bs_bs_banner

Journal of Family Therapy (2012) 34: 138–156

doi: 10.1111/j.1467-6427.2012.00587.x

The space of reflection: thirdness and triadic

relationships in family therapy

Carmel Flaskasa

This article explores ideas about threes and the triadic space of reflec-
tion. Early family therapy theory offered rich ideas about triadic rela-
tionship patterns. Contemporary practice frameworks have generated
reflective therapeutic practices and an attention to reflexivity. Mean-
while, a contemporary theory constellation allied with psychoanalysis
has studied the capacity for ‘thirdness’ in intersubjective relating, reflec-
tive functioning and mentalization, and the triadic space of thinking.
The article reviews this psychoanalytic theory, explores symbolic third-
ness alongside actual triadic relating, and maps an understanding of the
space of reflection as triadic and relational. Thinking about the space of
reflection as triadic and relational offers one way of orienting ourselves
to reflective processes and an inclusive frame for valuing both the
earlier family therapy attention to relationship patterns and the differ-
ent, current contemporary family therapy practices of reflection.

Keywords: reflection; family therapy practice theory; thirdness; triadic relationships;

reflexivity.

Introduction
This is a theoretical article about symbolic thirdness and actual threes,
and the relational space of reflection. What happens between threes is
a source of endless fascination that comes with the territory of family
therapy. Earlier family therapy theory offered rich descriptions of
triadic interactional patterns in the intimacy of family relationships, as
well as some explanations of these patterns. These understandings
have gone underground in contemporary theory, which has seen a shift
in emphasis to meaning and language from the earlier emphasis on
patterns of behaviour. A number of creative reflective practices have
been generated in these more contemporary theories (such as Milan

a
Associate Professor, Social Work, School of Social Sciences, University of New South
Wales, Sydney, 2052. E-mail: c.flaskas@unsw.edu.au.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice. Published by
Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA
02148, USA.
 Reflection, thirdness and triadic relationships 139
systemic, dialogical and narrative approaches) and the idea of reflex-
ivity has come to hold an important place in systemic thinking.
Meanwhile, in the same period as the development of contempo-
rary family therapy theory there have been theoretical developments
within psychoanalysis that may contribute to the systemic understand-
ing of reflective processes and reflexivity. These ideas include the
related concepts of mentalization and reflective functioning, and
understandings of thirdness. In terms of the latter, Ronald Britton’s
(1998, 2004) idea of the triangular space of thinking and Jessica
Benjamin’s (2004) ideas about thirdness and intersubjective relating
are particularly relevant.
My article is structured in three parts. The first part gives a discus-
sion of selected earlier systemic ideas about threes and triadic patterns
and theoretical interest in contemporary practice in reflexivity and
reflective processes. The second part maps the ideas of mentalization
and reflective functioning, and selected ideas about thirdness from
Britton and Benjamin. In the final part, I explore the resonance of
these ideas for understanding the space of reflection as triadic and
relational. This understanding potentially orients us more inclusively
to earlier ideas of practice, as well as offering a coherent way of
thinking about the range of reflective practices generated in the
different contemporary family therapy approaches.

Part 1. Selected earlier and current systemic practice

theory ideas
I will begin with a map of two sets of ideas produced in earlier and
contemporary family therapy practice theory respectively. In this first
part of the article, these earlier and current ideas will simply sit
side-by-side, in preparation for the later discussion.

Selected earlier systemic ideas about threes and triadic patterns

Triadic relationships are powerful in lived experience. They are pow-
erful in mediating intimate two-person relationships and they are
powerful in organizing and mediating larger family and relationship
constellations. Jay Haley (1989, p. 113) once estimated that even in an
unrealistically neat three-generational family of just eight people (two
children, two parents, four grandparents), there would be 56 triangles,
with each person involved in 21 triangles. The point of demarcation of
the first (and first-order) generation of family therapy practice theory

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
140 Carmel Flaskas
from other psychotherapies of the time was the move not just beyond
individual ‘one-person’ understandings of human experience, but also
beyond dyadic ‘two-person’ understandings of relationships, toward
understandings of what happens between threes – and then, through
interlocking threes, to what happens in wider systems. A number of
ways of thinking about threes was generated in the first-order
approaches (see, for example, Bowen, 1978; Guerin et al., 1996;
Hayley, 1989; James, 1989; Minuchin, 1974; Minuchin et al., 1978).
Structural therapy and strategic therapy, the dominant US first
generation of family therapy models, both attended to the ‘now’ of
family relationship patterns and privileged behaviour in thinking
about (and working with) what happens between people in intimate
relationships. The two frameworks were strongly related historically
and both were angled towards the specificity of family patterns in
particular cultural contexts. Though not in vogue in current theory,
the ideas about threes and triangles from structural family therapy
remain part of many family therapists’ practice repertoires. These ideas
include the way in which triangles can detour conflict between two or
more people or protect relationships, the idea of shifting dyadic and
triadic alliances, and cross-generational triangles and coalitions. The
pattern of triangulation, where one person comes to be located as a
third point of a triangle that serves the interests of the relationship of
the other two, has recently been named by Rudi Dallos as an enduringly
useful structural idea, which potentially forms a bridge to current
attachment theory (Dallos, 2010). Jay Hayley’s development of strate-
gic therapy showed his first home within structural therapy, as he
extended the idea of cross-generational coalitions to ideas about hier-
archy and behavioural sequences (see Hayley, 1989).
Concepts in psychotherapy attempt to describe processes, but they
can also attempt to explain them, and in the process they also con-
struct the manner in which we then orient and relate to particular
aspects of human experience. Sally Young (2010, p. 93) notes that the
concept of triangulation is ‘perhaps the systemic version of the chal-
lenges of the oedipal complex’. She also notes that while structural
ideas focused pragmatically on problematic triangles, Murray
Bowen’s theory places triangles and triangulation at the centre of
ordinary human relating in families and groups. Bowen’s approach
was also developed in the USA (see Bowen, 1978). His ideas have an
explanatory as well as descriptive force about triangular relational
patterns, while his background as a psychoanalyst flavours his theory
even as it significantly departs from the psychoanalysis of his time.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 141
Thus, though it is true that Bowen places triangles and triangulation
as an inevitable part of human relating, his understandings about
triangles locate them primarily in relation to dyadic relationships, so
that what happens in triangles is theorized in terms of what happens
in the intimacy and intensity of relationships between twos. In this
sense, while Bowen’s ideas are undeniably systemic in terms of theo-
rizing contexts and patterns, his theory places two-person relation-
ships at the core of human relating.
Bowen argued that intimacy in relationship generates anxiety –
anxiety that the self will be lost in the relationship, or that relationship
will be lost in the need to hold onto the self. Three seemingly eternal
patterns are generated by this anxiety – fusion in a relationship, in
which the self becomes sacrificed for the intimate connection; rela-
tionship cut-off, in which an intimate connection is sacrificed to pre-
serve the self and triangulation, in which a third person (or point) is
used to mediate and stabilize the anxiety in the two-person relation-
ship. The emotional task of differentiation – containing anxiety,
holding onto a sense of self while being able to stay connected in
intimate relating to another, without drawing in a third – becomes
foundational to intimate relating in the Bowen framework (see, for
example, Bowen, 1978 and the descriptions given in Brown, 2008;
Papero, 2000; Titelman, 1998).

Selected current systemic practice theory ideas: reflexivity and

reflective practices
The ideas that have just been laid out emerged in 1960–1970. The
second generation of practice theory emerged during the 1980s, a
transitional decade in family therapy between first-order and second-
order approaches, and between modernism and postmodernism. The
Milan systemic and narrative frameworks and the dialogical
approach1 all made a definitive shift from a focus on behavioural
patterns to meaning and language (Flaskas, 2010). These three con-
temporary theories of practice have a shared interest in reflexivity and
reflective practices as part of a landscape shaped by second-order and
postmodernist sensibilities.

1
The dialogical approach in family therapy practice theory includes the contributions of
Harlene Anderson (1997), Tom Andersen et al. (1987, 1991) and the work of Jaako Seikkula
(1993, 2008; Seikkula and Arnkil, 2006; Seikkula and Trimble, 2003) and Peter Rober (2002,
2005). Dialogue is placed at the centre of the practice theory of this group. See Flaskas (2011)
for a discussion of the emergence of this approach in family therapy.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
142 Carmel Flaskas
There are different layers to the idea of reflexivity and the practices
it generates. Inga-Britt Krause (2012a) gives a fascinating analysis of
the history and nuances of reflexivity in the context of tracking
culture and subjectivity. She notes that a concern with reflexivity has
been present throughout the history of family therapy theory, albeit
one expressed in different ideas – here she includes the earlier sys-
temic ideas of schismogenesis, feedback, ecology and joining, as well
as the later attention to curiosity, circularity, observing systems, reflect-
ing teams and dialogue. She suggests that
these are all ways in which systemic psychotherapists have captured the
notion that differences, which make up a thought, a feeling, a meaning,
an action, a relationship, a dialogue, a communication, a pattern or a
process, are turned back or turn back on the subject or subjects in such
a way that the relationship, dialogue, communication, thought, action
etc. is maintained or changed. (Krause, 2012a, p. xxvi)
In another recent illuminating discussion on the relationship between
reflexivity and reflective practice, Mary Donovan singles out Karl
Tomm’s 1980s elaboration of a reflexive process as one in which ‘one
is both performing, and at the same time an audience to one’s per-
formance’, as well as the idea of the ‘making of connections between
different levels of meaning in the family system’ (Donovan, 2009, p.
158). Donovan’s discussion of Tomm’s work draws attention to two
layers in the thinking about reflexivity. The first layer addresses self-
reflexivity on the part of the therapist, which John Burnham also
describes as
the process in which a therapist makes, takes, or grasps an opportunity
to observe, listen to, and question the effects of their practice, then use
their responses ... to decide ‘how to go on’. (Burnham, 2005, p. 3)
The second layer addresses the recursiveness of different levels of
experience in human relating, which in the therapeutic process
becomes another way of orienting to the family’s experience.
A third layer addresses the process in the therapeutic relationship,
which encompasses self-reflexivity on the part of the therapist and the
spin-offs in terms of how you might think about your own listening
and talking in dialogue with clients but with the additional idea of
therapeutic relating emerging in the space between therapist and
family (Flaskas, 2005). In the dialogical tradition this is the space of
dialogue. The third layer in the idea of reflexivity, then, is the rela-
tionship generated in the space between, which challenges a two-way
linear thinking about the therapeutic relationship.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 143
From even this brief description of reflexivity, one is aware of the
reach of second-order ideas and postmodernism. It is also easy to be
aware of the relationship between the orientation to reflexivity and
the momentum for the development of reflecting practices. Thinking
about reflexivity takes us directly to the territory of reflection and to
a whole range of Milan systemic, dialogical and narrative practices
that have emerged over the last two decades.
In Milan systemic theory practices of reflection are embedded in
circularity, circular questioning, curiosity, end-of-session opinions and
the three-way relational form of Milan hypothesizing, as well as Paolo
Bertrando’s more recent idea of hypothesizing-as-dialogue (Ber-
trando, 2007). From the dialogical approach, one would include the
practices of listening and being-with in Harlene Anderson’s earlier
collaborative language systems therapy (see Anderson, 1997) and in
the current work of Jaako Seikkula et al. (see, for example, Seikkula,
1993, 2008; Seikkula and Arnkil, 2006; Seikkula and Trimble, 2003),
and Peter Rober (see, for example, 2002, 2005). The foundational
contribution of Tom Andersen and colleagues in generating reflecting
practices, including the reflecting team, cannot be underestimated, as
these developments influenced the entire systemic field (Andersen,
1987, 1991, 2007). Krause chooses an interesting definition by
Andersen about his use of the term ‘reflection’:
The new format became known as the reflecting team. We thought of the
French meaning of the word, not of the English one, which in our
understanding comes close to replication. The French ‘réflexion’, having
the same meaning as the Norwegian ‘refleksjon’, means: something heard
is taken in and thought about before a response is given. (Andersen,
1991, pp. 12–13, quoted in Krause, 2012b, p. 6)

Meanwhile, a number of creative reflective practices have also

emerged from the developments of narrative therapy, mainly struc-
tured in the form of witnessing – from witnessing using letters and
other documents, to rites-of-passage and definitional ceremonies, to
witnessing groups in therapy and community work, through to
‘taking-it-back’ practices (Madigan, 2007; White, 2000, 2007; White
and Epston, 1990).
Yet while the form of these particular practices may be specific to
the tradition of family therapy, the process of reflecting about the self
and the other and relationship is generic to all psychotherapy, for all
psychotherapy is a process of making meaning about experience. I
will turn now to ideas that have been generated in psychoanalysis and

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
144 Carmel Flaskas
attachment theory that speak to the territory of reflection and reflec-
tive processes. It should come as no surprise that these ideas emerged
in roughly the same time period – from the 1980s to the time of
writing: they breathe the same air of the latter part of the twentieth
century and a wider environment of postmodernist sensibilities.

Part 2. Selected ideas allied with psychoanalysis

In a recent collection exploring contemporary integrations of psycho-
analytic ideas in the systemic field, the introduction (Flaskas and
Pocock, 2009, pp. xix–xx) notes that no fewer than eight out of the
twelve chapters draw on ideas of mentalization from Peter Fonagy and
colleagues and that the related theory of reflective functioning, as well
as particular ideas from Jessica Benjamin and Ronald Britton, all
feature a number of times. I will draw from this range, with a view to
layering ideas about thirdness that can then be taken back to systemic
concerns.

Mentalization and reflective functioning2

The concepts of mentalization and reflective functioning track the
same territory of human experience and lie in the borderland
between psychoanalysis and attachment theory, emerging concur-
rently in both fields (Fonagy, 2001). Knowledge about the process of
mentalizing has been generated in the empirical research and theory
of psychoanalyst Peter Fonagy and colleagues (Bateman and Fonagy,
2006; Fonagy, 2001; Fonagy et al., 1991). Mentalizing describes the
process of thinking about and understanding human behaviour (our
own and other people’s) in terms of ‘mental states’, which presupposes
that we are thinking about the ‘mind’ in understanding human expe-
rience and in mediating our relationship with the world. The idea of
‘mental states’ is framed broadly, as Fonagy notes that mental states
may relate to beliefs, attitudes, hopes, knowledge or even desires or
plans (Fonagy, 2001, p. 165). The content of how we might represent
our own or other people’s minds and mental states with respect to our
behaviour-in-relationships involves both meaning and emotion. In
this sense, it is artificial to separate out the individual ‘mind’ from

2
This discussion of mentalization and reflective functioning draws on fuller discussions in
Flaskas (2002, 2009).

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 145
representations of the relational world within which the behaviour
and relating and ‘mind’ is contextualized.
Mentalizing is not a cognitive activity and, as Bateman and Fonagy
(2006, p. 1) stress, it is a ‘mostly preconscious, imaginative mental
activity’ – we spontaneously imagine what others may be thinking or
feeling, or what we ourselves may have been thinking and feeling in a
particular situation or interaction or relationship. It is also a symbolic
process because mentalizing is all about our representations, and not
in the realm of mirroring concrete ‘facts’. Mentalizing may be thought
of as a prerequisite to the ability to reflect on our relationships and on
other people’s involvement in relationships as well as our own involve-
ment in relationships. In this way, mentalization, and being able to
mentalize, is central to the processes of psychological and relational
meaning-making about the self, the other and relationships, and goes
alongside the ability to differentiate between the self and the other
(Flaskas, 2009).
The idea of reflective functioning has emerged more squarely in
attachment theory and research and it fits closely with mentalization,
while being angled more directly towards the conditions of human
development. Within the context of attachment research, reflective
functioning has been used to describe a capacity that emerges in the
context of the ordinary relational process between parents and babies
or small children: parents wonder about their babies and they keep
trying to make some sense of the baby’s experience which is not yet in
the world of worded language, and they respond from that wonder-
ing in the way they relate to and physically hold and care for the baby.
Reflective functioning is all about this capacity to wonder about the
other and to hold the other in mind in the way that we then relate to
that other. The idea here is that the capacity to think about the
other-as-other and about our own self-in-relationship emerges in the
developmental context of a relationship and in which someone is able
to think about us and hold us in mind (Flaskas, 2002, 2009).
The focus of this article does not allow a detour to other intersec-
tions that relate to the theory of mentalization and reflective function-
ing. However, I will point to the significance of Wilfred Bion’s theory
about thinking, which converges quite powerfully with the develop-
mental understandings of reflective functioning and resonates
strongly with the idea of mentalizing, while also having significant
points of difference to it (see, for example, the discussion in Holmes,
2006). I will also point to the research on the development of narra-
tive coherence, which would seem to be one of the most powerful

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
146 Carmel Flaskas
transformative factors in intergenerational patterns of attachment as
well as a significant aspect of the development of resilience (Main,
1991; Walsh, 2006). The process of constructing a coherent narrative
about one’s own experience of early relationships would seem to rely
on the capacity of reflective functioning.

On thirdness and intersubjectivity

We come then to selected ideas about thirdness and intersubjectivity
in psychoanalytic theory; the last set of ideas to be explored before
returning to family therapy theory. Contemporary ideas of thirdness
have in their genealogy Freud’s idea of the Oedipus complex: the
powerful description of the relationship between a small child, mother
and father, and the patterns of unconscious phantasy and meaning
that foundationally shape intra-psychic experience and relating. Also
part of this genealogy is Donald Winnicott’s notion of the third as the
transitional/potential space between, which allows us the freedom of
symbolic play. In contemporary post-Kleinian and intersubjective
traditions, theorizing the oedipal triangle has been transformed into
less concrete understandings of what might constitute foundational
threes, and the importance of symbolic thirdness in thinking about
and relating to the self and the other has emerged. There is a rich field
of theory about thirdness (see, for example, the overview given in
Diamond, 2007) and I will select just two areas of this theory – Ronald
Britton’s concept of the triangular space of thinking, and selected
themes on thirdness from Jessica Benjamin’s work in intersubjectivity
theory.
Ronald Britton is a British analyst in the post-Kleinian and Bion
tradition. He takes the oedipal situation – a triadic relationship in
which one person has a relationship with each of two other people,
while any twosome in that triangle also have a relationship that is
separate and apart from the third person. It is this formative rela-
tional context that provides what Britton calls a ‘limiting boundary’
for the internal world, a limiting boundary that simultaneously allows
individuals the potential freedom of what he calls the triangular space
of thinking. This triangular space provides ‘the possibility of being a
participant in a relationship and observed by a third person as well as
being an observer of a relationship between two people’ (Britton,
2004, p. 47). Britton links this to the experience of the self as both
subject and object – we are able to be immersed in experience in a
subjective sense, while still being able to have a relationship to that

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 147
experience. In other words, we develop the human capacity to relate
to ourselves as both subject and as object, in the sense of being able to
witness ourselves in the experience while still being fully (subjectively)
involved in that experience.
In Britton’s thinking the lived relational context of threes, then,
provides the conditions of freedom to think and reflect on the self and
others and the self-in-relationship. But it also provides the conditions
of distinguishing between a belief or a thought you might have and a
fact. Britton writes:
I suggest that the recognition that one has a belief rather than that one is
in possession of a fact requires what I describe as a triangular psychic space
– a third position in mental space is needed from which the subjective self
can be observed having a relationship with an idea. (Britton, 1998, p. 13,
italics in original)

The triangular space of thinking allows reflection to be known as (just)

reflection, because part of reflective thinking is having a sense of the
relationship between the ‘I’ who is thinking and the thoughts that I
am having. The self-reflexivity of being able to wonder about your
relationship to your own thoughts would, in narrative terms, be one of
the conditions of fluid and open narratives.
Jessica Benjamin is an American analyst, and her understandings of
thirdness lie in the intersubjective and relational tradition of psychoa-
nalysis. Stephen Frosh notes that Benjamin’s work is ‘rooted in
feminist scholarship and critical theory, and hence oriented toward
emancipatory practice’ (Frosh, 2009, pp. 188–189). In Benjamin’s
important article ‘Beyond doer and done to: an intersubjective view of
thirdness’ (Benjamin, 2004), there are a number of points of reso-
nance with contemporary systemic theory. This article is pitched
towards exploring the conditions of transforming linear complemen-
tarity in relationships-between-two and of creating and finding the
space of mutual influence and co-creation in a relationship, which in
itself becomes the co-created ‘shared third’. One hears the strong
resonance between her work and systemic concerns when she writes:
The co-created third has the transitional quality of being both invented
and discovered. To the question of ‘Who created this pattern, you or I,’
the paradoxical answer is “Both and neither’. (Benjamin, 2004, p. 18)

Benjamin traces the legacy of Donald Winnicott’s theory of potential

or transitional space in her ideas, which shows directly in her use of
transitional space and in the idea of the shared third of relating being

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
148 Carmel Flaskas
invented and discovered. Winnicott’s legacy is also evident in her
theory of the intersubjective process of ‘surrendering’ to the third.
Here Benjamin explores the third as the intersubjective mental space
that allows a meeting with, or ‘recognition’ of the subjectivity of the
other. Frosh (2009) notes the potential significance of this theory for
understandings of culture, as Benjamin speaks of recognition as
‘being able to sustain connectedness to the mind of the other’s mind
while accepting ... separateness and difference’ (Benjamin, 2004, p. 2).
She explores the process of ‘surrendering’ in the process of this
recognition, by referring to the analyst Emmanuel Ghent’s idea of
surrendering. She says as she outlines her idea in relation to that of
Ghent:
The crucial point was that surrender is not to someone. From this point
follows the distinction between giving in or giving over to someone.... And
letting go in to being with them. I take this to mean that surrender
requires a third, that we follow some principle or process that mediates
between self and other. (Benjamin, 2004, p. 2, italics in original).
While dialogical practice theory comes to family therapy via a very
different theoretical heritage, one is aware of shared territory in
attending to the intersubjective process of being-with, which, in Ben-
jamin’s work is theorized in terms of both needing and creating
thirdness, and in dialogical theory generates the therapeutic practices
of being-with/‘withness’ (Hoffman, 2007).
Part 3. Reflection as a triadic relational space: inclusive
orientations to systemic theory and practices
How might these psychoanalytic theories be used in thinking about
systemic family therapy theory? This final part of the article is divided
in three parts: firstly, noting theory resonances and shared territories
and the fit of the idea of the triadic relational space of reflection;
second, noting the opportunities of the therapeutic space of actual
threes created in the environments of family therapy practice and, in
conclusion, drawing out the way in which the idea of reflection as a
triadic relational space offers one vantage point for an inclusive ori-
entation to both earlier theory of practice and current differences in
family therapy theory.

Resonance, shared territory and the triadic relational space of reflection

There are strong points of resonance and connection in the different
generations of family therapy practice theory; between the earlier

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 149
ideas, for example, of triangulation and differentiation, and between
the reflective practices in Milan, dialogical and narrative theory and
the shared emphasis on reflexivity. However, there has been a notable
lack of connection between the earlier theory on threes and our later
theory; a disjunction that gave rise to this article. Meanwhile, in terms
of the resonance and connection in the selected current ideas from
attachment and psychoanalysis, mentalization and reflective function-
ing are closely related concepts, and rest in a collection of ideas of
thirdness in relation to theory of mind and intersubjective processes,
including Britton’s idea of the triangular space of thinking and
Benjamin’s understandings of thirdness.
Links have been drawn in recent contributions between family
therapy theory and the particular psychoanalytic ideas presented in
this article. Mary Donovan has highlighted both the shared challenges
in systemic and psychoanalytic practice and the resonance and differ-
ence between reflecting processes and reflective functioning
(Donovan, 2009). In her extended explorations of culture and self,
Inga-Britt Krause has used the theory of mentalization (2002) and
recently drawn on Britton’s theory of thirdness and the development
of mind (Krause, 2009). The theory of mentalization has been part
of the work of Astri Johnsen, Rolf Sundet and Vigdis Wie Torsteinsson
in their integration of developmental theory in systemic practice
(Johnsen et al., 2003; Sundet and Torsteinsson, 2009). My own work
on meaning-making and the unconscious has also connected with
reflective functioning and mentalization (Flaskas, 2009). Benjamin’s
understandings of intersubjective processes have been addressed by
Stephen Frosh (2009), again with respect to culture. David Pocock
(2009, 2010) has drawn her ideas into his discussions of emotional
systems and the regulation of affect. Other practice and theory inter-
sections with thirdness have been offered by Jeremy Woodcock (2009)
and Paolo Bertrando (2002).
Here in this article, I will simply note the richness of the potential
intersections across the chosen sets of ideas from family therapy and
psychoanalysis. Though the theory domains are very different, it is the
attention to the common territory of the experience of relating and
being in relationships that gives the different sets of ideas such rich
resonance with each other. The exploration of a particular constella-
tion of theory in attachment and psychoanalysis leads to ideas about
the capacity for symbolic thinking about the self and the other in
relationship as triadic. This rests developmentally in actual relation-
ships. The argument is that triadic relationships create the conditions

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
150 Carmel Flaskas
in which the ‘I’ who is thinking and the thoughts that I am thinking
allow for a form of intersubjective relating that potentially meets
separateness and difference in the recognition of the other. This
constellation of ideas generates an understanding of the space of
reflection (and reflexivity) as triadic and relational. To think of the
space of reflection as triadic and relational makes a good fit with the
contemporary shaping of reflexivity and reflective practice in family
therapy theory and also holds the connection between lived experi-
ence in actual threes and reflexivity.

Family therapy practice: the beauty of the therapeutic space of actual threes
We return, then, to the lived experience of actual threes. The momen-
tum of psychoanalysis is to generate understandings of the intra-
psychic triadic space for reflection, albeit in the developmental
context of actual threes. Psychoanalytic practice is in part about invok-
ing the relational conditions of this triadic space in the intensity of the
dyadic analytic relationship. In the tradition of family therapy, we
construct and use a different kind of therapeutic relational space. As
soon as you have more than two people in the room (the client and the
therapist), then you have an immediate experience of (actual) threes
in the therapy itself. This shapes a different environment for the
therapeutic relationship and for therapeutic work, and offers different
opportunities for inviting reflection. Indeed, the history of family
therapy rests on the radical shift in practice of convening the family as
the venue of therapeutic change and forging theory to suit this dif-
ferent kind of therapy environment.
Early family therapy practice actively used people in the room.
Reflecting practices in contemporary family therapy exploit the same
potential by using the immediate lived experience of threes, inviting
people in the room to reflect (symbolically) on the self and others in a
relational context. Immediate (actual) threes are used to invite the
(symbolic) triadic space of reflection. The practices of circular question-
ing, reflecting teams and witnessing groups invite people to sit back just
a little and have their own thoughts about the descriptions and
responses other people are giving to their own experience. So a mother
listens to her partner and daughter talking about her situation, or the
therapist and the family sit back and listen to the discussion of a
reflecting team, or clients sit back and hear other people speak to the
resonance of the clients’ experience with their own lived experience, or
hear the response of friends and family to their letters documenting

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 151
their hopes for change. My point here is that these contemporary
practices create an environment for reflection by actively invoking the
here-and-now lived experience of threes and thirdness.
The beauty of family therapy as a way of working is precisely that at
least three people are convened in the therapy room. The therapeutic
relational space of three (or four, or five or more) offers different
possibilities for immediate and creative invitations to relational reflec-
tion. Having written these two sentences, I feel foolish, for what I have
just said is quite obvious, and yet it is sometimes lost. Systemic psy-
chotherapy has by now an established tradition of work with single
clients as well as with larger systems (Boscolo and Bertrando, 1996;
Hedges, 2005; Campbell et al., 1991). Context and relationship are
the enduring parameters of the systemic field and systemic theory
provides a foundation for work with individuals and organizations
as well as families. However, saying this does not detract from the
specificity of the therapeutic environment of working directly with
families and relationships. It would be a shame not to keep this
environment in the front of our heads in discussions of the theory of
family therapy practice, because creative therapeutic practices emerge
in the specificity of the therapeutic environment and not so much in
the abstractness of theory.

A vantage point for an inclusive orientation to family therapy practices

and theory
The concluding point comes back to the regret voiced at the begin-
ning of this article about the loss of earlier knowledge, combined with
a curiosity about the disjunction between first-generation and second-
generation theory with respect to the interest in threes. Across differ-
ent approaches, first-generation theory looked for and thought about
patterns of threes and triangles in the lived experience of intimate
relationship and in wider contextual relationships. In current practice
we are no longer committed to reconfigure triangles or to disentangle
dyadic relationships from triadic diversions, or to reconfigure genera-
tional boundaries in the face of cross-generational coalitions. While
the heritage of the practice interest in triangles is most embedded in
the Milan systemic principles of hypothesizing and circularity, the
current shared interest of the current Milan systemic, dialogical and
narrative approaches centres on reflective practices and reflexivity.
Meanwhile, psychoanalytic and attachment thinking has generated
understandings of reflection as a triadic and relational space.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
152 Carmel Flaskas
To think about reflection as a triadic and relational space offers a
vantage point for an inclusive orientation to a whole range of family
therapy practices and theory, and potentially re-asserts the impor-
tance of creativity in the therapy room with respect to practices of
threes and thirdness. Actively looking for and thinking about patterns
of triangles in couple and family relationships can still speak power-
fully to lived experience, and it remains an easy way of inviting
discussion of relationships beyond individual positioning. Actively
looking for and thinking about the experience of triangles within and
outside the family also provides another pathway to inviting a triadic
reflective space about the self and a relationship, and thinking about
the self and a relationship. Curiosity about the family’s experience of
actual threes links directly to contemporary theory because it becomes
another (and often more immediate) way of inviting and invoking the
triadic relational space of reflection.
Practices of reflection, the hallmark of contemporary approaches,
tend to rely heavily on language. Yet in the therapy room, family
therapists often use additional creative and playful ways of making it
easier for people to find language about relationships and the self-in-
relationships and to enter the space of relational reflection. As exam-
ples, I could think of John Burnham’s use of ‘paper play’ in depicting
relational constellations,3 or Peter Rober’s use of blocks in sessions or
postcards to invite thinking about relationships,4 or Justine Van
Lawick’s use of sculpting in her group work with couples.5 I could
point to the creativity of Jim Wilson’s work (Wilson, 2007) with chil-
dren and families and I constantly admire the largely unwritten prac-
tice wisdom of colleagues in their ‘in the room’ work with children and
young people. In fact, I have no difficulty in conjuring up a fantasy of
a huge group confession of family therapists, coming out about their
use of ‘out-of-date’ practices involving artwork, drawing, playful
extended hypothetical imaginings, and switching seats. Both in their
impetus and effect these creative practices could be thought of as an
invitation to reflective space – a space that can sometimes be more

3
As seen in John Burnham’s 2009 workshop, From paper work to paper play, hosted by
the NSW Association of Family Therapy, held in Sydney.
4
As seen in Peter Rober’s class presentations in the Master of Couple and Family Therapy
Program, University of New South Wales.
5
As seen in a video segment of Justine Van Lawick’s group work, presented as part of a
panel, On creating dialogical space in family therapy, with panel members John Shotter, Justine
Van Lawick, Peter Rober and Jim Wilson, 7th Congress of the European Association of Family
Therapy, held in Paris October 2010.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 153
easily invoked through points of focus (the blocks, the paper, the
drawing) that simultaneously allow us to distance ourselves from the
self while also connecting us to reflection on the self and its relation-
ships. To think of these creative practices as inviting and invoking the
(triadic relational) space of reflection is to locate them squarely within
the aims of our contemporary theory of practice, and also offers us the
freedom to use practice ideas from the first generation of family
therapy.
In short, the conclusion of my theory exploration is this: current
theory from psychoanalysis and attachment research generates an
understanding of the space of reflection as triadic and relational. This
understanding has the capacity to frame the commonality of the
emphasis on reflecting practices and reflexivity in our different con-
temporary theories of practice. It also offers one way of re-engaging
with earlier family therapy ideas addressing threes and triangles and
using a fuller range of creative ‘in the room’ practices. There are
many harmonious practice invitations to triadic relational reflection.

Acknowledgements
I would like to acknowledge and thank the contributors to the collec-
tion Systems and Psychoanalysis: Contemporary Integrations in Family
Therapy (Flaskas and Pocock 2009) and my co-editor David Pocock.
Reading their combined ideas sparked the connections that led to this
article.

References
Andersen, T. (1987) The reflecting team: dialogues and meta-dialogue in clinical
work. Family Process, 26: 415–426.
Andersen, T. (ed.) (1991) The reflecting team. New York: W.W. Norton.
Andersen, T. (2007) Crossroads: Tom Andersen in conversation with Pers Jensen.
In H. Anderson and P. Jensen (eds) Innovations in the Reflecting Process: The
Inspiration of Tom Andersen (pp. 158–174). London: Karnac.
Anderson, H. (1997) Conversation, Language and Possibilities: A Postmodern Approach
to Therapy. New York: Basic Books.
Bateman, A. and Fonagy, P. (2006) Mentalization-based Treatment for Borderline
Personality Disorder. Oxford: Oxford University Press.
Benjamin, J. (2004) Beyond doer and done to: an intersubjective view of third-
ness. Psychoanalytic Quarterly, 73: 5–46.
Bertrando, P. (2002) The presence of the third party: systemic therapy and
transference analysis. Journal of Family Therapy, 24: 351–401.
Bertrando, P. (2007) The Dialogical Therapist: Dialogue in Systemic Practice. London:
Karnac.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
154 Carmel Flaskas
Bowen, M. (1978) Family Therapy in Clinical Practice. New York: Jason Aronson.
Boscolo, L. and Bertrando, P. (1996) Systemic Therapy with Individuals. London:
Karnac.
Britton, R. (1998) Belief and Imagination. London: Routledge.
Britton, R. (2004) Subjectivity, objectivity and triangular space. Psychoanalytic
Quarterly, 73: 47–61.
Brown, J. (2008) We don’t need your help, but will you please fix our children?
Australian and New Zealand Journal of Family Therapy, 29: 61–69.
Burnham, J. (2005) Relational reflexivity: a tool for socially constructing thera-
peutic relationships. In C. Flaskas, B. Mason and A. Perlesz (eds) The Space
Between: Experience, Context and Process in the Therapeutic Relationship (pp. 1–17).
London: Karnac. K. Papadopoulos and J. Byng-Hall (eds) Multiple Voices:
Narrative in Systemic Family Psychotherapy (pp. 103–124). London: Duckworth.
Campbell, D., Draper, R. and Huffington C. (eds) (1991) A Systemic Approach to
Consultation. London: Karnac.
Dallos R. (2010) Attachment and systems theory: does the concept of triangulation
offer a useful bridge? Context, 110: 7–11.
Diamond, M. A. (2007) Organizational change and the analytic third: locating and
attending to the unconscious organizational psychodynamics. Psychoanalysis,
Culture and Society, 12: 142–164.
Donovan, M. (2009) Reflecting processes and reflective functioning: shared con-
cerns and challenges in systemic and psychoanalytic therapeutic practice. In C.
Flaskas and D. Pocock (eds) Systems and Psychoanalysis: Contemporary Integrations
in Family Therapy (pp. 149–165). London: Karnac.
Flaskas, C. (2002) Family Therapy Beyond Postmodernism: Practice Challenges Theory.
Hove: Brunner-Routledge.
Flaskas, C. (2005) Introduction: orienting to therapeutic relationships and the
space between. In C. Flaskas, B. Mason and A. Perlesz (eds) The Space Between:
Experience, Context and Process in the Therapeutic Relationship. London: Karnac.
Flaskas, C. (2009) Narrative, meaning-making, and the unconscious. In C. Flaskas
and D. Pocock (eds) Systems and Psychoanalysis: Contemporary Integrations in Family
Therapy (pp. 3–20). London: Karnac.
Flaskas, C. (2010) Frameworks for practice in the systemic field: Part 1:Continui-
ties and transitions in family therapy knowledge. Australian and New Zealand
Journal of Family Therapy, 31: 232–247.
Flaskas, C. (2011) Frameworks for practice in the systemic field: Part 2: Contem-
porary frameworks in family therapy. Australian and New Zealand Journal of
Family Therapy, 32: 87–108.
Flaskas, C. and Pocock, D. (2009) Introduction. In C. Flaskas and D. Pocock (eds)
Systems and Psychoanalysis: Contemporary Integrations in Family Therapy (pp. xvii–
xxvi). London: Karnac.
Fonagy, P. (2001) Attachment Theory and Psychoanalysis. New York: Other Books.
Fonagy P., Steele M., Steele H., Moran, G. and Higgitt, A. (1991) The capacity for
understanding mental states: the reflective self in parent and child and its
significance for security of attachment. Infant Mental Health Journal, 13: 200–
217.
Frosh, S. (2009) What does the Other want? In C. Flaskas and D. Pocock (eds)
Systems and Psychoanalysis: Contemporary Integrations in Family Therapy (pp. 185–
201). London: Karnac.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
 Reflection, thirdness and triadic relationships 155
Guerin, P. J., Fogarty, T. F., Fay L. F. and Kautto, J. G. (1996) Working with
Relationship Triangles: The One-Two-Three of Psychotherapy. New York: Guilford.
Haley, J. (1989) Problem Solving Therapy, 2nd edn. Jossey Bass, CA: San Francisco.
Hedges, F. (2005) An Introduction to Systemic Therapy with Individuals: A Social
Constructionist Approach. Basingstok: Palgrave Macmillan.
Hoffman, L. (2007) Practising ‘withness’: a human art. In H. Anderson and P.
Jensen (eds) Innovations in the Reflecting Process: The Inspiration of Tom Andersen
(pp. 3–15). London: Karnac.
Holmes, J. (2006) Mentalizing from a psychoanalytic perspective: What’s new? In
J. G. Allen and P. Fonagy (eds) Handbook of Mentalization-based Treatment (pp.
31–50). Chichester: John Wiley and Sons.
James, K. (1989) When twos are really threes: the triangular dance in couple
conflict. Australian and New Zealand Journal of Family Therapy, 10: 179–186.
Johnsen, A., Sundet, R. and Torsteinsson, V. W. (2003) Self in Relationships: Per-
spectives on Family Therapy from Developmental Psychology. London: Karnac.
Krause, I.-B. (2002) Culture and System in Family Therapy. London: Karnac.
Krause, I.-B. (2009) In the thick of culture: systemic and psychoanalytic ideas. In
C. Flaskas and D. Pocock (eds) Systems and Psychoanalysis: Contemporary Integra-
tions in Family Therapy (pp. 167–183). London: Karnac.
Krause, I.-B. (ed) (2012a) Culture and Reflexivity in Systemic Psychotherapy: Mutual
Perspectives. London: Karnac
Krause, I.-B. (2012b) Culture and the reflexive subject in psychotherapy. In I.-B.
Krause (ed.) Culture and Reflexivity in Systemic Psychotherapy: Mutual Perspectives
(pp. 1–38). London: Karnac.
Madigan, S. (2007) Anticipating hope within written and naming domains of
despair. In C. Flaskas, I. McCarthy and J. Sheehan (eds) Hope and Despair in
Narrative and Family Therapy (pp. 100–112). Hove: Routledge.
Main, M. (1991) Metacognitive knowledge, metacognitive monitoring, and sin-
gular (coherent) vs. multiple (incoherent) model of attachment: findings and
directions for further research. In C. M. Parkes, J. Stevenson-Hinde and P.
Marris (eds) Attachment Across the Life Cycle (pp. 127–159). London: Routledge.
Minuchin, S. (1974) Families and Family Therapy. Cambridge, MA.: Harvard Uni-
versity Press.
Minuchin, S., Rosman, B. L. and Baker L. (1978) Psychosomatic Families: Anorexia
Nervosa in Context. Cambridge, MA Harvard University Press.
Papero, D. V. (2000) The Bowen theory. In A. M. Horne (ed.) Family Counselling
and Therapy (3rd edn) (pp. 272–299). Itasca, IL: Peacock.
Pocock, D. (2009) Working with emotional systems: four new maps. In C. Flaskas
and D. Pocock (eds) Systems and Psychoanalysis: Contemporary Integrations in Family
Therapy (pp. 129–146). London: Karnac.
Pocock, D. (2010) Emotions as ecosystemic adaptions. Journal of Family Therapy, 32:
362–378.
Rober, P. (2002) Constructive hypothesizing, dialogic understanding and the
therapist’s inner conversations: some ideas about knowing and not-knowing in
the family therapy session. Journal of Marital and Family Therapy, 28: 467–478.
Rober, P. (2005) The therapist’s self in dialogical therapy: some ideas about
not-knowing and the therapist’s inner conversation. Family Process, 44: 461–475.
Seikkula, J. (1993) The aim of the work is to generate dialogue: Bakhtin and
Vygotsky in family session. Human Systems, 4: 33–48.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice
156 Carmel Flaskas
Seikkula, J. (2008) Inner and outer voices in the present moment of family and
network therapy. Journal of Family Therapy, 30: 478–491.
Seikkula, J. and Arnkil, T. E. (2006) Dialogical Meetings in Social Networks. London:
Karnac.
Seikkula, J. and Trimble, D. (2003) Healing elements of therapeutic conversation:
dialogue as an embodiment of love. Family Process, 44: 461–475.
Sundet, R. and Torsteinsson, V. W. (2009) Thinking through togetherness: devel-
opmental metaphors and systemic thinking. In C. Flaskas and D. Pocock (eds)
Systems and Psychoanalysis: Contemporary Integrations in Family Therapy (pp. 111–
127). London: Karnac.
Titelman, P. (ed.) (1998) Clinical Applications of Bowen Family Systems Theory. New
York: Haworth Press.
Walsh, F. (2006) Strengthening Family Resilience (2nd edn). New York: Guilford.
White, M. (2000) Reflecting teamwork as definitional ceremony revisited. In M.
White (ed.) Reflections on Narrative Practice: Essays and Interview (pp. 59–85).
Adelaide: Dulwich Centre Publications.
White, M. (2007) Maps of Narrative Practice. New York: W.W.Norton.
White, M. and Epston, D. (1990) Narrative Means to Therapeutic Ends. New York:
W.W. Norton.
Wilson, J. (2007) The Performance of Practice: Enhancing the Repertoire of Therapy with
Children and Families. London: Karnac.
Woodcock, J. (2009) Love and hate and the oedipal myth: the perfect bridge
between the systemic and the psychoanalytic. In C. Flaskas and D. Pocock (eds)
Systems and Psychoanalysis: Contemporary Integrations in Family Therapy (pp. 129–
146). London: Karnac.
Young, S. (2010) Two’s company, three’s a crowd: revisiting triangles in family
therapy. Australian and New Zealand Journal of Family Therapy, 31: 92–99.

© 2012 The Author

Journal of Family Therapy © 2012 The Association for Family Therapy and Systemic Practice