What does “containment” mean and why use this word as an alternative to or in conjunction with "holding"?


Chambers Twentieth Century Dictionary offers the following definition of containment: act or policy of preventing the spread beyond certain limits of a power or influence regarded as hostile, by means other than war. The Oxford English Dictionary suggests, holding (!), restraint, the action of preventing a hostile force etc. from expanding into other areas. In considering these definitions of the word it is apparent that most indicate some form of limitation or boundary. It is for this reason, I believe, that containment is more readily thought of in terms of physical boundaries (walls, fences, etc.) or in terms of proscribed behaviours or activities (rules, regulations, controls, etc.), than in other terms.


Instances of these uses of containment and probably their more common usage in residential social work with children for example, are those of setting boundaries (rules) around activities and behaviours, or of restricting liberty in the case of detention in secure units. Brookes (1991), however, identifies a similar limited view of containment in the worker-client relationship where she writes,


"It can refer to the necessity of firm and reliable boundaries to the relationship, and the need for a clear definition of the nature of the work in the mind of the helping professional. This sense of containment is only loosely linked to Bion's usage of the term, and has more to do with Winnicott's concept of a reliable and resilient enough 'holding environment' ... ".


(Brookes 1991: 137)




However, these are practical but limited uses of containment and it is not necessarily helpful to think of containment in such concrete ways. They do not convey the idea of a "relationship" which Bion (1962) proposed. In particular, Bion (1970) described a relationship between a "container" and that which was "contained" within the container.




The Container-Contained Relationship


Britton (1992) writes of how Bion saw the 'container-contained' relationship as pre-determined. That is, the first expression of this relationship was in the mother/infant relationship and its influence affected all subsequent situations.


"... the mother if she was receptive to the infant's state of mind and capable of allowing it to be evoked in herself, could process it in such a way that in an identifiable form she could attend to it in the infant".


(Britton 1992: 105)




In other words the parent was able to receive the projected anxieties, feelings and emotions of the child, tolerate them and communicate to the child that those feelings were able to be tolerated by the child himself. Bion's work was originally directed at the world of psychoanalysis. Britton argues that the benefit for psychoanalysis of Bion's work was, "... his notion of the transformation of experience through a process of 'containment', (Britton 1992:105). Britton describes how Bion thought that his patient,


“...took steps to give effect to his phantasy - e.g., that it was not the patient who was afraid, angry, helpless, despairing, impotent, or whatever, but that it was the analyst who should experience himself in that way ...".


(ibid.: 105)


A personal experience of this theory is that in the course of working on this paper, my newborn son at five weeks old went through a phase of crying, screaming and being generally very unhappy. I can remember feeling equally tired, angry, depressed and at times impotent in terms of dealing with those feelings in my son such that I could not tolerate those feelings in myself or help him with his. My inability at that moment to tolerate my son’s distress can be seen as a failure of containment.


Collie (1996) provides a similar account of projective identification and the containment of such projected feelings by workers in residential therapeutic settings with adolescents.


Britton (1992) utilises the word sanctuary to convey the idea of the "container" and the word meaning to represent that which is "contained".


Citing Bion (1962), Britton says,


"The 'contained' gives meaning to the context which contains it. The 'container' on the other hand gives shape and secure boundaries to that which it enshrines".


(Britton 1992: 104)


In this context there is a link with the task of social work. For example, England (1986) proposes in his view of the nature of social work cited in Chapter 1 that the social worker aims to understand the meaning of a person's experience and to communicate that understanding to the client. Britton, citing Bion's own description of work with a patient says,


"Meaning was felt by my patient to be something which would provide a desperately needed coherence to her thoughts, whose disconnectedness persecuted her"

(Britton 1992:103)




He goes on to say that if sanctuary is missing, i.e., there is no safety or security, then the patient has a sense of "falling for ever" or "that there is no floor to the world". If meaning is lost then there is internal fragmentation and incoherence. Clearly this is an image without limitation or boundary.


Britton describes two forms of the container-contained relationship. A 'symbiotic' relationship between the container and the contained leads to a mutually enhancing development of personality from new experiences. A 'parasitic' relationship means that the container is disrupted by new experience. i.e., the sense of continuity is disintegrated and subjectively experienced as fragmentation. So, Britton argues, in this second case in order to preserve some sense of continuity, "... all change must be resisted and no new experience allowed to emerge". (Britton 1992:112). An example of this phenomenon is that of Bobbie’s defensiveness against new experiences described in Chapter 4.


The result of such resistance Britton writes can be one of two 'catastrophic' alternatives - incarceration or fragmentation. The foregoing describes a “persecutory view” of containment. The alternative is an “ideal fit between container and contained”, i.e., perfect understanding.


Finally Britton writes, in reality there is tension between the container and the contained within ourselves, within our families, and within our institutions. This is what we live with. With reference to these three areas identified by Britton (ibid.:112) and using examples which move from the psycho-therapeutic to the social work context I will illustrate some of the relevance of containment to social work.


Containment within Ourselves


Schneider (1996) describes the arrival of a new patient with a 'borderline personality disorder' to a psychotherapy group. Schneider takes the view that the presence of a borderline patient can be both threatening and potentially helpful to the group. Threatening, in the sense that the borderline patient slows down the process of integration within the group and also often acts out what the more heavily defended patients feel but cannot express. Helpful, in the sense that the presence of the borderline patient may enhance the emotional feeling content and expression within the group by raising tensions and demands. All of these processes contribute to a raising of anxiety levels which can lead to patients leaving the group and therefore the disintegration of the group. Thus, the need for the group facilitator to manage these potentially disintegrating factors becomes crucial to the success of the group. Schneider describes how she needed to maintain a role as active leader of the group for a considerable length of time before the group was able to feel secure. She writes,


"This was done in order to contain and hold both group and borderline patient until such time as both felt sufficiently secure ... . (The) process whereby the group acts as the container did not come into play until much later. It was therefore left to me to provide the containing function during the first year".


(Schneider 1996: 128-129)




Throughout her paper, Schneider uses the terms "holding" and "containment" together. However, it is clear that she feels she performed a more containing function in the early stages of conducting the group and exercised a more holding function later on as the group began to contain its own anxieties. In ending her paper Schneider describes the establishment of a 'therapeutic environment' through a process of holding.


"Winnicott's (1960) theory of the parent-infant relationship offered me a framework in which to understand the consequences of my borderline patient's early childhood experiences and helped me to provide a corrective experience. His concept of holding enabled me to understand the need to create a safe environment, during the early phases, such that the group and the borderline patient can be held long enough for them to reach a point where they can take over the therapeutic containing functions".


(Schneider 1996: 133)




From this account, the containment which Schneider writes about was something she contributed from within her own individual resources at the beginning. Later the containment offered by the group for itself was much nearer to the idea of the holding environment described in the previous chapter. The initial process required establishing and sustaining relationships with the patient and the group.


Containment within Families


Two papers illustrate the use of containment in work with families. Brookes (1991) writes about the use of containment in marital work and Johns (1995) writes about child protection work involving containment. In the first example containment appears to be more about the containment of destructive forces from within, while the second example is rather more containment in the sense of protection from external destructive forces. In both cases however the idea is one of an open system at work as was described by Shapiro and Carr (1991) in the previous chapter.


Marital Therapy:


Brookes (1991) has written about her experience of Bion's (1970) concept of 'containment' and the 'container-contained' relationship in marital therapy. In her introduction Brookes describes a couple for whom,


"... their marriage was no longer working for them as a container in Bion's sense, and they wanted it held and thought about in the therapeutic container of marital work".


(Brookes 1991: 133)




Brookes goes on to say that there is a clear analogy between marriage and Bion's 'container-contained' relationship and she discusses the phenomena of "unconscious fit" and "shared fantasy" in the marriage relationship. In the first case, that of "unconscious fit", these are the attributes of the partner which match a need in the other partner but of which neither partner is consciously aware. In the second case, "shared fantasy", both partners perceive a particular attribute in the other which they believe will work well together for them. Both phenomena can exist side by side and each partner is seen as being able to contain and tolerate if not also modify destructive feelings in the other.


"Marriage ... can be a 'commensal' relationship in which both participants grow emotionally. Neither partner, but the marriage, the relationship between two partners, is seen as 'the container', which may, or may not have transformational qualities".


(Brookes 1991: 138)




However, if the unconscious fit and the shared fantasy fail to satisfy each of the partner’s expectations, anxieties result and the marriage relationship fails to contain those anxieties. Both partners can project destructive feelings into the other, as an infant may do with its mother, but in this case they are intolerable feelings. These ideas link with Bion’s concept of a “symbiotic” and a “parasitic” containing relationship.


Child Protection:


In the second example of containment in family work Johns (1995) describes the case of a client with whom she was involved because of child protection concerns. Elsa, a young mother with four children under six, resented the concerns expressed by other parents in the neighbourhood about her care of her children. Elsa saw the expression of these concerns as "intrusions" into her life. She believed that her children should be allowed to grow up without any "interference" by their parents. In other words she had what Johns calls a laissez-faire approach to childrearing. Although Elsa's care of her youngest child was always felt to be "good enough" Elsa appeared to "switch off" from her older children when they started to exercise independence of thought and action. She also "switched off" from professionals who had tried to work with her on this problem, which they felt was contributing to neglect of the children and on several occasions exposing them to physical danger because of a lack of supervision and safety. An alternative to the notion of “switching off” might be that of “letting go” in an emotional sense.


"While Elsa appeared to 'hold' or 'contain' her babies, both physically and emotionally, during their early infancy, it seemed that as soon as they began to separate they were, in effect, dropped from her mind. This led to the situation in which they explored the environment without adequate supervision and were exposed to dangers. What for me was most perplexing was how and why Elsa suddenly stopped holding her children in her mind and became apparently oblivious to their needs for safety and containment".


(Johns 1995: 35)


Elsa's capacity to offer containment to her children was seen as being the result of her own emotional fragility and the lack of external family support. The lack of external supports merely reflected her lack of emotional resilience internally.


Applying the theory to the worker-parent relationship in this case Johns asserts that what Elsa clearly required was what Winnicott called 'holding' and what Bion called 'containment' in order that she might develop these capacities more effectively for her own children. However, Elsa appeared to have as much difficulty in allowing herself to be held as she did in holding her own children in mind and her ambivalence about the role of the social worker as a helper persisted. Elsa's level of trust in the social worker as a potential helper rather than a persecutor was built up only after a considerable length of time. In the end a sufficient change was established by Elsa to the extent that she could accept some support and so begin to support and protect her own children sufficiently.


Like Schneider (op.cit.) Johns argues that the change in Elsa arose from the long-term relationship she had established with Elsa, which provided her with containment. The relationship enabled Elsa to introject a benign sense of being helped rather than one of complete abandonment (falling for ever) or total intrusion (fragmentation).


Containment within Institutions


Two papers look at the issue of containment within institutional settings. Parker (1989) describes the use of containment in a residential setting while Downes at al (1996) look at containment in the context of social work and training.


Residential Care:


Parker (1989) uses the word "asylum" to describe the environment created in one therapeutic community for children, (New Barns). In its original sense “asylum” meant a place of refuge and protection from the external world and one can compare this with Britton's (1992) use of the word "sanctuary" which may represent more an idea of refuge than of active protection. Both, however, are similar. In the context of residential child care Parker focuses on containment in the sense of boundary setting as well as in the context of helping others contain anxiety and this aspect of containment he writes, is much nearer to the protective "father role" idea of Winnicott (1950). Parker (op.cit.) also writes of the community providing "holding" for the staff as well as for the children and describes the therapeutic community as providing containment "in a real sense".


"By this I mean not only the capacity to deal with an outburst or an explosion from the child but also the day to day provision of 'containing' or 'holding' experiences for the child".


(Parker 1989: 15)




Once again, the difficulty with this description is that it does not really distinguish between "holding" and "containment". Parker also describes some of the children who come to the therapeutic community as being "uncontainable". I prefer the phrase "in need of containment" as "uncontainable" has a rather foreboding sense of finality about it, as if nothing will contain these children.


Social Work and Training:


The aim of the paper by Downes et al (1996) is to examine the relevance of "holding" and "containment" to the maintenance of successful engagement by practitioners with their professional task, with their organisation as well as with their clients in the context of major organisational change. Like Hadley and Clough (1996), Downes et al describe similar feelings of loss and change, as experienced by social workers and care managers, resulting from the Community Care implementation within their social services department. Downes et al (op.cit.) describe the setting up of a training module to assist workers coping with these feelings from which they could learn and make use of in their practice. The necessity of using containment as a mechanism for refuge and protection is apparent in this short extract from their paper.


"The notion that local authorities are now open to external market forces and central government financial regulations means that restructuring has to take into account the management of external uncertainty, as well as interdependence of staff within social services. The purchaser/provider arrangement internally, and collaboration with health services and the private sector are clear examples of this."


(Downes et al 1996: 26)




This quotation seems to imply perceived threats from both inside and outside the organisation. The local authorities are described as being "open to external forces" and yet set around by increasing "central government regulations" but the end result is one of "uncertainty" for those concerned. The external attempts at containment of the care management task through rules and regulations appear to demand more attention than the needs of workers' anxieties to be contained by their organisations. Drawing once again on Britton’s (1992) analysis, the anxiety of workers stems from fragmentation and a lack of coherence. Meaning is lost.


One of the views put forward by Brookes (op.cit.) is that the use of containment in a social work setting very often does not take sufficient account of the effects of institutional anxiety in its execution and I have made reference to this in Chapter 1. The splitting of the task into assessing, purchasing and provision inevitably means that no one worker and no single organisation or department remains in contact with the whole of the client's experience. It is difficult to imagine how this might be experienced by a client and especially a child. For those in long term need the emotional needs become obscured or denied. The totality of the child's experience is fragmented and uncontained.




Containment in a loose, practical and superficial sense can be about physical boundaries (walls, fences, etc.) and it can mean boundary setting (rules, regulations, controls, etc.). Boundaries are important, especially for boundary-less people, but they tend to be inflexible things when compared with relationships. Damaged relationships can also be inflexible as in the case example of marital work given by Brookes (1991) but normal, healthy, stable and sustained relationships allow for flexibility, growth and change.


In a therapeutic context containment is about a relationship. A relationship between the container and the contained or between the helper and the helped or the healer and the sick. The idea of containment as an open, fluid system or as a holding environment gives greater emphasis to its potential to heal damaged people and damaged relationships by creating coherence and tolerance. The “sanctuary” or “asylum” offered by the relationship is not a physical structure nor a rigid modus operandi but a reliable, continuing and uninterrupted emotional relationship - a modus vivendi.