Reciprocal Roles: Caught in the Crossfire

Elia, I., 1998. Reciprocal Roles: Caught in the Crossfire. Reformulation, ACAT News Autumn, p.x.

Reciprocal Roles: Caught in the Crossfire

Irene Elia

Two and a half years ago, I had never heard of reciprocal roles (RRs); now I lie in bed in the morning worrying about how they may be messing up my life and the lives of others. These notes on my voyage of discovery may interest other trainees and seasoned therapists too, because the understandings and misunderstandings, fears and insights experienced by a totally naïve trainee (one who was not a therapist before) are in many ways similar to those of a totally naïve patient.


During the two years of Basic Training, I saw eight patients in the Psychotherapy Department of Addenbrooke’s Hospital in Cambridge. Nobody dropped out, nobody missed a session (except by prearrangement), everybody received a reformulation, and a diagram in which RRs were discussed, and everybody received and wrote a goodbye letter. I was stunned, therefore, after the completion of about 2/3 of these cases, when my then supervisor said that nothing much had happened to these people. They had been “catted” and remoralized she said, but I had not been able to make them do anything! I wondered what apart from attending sessions, hearing, adjusting and accepting the reformulation, collaborating on the naming of the RRs, and the drawing of the SDR, noticing their TPPs operating in personal interactions, and saying goodbye, they were supposed to do.

Within this same period of time, I was also working on an MSc project to find out if the CAT community thought that lists of words or paired lists of words would help us name RRs. Eva Burns-Lundgren was working on her Advanced Course thesis on the possibility of reliable, early identification of RRs. Her work showed that it was possible for experienced therapists to make such identifications within the first session and that a cluster of RRs were being identified repeatedly: ideally caring-ideally cared for, controlling-controlled, rejecting-rejected, contemptuous-contemptible, neglecting-neglected, abandoning-abandoned, etc (see her thesis).

I found that many trainees found RR identification an area of uncertainty. Both trainees and therapists thought that word lists could help explain RRs to trainees. Therapists preferred lists of paired words, such as the one produced by a group of CAT trainers in 1996 (see my thesis). But whether or not lists might be used, whether or not RRs were easy or difficult to name reliably and validly, the focus in training upon their use in the therapeutic situation was unavoidable. It was a case of: Do you know what a spanner is? Never mind, this is one. Is it a well-made spanner? Never mind, use it!

I continued to learn what it was therapists needed to help patients do: see their RRs operating in session. Hopefully, this would allow them to “click” with insight rather than “crack” with shame and denial. But the gulf between the inexperienced trainee who thought lists ‘night be helpful and the experienced therapists who seemed to be able to know a person’s RRs at a glance was as wide as ever. Some therapists even wrote at the bottom of the pink questionnaires, which I mailed out for my study that they never used RRs at all and preferred to work with the transference and countertransference, behaviours that were still mysterious, “unproven”, and invisible to me. I suspected that exploring this dark domain with my patients was probably going to be the key to my getting them to do something that would help them significantly, change them for the better. There were, however, all sorts of half-seen obstacles in this territory, and I wondered if I would ever negotiate it.

The role a therapist had to play seemed contradictory to me: avoid bruising, confusing, and losing people, build up a working alliance, but also avoid being the ideal carer your patient might be wishing for. It seemed that to create rapport you had to play some reciprocal role with your patient. The only way to monitor this seemed to be to make an intentional choice about what role you would play, not just be unwittingly sucked into a role they wanted you to play. While this seemed rather unnatural to me, my CAT therapist tried to show me that it was benign and beneficial.

For example, at the end of therapy, she said that she had decided to allow me to control the therapy to give me the positive experience of being in control, of feeling I could make someone do something. This would be in contrast to my supervisor’s statement, which while true, had made me realise just how powerless and controlled 1 was: 1 couldn’t make patients do anything; they just bounced me from one reciprocal role to another without either of us realising it, and while they said they felt better in the end, they probably were not. T was grateful, therefore, to my therapist for her good intention of boosting my confidence by letting me take control. I must admit I was a bit confused because I had not realised that 1 had been allowed to control the therapy until she told me, and then when she did, I wasn’t sure I should feel good. After all, someone allowing you to take control is not all that
self-enhancing. I took it all to mean that this was a way of teaching me how to conduct myself as a therapist: decide what role to play toward a particular patient, allow them to respond, and then point out to them how you could arrange this response because you understood their RRs so well.

I also learned from my therapist, that my other RR, which she called “making the best of you-making the best of me” was probably keeping me on the “controlled” end of “controlling-controlled” in life, and as a therapist. It was my tendency to “normalise”, “live and let live”, “look for the silver lining”, which had made my early life liveable, that might block confrontations with patients. And these were the all important confrontations in which 1 would not be able to sugar-coat what I saw happening and would have to say straight out what a patient was doing, thereby, hopefully, getting them to do something better.


Paradoxically, the “sensitivity” engendered by years of looking for the best in people was not a quality to be shelved completely. I found it helped a lot in collaborative discussions after reformulation and in developing the RRs. Patients felt I was on their side and saw the positive potential in them, which meant the “naming, blaming, shaming” aspect of writing down RRs did not become a problem. When there was defensiveness, worry about labelling, or blaming parents, the sympathy that came from my deeply etched RR “making the best of you-of me” rescued the situation.

This was not, however, what happened in the case of a patient I shall call Benny, a 33 year old, single, white graduate, rebuilding his life after three years of serious illness. Strangely and excitingly, my new supervisor and one of my cosupervisees, highly experienced in working with transference and countertransference, were not satisfied with my patient’s acceptance of his RR, which was “critical and unpredictable-striving and uncertain”. They encouraged me to role-play with him the situations in which he had avoided expressing his feelings by creating longwinded extenuating explanations for those who provoked him or for his inability to say how he felt. And role-play we did, to the point of his getting more frustrated and angry with what I was trying to show him and to his missing more and more sessions. I was quite taken aback when the supervision group held that this was alright; I had put the ball in the patient’s court, and he could realise and respond, or he could leave. The choice was his. That seemed so stark to me. It seemed to mean that as a therapist you were claiming to be absolutely right, and, well, that seemed to mean the patient was wrong or at least “in the dark”. I felt very sorry for my patient, and then I felt very sorry for being sorry, because I was not supposed to be, if I were right!

It was during this year that 1 took the impression that therapists and patients were divided into two camps, with many of the therapists I met seeming to act like “winners” with respect to their patients who were the “losers”. In many ways the therapists seemed subtly (and sometimes not so subtly) to be blaming, criticising, and controlling their patients. I tried to believe it was just the impression of a relative outsider to the field of therapy, and, anyway, even if it was true, they must be behaving this way to benefit the patient.

The two camps idea could not have been more clearly illustrated than by Benny’s case, because when I read the letter of reformulation in supervision and then to Benny, I found that he accepted, approved, thought accurate and helpful the exact portions of the letter that the supervision group thought could be left out. And he rejected the parts that the group thought were “spot on”.

Sadly, I think for Benny, the issue of what was his most important, operant RR got lost in the crossfire of this patient-therapist divide. Too much time was wasted in wondering if the patient was as benign and good-natured a victim as he and I thought or if he were hiding the horrible side of his nature: patient “bad”, therapist “right/vindicated”.

It was not until I role-played Benny in a training session that I realised what he was doing. It was a simple RR of “blaming-blamed” enacted with his father (who blamed him and others a lot), with doctors (whom Benny blamed a lot), and with himself (whom he blamed if he caught himself in the unchristian act of blaming someone!). Me he blamed partially, but also thanked; he mostly feared that what I was doing in the parts of the letter he did not like, and in the subsequent role-plays, was blaming him. I had missed the entire issue of how his unpermitted anger was habitually expressed through roundabout blaming until I “became” Benny. Possibly, I had felt that the feedback in supervision laid too much blame on him, not because that feedback was inaccurate but because it seemed to be expressed with a force bordering on anger and contempt.

Therefore, by the time I realised Benny’s correct RR, I had already written the goodbye letter, which, not surprisingly, the supervision group found useful from A to B, and Benny found acceptable and useful from B to C.


The reciprocity of the therapist-patient camps still seems to me something to be wary of I think we must not only avoid playing out “admiring-admired” with our patients, but also eschew the more seductive “contemptuous-contemptible”. It is tempting to relax into a stance of well-earned contempt toward patients. And I am not speaking about responsive anger or contempt during a challenging moment of therapy; I refer to a general feeling toward patients as a group. I think that individual patients will always pick up such a stance, even if a therapist thinks it is hidden or is justified as a mobiliser for therapeutic interaction. And if a patient picks up this contempt, even if s/he
is unable to name it, it will, I think, negate the implicit pledge of the therapist.

Although, at the beginning of training, I thought that RRs were difficult to identify, I begin to be convinced that one can see the outline of the psychological skeleton almost as easily and quickly as you can see that of the physical one. However, I hesitate to trust this conviction, knowing that I have consumed a strong course of learning, and, like a potent medicine, CAT’s central construct (inner parent-inner child; a core reciprocal dialogue) may be affecting my perception. I still respect my first feeling that it is a struggle to know someone else. Maybe this feeling will keep me from being confused by the “crossfire” and on the side of the patient.

Irene Elia

Petition to NHS England - The Case for Funding Training in the NHS 2021 Alert!
ACAT's online payment system has been updated - click for more information

Full Reference

Elia, I., 1998. Reciprocal Roles: Caught in the Crossfire. Reformulation, ACAT News Autumn, p.x.

Search the Bibliography

Type in your search terms. If you want to search for results that match ALL of your keywords you can list them with commas between them; e.g., "borderline,adolescent", which will bring back results that have BOTH keywords mentioned in the title or author data.

Related Articles

The Neurotic Neighbourhood
Elia, I., 2002. The Neurotic Neighbourhood. Reformulation, Autumn, pp.18-19.

What Happens After “Happy Ever After”?
Jenaway, A., 2010. What Happens After “Happy Ever After”?. Reformulation, Summer, pp.8-9.

Musings on Doing CAT with Couples
Gray, M., 2006. Musings on Doing CAT with Couples. Reformulation, Winter, pp.29-31.

Looking at a Therapeutic Relationship In CAT
Raymond, C., 2008. Looking at a Therapeutic Relationship In CAT. Reformulation, Winter, pp.26-27.

A Case Example of CAT with a Person who Experiences Symptoms of Psychosis
Dr Sara Morgan, 2018. A Case Example of CAT with a Person who Experiences Symptoms of Psychosis. Reformulation, Winter, pp.35-38.

Other Articles in the Same Issue

Reciprocal Roles: Caught in the Crossfire
Elia, I., 1998. Reciprocal Roles: Caught in the Crossfire. Reformulation, ACAT News Autumn, p.x.

SDR's for Beginners
Boa, C., 1998. SDR's for Beginners. Reformulation, ACAT News Autumn, p.x.

Sequential Diagrams, Reflections and Suggested Revisions
Ryle, A., 1998. Sequential Diagrams, Reflections and Suggested Revisions. Reformulation, ACAT News Autumn, p.x.


This site has recently been updated to be Mobile Friendly. We are working through the pages to check everything is working properly. If you spot a problem please email and we'll look into it. Thank you.