theme under
discussion. If the team manages to generate
a deliberating type of atmosphere, allowing
different, even contradictory
voices to be presented, the network makes
it possible to construct narratives of restitution
or reparation (Stern, Doolan, Stables,
et al., 1999). As Trimble (2000) puts
it, when comparing the dialogical approach
to the ideas of network therapy,
"restoration of trust in soothing interpersonal
emotional regulation makes it possible
to allow others to affect us in dialogical
relationships" (p.15). This may be one
aspect of the process in which the patient
and his/her social network can begin to
construct new words for their problems.
In general, the idea for the team during
the meeting is to allow the network to
take the lead and, by responding to each
utterance in a dialogical way, to promote
the building of new understanding (Bakhtin,
1984; Voloshinov, 1996). Dialogue becomes
both the goal and the specific way
of being in language in the therapy. Instead
of primarily trying to change the
patient (e.g., a rapid removal of the psychotic
symptoms), or the family (e.g., a
new interactional style within the family
system), the main therapeutic efforts occur
in the area between the team (and
other parties) and the family or social network
present. Building up a dialogical
rather than a monological dialogue means
thinking more about how to answer the
utterances produced by the patient and
the family. It means being present in the
actual conversation. In systemic family
therapy, the team employs a tactic, such
as circular questioning, through which a
change in the family system can be initiated.
It is not essential that every utterance
be answered, because the primary
192 / FAMILY PROCESS
focus may be outside the actual theme
under discussion.
In OD, the "tactic" is to build up dialogical
discourse. In the dialogue, new understanding
starts to emerge as a social,
shared phenomenon. The individuals
present at the meeting are speaking
about their most difficult experiences. In
the dialogue, the goal is to capture the
behavior of the patients as one dimension
of their life context, which often means
that even very odd types of behavior start
to seem more normal. One element of this
normalizing discourse is that those aspects
of the patients' hallucinations or delusions
that are mixed in with the real
incidents in their lives are highlighted.
The difficult reality can be shared, and
thus new resources become available.
What first occurs in outer dialogue in the
social domain may thereafter evaporate
into an inner dialogue. Vygotsky speaks
of the zone of proximal development (Vygotsky,
1978) in the child. This idea can
be used to describe the psychotherapeutic
situation as well (Leiman & Stiles, 2001).
This may be one explanation as to why
psychotic patients frequently are able to
participate in the conversation in the first
meetings without psychotic experiences
(Alanen, 1997).
One way to respond is to have reflective
conversation3
(Andersen, 1995) among
the team members--by the very same
professionals who are conducting the interview.
No specific reflective team is
formed, but the team members change
positions in a flexible way, from constructing
questions and comments to having
reflective conversation with other
team members. Sometimes this presupposes
that the team asks permission to do
so: "I wonder if you could wait a moment
so that we could discuss among ourselves
what we are beginning to think. I wish
that you would sit quietly, and listen if
you want to. Afterwards, we will ask for
your comments on our discussion." Usually,
the family and the other part of the
social network listen very carefully to
what the professionals say about their
problems. Reflective discussion has a specific
task. Since the main idea is to construct
treatment plans in these conversations,
everything is transparent. Decisions
concerning hospitalizations, motivation
for medication, and use of individual psychotherapy
are examples of the content.
Discussions are aimed at opening up a
variety of alternatives for decisions. In
the case of a decision for involuntary
treatment, for instance, it seems to be
important that different opinions and
even disagreement over the decision can
be openly expressed and discussed.
Guiding principles in Open Dialogues
Several studies have provided information
on the course and results of treatment
(Kera¨nen, 1992; Seikkula, 1991,
1994). There are also qualitative analyses
of dialogues in treatment meetings
(Haarakangas, 1997; Holma, 1999). The
results are promising. The incidence of
schizophrenia, according to DSM-III-R,
has decreased, since 1985 through 1994,
from 33 to 7/100,000 inhabitants (Aaltonen,
Seikkula, Alakare, et al., 1997). In
an ongoing study of first-episode psychotic
patients, the need for hospitalization
decreased and the use of neuroleptic
medication could be compensated for by
using anxiolytics at the outset, so that
26% of the 80 patients used neuroleptics
during the two-year followup period
(Alakare, 1999; Seikkula, Alakare, Aal-
3
Reflective conversation means making it possible
to shift between outer (talking) and inner (listening)
dialogues. Andersen created a reflective
team in which, at one point in the family interview,
the interviewer, together with the family, could
start to listen to the team member's reflection about
what they had heard during the interview. After
this, the family had an opportunity to give their
comments if they wished.
SEIKKULA, ARNKIL, & ERIKSSON / 193
Fam. Proc., Vol. 42, Summer, 2003
tonen, 1999, 2000). This did not lead to
poorer outcomes: 83% of the patients had
returned to their jobs or studies, or were
job-seeking, and 77% did not have residual
psychotic symptoms. One reason for
quite good prognoses was suggested to be
the fact that the duration of untreated
psychosis was only 3.6 months in Western
Lapland, where the network-centered
system had enabled easy contacts to psychiatric
care and an immediate start of
treatment (Seikkula, Alakare, & Aaltonen,
2001). In many other studies, the
duration of untreated psychosis has been
noted to vary approximately between 12
months (Loebel, Lieberman, Alvir, et al.,
1992) and 3 years (Larsen, Johannessen,
& Opjordsmoen, 1998).
As an outcome of research programs
and psychotherapy training, seven main
principles of OD have been established.
1. Immediate help. The units arrange the
first meeting within 24 hours of the
first contact, made by the patient, a
relative, or a referral agency. In addition,
a 24-hour crisis service is set up.
2. Social-network perspective. The patients,
their families, and other key
members of their social network are
always invited to the first meetings to
mobilize support for the patient and
family. The other key members may be
other authorities, including employment
agencies and health insurance
agencies in support of vocational rehabilitation,
colleagues or the head of the
patienťs workplace, neighbors, or
friends.
3. Flexibility and mobility. These are
guaranteed by adapting the treatment
response to the specific and changing
needs of each case, using therapeutic
methods that best suit the case. The
treatment meetings are organized at
the patienťs home, given the approval
of the family.
4. Responsibility. Whoever received the
first contact is responsible for organizing
the first meeting, in which the
treatment decision is made. The team
takes charge of the entire treatment.
5. Psychological continuity. The team
takes responsibility for the treatment
for as long as needed in both the outpatient
and inpatient settings.
6. Tolerance of uncertainty. Tolerance is
strengthened by building up a safe
enough relationship for the joint process.
In psychotic crises, the development
of an adequate sense of security
requires meeting every day for at least
the first 10 to 12 days. Premature conclusions
and hasty treatment decisions
are avoided.
7. Dialogism. The focus is primarily on
promoting dialogue, and secondarily,
on promoting change in the patient or
family.
ANTICIPATION DIALOGUES
Anticipation Dialogues (AD) consists of
a set of methods that have been developed
in successive research and development
projects throughout the 1990s, organized
by Stakes (National Research and Development
Center for Welfare and Health,
Finland) in collaboration with several
Finnish cities (Arnkil, 1991a, b, 1992;
Arnkil & Eriksson, 1994, 1995, 1996;
Arnkil, Eriksson, & Arnkil, 2000). The
general goal of these projects has been to
develop resource-centered methods, a network-oriented
work approach, and service
structures that transcend sector boundaries.
The goal was to develop psychosocial
child and family services, especially
in multiproblem situations and in preventive
work involving social and health services
and other networking actors. An effort
was made to improve the quality of
the work performed with clients by developing
dialogic methods that showed consideration
for the clients and their personal
networks and work practices that
supported their resources.
194 / FAMILY PROCESS
Although ODs and ADs are close relatives
and are both network-oriented and
dialogic, their working procedures are
very different. Two ways of engaging in
AD are presented in the following section:
recalling the good future and multiprofessional
anticipation dialogue. The methods
are very structured; an independent consultant
takes the lead for outer dialogue
that occurs in the form of talks that
should not be interrupted.
Case: Recalling the good future
A social worker, worried about the situation
of a young child in a family, summoned
a network meeting with the consent
of the parents. The parents decided
not to bring the child, but both brought
their mothers. Also present were the various
professionals (about 10 in all) who
were involved in helping and controlling
the family and who had been invited with
the parents` consent. The gathering
brought together the stakeholders who
had been working separately for years,
doing their best, but increasingly worried
about the chilďs situation and also dissatisfied
with what the others had been doing.
Authors TEA and EE were invited to
facilitate the dialogue. They explained the
idea that speaking and listening are separated
to allow rich inner dialogues and
that the facilitators interview the family
members and professionals about a good
future. They went on to explain that this
is done in order to bring to the fore points
critical for a plan of joint action to make a
good future come true. In other words, the
platform is for negotiating clarity and coordination,
not for decision-making. The
facilitators arranged the larger family
(parents plus grandmothers) around one
table and the professionals around a second
one. One facilitator joined the family;
the other joined the professionals. The extended
family group was interviewed
first.
The first facilitator asked the parents
and grandparents, each in turn, to think
aloud: "Leťs assume a year has passed.
Matters are well in the family now; the
chilďs situation is better. How are things
now, from your point of view?" "What are
you particularly happy about?" The facilitator
assisted the families in taking off
into the future by asking about concrete
everyday matters. Every now and then
the facilitator quoted in a summarizing
manner what had just been said, carefully
adhering to the words used by the family.
He inquired, "Have I heard you correctly,
in that you said. . .?" This was to help
those involved to reflect by echoing what
they had said in the course of thinking
aloud and to underline that the intent, in
the session, was to listen keenly to every-
one.
After hearing each family member's
views on the good future, they were
asked, still supposing a year had passed:
"What did you do to make this good development
possible, and who helped you, and
how?" Through this question the activity
and support network was outlined.
The third round, at the family table,
was facilitated by the question: "What
were you worried about a year ago, and
what lessened your worries?" In this way,
present worries can be approached from a
less stressful future viewpoint. The process
of voicing reflections was aided by the
facilitator's summarizing quotations, and
by his inquiries on whether or not the
family members` views had been correctly
heard. The family's views of the good near
future now became the fixing points for a
constructive plan of action. The basic elements
of the good situation were written
down for all to see. The family members
were asked to correct the notes if needed.
The second facilitator began to interview
the professionals. One after another
they were asked two questions while the
others, including the family, listened to
the questions: "As you heard, things are
SEIKKULA, ARNKIL, & ERIKSSON / 195
Fam. Proc., Vol. 42, Summer, 2003
going well in the family now that a year
has passed. What did you do to support
these good developments?" "What were
you worried about a year ago, and what
lessened your worries?" The process of the
professionals voicing their reflections was
also aided by short quotations. After this
round, the professionals views on helpful
measures were written down (and corrected
if they so wished). A discussion
followed on whether or not all present
could commit themselves to the emerging
plan. Agreements on the next steps and
followup sessions were then made. The
session ended with the participants deciding
who would coordinate further actions.
Isomorphic interactions
Immediate feedback from such dialogues
has been very positive, for both
clients and professionals. Family members
report that they find it a relief to
imagine a less burdensome future in the
presence of professionals who normally
focus on problems and shortcomings. In
the sessions described above, all the participants
gather impressions at the same
time--and know that all others hear what
they are hearing. This appears to be quite
an intervention in itself. It is more usual
for the counterparts in a multiagency network
to meet the clients separately,
guessing what the others are doing. It is
not unusual for the interaction in the professional
network to replicate fundamental
figurations in the personal networks.
On the one hand, (Arnkil & Eriksson,
1995) professionals' activities mirrors clients'
activities for positive reasons: professionals
identify themselves emotionally
with their clients and are thus able to
understand them more comprehensively
than by mere cognitive means. On the
other hand, through such empathizing,
professional work is constantly at risk of
becoming more of the same, i.e., professionals
may replicate among themselvesand
with the client­the fundamental figurations
in the families and in the personal
networks. It is not exceptional that
in split situations, the professional network
becomes divided, or that systems of
blame or secrecy prevail among and
within agencies. Client and helper figurations
may become isomorphic (Schwartzman
& Kneifel, 1985). This seldom promotes
change. It is precisely in and for
such stuck and worry-laden situations
that the Anticipation Dialogues were
originally developed.
In the above example, the method of
Recalling the Future was applied. This
seems to be the most powerful method in
the set of ADs. As can be seen, the facilitators
(network consultants) take the lead
in turn and organize the external dialogue
as a sequence of uninterrupted
speeches. The facilitators move the monological
thought experiments along with
their questions and help them to develop
into dialogues. It is essential that the facilitators
do not attempt to solve the case
and that they refrain from giving advice
(even if they are tempted to do so and, in
our experience, often are). They must
stick to facilitating the process of dialogue.
This is the most valuable contribution
that they can make to the involved
network.
Merging sources
The basic sources of inspiration for the
method are clearly visible. Like Andersen
(1990) and his team, AD facilitators separate
speaking and listening to create
room for rich dialogues. Like Seikkula
(1991), we too organize network meetings
of involved stakeholders. And, as deShazer
(1991) does, we approach present problems
from future solutions. We have
found it necessary, however, to also ask
directly about the worries. If present worries
are not addressed in the joint dialogues,
they seem to reappear in backtalk,
outside the dialogues, and may dominate
the view. To enhance commitment, it is
196 / FAMILY PROCESS
necessary to encourage credible hope. The
first question (assuming that things are
well after the year has passed) carries
certain euphoric elements. The third
question (what were you worried about a
year ago?) is, in a way, a realizing question,
but offers an opportunity to handle
today's problems from the perspective of
possible solutions. The AD set of methods,
and Recalling the Future as one of them,
have been developed in and for situations
that are susceptible as isomorphic developments.
The facilitator's active role is in
curbing cycles of blame, domination, etc.,
by (a) helping to construct the family's
good future (instead of presenting demands
for change) as the fixing point of
plans, (b) promoting polyphony by conducting
an external dialogue through interview
while comment is restrained (instead
of allowing dominance of defining
and watertight views with no openings for
change), and (c) encouraging subjectivity
by interviewing each participant concerning
personal viewpoints (instead of allowing
others to attempt to dictate how family
members should think or act). The goal
of ADs is to help stakeholders find coordination
in a network that they cannot control
either directly or unilaterally. The
basis for coordination is sought in the lifeworld
of the clients instead of the professionals'
specialized tasks.
You can change only your own activity
In our projects, we initially started with
multiprofessional meetings--without clients
present. If a professional found him/
herself constantly worried about a child
(or, in elderly care, a senior citizen) and
found that the professional network was
not making progress, i.e., was unable to
proceed or was in great disagreement,
even if everyone was doing their best to
carry out their responsibilities, she/he
summoned the involved professionals for
a clarifying session. We (TEA & EE) were
called in to act as facilitators. We organized
a multiprofessional anticipation dialogue,
suggested that speaking and listening
be separated, and asked three
questions of each participant while the
others listened: (a) What would happen if
you did nothing (in the given case)?, (b)
What could you do (to help)?, and (c) What
would happen if you did that? The professionals
were thus encouraged to focus on
themselves, instead of the clients, and on
intended and unintended consequences of
their actions instead of objective problem
definitions. In fact, no problem definitions
were made. This shift in gaze was intended
to promote curiosity; curiosity in
turn endeavored to encourage experimentation.
Experimentation was intended to
pave ways out of impasse situations in
which counterparts usually expect others
to change.
No overall descriptions or full-scale plans
were made. At the end of the sessions, a
simple agreement was made on who would
do what with whom next. Typically, a large
proportion of the network would decide that
they would step back and follow up. The
participants seemed surprised and relieved
to realize that the only factor they could
directly change was their own activity, in
which, at long last, they had a realistic target
for change. The shift from attempts to
change clients, families, and neighboring
agencies was marked. Work regained an
experimental nature, multiproblem situations
became interesting, and there was
room for new influences.
It appears that abstaining from defining
a common problem promotes multiprofessional
collaboration. That there is
no objective picture of a situation and that
each observer has his/her perspective in
the observing system became very clear in
the sessions described above. Commonly
occurring joint problem definitions are, in
our experience, not only futile, but hinder
collaboration. As Anderson, Goolishian, &
Winderman (1986) point out, problem definitions
tend to capture participants into
SEIKKULA, ARNKIL, & ERIKSSON / 197
Fam. Proc., Vol. 42, Summer, 2003
problem talk and reproduce problemdetermined
systems. Moreover, there are
no common problems to be defined. The
shift in focusing from defining underlying
problems to anticipating possible consequences
of one's actions seems, in our experience,
to keep professional duels at
bay. Furthermore, if problems are seen as
problems of activity and not as characteristics
or qualities of a person, family, professional,
agency, or the like, it becomes
very clear that each player has his/her
own activity problem. In an interesting
way, asking involved specialists to anticipate
makes them equals. No one can say
with certainty what will happen if this or
that or nothing is done. Increasing subjectivity
to the extent that one realizes that
objectivity comprises an endless polyphony
of subjectivities appears to pave the
way to the postmodern expertise with
fewer fantasies of control and a high tolerance
of uncertainty--and curiosity toward
others subjectivity. In our experience,
professionals listen with great interest
to each other's anticipations and
seem to appreciate the expert professional
knowledge and contextual wisdom
("knowing from within" the relationship,
as Shotter [1993] calls it), that each specialist
can bring to bear.
Anticipating the consequences of noninterventions
aroused the professionals' curiosity
toward the clients' personal networks.
Would someone else help if the
professionals did nothing? Again and
again the professionals were surprised
how little they knew about these potential
resources and became curious about
them. The step toward inviting personal
networks was no longer a giant leap.
Guiding principles in Anticipation
Dialogues
Apart from the future recalling and
multiprofessional anticipation methods,
the AD set includes variations designed to
suit planning tasks of preventive work.
We will not discuss them here. They are,
in short, methods for working out joint
themes, organization, and tasks through
dialogues, and they greatly resemble the
methods described above (for a detailed
account, see Arnkil, Eriksson, & Arnkil,
2000).
It seems that the sectored and specialized
professional system is desperately in
need of intermediaries. ADs are usually
one-time consultations, the main purpose
of which is to clarify complex situations
and promote change by producing inner
dialogues as much as possible with everyone
present. Experiences in these consultations
have been good. Feedback from
both clients and professionals has been
very positive. Their experience is that
clarity and hope have clearly increased
during these sessions even in very complicated
situations.
Based on development work, feedback,
and training experiences, eight main
principles of AD have been established.
1. Subjectivity. Each participant is encouraged
to elaborate his/her own
point of view instead of trying to represent
the overall picture. Already, the
settings emphasize that the overall
picture comprises a multitude of subjective
pictures. Gaining more understanding
of others' points of view and
positions can lead to a better understanding
of the interactive and interpreting
network in which one is embedded.
A transition from objective
problems to subjective concerns is cen-
tral.
2. Emphasis on the reciprocal character
of professional work. In the conventional
expertise it was acceptable for
the expert to say (reproducing a topdown
setting): "You have this or that
problem." The expression "I am worried
about this or that and my possibility
to help" suggests, in principle, a
professionaľs request for help from the
198 / FAMILY PROCESS
clients and their personal networks.
They are potential providers of problem-solving
resources (whereas they
tended to be potential sources of disturbance
or problem-sustainment in
the conventional expertise).
3. Polyphony. The principle of subjectivity
is tightly linked with the principle
of polyphony. As Castells (2000) points
out, networks have no centers because
each link or participant is the center of
his/her network, and subjectivity is a
way to make sense of multisubjective
systems. Networks are also fundamentally
uncontrollable--too complex for
unilateral control.
4. High tolerance of uncertainty. Because
networks are fundamentally uncontrollable,
a high tolerance for uncertainty
is called for. Polyphony in dialogue
enhances tolerance of uncer-
tainty.
5. Dialogism. Dialogues can provide enriching
impressions of the multisubjective
systems in which one is embedded
and understanding not available for
detached actors.
6. Experiments in thought and action.
Subjective anticipation brings to light
the fundamentally experimental nature
of all activity. Confident predictions
belong to conventional expertise
whereas postmodern expertise is based
on tentative anticipations and sensitivity
to the fact that actions have both
intended and unintended consequences.
7. Future perspective as the basis for coordination.
Very little emphasis is put
on the past. Instead, the future of the
life-world of the clients serves as the
platform for coordinating activity.
8. Facilitation. The Fordist system of professional
help creates a great need for
facilitation. Facilitating both horizontally
(e.g., between professions, over
sector boundaries) and vertically (e.g.,
between managers and personnel, professionals,
clients, and personal networks)
puts into action the seven preceding
principles.
DISCUSSION
Both approaches are dialogic and network-oriented.
On the surface, they are
almost opposites. ADs are very structured,
not open. The facilitator­the network
consultant­conducts the dialogues.
Strikingly, ADs are unsuited for those crisis
situations in which ODs function best.
ODs appear to be helpful in psychotic situations
where ADs do not appear advisable.
Conversely, ADs yield the best results
in open-care muddles, which are not
the basic territory of ODs. The authors of
this article became curious about the
complementarity of the approaches. Do
they, through contrast and comparison,
reveal something essential about the contexts
in which they are applied?
One would expect that severely split or
psychotic situations, for which ODs are
well suited, would call for methods more
structured than lighter cases and vice
versa. Curiously, the less-structured ODs
are an intervention suited to highly structured
care contexts, e.g., mental hospitals.
ADs, on the other hand, are at home
in outpatient care, where the professions
of various agencies meet each other as
well as common clients in a diffuse noman's
land. The cases dealt with in the
open care of child welfare are by no means
lighter than the psychotic cases encountered
in adult psychiatry. The big difference
is in how the treatment filters
through the professional system. In adult
psychiatry, the serious cases should, in
principle, be dealt with in highly specialized­and
institutionalized­units right
from the beginning. In Finnish child protection,
the ultimate responsibility is in
the open-care system--and with child
welfare social workers. In adult psychiatry,
one encounters strongly structured
modes of service; in child protection, one
encounters institutional vagueness.
SEIKKULA, ARNKIL, & ERIKSSON / 199
Fam. Proc., Vol. 42, Summer, 2003
ODs and ADs share common ground in
dialogism, polyphony, and social constructionism.
The authors of this article
are intrigued by the possibility that they
are approaching something post-Fordist
from different angles and in different contexts.
What becomes essential seems no
longer to be the intervention itself or the
therapeutic method itself. Rather, what is
essential is seeing the polyphonic nature
of our clients' reality. In this respect language­aiming
for a dialogical conversation
in both approaches as a specific form
of being in language­as the focus of treatment
makes the practical forms of different
approaches secondary. We can make
use of ODs, especially in the heaviest crisis
situations and of ADs in impasse cases
encountered in social care, while making
use of our understanding that the resources
for recovery are in the dialogue
itself.
The basis for attempts to help professionally
is formed by listening carefully to
what the client, members of the family,
and those of the personal networks have
to say about the good future, in the case of
ADs, and in finding expressions for terrifying
experiences in the ODs. The expert
system is reminded that the basis for help
is in the comprehensive life-world, not in
the specialized fragments as seen in the
system-world. Already, the setting emphasizes
that the top-down approach of
the conventional expertise must be
turned around. Specialized expertise,
however, does not become obsolete. It can
be made a more fruitful element of the
whole if the fixing point is the life-world
instead of one's professional slice. At least
in the Nordic countries, the second phase
of modernization appears to challenge the
top-down expertise carried out by singular
professionals.
REFERENCES
Aaltonen, J., Seikkula, J., Alakare, B., Haarakangas,
K., Kera¨nen, J., & Sutela, M. (1997,
October). Western Lapland project: A comprehensive
family- and network-centered
community psychiatric project. ISPS, Abstracts
and Lectures, 12-16, 124.
Alanen, Y.O. (1997). Schizophrenia: Its origins
and need-adapted-treatment. London: Karnac
Books.
Abbott, A. (1988). The system of professions.
An essay on the division of expert labor. Chicago:
Michigan Press.
Alakare, B. (1999). Neuroleptila¨a¨kitys ja dialoginen
hoitomalli akuutin psykoosin hoidossa.
In K. Haarakangas & J. Seikkula
(Eds.), Psykoosi -- uuteen hoitoka¨yta¨nto¨o¨n
[Psychosis -- towards a new approach]
(pp.123-145). Helsinki: Kirjayhtyma¨.
Andersen, T. (1990). The reflecting team: Dialogues
and dialogues about dialogues. New
York: Norton.
Andersen, T. (1995). Reflecting processes. Acts
of informing and forming. In S. Friedman
(Ed.), The reflective team in action (pp. 11-
37). New York: Guilford Publication.
Anderson, H. (1997). Conversation, language
and possibilities: A postmodern approach to
therapy. New York: Basic Books.
Anderson, H., & Goolishian, H.A. (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., & Winderman,
L. (1986). Problem-determined systems: Towards
transformation in family therapy.
Journal of Strategic and Systemic Therapies,
5, 1-14.
Arnkil, E. (1991a). Social work and the systems
of boundary. Suggestions for conceptual
work. In E. Arnkil (Ed.), Keita¨ muita
ta¨ssa¨ on mukana? Viisi artikkelia verkostoista
[Who else is involved? Five articles on
networks] (pp. 97-120). Sosiaali- ja terveyshallitus.
Raportteja 23. Helsinki: VAPK-
kustannus.
Arnkil, E. (1991b). Peileja¨. Hypoteeseja sosiaalityo¨n
ristiriidoista ja kehitysvyo¨hykkeesta¨
[Mirrors. Hypotheses on the contradictions
and developmental zone of social
work]. Sosiaali- ja terveyshallituksen tutkimuksia
5. Helsinki: VAPK-kustannus.
Arnkil, E. (1992). Sosiaalityo¨n rajasysteemit
ja kehitysvyo¨hyke [The systems of boundary
200 / FAMILY PROCESS
and the developmental zone of social work.].
Jyva¨skyla¨ Studies in Education, Psychology
and Social Research, 85. Jyva¨skyla¨: University
of Jyva¨skyla¨.
Arnkil, E., & Eriksson, E. (1994). Deficiencies,
resources and curiosity. In S. Ha¨nninen
(Ed.), Silence, discourse and deprivation (pp.
134-148). Stakes: Research Reports 43. Jyva¨skyla¨:
Gummerus.
Arnkil, T. E., & Eriksson, E. (1995). Mukaan
meneminen ja toisin toimiminen. Nuorisopoliklinikka
verkostoissaan [Feeling alike and
acting differently. The adolescent clinic in
its networks]. Stakes: Tutkimuksia. Saarija¨rvi:
Gummerus.
Arnkil, T. E. & Eriksson, E. (1996). Kenelle ja¨a¨
kontrollin Musta Pekka kortti? Sosiaalitoimisto
verkostoissaan [Who is going to have
the Old Maid Card of the control game? Social
welfare office in its networks]. Stakes:
Tutkimuksia 63. Saarija¨rvi: Gummerus.
Arnkil, T.E., Eriksson, E., & Arnkil, R. (2000).
Anticipation dialogues. Vertical and horizontal
dialogue methods for strategic management
and smart networking. Helsinki:
Stakes Themes from Finland 3.
Attneave, C. (1990). Core network intervention:
An emerging paradigm. Journal of
Strategic and Systemic Therapies, 9, 3-10.
Bakhtin, M. (1984). Theory and history of literature:
Vol. 8. Problems of Dostojevskij's poetics.
Minneapolis, MN: Manchester University
Press.
Beck, U. (1992). Risk society: Toward a new
modernity. London: Sage.
Beck, U., Giddens, A., & Lash, S. (1994). Reflexive
modernization: Politics, tradition and
aesthetics in the modern social order. Cambridge:
Polity Press.
Boscolo, L., & Bertrando, P. (1993). Times of
time. New York: Norton.
Bourdieu, P. (1994). Language and symbolic
power. Cambridge, MA: Harvard University
Press.
Castells, M. (2000). Materials for an exploratory
theory of the network society. British
Journal of Sociology, 5, 5-25.
deShazer, S. (1991). Putting difference to work.
New York: Norton.
Foucault, M. (1977). Discipline and punish:
The birth of the prison. London: Allen Lane.
Garrison, J. (1981). Clinical construction of
action social network. International Journal
of Family Therapy, 3, 258-267.
Gergen, K., & McNamee, S. (2000). From disordering
discourse to transformative dialogue.
In R. Neimeyer & J. Raskin (Eds.),
Constructions of disorders (pp. 333- 349).
Washington DC: American Psychological
Association.
Giddens, A. (1979). Central problems in social
theory: Action, structure and contradiction
in social analysis. Cambridge: Polity.
Giddens, A. (1990). The consequences of modernity.
Cambridge: Polity.
Goodman, L., Rosenberg, S., Mueser, K., &
Drake, R. (1997). Physical and sexual assault
history in women with serious mental
illness: Prevalence, correlates, treatment,
and future research directions. Schizophrenia
Bulletin, 23, 685-696.
Gramsci, A. (1971). Selections from the prison
notebooks. New York: Columbia University
Press.
Haarakangas, K. (1997). The voices in treatment
meeting: A dialogical analysis of the
treatment meeting conversations in familycentered
psychiatric treatment process in
regard to the team activity. English Summary
(pp. 119-126). Jyva¨skyla¨ Studies in
Education, Psychology and Social Research,
130. Jyva¨skyla¨: University of Jyva¨skyla¨.
Habermas, J. (1984). The theory of communicative
action: Vol. 1. Reason and the rationalization
of the society. Boston: Beacon
Press.
Hirsch, J. (1985). Turvavaltio. Saksan malli,
sen kriisi ja uudet yhteiskunnalliset liikkeet
[The safeguard state. The German model,
its crisis and new social movements]. Jyva¨skyla¨:
Gummerus.
Holma, J. (1999). The search for a narrative:
Investigating acute psychosis and the needadapted
treatment model from the narrative
viewpoint. Jyva¨skyla¨ Studies in Education,
Psychology and Social Research, 150. Jyva¨skyla¨:
University of Jyva¨skyla¨.
Imber-Black, E. (1988). Families and larger
systems: A family therapist`s guide through
the labyrinth. New York: Guilford.
Karon, B. (1999). The tragedy of schizophrenia.
The General Psychologist, 32, 3-14.
Kemenoff, S., Jachimczyk, J., & Fussner, A.
(1999). Structural family therapy. In D.
SEIKKULA, ARNKIL, & ERIKSSON / 201
Fam. Proc., Vol. 42, Summer, 2003
Lawson & F. Prevatt (Eds.), Casebook in
family therapy (pp. 111-145). Toronto:
Brooks/Cole Wadsworth.
Kera¨nen, J. (1992). The choice between outpatient
and inpatient treatment in a familycentered
psychiatric treatment system. English
summary (pp. 124-129). Jyva¨skyla¨
Studies in Education, Psychology and Social
Research, 93. Jyva¨skyla¨: University of Jy-
va¨skyla¨.
Klefbeck, J., Bergerhed, E., Forsberg, G., Hultkranz-Jeppson,
A., & Marklund, K. (1988).
Na¨tverksarbete i multiproblemfamiljer [Networking
in multiproblem families]. Botkyrka:
Botkyrka kommun.
Larsen T., Johannessen, J., & Opjordsmoen,
S.(1998). First-episode schizophrenia with
long duration of untreated psychosis. British
Journal of Psychiatry, 172 (Suppl. 32),
45-52.
Lehtonen, H. (1986). Sosiaalivaltion ja perheen
historiallisista suhteista [On the historical
relations between the social state and
the family]. Vammala: Sosiaalipolitiikka
1985.
Leiman, P., & Stiles, W. (2001). Dialogical sequence
analysis and the zone of proximal
development as conceptual enhancement to
the assimilation model: The case of Jan revisited.
Psychotherapy Research, 11, 311-
330.
Loebel, A., Lieberman, J., Alvir, J., Mayerhoffer,
D., Geisler, S., & Szymanski, S. (1992).
Duration of psychosis and outcome in firstepisode
schizophrenia. American Journal of
Psychiatry, 149, 1183-1188.
Luhmann, N. (1989). Ecological communication.
Padstow: Polity Press.
Lyotard, J.-F. (1979). The postmodern condition.
Manchester, UK: Manchester University
Press.
Pattison, E., & Pattison, M. (1981) Analysis of
a schizophrenic psychosocial network.
Schizophrenia Bulletin, 7, 135-143.
Penn, P. (1998). Rape flashbacks: Constructing
a new narrative. Family Process, 37,
299-310.
Schwartzman, H., & Kneifel, A. (1985). How
the child care system replicates family patterns.
In. J. Schwartzman (Ed.), Families
and other systems (pp. 87-107). New York:
Guilford.
Seikkula, J. (1991). Family-hospital boundary
system in the social network. English summary
(pp. 227-232). Jyva¨skyla¨ Studies in
Education, Psychology and Social Research,
80. Jyva¨skyla¨: University of Jyva¨skyla¨.
Seikkula, J. (1994). When the boundary opens:
Family and hospital in co-evolution. Journal
of Family Therapy, 16, 401-414.
Seikkula, J., Alakare, B., & Aaltonen, J.
(1999). Potilaat sosiaalisissa verkostoissaan
- kahden vuoden seurantatutkimus akuutin
psykoosin kotihoidosta. In K. Haarakangas
& J. Seikkula (Eds.), Psykoosi -- uuteen
hoitoka¨yta¨nto¨o¨n [Psychosis -- towards a
new approach] (pp 107-122). Helsinki: Kir-
jayhtyma¨.
Seikkula, J., Alakare, B., & Aaltonen, J.
(2000). A two-year follow-up on open dialogue
treatment in first episode psychosis:
Need for hospitalization and neuroleptic
medication decreases. [Published in Russian;
English manuscript from the authors
or www.talkingcure.com]. Social and Clinical
Psychiatry, 10, 20-29.
Seikkula, J., Alakare, B., & Aaltonen, J.
(2001). Open dialogue in first-episode psychosis
II: A comparison of good and poor
outcome cases. Journal of Constructivist
Psychology, 14, 267-284.
Seikkula, J., Aaltonen, J., Alakare, B., Haarakangas,
K., Kera¨nen, J., & Sutela, M. (1995).
Treating psychosis by means of open dialogue.
In S. Friedman (Ed.), The Reflective
team in action (pp. 62-80). New York: Guilford
Publication.
Selvini-Palazzoli, M., Boscolo, L., Cecchin, C.,
& Prata, G. (1978). Paradox and counterparadox.
New York: Jason Aronson.
Shotter, J. (1993). Conversational realities:
Constructing life through language. London:
Sage.
Sluzki, C. (2000). Social networks and the elderly:
Conceptual and clinical issues, and a
family consultation. Family Process, 39,
271-284.
Smith, G., & Nylund, D. (1997). Narrative
therapies with children and adolescents.
New York: Guilford Press.
Speck, R., & Attneave, C. (1973). Family networks.
New York: Pantheon Books.
202 / FAMILY PROCESS
Stern, S., Doolan, M., Stables, E., Szmukler,
G.L., & Eisler, I. (1999). Disruption and reconstruction:
Narrative insights into the experience
of family members caring for a relative
diagnosed with serious mental illness.
Family Process, 38, 353-369.
Trimble, D. (1996-1997). Network therapy.
AFTA Newsletter: 26-30.
Trimble, D. (2000). Emotion and voice in network
therapy. Netletter, 7(1), 11-16.
Trimble, D., Kliman, J., Villapiano, A., &
Beckett, W. (1984). Follow up of a full-scale
network assembly. International Journal of
Family Therapy, 6, 102-113.
van der Kolk, B., & Fisher, 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. Cambridge, MA: Harvard
University Press.
Vygotsky, L. (1978). Mind in society: The development
of higher psychological processes.
Cambridge, MA: Harvard University Press.
SEIKKULA, ARNKIL, & ERIKSSON / 203
Fam. Proc., Vol. 42, Summer, 2003