RED BALLOON: APPROACHING DREAMS AS
Training and Research Institute for Systemic Psychotherapy and Private Practice
In this article, dreams are seen as stories within a self-narrative. Dream stories, like all other
stories, are told in an eﬀort to make sense of experiences. Here, dream content is linked to
current concerns, some aspects of which are not given voice in waking. Dreams depict
restricting themes but also openings in self-narratives. Several examples are provided of how
dreams can be linked to early, middle, and late therapy phases associated with recognizing,
challenging, revising, and maintaining a revising stance. It is further suggested that dream
stories can be used to trace, facilitate, and evaluate the process of reconstructing self-
narratives. Finally, a number of therapeutic interventions are brieﬂy presented to facilitate the
work of narrative-informed family therapists working with individuals, families, and groups.
Father: Well-dreams are bits and pieces of the stuff of which we are made ...Or
something like it.
Daughter: All right. But how are dreams put together?
(Bateson, 2000, p. 50)
Dreams in Systemic and Family Therapy Approaches
Early in her therapy, a young woman dreamed that she was holding on to the rope of a
red balloon together with a group of other people, each holding on to a separate rope. She
enjoyed the whole experience, but was thinking that she could not hold on to her rope anymore
and would be the ﬁrst to fall into the sea underneath. She fell and all the ropes appeared
around her body in knots. She was trying to free herself. The unknotted edges of the ropes
could be seen reaching far to the horizon.
Systemic and family therapy approaches have no theory of dreams and do not suggest their
own ways of working with dream material. With the exception of psychoanalytic family thera-
pists who apply dream analysis to couples and families (see Buchholz, 1990; Scharﬀ, 1992),
other therapists in the family therapy tradition unavoidably borrow ideas and techniques from
various approaches. For instance, Feixas, Cunillera, and Mateu (1990) take the constructivist
perspective of George Kelly as their theoretical background for the therapeutic use of dreams.
Andrews, Clark, and Zinker (1988) use techniques from gestalt, and Sanders (1994) takes a
pragmatic perspective that sees dream content in relation to family problems. Kaplan, Saay-
man, and Faber (1981) adopt a Jungian rationale for their empirical study associating the con-
tent of family dreams with the degree of functionality in families.
In this article, dreams are seen as stories within a self-narrative aiming, like all stories of
the kind, to organize and make sense of personal experiences (see also Knudson, Adame, &
Finocan, 2006). This view oﬀers a theoretical background to the use of dreams that is compati-
ble with the narrative turn in the ﬁeld of family therapy (see Hoﬀman, 2002).
Athena Androutsopoulou, PhD, is a clinical psychologist and psychotherapist in private practice (www.
androutsopoulou.gr). She is also co-director and supervisor at the Training and Research Institute for Systemic
Address correspondence to Dr. Athena Androutsopoulou, Patision 130, GR-112 57, Greece; E-mail: athena@
Journal of Marital and Family Therapy
October 2011, Vol. 37, No. 4, 479–490
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 479
A Narrative Turn
As part of the narrative turn, some family therapists have embraced particular models such
as White’s ‘‘Narrative Therapy,’’ and others, including the author of this article, have chosen
to inform their work by a narrative metaphor while continuing to practice within the family
therapy model in which they have been trained (see Androutsopoulou, 2005b). Zimmerman and
Dickerson (1994) say the link between narrative therapies and family therapy is in relationships.
They explain that using a narrative metaphor means a shift from people’s relationships as the
object of therapy to the stories about the relationships as the object of therapy. Parry (1991)
points out that our stories are always connected to the stories of others and to larger stories, a
factor that is central to family therapists’ adopting a narrative metaphor but is sometimes over-
looked by those who practice narrative therapy. Penn and Frankfurt (1994) give examples from
their own family therapy practice of how changes in one individual can lead to changes in the
rest of the family based on a narrative rationale. So, postmodern family therapists have come
to refer to family therapy as a way of thinking rather than a way of doing (e.g., only family
sessions; Anderson, 1999; Parry & Doan, 1994). I choose the term family-oriented therapy to
describe a practice where the focus may be the individual, the family, or a group.
In this article, the reader will ﬁnd selected ways of working with dreams that have been
found useful in my family-oriented, narrative-informed clinical practice.
Phenomenological View of Dreams
Dream agent or ‘‘mind chat?’’ Freudian theory views dream symbolization as a ‘‘defensive
maneuver designed to conceal meaning’’ (Rycroft, 1991, p. x). Jungian theory regards dreams
as messages from a wider self. Both of the above theories and contemporary psychodynamic
approaches treat dreams as creations of a dream agent who endows them with meaning. This
meaning is interpreted in dream work.
From a phenomenological perspective, ‘‘dream phenomena are [...]always just what
they are as they shine forth; they are uncovering and unveiling and never a covering up or a
veiling of psychic content’’ (Boss, 1958, p. 262; see also Craig & Walsh, 1993; Domhoﬀ, 1999,
Dreams are the continuation of waking thinking or ‘‘mind chat’’ (Kilroe, 2000a). We con-
tinuously process experiences, but in our sleep, we do that without conscious awareness and
without input from the external world (Hill, 2004). In that sense, the concept of a dream agent
is not necessary and can even be misleading. In support of this idea, Mazzoni and Loftus
(1998) have empirically shown that people can be manipulated to think that they have actually
experienced what they have only dreamed or, worse, that they can be manipulated to believe
that their dream contents reveal forgotten childhood experiences. No dream agent was involved
in those cases, only the manipulative intention of therapists or researchers.
Narrative View of Dreams
Constructing a self-narrative. From the social constructionist and narrative viewpoints, the
‘‘self’’ equals to a continuous construction of a self-narrative, aiming to secure a sense of histor-
ical continuity, directionality, and coherence among what often appears to be loosely connected
‘‘selves’’ that may seem to act differently depending on the circumstances. The process of con-
structing, deconstructing, and reconstructing a self-narrative is seen as a lifelong struggle.
To illustrate the process of constructing and reconstructing a self-narrative, one may think
of the writing of an autobiography, which follows certain literal ‘‘rules’’ (i.e., chronological and
causal progression of events, plot, and outcome), but which is also constantly unfolding. As it
would happen in ﬁction, even chapters that have already been written (life episodes) are being
continuously modiﬁed so that the various incidents ‘‘ﬁt’’ well and the whole story makes better
sense (see Eakin, 1992). Persons—being reﬂective and able for internal dialogue—simulta-
neously place themselves in the role of ‘‘author’’ and ‘‘reader’’ (or ‘‘audience’’) so that anything
written or said is edited and revised almost at the same time. In that process, (signiﬁcant) oth-
ers play also the role of co-authors and ‘‘audience,’’ and their presence—even if only internal-
ized—shapes the content of the self-narrative to a considerable degree. Life stories are also
shaped by and told to various other audiences. In the words of White and Epston (1990),
480 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011
‘‘experience must be storied and it is this storying that determines the meaning ascribed to
experiences’’ (p. 10).
Self-narratives in trouble. White and Epston (1990) mention that ‘‘persons experience prob-
lems [ . . . ] when the narratives in which they are storying their experience, and ⁄or in which they
are having their experiences storied by others, do not suﬃciently represent their lived experi-
ence.’’ Aspects of experience ‘‘that fall outside the dominant story provide a rich and fertile
source for the generation, or regeneration, of alternative stories’’ (p. 15). Papp and Imber-Black
(1996) talk of a central restricting theme in persons’ narratives and propose that therapists and
family members work together to recognize, develop, question, challenge, and ultimately change
such a theme, ‘‘a gradual and evolving process that takes place over time’’ (p. 7).
Dreams as stories. In an often used quote by Hardy (1968), ‘‘we dream in narrative, day-
dream in narrative, remember, anticipate, hope, despair, believe, doubt, plan, revise, criticize,
construct, gossip, learn and love by narrative’’ (p. 5). Sarbin (1986) states that ‘‘our dreams are
experienced as stories, as dramatic encounters, often with mythic shadings’’ (p. 11). Persons
continue their eﬀort to make sense of experiences while dreaming by creating and telling stories
to an audience that in the case of dreams consists of the self only, as we do in self-talk while
awake. By reporting a dream to an audience (relative, friend, and therapist), one situates the
self-narrative in its social context.
Many different ideas exist as to what constitutes a story or narrative. Here, I rely on the
notion of story grammar (McLeod, 1997) and, more speciﬁcally, structure, to present dreams as
Jung (1970) had observed that many dreams are reported with a structure that resembles
that of classical Greek drama: exposition (the setting, including time, place, actors, and situa-
tion), peripeteia (development of the plot, something begins to happen or change), crisis (culmi-
nation, something critical is happening or changing), lysis ⁄catastrophe (resolution or disaster,
how the story ends). In dreams, the last part is sometimes absent or appears pending. Contem-
porary theorists and researchers make use of this or similar structures in their dream studies.
Kilroe (2000a) found it useful in her study of ‘‘the dream as text, the dream as narrative.’’ Cari-
ola (2008) analyzed dream narratives applying the general narrative framework of Labov and
Waletsky (1967), which has many similarities to the above.
In the example of the red balloon dream, this structure may be applied as follows. (The
dream is here written in the ﬁrst person and in present tense, a useful way to report a dream): I
am holding on to the rope of a red balloon together with a group of other people, each holding on
to their own rope (exposition). I am thinking that I cannot hold on to my rope anymore and will
be the ﬁrst to fall into the sea underneath (peripeteia). I fall and all the ropes appear around my
body in knots (crisis). I am trying to free myself. I can see the unknotted edges of the ropes
reaching far to the horizon (pending).
MAKING SENSE OF DREAMS
The language of dreams is personal and dream ‘‘interpretation’’ should make sense to the
dreamer herself. It is possible, however, to ‘‘interpret’’ or, in Kramer’s (1991, 1993) term, to
‘‘translate’’ a dream without the contribution of the dreamer, although this work is not a sub-
stitute for the collaborative work needed in therapy. ‘‘Dream translation’’ is possible because
the language of the dream is not just personal but also collective, in the sense that personal lin-
guistic codes reﬂect wider social codes or ‘‘extrinsic structures’’ (States, 1994).
According to Lakoﬀ (1993), most dream symbolism makes use of the everyday idioms and
metaphor system. Bateson (2000) writes that when a metaphor is being used in waking, ‘‘it is
marked by the insertion of the words as if or like,’’ but in dreaming ‘‘(as in art) there are no
markers to indicate to the conscious mind that the message material is metaphoric’’ (p. 140).
Kilroe (2000b) points out that our linguistic minds create dream puns all based on homonymy
(a word substitutes another with the same sound) or polysemy (one meaning of a word substi-
tutes another meaning of the same word), and this way, they can represent abstract thoughts in
a concrete form.
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 481
In the red balloon dream, images may be translated into idioms. We can assume that the
young woman is thinking while dreaming that she is ‘‘up in the air,’’ that ‘‘she does not know
the ropes,’’ and that she will ﬁnd herself ‘‘at sea’’ and will ‘‘get into knots.’’ In the dream story,
adverbs are missing and we may attempt to ﬁll them in to understand cause and effect (e.g.,
when . . . then . . . ) or hypothetical statements (if . . . then . . . ). Use of the drama structure,
as previously described, and knowing the dreamer personally can help this task of ﬁlling in
adverbs. For instance, ‘‘when I am in a group of other people I feel I am up in the air, that I
do not know my ropes and my position is at risk. I do not say anything to anyone. When I
abruptly ﬁnd myself alone I am still at sea, and I get into knots trying to ﬁnd a way out. I am
hopeful for the future, though it seems too far away.’’
From a narrative perspective, dream stories can never be given a deﬁnite or ﬁnal ‘‘interpre-
tation.’’ Interpretation is best understood as a tentative effort for meaning making. This mean-
ing will always depend on who approaches the story, for what purpose, in what mood, and so
on. The dreamer herself will give it different meanings at different times and will even ‘‘forget’’
or invent parts of the dream if that serves her life-lasting, ongoing effort to create coherent
self-narratives (see Androutsopoulou, Thanopoulou, Economou, & Baﬁti, 2004) to match the
changing circumstances of any given moment (diﬀerent audience, diﬀerent occurrences, etc.;
Dreams and Current Concerns
When the memory of a dream is intense, and given the continuity of our meaning-making
experience from waking to dreaming (‘‘mind chat’’ continues), one can assume that the person
is trying to make sense of a very emotionally signiﬁcant concern. The degree of emotional
signiﬁcance may not be fully realized in waking awareness, as shown in an interesting study
by Deslauriers and Cordts (1995). Participants completed an open-ended questionnaire that
initially asked them to write a list of their most important current concerns, choose one as the
most signiﬁcant to provide a detailed description, and give an example. They were then asked
to pay attention to their dreams during the following week and report any dream that appeared
relevant to their list of concerns.
Dream reports indicated that while asleep, participants were not occupied with their chosen
most signiﬁcant issue, but with the other issues that they thought were relatively less signiﬁcant
while awake. It is possible that the type of processing occurring while dreaming allows for
‘‘warded off areas of feeling’’ (McLeod, 1997) to be noticed, together with the concerns that
are associated with them. Recent empirical studies have revealed that ‘‘dreams require the con-
certed activation of a tight network of brain mechanisms responsible for instinctual behaviors,
emotion, long-term memory, and for visual perception, with simultaneous de-activation of
mechanisms responsible for reality monitoring and goal-directed activity’’ (Solms, 2000, p. 619;
see also Solms, 1997).
From the Deslauriers and Cordts study, we can infer that when we try to make sense of
our dreams, we need to trace concerns that may have gone unnoticed because of warded-off
feelings and put them into words. The red balloon dream offers an example of tracing concerns
back to the therapeutic process itself. In the days prior to the dream, the young woman was
encouraged by her therapist to ‘‘hold on’’ to the group and try to share some of her fears and
uncertainties with them. The dream of the red balloon may have been a metaphor for the psy-
chotherapy group members as holding on to the same balloon and also to her sense that she
lacks the skills or capacities that others possess. In her dream, she obviously thought that she
could not ‘‘hold on.’’ She dropped out of therapy soon after without discussing any of her con-
cerns, but came back to the group 6 months later to ‘‘untie her knots,’’ as she put it. Of course,
the dream may also have been a metaphor for her family, a dimension we will add later.
Revising Dream Stories
Themes in dream stories are being constantly revised, and this may be obvious in a series
of dreams of the same night or over time. Boss (1958) presented a series of over 800 dreams
reported by one of his clients during 3 years of therapy. He was an engineer in his 40s with
symptoms of depression and sexual impotency. His dreams developed from depicting only
482 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011
machinery and material objects to plants, insects, small animals, big mammals, and ﬁnally a
human being, a woman he became romantically involved with (see also Boss, 1978). There are
also many similarities in the personal codes of the dreamer’s dreams, either in the themes and
plots or in the metaphors used. In the later stage of her therapy, the red balloon dreamer saw
that she was conﬁdently preparing her own red balloon to get oﬀ the ground for a contest and
thinking that she would win.
There are times when the revising of a dream story appears stuck. An extreme example is
the case of people with PTSD where a traumatic scene repeats itself exactly as it has occurred.
The process of making sense of this concern has obviously been stuck even though the nature
of the concern is obvious. A less extreme and more common example is recurrent dreams. The
narrative theme repeats itself but one can usually notice slight changes and some development
in the meaning-making effort. A therapist can offer substantial help in these cases. In my expe-
rience, when concerns are storied in waking, dream stories become unstuck and recurrent
dreams often disappear.
DREAMS IN THERAPY
The Context of Therapy Practice
The work on dreams presented here aims to facilitate the practice of any narrative-
informed family therapist working with individuals, families, and groups. Information on my
own practice is provided only as the context in which these ideas were developed. The practice
combines individual and family sessions, but work in groups of individuals is preferred, espe-
cially for long-term therapy. Such groups function as small communities. Borrowing the grow-
ing conviction of Hoﬀman (2002), ‘‘eﬀective therapy is not individual but communal, even
when the individual is the focus of work’’ (p. 237).
Groups are open and heterogeneous, usually headed by two co-therapists, and meet weekly
(Androutsopoulou, 2005b). Members function as both participants and active audiences. As
participants, they take turns in recounting events that most troubled them (personal, familial,
and work-related) or thoughts they had during the week. As active audiences, they propose
alternative stories based on their own life experiences, validate new stories, and oﬀer compan-
ionship during the long ‘‘journey’’ as equals.
Therapy Phases and Dreams
In this section, I put forward the idea that dream stories are indicative of the therapeutic
processes in the macro-phases of long-term therapy (see also Androutsopoulou, 2005a). Dreams
can be linked to early, middle, and late therapy phases associated with recognizing, challenging,
revising, and maintaining a revising stance. These macro-phases are not as separate as one
might think. In fact, it is important that the new, alternative stories appear as evolving from
and yet containing elements of the old, ‘‘familiar’’ stories (Sluzki, 1992). Monitoring with the
client how a dream theme is developing, and how this theme relates to his or her self-narrative
in general, helps clients see the connection between old and new stories. It also reduces the
anxiety that is often felt, especially in the middle phases. More about this point will be
discussed in the applications section.
Initial dreams: Building a relationship. Initial dreams are often a prologue to the recogni-
tion of restricting themes, as emerging also in the therapeutic context. The therapist is initially
a stranger to the client and clients may have many feelings regarding the work ahead, often
feelings of fear. In clients’ initial dreams, fears are expressed in metaphorical ways: (a) fear that
the therapist will not be on the client’s side or that the client will be excluded (e.g., a young
woman dreamed that she was late for a family session, and when she walked into the room,
the therapist and her parents were whispering in a corner); (b) fear of the therapist seeing
things about the client’s personal life without the client exerting any control (e.g., a man
dreamed that his therapist was having a peep at him through the window); (c) fear that changes
may not be applied in real life or that the client will end up depending on the therapist (e.g., a
woman dreamed that she entered a hothouse to protect herself from the cold, but her feet got
stuck in the mud); (d) fear that the client does not have the necessary strength or capacity to
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 483
endure the process, as with the dreamer of the red balloon, who saw that she was participating
in a TV reality game with zero chances of winning. The theme of incompetence ⁄competence
ran through most of her later dreams. Other fears of clients include that of exposure, that noth-
ing will change, or that more harm will come out of it than good. Reporting such dreams offers
opportunities for these feelings to be expressed openly, for misunderstanding to be cleared up,
and for a relationship to be built.
Early phase: Recognizing restricting themes. A variety of restricting themes may be recog-
nized in a client’s self-narrative. Stories are organized around dimensions of time, space, causal-
ity, interactions, values, and telling (Sluzki, 1992), all of which are associated with restricting
narratives. For instance, clients usually attribute their psychological and ⁄or family problems to
external factors (‘‘that boy has always been trouble’’) and ⁄or to internal factors (‘‘I am a bad
mother’’; causality), and therapy aims at enhancing a more relational way of thinking. The
dreamer of the red balloon presented the self as incompetent (telling), and the reconstruction of
the self-narrative would need to achieve a shift toward competence.
A therapist and client may be working with many restricting themes in mind, all of which
affect the ways that the client relates to the self and to others. A variety of such themes may be
elicited from one or more dreams of the same dreamer. For instance, other themes that may be
recognized in the dream stories of the red balloon dreamer include that of being passive (vic-
tim) versus active (agent; telling), and the theme of feeling legitimate or illegitimate in her need
for privacy (values).
Middle phases: Challenging and revising restricting themes. Clients are encouraged to
deconstruct and reconstruct the self-narrative. As Gergen (1994) has put it, ‘‘the more capable
we are in constructing and reconstructing our self-narrative, the more broadly capable we are
in eﬀective relationships’’ (pp. 202–203). The self is seen as consisting of many parts or voices
(Penn & Frankfurt, 1994). Internalized parental and societal voices are associated with restrict-
ing themes. These voices are challenged and alternative voices (including those of the therapist)
are strengthened to encourage an internal dialogue in the place of what sounds like an internal
monologue. Ultimately, the person is encouraged to ﬁnd ‘‘her’’ voice, an ‘‘authorial voice’’
(Parry, 1991) that decides how to act or which character to be in a particular context (see also
As a result of the deconstruction process and the challenging of restricting themes, clients
experience an unpleasant sense of aching void. Dreamers may ﬁnd themselves in an abandoned
ship in the middle of the ocean, or in between an old and a new house. One client dreamed that
she was trapped in the ventilation system of the psychotherapy center! In the process of recon-
struction, where restricting themes are revised, clients often experience at ﬁrst a sense of anxiety
about ‘‘who they are becoming.’’ For example, a dreamer was negotiating with her plastic
surgeon the extent of her face lift. Clients also wonder to what extent and direction these
changes will affect their relationships, as with the dreamer of the red balloon and her attempts
for privacy. She dreamed that she was living in a tree house and was tidying it up to make it
cozy. She felt guilty for not wishing to join a festivity that was taking place in the main house.
Notice the similarities between a balloon and a tree house!
Toward the end of this phase, clients report a sense of freedom. For example, a dreamer
realized that some wires had been planted into his head, same as those found in the head of his
godmother. He was afraid of the pain he might experience if he pulled them out, but he did so
and felt an immense sense of freedom. It was at this stage that the dreamer of the red balloon
had the dream of conﬁdently preparing her own balloon to get oﬀ the ground for a contest and
thinking that she would win. Following that, she recently told of a dream where she was ﬂying
and experiencing freedom.
Late phase: Maintaining a revising stance. In the late phase, it is hoped that clients have
learned to continuously revise stories within their self-narrative or to make sense of experiences
in a way that allows them to tell satisfactory stories regarding relationships, both with the self
and with others. The voice of the therapist has been internalized and has become part of the
internal dialogue. A dreamer saw that his therapist’s practice was situated in the dreamer’s
home. The dreamer could open the door and consult him at any time. Another dreamer saw
that her therapist was a doctor who had taught her to self-examine and was needed no more.
484 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011
Beyond therapy: Saying goodbye. Through dream reports, clients often point to topics that
they will probably be working through on their own after therapy completion. In other words,
they point to aspects of restricting themes that are persisting. A few days before completing
therapy, a client dreamed that she was saying goodbye to group members of her therapy group.
They were all gathered for a celebration. She then walked into a shopping mall but could not
ﬁnd the ﬂoor she was looking for and would not ask anyone for help. Upon sharing her dream,
group members warned her not to go back to her old ways of never sharing her difﬁculties.
The therapist approaches dream stories with no preconceptions about the meaning of met-
aphors, but from the basis that dream content is linked to current concerns, some aspects of
which are not given voice in waking. Dream content depicts restricting themes but also open-
ings in self-narratives. The metaphorical language of dreams makes the personal meaning and
implications of such themes more obvious to both clients and therapists.
Dream material comes up spontaneously in a therapy session and is discussed as any other
story, with the therapist’s input limited to questioning and making tentative suggestions. Partic-
ularly relevant for the process of working with dream stories is the ‘‘blueprint of the therapeutic
encounter’’ provided by Sluzki (1992), describing a session process as episodes of ‘‘framing the
encounter,’’ ‘‘eliciting and enacting the dominant stories’’ (restricting themes), ‘‘favoring alter-
native stories or relations between stories,’’ and ‘‘enhancing the new stories and anchoring the
new stories.’’ I see these episodes as micro-phases within a session, as opposed to the psycho-
therapy macro-phases described above.
Theoretical Issues in Working With Dreams
Progress. We saw that dreams can be linked to early, middle, and late therapy macro-
phases associated with recognizing, challenging, revising, and maintaining a revising stance. It
is further suggested that dream stories can be used to trace, facilitate, and evaluate the process
of reconstructing self-narratives. A detailed ‘‘analysis’’ of a dream is not the aim. Elements of
the dream may be kept in mind to be used when they appear relevant to a dialogue. It is
important that the therapist’s tentative suggestions derive directly from the client’s narratives to
prevent what Butler and Bird (2000) have referred to as harmful struggle in therapy and what
most therapists would call ‘‘resistance.’’ A dream example: In waking, a woman client refuses
to openly discuss her fears about her partner’s antisocial behaviors. At a particular moment,
though, she tells about a dream where her partner is hiding a corpse in a closet. The telling of
the dream indicates that the client may be more open to alternative stories. Sometimes clients
need encouragement and the therapist can bring up dream material to indicate how the self-
narrative is being reconstructed (where this dream story diﬀers from the previous).
At the time when the dreamer of the red balloon dreamed that she was conﬁdently prepar-
ing her own balloon to get oﬀ the ground, all three restricting themes recognized in her self-
narrative had been substantially revised: competency, agency, and legitimate privacy. But
revising continued further: in a dream following that, she was the protagonist of a commercial
and needed to dive into a swimming pool. The director decided to change the script and asked
her to dive into the sea instead. She refused adamantly, reminding him that she is pregnant
(also in real life) and saying that her priority from now on is to take care of herself and the
baby (agency). Remember the inevitable fall into deep waters in the early red balloon dream.
Other times dreams may come as an encouragement to the therapist when she is worried that
the client’s self-narrative is stuck.
Therapeutic relationship. The therapeutic relationship is an important factor in dream
work. Zack and Hill (1998) examined the relationship between type of dream and outcome of
single-session interpretation with an unfamiliar therapist. When the dreams were extremely
unpleasant, the outcome of the session was least satisfactory. The importance of the therapeutic
relationship is particularly evident in dreams where the therapist is also part of them. The client
may choose to report such a dream as a safe vehicle to express (a) how she feels about the ther-
apist and her role, (b) how she perceives and what she feels about the therapeutic procedure,
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 485
and (c) how she feels about her own ability to continue her therapy. This information should
be taken into account. Another dream example: A client saw her therapist dressed as a fairy
godmother inviting her to enter a ballroom. She was happy in the dream. In waking, her feel-
ings were also very positive toward her therapist and the therapeutic procedure, but how realis-
tic were they? Were changes felt as temporary and as subject to her therapist’s magic tricks or
did she trust herself in dealing with life challenges on her own? These types of questions are
put forward as part of the therapeutic dialogue.
Transference. Sometimes therapists choose to see their participation in the dream story as
an indication of ‘‘transference.’’ For instance, they may take the stance that the dream is never
really about them, but about parental ﬁgures. Andersen and Miranda (2000) have designed
experiments to investigate whether ‘‘transference’’ can be empirically supported. Among other
interesting things, they found that when encountering a new person who resembles a signiﬁcant
other, we tend to feel about that new person in the same way that we feel about the signiﬁcant
other. They also found that we experience a similar desire for emotional closeness or distance,
and we expect to be accepted or rejected in the same way that the signiﬁcant other is experi-
enced as treating us. From a narrative perspective, it seems that we tend to recruit unfamiliar
persons to participate in familiar relationship stories and expect them to play familiar roles.
In that frame, therapists constitute a safe place for rehearsing the expression of unsaid
emotions, even if these will eventually need to be expressed elsewhere. A client reported a
dream where her therapist was painting her nails while she was bleeding. She had been sexually
abused in childhood and her mother had ignored all signs of that. One can assume that she
was angry with her mother but did not express it. The dream may be depicting how she feels
toward the mother, but it also made her therapist think whether she needed more support and
attention from her, but could not express it.
Sometimes, therapists recruit clients to participate in their own dream stories. The ‘‘choice’’
of clients to recruit may be based on the commonality of restricting themes found in the stories
of both therapist and client. I offer as an example one of my dreams, where the red balloon
dreamer had sneaked into my bedroom and had opened my drawers. This was a scene bor-
rowed from my client’s waking stories involving her mother as the intruder, and it obviously
had to do with my own struggle with the theme of ‘‘privacy.’’
Dream Stories and Relating
Dream stories can be seen as metaphors of the ways clients relate not only to family mem-
bers, but also to other social groups (including group therapy members) and to the self.
Relating to the family. Dream stories indicate how clients struggle to make sense of family
relationships. The dreamer of the red balloon was accused by her mother for having her head
in the clouds and for having cut oﬀ family ties. The young woman loved her family but also felt
uncomfortable with them always functioning as one. She could not directly communicate her
need for privacy, which she also feared.
The way that family members metaphorically depict their relationship in their dreams pro-
vides valuable material to the family therapist. An empirical study by Kaplan et al. (1981),
mentioned earlier, revealed that a researcher familiar with the systemic approach—but with no
access to the members of families—was able to accurately evaluate the level of functionality of
ﬁve families on the basis of their dream series only (a total of 126 dreams). A family therapist
is in the privileged position of having access to both the family and their dream contents.
Relating to the self. Dream stories are also metaphors of the ways different parts of a per-
son relate in waking. This is a point we will now elaborate on. Here is a dream of a woman in
her 20s, very early in her therapy. It consists of two episodes.
•She is surprised to see her photograph on the cover of a magazine. She likes seeing
her picture on a cover but does not like the particular picture. She looks very serious
and unfashionable with brown hair that she dislikes. Her name is printed wrong on
•Back home, her father has already seen the magazine cover. He shows her previous
issues, most of which have a photograph of hers on the cover. She does not recognize
herself. Half the covers depict a woman in her 30s with dark hair and exotic beauty.
486 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011
The others depict a woman in her 40s, blond and elegant, who looks a bit like her.
She keeps staring at them.
Notice the way this woman is trying to make sense of herself, her self-image, and particu-
larly her femininity. Some parts of her appear strange. If she could connect to these parts she
would probably feel more comfortable and enjoy the attention from others. Her father is the
person who brings her in touch with these parts. When working with the dream, the young
woman mentioned that she would have wanted her father to play a more active role both in
the dream and in her waking in voicing her femininity, instead of always praising her for aca-
demic achievements. She wanted her father’s consent in becoming not only reﬁned (like the
elegant woman) but also more outgoing (like the exotic woman). The client told her father
about the dream and that brought them to a family session with a speciﬁc agenda.
Procedures in Working With Dreams
Feelings. Clients are generally encouraged to tell dreams that they thought were important
or that they remember intensely. The therapist’s interest in dreams encourages clients to remem-
ber and share this material spontaneously.
Clients are asked to tell their dreams in the present tense. While they narrate, the therapist
may keep in mind the dream story’s structure—exposition, peripeteia, crisis, lysis ⁄catastrophe,
or pending lysis—and ask about feelings that may be associated with each part. In the red bal-
loon dream, the dreamer reported no feelings at the exposition and peripeteia phase, relief at
the crisis phase, and slight hope at the resolution (pending) phase. This may be contrary to
what one would expect. Knowing the client and looking at metaphors and idioms can help the
therapist make tentative suggestions about aspects of the client’s experience that have not been
put into words. In the red balloon example, the translation of images into idioms (up in the
air, ties in knots, etc.) suggested that the client felt uncomfortable and uncertain. However,
these feelings were warded off. At the same time, hopefulness and optimism at the resolution
phase were emotions that one would, in fact, expect from a woman who did not give up and
who did return to her therapy after dropping out at the early phase. But the client could not
recognize any of these emotions as being part of her at the time. Recognizing emotions and
feeling the restricting nature of certain themes encourages clients to look for and sustain alter-
Revising techniques. The structure is also kept in mind when the therapist asks the dreamer
to think whether she would like to make any changes in the story. Some dreamers may choose
to change the end and others may choose to intervene earlier to change the story. It is impor-
tant to remind them that it is their dream and that they can retell the story in any way they
wish. According to Hill (2004), this task facilitates a sense of empowerment (or authorship),
especially in the case of nightmares and recurrent dreams (see also Cushway & Sewell, 1992),
and may also be a fun and creative way to get someone started in thinking of making changes
in waking. Hill presents this task also as a way of assessing readiness for change. Clients who
are unwilling to play with the dream may not be willing to make changes in their lives either.
In my view, this unwillingness reﬂects their pessimistic outlook to life, possibly their depression,
and it is also evident in exercises where clients are asked to write how the story of ﬁctional
characters will develop (Androutsopoulou, 2001b).
In group therapy sessions, members may provide their input to aid the process of generat-
ing alternative stories and of revising and to increase the sense of authorship. The therapist
asks group members (a) ‘‘what they would think and ⁄or feel if that were their own dream and
tell it in the ﬁrst person’’ (‘‘If that were my dream I would . . .’’). She can also ask them (b)
‘‘what part of the dream they might want to change and how they would choose to rewrite it,’’
perhaps by also including other dream characters who could assist them. Finally, the therapist
can ask the dreamer and the group members (c) ‘‘how they think the dream protagonist(s) will
live from now on.’’ From this brainstorming, the dreamer may then choose ideas that best ﬁt
her and share them with others.
In family sessions, all of the above techniques are also applicable. Alternatively, all family
members can be encouraged to participate in an exercise and take turns in sharing dreams.
In cases where one member tells a dream spontaneously, other family members may be
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 487
encouraged to share memories of similar dreams they had in the past. These techniques use
dream telling as a safe vehicle to express emotions that remain unsaid because they are thought
as potentially disruptive for relationships. Sharing dream stories is a way of giving each other
permission to talk about ‘‘dangerous topics’’ and of realizing similarities in the way they feel
Element of surprise. Very often dreamers have their own ideas about what a dream
‘‘means.’’ Unless they have been brought up in a cultural environment that sees dreams as
supernatural (divine and prophetic), they will normally explain dreams as associated with their
current concerns. Sometimes, the therapist or family ⁄group members may have nothing else to
add to what the client herself has already thought about the dream. Surprise is the key element.
When the client is surprised about some element of her dream, at least one aspect of her un-
noticed experiences has been noticed and an opening has been provided to new, alternative
stories. The dreamer of the red balloon was surprised in her later dreams at how conﬁdently
she was setting off for a contest and at how conﬁdently she was protesting against swimming in
the sea for shooting a commercial. In this latter dream, the diving into the sea while pregnant
was frightening, an emotion that was warded off in the ﬁrst dream where she fell from the sky
into deep waters.
Anchoring rituals. Titles summarize important issues for the client, and if put in a
sequence, they can indicate the process of revising themes and reconstructing a self-narrative.
Clients are encouraged to choose their own titles as a way of ‘‘anchoring new stories’’ (Sluzki,
1992). The dreamer of the magazine covers chose the title ‘‘various portraits of me,’’ which
eloquently summarized her understanding of what she was working on both in her waking and
in her dreams. Also, dream objects can be used as summaries and ⁄or reminders of a client’s
work in therapy. The dreamer of the red balloon bought such a balloon as a lampshade for her
newborn baby’s bedroom.
In the mind of most family therapists, dream work is mostly associated with modern indi-
vidualistic approaches to therapy, especially psychoanalytic. However, story-generating methods
and techniques are not modern or postmodern per se. Efran and Clarﬁeld (1992) have criticized
attempts to consider certain methods as only ‘‘appropriate’’ for therapists who wish their work
to be informed by constructionist ideas. Providing a coherent narrative framework for story-
generating methods and technique allows family therapists to combine an interest in narrative
with favorite ways of working. In this article, dream stories are seen as part of self-narratives
and as aiming to make sense of personal experiences. ‘‘Interpretation’’ is best understood as
tentative eﬀorts for meaning-making of the dream story.
Dream telling is a ‘‘safe’’ way of expressing concerns and accompanying emotions that are
left unnoticed and ⁄or warded off in waking. Dream stories are approached with no preconcep-
tions regarding the meaning of metaphors, but with the aim to help the dreamer link dreams to
current concerns and generate ideas on which aspects of his waking experience are not given
voice in waking.
In general, dream reports are used to supplement an ongoing therapeutic dialogue aiming
to recognize, challenge, and revise restricting themes. Recognizing themes is usually easier and
more creative when working with dream stories and metaphors. Therapists can link dreams
with therapy phases to pinpoint progress and to help clients monitor with them how self-narra-
tives are gradually being reconstructed. Narrative-informed family therapists working with
groups can recruit group members as ‘‘audiences’’ to help generate alternative stories and vali-
date new ones. Those working mostly with families can work with dreams in family sessions.
Family members share dreams and notice common themes in their dreams (how they may be
alike). They give each other permission to express ‘‘dangerous’’ emotions through the safety of
dream telling. The way that family members metaphorically depict their relationship in their
dreams provides valuable material. Finally, dreams reﬂect not only the ways that persons relate
to their social groups, but also the way that different parts within them relate to each other in
488 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011
Andersen, S. M., & Miranda, R. (2000). Transference: How past relationships emerge in the present. The Psy-
Anderson, H. (1999). Reimagining family therapy: Reﬂections on Minuchin’s invisible family. Journal of Marital
and Family Therapy,25, 1–8.
Andrews, J., Clark, D. J., & Zinker, J. C. (1988). Accessing transgenerational themes through dreamwork. Jour-
nal of Marital and Family Therapy,14, 15–27.
Androutsopoulou, A. (2001a). The self-characterization as narrative tool: Applications in therapy with individuals
and families. Family Process,40, 79–94.
Androutsopoulou, A. (2001b). Fiction as an aid to therapy: A narrative and family rationale for practice. Journal
of Family Therapy,23, 278–295.
Androutsopoulou, A. (2005a). The innocent language of our dreams. Athens: Hellinika Grammata [in Greek].
Androutsopoulou, A. (2005b). Themes in nested stories: A case in family-oriented synthetic therapy. Journal of
Systemic Therapies,24, 43–59.
Androutsopoulou, A., Thanopoulou, J., Economou, E., & Baﬁti, S. (2004). Forming criteria for assessing the
coherence of clients’ life stories: A narrative study. Journal of Family Therapy,26, 383–405.
Bateson, G. (2000). Steps to an ecology of mind: Collected essays in anthropology, psychiatry, evolution and episte-
mology. Chicago: University of Chicago Press.
Boss, M. (1958). The analysis of dreams. New York: Philosophical Library.
Boss, M. (1978). I dreamt last night. . . . New York: Gardner Press.
Buchholz, M. B. (1990). Using dreams in family therapy. Journal of Family Therapy,12, 387–396.
Butler, M. H., & Bird, M. H. (2000). Narrative and interactional process for preventing harmful struggle in ther-
apy: An integrative empirical model. Journal of Marital and Family Therapy,26, 123–142.
Cariola, L. A. (2008). A structural and functional analysis of dream narratives. Dreaming,18, 16–26.
Craig, E., & Walsh, S. J. (1993). Phenomenological challenges for the clinical use of dreams. In G. Delaney (Ed.),
New directions in dream interpretation (pp. 103–154). Albany: State University of New York.
Cushway, D., & Sewell, R. (1992). Counselling with dreams and nightmares. London: Sage.
Deslauriers, D., & Cordts, J. (1995). Dreams and current concerns: A narrative co-constitutive approach. Dream-
Domhoﬀ, G. W. (1999). Drawing theoretical implications from descriptive empirical ﬁndings on dream content.
Domhoﬀ, G. W. (2003). The scientiﬁc study of dreams. Washington, DC: American Psychological Association.
Eakin, P. J. (1992). Touching the world: Reference in autobiography. Princeton, NJ: Princeton University Press.
Efran, J. S., & Clarﬁeld, L. E. (1992). Constructionist therapy: Sense and nonsense. In S. McNamee & K. J. Ger-
gen (Eds.), Therapy as social construction (pp. 200–217). London: Sage.
Feixas, G., Cunillera, C., & Mateu, C. (1990). Dream analysis in a systemic therapy case: A constructivist
approach. Journal of Strategic and Systemic Therapies,9, 55–65.
Gergen, K. J. (1994). Realities and relationships: Soundings in social construction. Cambridge, MA: Harvard Uni-
Hardy, B. (1968). Towards a poetics of ﬁction: An approach through narrative. Novel,2, 5–14.
Hill, C. E. (Ed.). (2004). Dream work in therapy: Facilitating exploration, insight and action. Washington, DC:
American Psychological Association.
Hoﬀman, L. (2002). Family therapy: An intimate history. New York: Norton.
Jung, C. G. (1970). Structure and dynamics of the psyche. (Collected Works of C. G. Jung, Vol. 8). Princeton, NJ:
Princeton University Press.
Kaplan, J., Saayman, G. S., & Faber, P. A. (1981). An investigation of the use of the nocturnal dream
reports as diagnostic indices in the assessment of family problem solving. Journal of Family Therapy,3,
Kilroe, P. A. (2000a). The dream as text, the dream as narrative. Dreaming,10, 125–137.
Kilroe, P. A. (2000b). The dream pun: What is a play on words without words? Dreaming,10, 193–209.
Knudson, R. M., Adame, A. L., & Finocan, G. M. (2006). Signiﬁcant dreams: Repositioning the self-narrative.
Kramer, M. (1991). Dream translation: A non associative method for understanding the dream. Dreaming,1,
Kramer, M. (1993). Dream translation. In G. Delaney (Ed.), New directions in dream interpretation (pp. 155–194).
Albany: State University of New York.
Labov, W., & Waletsky, J. (1967). Narrative analysis: Oral versions of personal experience. In J. Helm (Ed.),
Essays on the verbal and visual arts (pp. 12–44). Seattle, WA: American Ethnological Society ⁄University of
October 2011 JOURNAL OF MARITAL AND FAMILY THERAPY 489
Lakoﬀ, G. (1993). How metaphor structures dreams: The theory of conceptual metaphor applied to dream analy-
sis. Dreaming,3, 77–98.
Mazzoni, G. A. L., & Loftus, E. (1998). Dreaming, believing and remembering. In J. de Rivera & T. Sarbin
(Eds.), Believed-in-imaginings: The narrative construction of reality (pp. 145–156). Washington, DC: Ameri-
can Psychological Association.
McLeod, J. (1997). Narrative and psychotherapy. London: Sage.
Papp, P., & Imber-Black, E. (1996). Family themes: Transmission and transformation. Family Process,35, 5–20.
Parry, A. (1991). A universe of stories. Family Process,30, 37–54.
Parry, A., & Doan, R. E. (1994). Story re-visions: Narrative therapy in the postmodern world. New York: Guilford
Penn, P., & Frankfurt, M. (1994). Creating a participant text: Writing, multiple voices, narrative multiplicity.
Family Process,33, 217–231.
Rycroft, C. (1991). The innocence of dreams. London: Hogarth Press.
Sanders, C. M. (1994). ‘We are the stuﬀ that dreams are made on’: The use of dreams in systemic therapy. Jour-
nal of Family Therapy,16, 367–381.
Sarbin, T. (1986). The narrative as a root metaphor for psychology. In T. Sarbin (Ed.), Narrative psychology: The
storied nature of human conduct (pp. 3–21). New York: Preager.
Scharﬀ, D. E. (1992). Reﬁning the object and reclaiming the self. New York: Jason Aronson.
Sluzki, C. (1992). Transformations: A blueprint for narrative changes in therapy. Family Process,31, 217–230.
Solms, M. (1997). The neuropsychology of dreams. Mahwah, NJ: Lawrence Erlbaum.
Solms, M. (2000). Freudian dream theory today. The Psychologist,13, 618–619.
States, B. O. (1994). Authorship in dreams and ﬁctions. Dreaming,4, 237–253.
White, M., & Epston, D. (1990). Narrative means to therapeutic ends. New York: Norton.
Zack, J. S., & Hill, C. E. (1998). Predicting outcome of dream interpretation sessions by dream valence, dream
arousal, attitudes toward dreams, and waking life stress. Dreaming,8, 169–185.
Zimmerman, J. L., & Dickerson, V. C. (1994). Using a narrative metaphor: Implications for theory and clinical
practice. Family Process,33, 233–246.
490 JOURNAL OF MARITAL AND FAMILY THERAPY October 2011