An introduction to theory

An Introduction to Theory

We teach an approach to supervision for clinicians that is based on a particular attitude of mind, and a set of techniques that convey that attitude. This paper offers a brief description of the theoretical ideas behind the attitude and the techniques. We want to emphasise two things from the start. Firstly, the attitude of mind is more important than the theory. Some people instantly grasp the attitude but are bored by the theory (which is fine).  Others get very excited by the theory but don’t seem able to apply it in live supervision (which isn’t fine). So if you struggle to understand some of these theoretical ideas it may not matter. Equally, if you fall in love with them it still may not mean that you can supervise well. Secondly, we are applying ideas that we have learned through our training in family therapy. This mystifies some people since they wonder how on earth ideas from therapy with families can possibly be relevant to supervising doctors, dentists and other clinicians. However, the ideas have a history that is quite independent of family therapy, and these days they are increasingly being used elsewhere including management and education. You do not have to be a family therapist to understand them or apply them skilfully to supervision.

Our approach draws on two distinct but related fields of thought: systems theory and narrative studies. The following sections describe each of them in turn, with a linking section that describes a school of thought that in some ways bridged the two sets of ideas.

Systems theory

Systems ideas have been around since the middle of the twentieth century. They arose in many different disciplines including engineering, physics, cybernetics, biology and anthropology. They are associated with a number of names that have largely been forgotten outside specialist disciplines. These include Norbert Wiener, Heinz von Foerster and Ludwig von Bertalanffy. The best known of all the systemic thinkers was a man named Gregory Bateson. He was British but lived much of his life in California.

Bateson was something of a polymath. His essays covered a huge range of interests including evolution, political theory, religious mysticism, art and psychiatry. Unfortunately he was not a very clear writer and his arguments can be hard to follow but they can all be summed up by a single idea: everything in the world is ultimately connected with everything else, through a complex pattern of interactive loops that never really has any beginning or any end. Because of this fact, all that we can ever perceive of any phenomenon is only partial and provisional. Moreover, we ourselves as observers are really only a part of the pattern of interactive loops and can never really stand outside it and be entirely objective.

A typical example of Bateson’s thinking relates to the problem of schizophrenia. Rather than focussing on the individual person with schizophrenia, Bateson preferred to talk about ‘schizophrenic interaction’. While recognising that some people might have a genetic tendency towards schizophrenia, he pointed out that this could only be a small part of a much wider pattern. People behaving in a schizophrenic way would inevitably have an effect on everyone around them, and these effects would then have other effects. For example, family members might react by treating them as weird or dangerous, and this in turn might make them more likely to be so.

Or, in a more complex way, people who showed schizophrenic behaviour might be likely to choose partners who behaved in similar ways to themselves. This would result in children who would have an additional genetic tendency to the same kinds of behaviour but would also be nurtured in an environment where these kinds of interactions were more common. The social circle around them would then respond in turn by marginalising them, thus adding to their problems. If anyone caught up in such multiple interactions saw a psychiatrist, the problem might then be amplified further, especially if the psychiatrist dwelt only on the abnormal aspects of the person’s behaviour and emphasised the diagnosis rather than ever engaging in normal conversation with them.

Bateson argued that this kind of ‘systemic’ thinking was useful not just in clinical fields but in every area of human experience. Let me give a personal example. I regularly visit a café near my workplace where there is a young black waitress whom I used to find very bad tempered. I used to respond in kind by giving her fairly curt requests and not leaving tips. One day I noticed her laughing with a group of young black customers and I began to wonder if she was grumpy with me because she expected me as an older white man to behave exactly as I was doing: namely being curt and ungenerous. I altered my behaviour and of course she did too. Hopefully we both learned from this, in a way that may make a small contribution to lessening racism, sexism and ageism more widely. Of course we are likely to have less impact on the wider social contexts that determined her previous behaviour and mine, or on the enduring effect of colonialism on black and white people, nor indeed on a range of wider contexts including the relations between males and females, or between in-groups and out-groups across all human cultures. But you never know.

Bateson was not himself a clinician but he worked for a time with psychologists and psychiatrists. He was particularly influential on a group of people who became the founders of family therapy in the 1950s and 60s. These people started to use his ideas not just with schizophrenia but with alcoholism, behaviour problems in childhood, marital discord and a host of other problems. Instead of seeing any problem as ‘belonging’ to a single individual, they started to focus on how people interacted with each other and how this could make any problem far worse – or far better. They would see patients together with their close relatives, and work with the whole family system to try and understand and help what was going on.

Although family therapy has changed in many ways since its earliest days, family therapists continue to use Bateson’s ideas. In particular, they tend not to make interpretations about the ‘cause’ of a problem, nor to give advice about how to deal with it. Instead, they ask questions in order to stimulate everyone’s interest in the nature of the problem, how it arose, and what is keeping it going. They hope that by  thinking about such questions, everyone involved may become more aware of their own contribution towards the situation in the ‘here and now.’ By working in this way, they aim to help people question the objectivity of their own fixed judgements and labels, and to explore new ways of seeing the world around them and their part in it. It is also from Bateson that we take the idea of “the difference that makes a difference”. This informs our thinking about how to help people think about change. People need to do something different but not too different so we ask about the smallest steps they can take that might be useful.

In the context of supervising healthcare professionals, we find that Bateson’s thinking, and systemic ideas generally, are helpful in all kinds of ways. They can make people aware of how any problem may only become a problem in the context of human interactions, and how those interactions can contribute to it, or make it better. Systemic ideas can also help people to see that any understanding of a problem can only ever be partial or temporary, and that solutions – or resolutions – can only really be generated by the parties involved. The role of supervisors is therefore to be curious and sympathetic but also to be challenging – in the sense that they will never simply accept the supervisee’s account as the only possible description of what is going on, or as the ‘truth’ of the matter.

The Milan team: a bridge from systems to narratives

Like every other branch of psychology, systemic thinking and family therapy have given rise to many different schools of thought. However, the followers of Bateson who have most influenced us are a group of Italians  known as the Milan team. These were four psychiatrists who were also psychoanalysts. In the 1970s they became frustrated and disaffected by some aspects of psychoanalysis including its emphasis on the individual and its apparent certainty about the mind and how it works. Using principles derived from Bateson, they developed a way of working with families (and later with individuals, and then in supervision) that depended almost entirely on using questions to open up new ways of thinking for their clients. Eventually they proposed that their approach could be understood in terms of three simple guidelines: hypothesising, circularity and neutrality.

When they talked about hypothesising, the Milan team were trying to draw attention to the fact that it is quite impossible not to form ideas in your mind about causes, reasons, explanations and interpretations for anything you hear about. However there are two quite different ways of responding to these ideas. On the one hand you can assume that your own ideas are right and to try and persuade other people of this. On the other hand, you can regard these ideas simply as different descriptions of
what is going on, and then to try and find out if these descriptions are of any interest or use to the other person.

The conversion of hypotheses into questions is one of the key skills of systemic questioning. It not only involves identifying what you are thinking in the first place, but it also includes the discipline of becoming sceptical about your own ideas at the same time, and then asking a question that gives no hint of your opinion.  To use the example I gave earlier of the ‘bad-tempered’ waitress, somebody listening to me talking about her might quite reasonably form the idea that I was being unfair or prejudiced in my description of her. In normal circumstances they might just tell me so – but at the risk of offending me and therefore not helping me to think about my behaviour. If they questioned me on Milan principles instead, they would ask me something like: ‘Does she seem grumpy with everyone or have you ever seen her behave in a different way?’ Ideally, they would ask this in a way that showed genuine curiosity and no hint of criticism of me, making me more inclined to reflect on what had happened from an interactional perspective.

The Milan team’s next guideline of circularity covers the idea that the person doing the questioning in a systemic interview (whether in a consultation or when doing supervision) should always note in careful detail what the response is to each question, and use this to frame the questions that follow. This involves a willingness to ‘go with the flow’ of a conversation even if it is going in a quite different direction from the expected one. One of the necessary skills for the interviewer here is what the Milan team called ‘not being wedded to your hypotheses’. This implies the ability to respond with equal interest whether or not the ideas in one’s own mind are confirmed.

Taking the ‘grumpy waitress’ example once more, I might conceivably respond to the interviewer’s question by asserting stoutly that I had only ever seen her being grumpy, whoever she dealt with. A good interviewer would take this at face value and move on immediately to a different hypothesis and a different question, for example: ‘Can you imagine any situation outside her work where she might not be so grumpy?’ Another possibility is that I might confess that I had indeed seen the waitress behave cheerfully with young black people, opening the way for a further question like: ‘What explanation do you have for why she behaves differently to you?’. The Milan team’s view, and our own experience, is that such a question would be far more likely to induce me to explore my own set ideas and alter them.

The Milan team’s third guideline – neutrality – really flows from the previous two. It expresses the idea that interviewers should constantly maintain an open, tolerant stance that allows their client or patient the maximum possible space, unimpeded by the intrusive beliefs or prejudices of the interviewer. The Milan team were at pains to emphasise that this did not mean that interviewers should have no beliefs and prejudices of there own. Nor did they ever rule out the possibility of situations (including dangerous or life-threatening ones) where it was legitimate and ethical to declare these. What they did argue, however, was that clinicians very often found themselves in situations where they could do more harm by inappropriate certainty than by carefully considered neutrality.

In time, one member of the Milan team named Gianfranco Cecchin wrote a further paper in which he boiled down the approach of the team into one word: Curiosity. If one felt and expressed adequate curiosity, he suggested, everything else necessary for a systemic interview would follow automatically. This would not only include a helpful exploration of the nature and content of the problem, but also the client’s response to the interview itself.  (‘How is this conversation going for you? How helpful are you finding it? Are there any other questions I should have asked you? Am I getting the balance of questions to advice about right? Am I showing any prejudices that are getting in the way of your thinking?’ and so on.)

In our trainings in supervision skills we have used the ideas of the Milan team in all sorts of ways, but the most important of these are probably the stress we place on attentiveness to language, and on following feedback. In our experience many doctors are quite empathic and sensitive to the general tone of feeling in a conversation but they may have inadequate skills in noticing the tiny, giveaway, words and phrases that can act as cues for curiosity and helpful questions (eg the word ‘always’ in the expression ‘always grumpy’). Equally, they may never have been trained to be sufficiently aware of their own certainties, so that they are inclined to plough on with a particular, pre-determined line of questioning even when every response is indicating that it would be better to pursue a different set of ideas.

One specific point worth making here is whether it is useful to impart a list of systemic questions (sometimes referred to as ‘circular’ questions) that are useful in many different situations. The answer is probably ‘Yes and no’. We do sometimes give as set reading a famous paper by a follower of the Milan team named Karl Tomm, who systematised their approach to questioning (although they repudiated his system as too rigid and militating against spontaneity). From time to time we also ask groups to generate their own list of ‘favourite effective questions’. However on the whole we regard such approaches to questioning as essentially anti-systemic. By definition, a set of prepared questions cannot possibly relate to the specific language cues given by individuals in particular conversations. However, when they are learning this technique some people find it very helpful to have a few “favourite questions” handy to refer to in their consultations or supervision conversations.

The Milan team never explicitly described themselves as narrative therapists or narrative practitioners. However, in their preoccupation with language and its importance they were very much in accord with a rising interest in narrative that was also emerging around the same time.

Narrative ideas

The narrative movement that emerged in the 1980s was entirely distinct from the world of systemic thinking. However, like systems theory, it emerged in a whole range of different and apparently unrelated fields including the social sciences, philosophy and literary studies. Pioneers of narrative thinking included the psychologist Jerome Bruner, the literary critic Paul Ricoeur, the philosopher Charles Taylor and the Russian linguistician Mikhail Bakhtin – who had actually written much of his important work fifty years earlier but was now being rediscovered.

A narrative is simply a story. People within the narrative movement have taken all sorts of different theoretical positions, but they all basically share the same central ideas: human beings are story-telling creatures. This means that we make sense of our realities by telling each other stories (or bits of stories) and we experience our lives in ways that resemble stories – in other words with characters, plots, motives, suspense, beginnings and endings and so forth. Another feature that narrative thinkers generally have in common – and that Bakhtin emphasised – is that stories are made up not by single individuals but between them. Who a story is being told to (and where, and when, and why) is just as important as who is doing the telling.

This is not the place to explore the relationship between the narrative movement and other similar movements from the late twentieth century including post-modernism and social constructionism. Nor is it the place to examine the different approaches that narrative thinkers have taken to the philosophical question of whether stories approximate to something that really exists, or whether the only reality we can ever know consists of the stories that we tell ourselves and each other. What is important, however, is to notice the points of similarity and overlap with systemic thinking. These include a crucial emphasis on interaction (who is telling what to whom) and on context (how our story-telling is determined by our various identities and relationships in terms of family, culture, belief systems and so forth).

Narrative thinking has affected family therapists just as profoundly as systemic thinking. It is probably true to say that many or most family therapists working in the UK nowadays would probably describe themselves as working within a narrative framework as much as a systemic one. Narrative ideas have also affected other schools of psychology as well, including psychoanalysis. There is in fact an emerging consensus in many schools of psychological thought that people’s problems are changed not so much by helping them find the ‘real reason’ or the ‘best answer’ to their problems, but by helping them to find a coherent story that provides them with a satisfactory meaning for what they are going through.

In our work in supervision skills training, we particularly use the concept that clients usually bring a ‘stuck story’ to supervision. They may have told the story over and over again to themselves and to others, so that the story itself (often involving a sense of being helpless or overwhelmed) has become part of the problem. We promote the idea that thoughtful and sensitive questioning can invite people to retell their experiences to themselves in a different way. Quite often, for example, we notice that someone will start to present a problem in supervision with a phrase like ‘Well, it’s a very complicated story…’ Fifteen or twenty minutes later, they may actually say ‘I guess it’s really fairly simple and I’ve known all along what I ought to do’. People also sometimes start by presenting something for supervision that they think is fairly simple but the process of supervision develops something more complex. They may be entirely unconscious of how they have been helped to reconstruct their narrative in this way unless they review the process on video.

Contexts and power

From a practical point of view, good supervision often involves attention to contexts rather than content. Attention to context provides richer meaning to what is going on. It moves the conversation away from a homeostatic tendency towards a dynamic one, and from a ‘vertical’ plane to a ‘horizontal’ one. Questions about content tend to encourage the solidification of the narrative. Questions about context lead to more flexibility and negotiability.  In this respect, questions about different contexts can be particularly powerful: different perspectives, different backgrounds, different beliefs, different values.

Family is not the only context that defines us but is nearly always the most important and the most powerful. Most of our conversations take place there, and most of our narratives about ourselves and our lives are forged there. Both in clinical work and in supervision, family therapists work from the premise that the family and the individual are in a constant process of mutual redefinition. Many therapists work on their own geneograms as part of their training in order to understand the perspectives from which they think and work. In supervision, it is unusual to come across a case where inquiry into the identified patient’s family network is not relevant in some way. Inviting clients to reflect on what they know – or do not know – about the patient’s family context will very often produce new understanding.

Working as professionals, in many ways our teams are our families. These teams take many forms: partnerships, medical teams, primary care teams, practices and so forth. We and they mutually define our realities in the same way that families do. When patients and their literal families come into contact with us and our teams, we can often only make sense of this by understanding how things are happening across the whole network: ours and theirs. Many problems brought to supervision are not the problems of one practitioner but of several. Often they relate to how different practitioners are operating differently, or how different family members are interacting with different team members. Drawing the lens back to see the wider dance of family and team will provide a wider and often more helpful context for understanding the difficulties of a single consultation or a single relationship.

Power is a feature of every encounter involving doctors. It is power due to medical knowledge, power due to social position, and power due to privileges or perceived privileges (prescribing, referring, certification, advocacy etc). Cases brought to supervision often involve contexts of power that have been ignored or denied. In particular, doctors may see themselves as neutral or benign agents when it is clear that they are being seen in a very different way by the patient or by other players in the drama. Supervision also sometimes needs to address the ways in which doctors try not to acknowledge or own their power. It may also need to look at how they might use power ethically and transparently. In remedial work, it is nearly always important to be transparent about the power issues involved.

Culture can be understood in a fairly thin way (‘He is a Catholic’ ‘They are a Somali family’) It can also be understood in a thicker way: not in terms of who this person is in the practitioner’s eyes, but who the person is in their own eyes. For some people, the cultural identity or identities will provide the overarching context for who they are, while for others it may seem irrelevant. In supervision, it can be important to establish what is known and what is not known about the way that patients define and understand themselves in cultural terms and what this means to them. The same is true of such definitions as ‘refugee’, ‘second generation’, ‘immigrant’ and so on. One very effective reflexive question can be to ask practitioners who come from majority cultures: what does it mean to you to be white, British, part of the dominant culture?

Gender, like culture, is not a fact. It is a lens. It is one way in which difference can be examined. What does it mean to a male doctor that the patient is of the opposite sex, or of the same one? What seems to be the patient’s understanding of  the part that gender is playing in their lives, their problems, or the consultation? How might things have been different if one or both parties had been of the opposite sex, or of a different sexuality? Almost any consultation, and almost any problem, can be profitably examined under the lens of gender.

From a theoretical point of view, contexts are what give meaning to any utterance we ever make. The words we used are embedded in sentences that are embedded in conversations within relationships that take place in human networks that form cultures that are part of historical systems. Communication theorists talk of how these different levels are recursive: in other words, culture and relationships determine speech acts, but speech acts also contribute to the construction of relationships and culture. It is often hard to understand any utterance without clarifying the ‘higher levels’ of meaning within which it is being uttered.

This can be particularly important when muddles occur in communication (something that you can often pick up intuitively or somatically). The muddles may be the result of two people competing at a higher level in terms of their assumptions. The only way out of this may be to look for an even higher level eg a description of the muddle and why it is happening.

A common example of this is when a kind, thoughtful supervisor has to talk to a junior about a lapse of judgement or a poor decision. Since all their previous conversations have been positive and friendly, the supervisor starts out by trying to elicit the junior’s account of events, in the hope that the junior will spontaneously come out with an admission that things went wrong, and with an apology. However it sometimes happens that the junior does not realise anything was amiss, and describes the case as if nothing went wrong. The supervisor then starts asking more and more persistent questions until the trainee starts to feel first puzzled and then harassed. If the process goes on for long, the supervisor the junior may redefine the supervisor as somebody inconsistent, untrustworthy and a bully. This might have been avoided if the supervisor had offered a simple context marker: ‘Look, I know you’re normally a great doctor and most of the time all I have to do is to praise you and help you develop even further, but on this occasion I need to talk to you about something you’ve got wrong’. This statement hopefully preserves the  context of being a caring supervisor while making it clear that one of the roles of such a supervisor is also to talk about mistakes.

Our own synthesis

Our use of systemic and narrative ideas changes all the time. The emphasis we put on different elements of systemic and narrative ideas inevitably alters, in response to each course we teach. The account we give of our thinking also changes. We make use of ideas from other fields as well. For example, some psychoanalytical ideas (like  the word ‘unconscious’ in the previous paragraph) still inform our hypotheses and our language at times – not least because they are so embedded in popular thinking and in medical culture. However we also try to remain sceptical about our hypotheses and the language we use, and open to challenge in all our teaching and our ideas.

We recognise that our own particular area of interest – supervision for practising doctors, dentists and other health professionals – requires a theoretical approach that draws on systemic and narrative thinking but is not entirely dominated by it. Broken bones, strokes and death are not just words: they are real. We realise the need to balance the scepticism and relativism of systems theory and narrative ideas on the one hand, with an ethical sense of what is solid and non-negotiable in medicine and health care on the other hand. Having said that, we find time and again that what supervisees (and patients) find most helpful in an interviewing stance is the one thing that Cecchin said most characterised the systemic approach to helping people: Curiosity.


Andersen T. The general practitioner and consulting psychiatrist as a team with ‘stuck’ families. Family Systems Medicine 1987; 4: 468-481

Bateson G. Steps to an Ecology of Mind. New York: Ballantine, 1972

Cecchin G . Hypothesising, circularity, and neutrality revisited: an invitation to curiosity. Family Process 1987; 26: 405-413.

Launer J. Narrative-based Primary Care: A Practical Guide. Abingdon: Radcliffe, 2002.

Launer J. New Stories For Old: Narrative-based primary care in Great Britain. Families, Systems and Health 2006; 24: 336-344

Launer J and Halpern H. Reflective Practice and Clinical Supervision: an approach to promoting clinical supervision among general practitioners. Work Based Learning in Primary Care 2006; 4:69-72

Palazzolli Selvini M, Boscolo L, Cecchin G, Prata G. Hypothesising-circularity-neutrality: three guidelines for the conductor of the session. Family Process 1980; 19:3-12.

Tomm K (1988) Interventive interviewing: Part III. Intending to ask lineal, circular, strategic or reflexive questions? Family Process 27:1-15.