‘Not calling it a day’: Perspectives on ending therapy

Considerations on ending therapy from a relational perspective

Christina Moutsou

thumb_Dancing Imagination by deanfotos66_1024
Image: Thumb Dancing Imagination by deanfotos66_1024 (on Flickr)

A relational perspective
The inspiration for this paper came from a patient leaving her long-term therapy with me to begin some form of CBT treatment in the hope of ‘getting better’ quickly. (I favour the term ‘patient’ here over ‘client’ to emphasise the element of care in the therapeutic relationship over the contractual side of the relationship.) This patient had made significant progress during the two and a half years she worked with me in terms of her own objectives when she began the therapy. These were to do with her difficulty in sustaining close and committed relationships and her wish to change this pattern. She had experienced very unstable attachment figures in her early childhood and through the therapy with me, she had begun to see how unsafe it felt to allow herself to be close to others.

The reason she chose to leave was the emergence of a new symptom. This took the form of anxiety, at times severe, about having established a close relationship with a man for the first time in her life, a relationship that seemed to have the right ingredients to become long term and committed. The therapy focused then on how being in a relationship brought up an intense fear of loss for her. Ironically, it was precisely the progress she had made in allowing herself to feel close and attached to me and her boyfriend that her anxiety was expressing. Although the anxiety had begun before I had to suddenly take a break of four months for health reasons, I believed and repeatedly conveyed to her upon my return that her anxiety was exacerbated through my unexpected break and the issues around dependency and attachment that it had brought up for her. She was not ready to fully see the link and persevered in her wish to find a complete cure through seeking out a solution-focused therapy. She was hoping that by having CBT she would stop having the symptom as quickly as possible, but the problem was that in doing so, she would also ironically go against the mainstream CBT advice of exposing oneself to what triggered her fear – ie she would run away from being in an intimate therapeutic relationship.

The above vignette highlights in a rather crude way how the ending of therapy often connects with the resistance we all experience when issues around dependency are stirred up, as subjects of a Western socio-political and cultural arena that condemns dependency needs as pathological, and therefore perceives long-term, open-ended therapy, as pathological also. In her paper ‘Maternal Resistance’, Layton points out that our Western thinking along with psychoanalytic theory are underpinned by valuing separation and independence over dependency as the desired cultural ideals (2010:191-209). Such values, whether implicit or explicit, are bound to create resistance in both therapist and patient, the stronger the attachment becomes and the more it feels like a close relationship. In the introduction of the same edited book on therapy endings from a relational perspective, Salberg points out that it takes considerable courage to view endings from a relational point of view, as this undermines the sense of certainty and mastery of technique that an analyst may possess when endings are concerned. More precisely she says:

“When I entered my own analysis, I believed that I would be learning how to fully separate and individuate from my family of origin and complete what had felt, heretofore, to be an incomplete process. Staying attached had not felt like a welcomed option, because it felt shameful in its link to dependency issues.” (Salberg, 2010: xvii).

parents 1

It would be useful here to clarify what I mean by therapy from a relational perspective, as my definition is not synonymous with that of the above author, who views the relational move as more or less synonymous with adopting a ‘middle school’ analytic position (ibid 2010:xix). I largely agree with Loewenthal’s definition of the relational (2014:6) which highlights a threefold dimension of the term. As he points out, the relational prioritises the therapeutic relationship over questions of technique. Research has consistently demonstrated that a constructive relationship with the therapist is the most crucial factor in positive outcome of the therapy for the patient (Samuels, 2009:357–362). It also places the emphasis on the inter-subjective element of the therapeutic relationship – ie the therapist making herself available to the patient as a real person and not a blank screen. Finally, it allows for a more pluralistic and multimodality approach to therapeutic work. I also agree that the emphasis on the relationship introduces a bias in the process, which is to do with the assertion that the therapist values stable, long-term relationships over other possible exchanges between people but I would argue that such bias is explicit from the beginning of the therapy journey and mutually agreed between both parties. I see therapy as an inter-subjective dance between two people that has a real impact on the lives of both (Haberlin, 2014).

In her book on managing difficult endings in psychotherapy, Murdin points out that the ending of therapy is a very dangerous arena and that in her experience of sitting on complaints panels, often complaints are brought up by patients in relation to the unsuccessful ending of their therapy. She argues that the ending of therapy can bring up very painful and unresolved feelings for the therapist to do with loss, finality and death and that often therapists use the therapy process as a manic defence against such feelings, not wanting to face up to their own ageing, illness and losses (Murdin, 2015:89–104).

While I very much agree that the ending of therapy is one of the most difficult and dangerous issues to tackle and that in my experience it is both painful and often unsatisfactory, I question Murdin’s assertion that one of the therapist’s central tasks is to facilitate the ending of the therapy and the analysis of the patient’s as well as one’s own defences against endings and what they represent. This seems to derive from the traditional analytic position according to which the therapist masters a technique that facilitates the deconstruction of the patient’s defences. When it comes to questions of suffering, loss and death I would argue that such assumption is rather arrogant, as we are all equally helpless in relation to the above. In other words, if therapy is a significant relationship, why should the therapist be in charge of introducing, facilitating and managing its ending as an integral part of the therapy process? If we were to imagine a couple approaching the end of their lives after many years of close relationship, should one of them say to the other: ‘Darling, we know it too well that we are both going to die, and chances are that one of us will have to face up to the loss of the other. We‘d better work through an ending of our relationship now, so that we can process the pain of the impending separation and death’? Instead, elderly couples that manage a close relationship often talk about the fear of the other’s absence, the loss of the companionship and what that would feel like when it becomes unavoidable. In this paper therefore, I am asking the question of whether it is appropriate or desirable to see long term psychotherapy as a process that necessarily has to entail a planned ending. Such a question is inevitably intertwined with whether therapy aims at the alleviation of mental suffering or at reconsidering our suffering and helplessness in relation to other people.

parents 2

The End of Analysis
“The aim of analysis is to transform neurotic misery into everyday unhappiness” (Breuer & Freud,1893-95/1955:305).
In his 1939 paper, ‘Analysis Terminable and Interminable’, one of his later and consequently more mature and philosophical papers, Freud poses the question of what is the aim of a successful analysis and also, if analysis can ever be completely successfully terminated. The two criteria he sets out for the successful termination of analysis are: (1) The alleviation of symptoms, and (2) The dissolution of the transference. In his discussion of the above, he clearly finds both premises problematic. As many of Freud’s patients presented with so-called conversion symptoms (hysteria), it was a clear aim to alleviate such symptoms before the end of therapy. The same would apply for symptoms that patients present with nowadays, such as panic attacks, self-harming and so on. However, as Freud has infamously implied in the quote above, what are we to do with issues such as loneliness, isolation, illness, or bereavement? We can choose to see these problems as neurotic symptoms or we can choose to see them as part of the human condition and the world we live in.

The second premise, that of the dissolution of the transference, Freud also recognised as problematic as he said that the question of ending analysis, especially if brought up by the therapist, can create complications in the transference. Besides, transference can never be completely resolved. Despite his reputation for rigidity, Freud was more lax in his boundaries with patients than most therapists of any modality would be nowadays. He would occasionally go for walks with them, introduce them to members of his family or turn them into lifelong friends.

Knowing when to stop
In my mind, there are two reasons for which therapy should not necessarily end, unless either or both parties wish it to and unless there are external factors that impose an ending. All forms of therapy are ways in which one attends to one’s feelings and one’s thought processes. As people who have put themselves through long-term therapy know, such attentiveness to one’s own process never really stops once a person has engaged enough in the therapy. Freud recommends in his paper that analysts should begin a new analysis every five years or so, if they want to process successfully their own unconscious conflicts. Of course, there are other ways in which one can attend to one’s own process, a very good example of which is doing art. However, would anyone ever suggest to an athlete that rather than continue with one-to-one training, they should now find a way of exercising by themselves? It may be worth thinking therefore, that there are reasons for which we are collectively inclined to see therapy as something that has a beginning and an end rather than see it as an open-ended process. The reasons may not have only to do with a certain inclination to follow and mimic the medical model (Freud, 1912e/2002), but also with the idea that therapy is not like any other relationship, and therefore it would be somewhat insidious to see it as such.

parents 3

This brings me to my second point, which is also extensively explored by Adam Phillips (2006:21-41) in his paper ‘Talking nonsense and knowing when to stop.’ If we are to accept that therapy is a form of relationship in which progress is made through the engagement of the two people with each other in the room, then why should we see it as different to any other form of relationship, where the ending is dependent on whether the people in the relationship want to continue seeing each other or not? As Phillips points out, the end of seeing each other in a relationship and especially in the therapeutic one, does not signify an absolute ending, in that two people can continue thinking of each other and processing things about their relationship well after its ending (2006:22). As experienced therapists who have seen a number of people to the end of their long-term therapy know, progress is sometimes made well after the end of the therapy and also, the therapeutic relationship can move on in its absence.

Ordinarily, personal relationships we form in life often end for what we could call internal reasons, ie the two people fall out or they drift apart for what we could call external reasons; one of the two people can no longer meet because of illness, death or relocation. These are also common reasons for which long-term psychotherapy ends. Often, there are practical concerns or otherwise; the conversation that therapy can be when it works becomes too conflicted or too sterile. However, in the context of open-ended therapy, such reasons would be seen as an unsuccessful ending of the therapy, as Freud has highlighted in some of his case studies such as Dora’s (Freud, 1905e/1955, Mahony, 1996).

What is the aim of a successful analysis? A consideration of relationship and ‘Care’
It has recently been pointed out in the media and by philosophical thinkers that the pursuit of happiness, culturally portrayed as the aim of a fulfilling life, can actually lead to a sense of frustration and dissatisfaction with one’s life when it does not resemble the expected, ideal model (Van Deurzen, 2009). In the consulting room, we see an increasing number of people who get hooked to a virtual dream of happiness and success, imagining their friends’ and other people’s rich lives through daily peering at their pages in Facebook. These people get increasingly isolated and frustrated with their lives that cannot possibly compare to such imagined riches. Their pursuit of happiness focuses more and more on image and status as opposed to relationship and connection.

The CBT model, but also any model of short-term, solution-focused therapy, the only models promoted nowadays by the government, unwittingly reinforce the idea that we should be happy and that unhappiness is actually a medical symptom that we should tackle with treatment (King & Moutsou, 2010). But what about people who have endured severe trauma and losses in their lives or even those who are becoming unhappy precisely because they set themselves such high standards of fulfilment that they can never reach? One of my patients, somebody who has endured childhood losses, bereavement and life-threatening illness was put off therapy for years when she witnessed a therapist jot down ‘victim complex’ during the initial consultation after she narrated to her the tragic losses that were bringing her to therapy. So, the question arises of whether the remedy for tragic occurrence or persistent unhappiness is to treat it or perhaps to try to understand it and even to empathise with it.

If we are to accept that the human condition involves a significant degree of suffering, as we are all faced throughout our life with death, often with tragic circumstances and almost always with frustration, then the aim of any talking therapy should not be focused on the alleviation of unhappiness but on the question of how one positions oneself in relation to it. The first step would be to acknowledge and accept its existence. This is something that analytic therapies are best at doing. As Adam Phillips has pointed out, psychoanalysis is one of the few places in the modern world, where one can feel free to be sad and unhappy (Phillips, Guardian Interview, 2012). This can be valuable on its own.

However, to return to the question of the aim of a successful analysis, the legacy of the medical model through Freud has created confusion in analytic therapies about whether the focus should be on some form of cure or not. To the degree that we see human suffering as related to the human condition, then focusing on the lived experience and on ways of being in the world, as the emphasis has been within existential therapy, seems more appropriate. As Heidegger (1927) argues, we cannot exist in any other state than in inevitably being in the world with other people since our conception and birth. His concept of ‘care’ seems highly relevant to the question of what constitutes the outcome of a successful therapy. For Heidegger, ‘care’ is not about caring for others, but about accepting that we are all intertwined with each other in being in the world and therefore, we inevitably affect each other. The purpose therefore, of living in a meaningful way is to be authentic in relation to ourselves and other people (Van Deurzen, 2009), as opposed to the pursuit of some elusive happiness.

To rephrase the above, one could say that to engage in meaningful relationships with others, and ways of sustaining a conversation about what it is that matters in life, could be closer to the aim of a successful therapy. To do so, one would need to get away from a medical model of pathology and the alleviation of symptoms. Instead, a successful therapy would need to focus on the fact that what makes people feel unwell is the lack of authentic relationships in their lives and that the therapeutic one could constitute for some people an important beginning of being with others in the world.


Christina Moutsou is a psychoanalytic psychotherapist in private practice in Queen’s Park, NW London. She is also a supervisor of phenomenological research for New School of Psychotherapy and Counselling (NSPC). She is a Cambridge graduate with a PhD in Social Anthropology. Her co-edited book with Rosalind Mayo Mother and Other Stories: From Matricide to Maternal Subjectivities will be published by Routledge in 2016. Address for correspondence: [email protected]


Breuer, J. and Freud, S. (1895d) ‘Studies on Hysteria’, Penguin Freud Library, Volume 3.
Freud, S. (1905e), ‘Dora’. Penguin Freud Library, Volume 8.
Freud, S. (1912e/2002) ‘Advice to doctors on psychoanalytic treatment’, in Wild Analysis (2002). London: Penguin Classics.
Freud, S. (1937c/2002) ‘Analysis Terminable and interminable’, in Wild Analysis (2002). London: Penguin Classics.
Haberlin, J. (2014) ‘Beloved’, in Relational Psychotherapy, Psychoanalysis and Counselling, London and New York: Routledge.
Heidegger, M. (1927/1978) ‘From Being and Time’ in Basic Writings, London: Routledge.
King, L. and Moutsou, C. 2010 (eds), Rethinking Audit Cultures, Ross-on-Wye: PCCS Books.
Layton, L. (2010) ‘Maternal Resistance’ in Salberg, J. Good Enough Endings, New York, London: Routledge.
Loewenthal, D. and Samuels, A., (2014) Relational Psychotherapy, Psychoanalysis and Counselling, London and New York: Routledge.
Loewenthal, D. (2014) ‘The Magic of the Relational?’ In Relational Psychotherapy, Psychoanalysis and Counselling, London and New York: Routledge.
Mahony, P. J., Freud’s Dora, Hew Haven and London: Yale Press.
Murdin, L. (2015) Managing Difficult Endings in Psychotherapy, London: Karnac.
Phillips, A. (2006), Side Effects, London: Penguin Books.
Phillips, A. (2012), ‘A Life in Writing’ Interview, The Guardian (1 June 2012), England.
Samuels, A. (2009) ‘Trahison des clercs’: Professional and political notes on the inflation of psychoanalysis-Responding to Lesley Murdin, British journal of Psychotherapy, 25(3), 357-62.
Samuels, A. (2014) ‘Shadows of the Therapy Relationship’ in Relational Psychotherapy, Psychoanalysis and Counselling, London and New York: Routledge.
Salberg, J., 2010 (ed.) Good Enough Endings, New York, London: Routledge.
Van Deurzen, E. (2009) Psychotherapy and the Quest for Happiness, Los Angeles/ London/ New Delhi/ Singapore/ Washington DC: Sage.


Please enter your email address below to receive notifications about Contemporary Psychotherapy:

Something went wrong. Please check your entries and try again.


20 Lonsdale Road

In association with The Minster Centre, London.


Opinions expressed in this journal are solely those of the author(s).
Publication in Contemporary Psychotherapy does not imply endorsement of those views.


Copyright belongs to Contemporary Psychotherapy. Material may only be reproduced with written permission from the Editor. Authors may use their own material elsewhere after publication without permission. Production of single copies for personal use is allowed without special permission.

Scroll to Top