A rumour rather than a fact

In a tradition espoused by one of its key founders, Solution-Focused Brief Therapy (SFBT) has remained more of a rumour (a fuzzy bundle of ideas and practices) than a well-defined fact (Miller & de Shazer, 1998). Since the publication of the ground-breaking paper Brief Therapy: Focused Solution Development (de Shazer et al., 1986) there has been far-reaching evolution within the field. In the spirit of “rumouring” this evolution has been documented more by story-telling than through rigorous academic discourse. Practitioners have been inspired as much by conferences and conversations as by the written word and even that has been in the story-telling tradition: what to do rather than why to do it. This practical approach to learning and developing Solution-Focused skills has brought great benefit to the field which is now global, not only in its therapeutic mode (Franklin, Trepper, McCollum & Gingerich, 2011; Ratner, George & Iveson, 2012) but also in its applications to organisations (Jackson & McKergow, 2002; Berg & Szabo, 2005; Lueger & Korn, 2006; Iveson, George & Ratner, 2011). There is also firm evidence of the effectiveness of SFBT across a variety of contexts (see Macdonald, 2011 as well as Franklin et al, 2011 cited above). However, the absence of Solution-Focused accounts in the wider academic world, particularly in big-hitting fields such as psychiatry and psychology, places the field in a curious limbo when seen from the viewpoint of policy-makers and commissioners.

Solution-Focused Brief Therapy’s lack of a “scientific” psychological theory has placed it at a disadvantage within the medicalised mental health field within which most academic discourse takes place. In this paper we hope to initiate a fresh look at theory within the SF field; not a scientific “causal” theory but a theory based in that other equally long-lived academic endeavour to come to grips with the world, philosophy. We might even begin to see Solution-Focused Brief Therapy as a form of “clinical philosophy”, interested not so much in causes and cures as in influences and possibilities.

Development of Solution-Focused Brief Therapy

This conclusion to an early paper by the Milwaukee team is as true today as it was thirty years ago. Solution-Focused Brief Therapy continues to evolve but not as a single entity. It has become a growing collection of ideas and practices that share a common ancestor and consequent family resemblances as well as increasing differences. Its roots lie in a complex mesh of therapeutic theories and practices chief of which were the systemic and interactional approach of the Mental Research Institute (MRI) in Palo Alto, California (Weakland & Watzlawick, 1974) and the hypnotic approaches pioneered by Milton Erickson in Phoenix, Arizona (Haley, 1973). The MRI focused on determining repetitive patterns of problem behaviour and then creating interventions intended to disrupt those patterns so new possibilities could emerge. Many of these interventions were inspired by Erickson’s work which was almost always indirect — creating a new experience from which to build new ways of living. Erickson’s other major contribution was to focus therapeutic attention on the client’s own resources — searching for future possibilities within the client’s existing (though perhaps unrecognised) repertoire of behaviours.

Many Solution-Focused brief therapists have remained close to these early influences (best summarised in Steve de Shazer’s Keys to solution in brief therapy — de Shazer, 1985 — though the “blueprint” for most is probably de Shazer’s Clues: Investigating solutions in brief therapy — de Shazer, 1988). Though “the problem” is still necessary in this framework it is much less central. It is needed because “exceptions” are the preferred route to solutions with a future orientation being directed towards specific goals. In de Shazer’s next book, Putting difference to work, (de Shazer, 1991) he tried but could not quite manage to remove the notion of problem altogether. Instead the word is always written problem, in the sous-rature style of Heidegger and Derrida, in order to denote its irrelevance to the “solution”.

Leaving the problem behind

The next development, dispensing with the need even to know the client’s problem, came from a number of sources including the work of John Walter and Jane Peller (Walter & Peller, 1992) and Harry Korman and Martin Soderquist (Korman & Soderquist, 1994). At BRIEF, instead of asking “What brings you here?” — which elicits a problem account — they began to ask, “What are your best hopes from coming here?” — which invites the client to specify an outcome (George, Ratner & Iveson, 1999). With this question the client is freed from the need to describe a problem (though many clients still choose to do so). An associated development is that without knowing the problem it is impossible to ask for exceptions so the second question became the “miracle” or “tomorrow” question eliciting a description of the client’s preferred future. Exception questions are then replaced by finding instances of the miracle already happening (Iveson et al., 2011). The more detailed the description of the hoped for future the more likely were these instances to be uncovered. This greater attention to the client’s preferred future (as opposed to the dreaded future, harbingered by problem-defined past) appeared to have a therapeutic value in itself and this realisation led to further experimentation with the model.

Similar thinking about the connection of exceptions to the problem was also going on in the organisational consulting arena. Jackson & McKergow (2002) coined the term “Counters” to describe examples of the preferred future happening already, or in part, or sometimes, or even a little. They too were dissatisfied by the way “exceptions” kept the problem in the room, where a more focused conversation could be had by asking more specifically about elements of the preferred future already sometimes occurring. Jackson and McKergow’s version also includes strengths, skills, resources, co-operation and know-how relevant to the preferred future (as opposed to just not to do with the problem).

“Ockham’s Razor”

A guiding principle in de Shazer’s work and continued by BRIEF has been “Ockham’s Razor” — simply put, “don’t do more than you have to in order to achieve your desired end”. This requires us to test and continue to test the necessity of what we do. One example was to dispense with tasks. As clients’ progress did not seem to be conditional on task performance it was logical to test their necessity. BRIEF, therefore, decided to drop tasks and see what happened to outcomes. What happened (in an admittedly modestly-sized survey) was that the average number of sessions dropped and the outcomes remained the same (Shennan & Iveson, 2011).

As therapists became more skilled at eliciting descriptions of possible futures and the histories that could support them they also became more aware of their own insignificance: well-meaning interventions like tasks and even encouragement looked more like intrusions, interrupting rather than assisting the client’s progress. Similarly, when single-session therapy is a common occurrence, it is difficult to award much credit to a “special relationship”. Instead, the credit must go first to the client: whatever changes he makes tomorrow will have its setting already intact — if not in view — before the first session begins. The therapist’s part is to be skilled enough in the conversational process to help the client describe a possible future and uncover its potential history while not becoming a stakeholder in the client’s life. As will be seen later in the transcript section of this paper this is a deceptively simple-yet-not-easy task and one which requires acute attention to each of the client’s responses and quick decisions about what parts of each response to follow.

Why do we do what we do? Why does what we do work?

The apparent circularity of these two questions is a trap that this paper is seeking to avoid. The version of Solution-Focused Brief Therapy described above did not develop “because it works” but rather through a mixture of deliberate and accidental trial and error at BRIEF. The “because it works” came later when the next outcome study showed no change in effectiveness alongside a reduction in the average number of sessions. As we continue to experiment we can expect (and hope) that what we do continues to change and continues to work. What endangers this evolution is the question “Why does what we do work?” and the danger is to be found in the word “because”.

Once we fix on a theory (x happens because y — e.g. the client improved because the therapy raised his self-esteem) we, and those who subscribe to our theory, start putting the theory into practice (doing y to achieve x — e.g. raising the client’s self-esteem to achieve change). Practice then begins to follow the theory and the client ceases to be the “expert”. Positive Psychology has followed exactly this route (Seligman, 2011). The challenge for us is to keep theory subservient to practice, to what actually happens while at the same time answer, at least provisionally, the legitimate question Why does what we do work? This brings us to the vexing question of “theory”.

Theory, no theory and what kind of theory

It is sometimes said that SF practice has no theory — that it’s about finding what works for each client, whatever that turns out to be. We think this cannot be the whole story. There are many kinds of theory. The one being complained about by Steve de Shazer in the quote above is to do with explanations — explanations of how the client came to their present situation (and therefore what to do about it) and explanations of how change happens. Many therapy schools have theories like these — that change happens by changing thoughts, by addressing past fears, by “working through” negative feelings, and so on. This is theory of mechanism. Similar kinds of theory are found in the natural sciences, where iron rusts because of exposure to oxygen, diseases spread by infection of viruses and planets attract each other because of gravity.

In this kind of theory, knowing the theory helps us to get the results we want. So galvanising (protecting iron with zinc) helps prevent rusting (by keeping the water away from the iron), hand washing helps prevent the spread of infections (by removing bacteria). If we want to send a rocket to the moon, having a theory of gravity will help calculate the exact trajectory for the rocket to arrive in the right place, given the competing pulls of the moon and the earth.

SF practice is notable (though not unique) in eschewing this way of thinking. We do not claim to know how our clients get into difficulties nor what they need to do in order to get out of them. We assume that knowing “why” (especially given the number of competing “why?”s) will not help the client do “something other” nor can we know what that “something other” could be until the client develops it in practice. This not knowing position requires us to change our way of listening to what the client says. When we have a theory we process the client’s answers through that theory and ask questions which derive from the theory. If we have linear causal theory we might ask “When did it begin?” If we have a systemic theory we might ask “How does that affect your relationships?” There is then a danger that the theory begins to drive the conversation or, even worse, the client is shoe-horned into a fit with the theory. As soon as we start to think in these terms, whether from our training or our “hunch” about this client, we can only “listen with one ear”, the other ear being engaged in an internal conversation with the theory.

Theory as “what to pay attention to”

For most scientists a theory is about a mechanism, an explanation of how things work. This works well in the “molecule” fields such as physics but in the “meaning” fields such as therapy, where meanings are in a permanent state of being socially and publically constructed and reconstructed, theory cannot be separated from the feedback loop of practice. John Shotter (Shotter, 2005) points to theory-in-use by practitioners — the ways in which the practitioner has learned what to pay attention to and how to respond to what he is hearing. Though useful this is not an easy idea. It is useful because it provides a way for the field to study and discuss what we do and why we do it. It is hard because it is not an “A + B causes C” theory but a more inexact “process” theory in which every time A and B come together they rub up against each other in unpredictably different ways so that C never quite looks the same; our understanding of the relationships between A, B and C is always provisional, always needing adjustment. When we then factor in the “observer influence”, the fact that how we look at the As, Bs and Cs makes a difference to what we see (and how they each react to being seen), it is easy to understand why on-the-ground practitioners might decide to dispense with theorising altogether.

Fortunately, doing this “clinical philosophy” is easier than talking and writing about it. One practitioner can watch another at work and notice what aspects of their practice seems to generate positive effects. They might particularly notice a variation in the use of a particular technique such as the “miracle question” that leads to a different sort of conversation. When therapist and observer have their post-session conversation they will talk about this difference, consider how it changes their ideas about the therapeutic process, assess whether it can be generalised and made to fit with other clients and in subsequent weeks talk about the differences that have begun to emerge in the doing and thinking about therapy. This might all take place in formal clinical meetings or in snatches of corridor conversation: it is what we would call practice and theory in evolution and is associated with “good practice” whatever the model of therapy.

A very marked difference between Solution-Focused conversations and those of many other models is in the way of listening, as summarised by McKergow & Korman (2009). If a model is based on an explanatory theory with its own language and beliefs the therapist must both listen to the client’s words and “translate” them to the language of the theory. She also needs to seek the client’s cooperation with her theoretical position and one way to achieve this is to begin paraphrasing the client’s responses. If the client agrees with this slight change in the meaning given to his words then he feels heard and possibly understood in a new and engaging way. If he doesn’t agree then he can say so and the therapist will adjust her own words until a fit is found. (It is at this, usually very early, point that “manualised” therapeutic procedures begin to break down since it is impossible to manualise this process of adjustment.)

The “fit” SF practitioners seek is around a description of the client’s aspirations, not an understanding of the client’s problem, and for this they need to rely on the client’s language, since this will most accurately represent their aspirations. The microanalysis research of Janet Bavelas and colleagues (Korman, Bavelas & De Jong, 2013; Tomori & Bavelas, 2007) shows SF practitioners using the client’s words significantly more and introducing their own concepts significantly less than any other model studied. Hearing her own words being spoken back is another way the client will know that she is being listened to carefully.

Though therapists will try to listen to everything the client says they cannot respond to everything, they must select which part of a client’s answer will be most useful in constructing the next question. This is why a model is essential — we need to have a coherent framework for making these selections. We might listen to everything but we select very carefully what we pay attention to and it is this selection which shapes the conversation into one about past causes, present challenges or future possibilities. The model we propose here is based on description.

Three key elements in first therapy sessions

The somewhat pared-down version of Solution-Focused Brief Therapy at BRIEF consists of three questions, based on the assumption that every client, including those mandated to attend, have a good reason — a desired outcome — for being there.

There are many versions of these questions but what they share is a focus on description and only description. The broad description of an outcome, a more detailed description (perhaps beginning with a “miracle” or “tomorrow” question) and a description of past and present instances of the hoped-for future happening (usually summarised in a scale).

This process is exemplified in the case of Mary below. The therapist stays entirely within the realm of description, making no attempt to introduce any notion of his own about what Mary “needs to do”. Indeed, he works hard the whole time to maintain a neutrality towards what the client does tomorrow (a neutrality he would abandon only if he thought the client or anyone else might come to significant harm.)

Case example: Mary and the cuddle

This case concerns Mary, a woman in her mid-40s who attended BRIEF referred by her GP following depression and the GP’s concern at the risk of suicide.

Having established Mary’s hope that she wants to have a sense of peace and hope for the future, and to not be continually dragged back into the past (the contract), the therapist leads into the following “miracle” question (at five minutes into this particular session):

The client’s global answer is (non-verbally) accepted and then she is invited to think small. This is one of the most useful ways into a detailed description — locating it in a particular and familiar time and place. The client goes on to describe her breakfast, her drive and entry to the gym, her workout, her meeting with friends, lunch, reading a book and talking to her sister on the phone. These descriptions fall into three broad categories: what she notices about herself, how she appears to others and what happens between her and those others.

Twenty-five minutes into the session, the therapist invites a description of what he guesses is one of the more significant moments of the client’s day — the moment her partner discovers and responds to her “post-miracle” state. Her partner has left for work before she wakes so his discovery of the “miracle” will take place when he arrives home in the evening. As the most significant persons in each other’s lives this meeting will hold many possibilities. (This is an example of co-construction; it’s not a question of the client leading or therapist leading, the client has given the therapist information upon which he can act. If this doesn’t turn out to be a significant moment for the client, we can move on.) The description starts a few minutes before her partner’s arrival, once again with a scene-setting question.

Even before her partner gets home the relationship between them, what they do together is changing thus preparing the way for a different interaction.

This description of the cuddle takes about three minutes, considerably longer than the event itself is likely to be. During the description, a visible change takes place on the client’s face, in her tone of voice which suggests that the description is evoking some sort of felt experience. This is not an “accidental” description. Such detail does not come without careful scene-setting which helps place the client’s future within her everyday routines.

A little time is spent on the post-cuddle moment and then on to the next “scene”:

Experience and description

To simplify (or more likely mangle and misrepresent) Wittgenstein, from whom de Shazer drew much inspiration, conversations will generally include expressions of feelings, descriptions of actions and explanations of both. For Wittgenstein description was the most clear. Descriptions need to be of something that’s open and visible, in order that the description can be seen and agreed to be accurate (Wittgenstein, 1953). This means “staying on the surface” in Steve de Shazer’s terms (de Shazer, 1991) — talking about what the client does within and in response to their surroundings.

The descriptions we are seeking in the therapy room are innocuous looking everyday mundane descriptions of normal events — either in the client’s possible preferred future, in the present or in the past. It is clear from the example that the therapist does not ignore the client’s “inner world” of emotions: he frequently asks questions such as “Would you be pleased…?” but this inner or private experience of pleasure is then translated into the public arena of described actions (“I probably would maybe just stand there with him”). Throughout, the focus is on description rather than explanation.

Reaching out — embodied and enactive cognition

Mary’s apparent emotional experience, coming with her description, adds weight to the idea that the mind is “embodied” rather than held within the confines of the skull, and that cognition is “enactive”, comprising our interactions with the world rather than computing all the messages coming from the world. Theories of embodied and enactive cognition have been gaining ground in both psychological and philosophical disciplines (Clark & Chalmers, 1998; Varela, Thompson & Rosch, 1991) and the implications for therapists are still becoming clear. The writing seems to be on the wall for the body/mind separation which has allowed us to imagine, for instance, that it will one day become possible to “know” ourselves fully by simply understanding how the brain works.

Enactive cognition, see for example Hutto and Myin (2013), challenges the conventional view of the mind as some kind of computer, taking in information to process and produce behaviour. Instead they see thinking as just one part of a cognitive process that engages the whole person. In a similar fashion to Wittgenstein (Moyal-Sharrock, 2013), enactivists propose that the mind has no independent mental function so going in search of desires and beliefs in a skull-bound mind is a fool’s errand. People, not minds or brains, believe and desire things, and they do this in their actions and interactions with the world, including other people. Experience, in the enactive account, is not an outcome of cognitive processes, it is the way in which we as “whole persons” work directly with the world.

It is this notion that BRIEF has bumped into by its application of Ockham’s Razor. As therapists at BRIEF concentrated more on description and less on action they became aware of unexpected “in-session” changes in the client’s way of being in the room. Not just changes of mood associated with an outcome-focused conversation but an entire change in the client’s described experience of themselves in the moment. In these cases it seemed that the description of a preferred future led to the actual experience of (something akin to) that future rather than a “cognitive map” of something yet to happen.

When Mary describes the future she would like to have, even though she sees no possibility of it happening, the description, because it is so detailed, is no longer just an imagined possibility, it becomes an experience in itself. Mary does not just describe a possible future, she experiences that future and so becomes a person with those experiences, a person with hope who does not have to be “sucked back into the past”. It is possible, therefore, that the experience of co-creating a detailed description is a potent therapeutic intervention in itself, the conversation being the thing rather than “about” the thing.

Description and Narrative

Humans are story-tellers, we like to “join the dots” between our experiences and create “narratives” that somehow make sense of our lives. These narratives then influence our expectations and consequently our ambitions for the future. Our capacity for story-telling has been the subject of much philosophical debate and theorising.

At one end of the spectrum “Strong Narrativism” (in its simplest form) argues that we construct the “self” through the stories we tell and the self can therefore be reformed by changing the stories. The philosopher Anthony Rudd (Rudd, 2012) argues that the self “only comes to exist through its being narrated” (Rudd, 2012, p. 1). Changing the narration must therefore change the self. It is this idea that lies at the heart of Narrative Therapy (White & Epston, 1990). A more modest narrativism is proposed by Dan Hutto, who says the accounts we give of our lives, our narratives, have a more metaphorical function “a natural form of self-understanding and self-shaping” (Hutto, 2014), which brings the infinite complexity of our lives to a more manageable size. What aspects of our experience we choose to put with our life story will undoubtedly influence the life we lead but that life will not be determined solely by the story. The story isn’t the person and the person isn’t the story; the story just a vehicle for making sense of life and our place in life. Thus the more elements allowed into the story of our past the more possibilities we are likely to see in the future. Exceptions (to the problem story) and instances (of the preferred future story) both add new elements from which “self-understanding and self-shaping” can be drawn.

What is crucial in this process is that the therapist remains neutral about the future steps the client might choose to take. Any attempt, however subtle, to direct the client towards action is likely to be experienced as a form of expropriation: using the client’s ideas to feed the (good) intentions of the therapist. Only by staying with description can this neutrality be maintained and the client be left fully in charge of her life. Similarly, the therapist is not out to create an emotional experience, to make this an aim would be to assume that this is right for the client. The emotional experience that might arise from a description can be best seen as a bonus — one of the many ways SFBT influences lives and one particularly associated with rapid change.

Conclusions

This paper began with the idea of finding a theoretical home for Solution-Focused Brief Therapy (and its offshoots) but one that did not constrain or direct the continuing development of the model. The best we have been able to do is to follow the habit of the hermit crab and find a home that fits but does not dictate and one which can be exchanged for another as our practice and our thinking about practice evolves.

The new “home” closely resembles the one provided by Wittgenstein to de Shazer’s early ideas but have expanded it with ideas from current developments in philosophy which offer not causal explanations but possible patterns of influence: when A and B come together something like a C often appears. Or, more specifically, when we ask questions about a client’s hoped-for future we think that their answers set off different thoughts, emotions and actions which lead them to have richer ways of seeing themselves within their life: a richer history from which to select a view of their past and a wider selection of possibilities in their future. We also have to admit that we have painted a somewhat caricatured picture of psychological “causal” theories. They are not homogeneous entities and there are many crossovers between psychology and philosophy as well as between the growing number of resource-oriented therapies.

In the end we hope simply to have shown that there is as much intellectual legitimacy as there is pragmatism behind Solution-Focused practice and that this form of theory supports the continued development of a model which provides no way of knowing what any client should do next. Our theoretical exploration grows out of the practice as seen in the “Mary” transcript and so remains practice-led. If we were bold enough to imagine Steve de Shazer’s response to these developments we might expect him to be more than satisfied that the “facts” as we know them today are as fully supported by the “theories” of Wittgenstein as the “facts” he and his team discovered thirty years ago: