search for


Holding and Its Clinical Vignettes
Psychoanal 2022;33:31-35
Published online April 30, 2022;
© 2022 Korean Association of Psychoanalysis.

Hongsuk Jang

Dr. Jang’s Psychiatric Clinic, Seoul, Korea
Hongsuk Jang, MD, PhD
Dr. Jang’s Psychiatric Clinic, 510 Seolleung-ro, Gangnam-gu, Seoul 06162, Korea
Tel: +82-2-565-9889, Fax: +82-2-565-9809, E-mail:

Earlier draft of this paper was presented at the IPA Asia-Pacific Conference–Taipei 2017.
Received March 16, 2018; Revised March 14, 2022; Accepted March 14, 2022.
cc This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Contrary to Freud, who understood childhood state of mind through observation and understanding of adult patients, Winnicott, as a pediatrician, directly observed numerous children, mothers, and their interactions in his practice and postulated his theory of childhood state of mind as either healthy or pathologic. Among his contributions to psychoanalysis, ‘holding’ and ‘good enough mother’ are concepts of interest in this paper. Winnicott, as well as other theorists after S. Freud has been known for paving the way into understanding and treating so-called ‘sicker’ patients such as borderline patients. In this paper, I illustrate and discuss a couple of clinical vignettes to understand and help the patient in an emotionally overwhelmed state. Particularly I highlight the differentiation of the real and fantasy and/or mentalization of the patient through the proper clinical application of the ‘holding’.
Keywords : Holding; Mentalization; Reality testing; Fantasy.

One day when I was struggling to describe “What holding is like in a clinical sense,” I was driving a car with my young son at the passenger’s seat through a riverside park of Seoul at night. There was not a soul in the park, and it looked deserted. I said, “It would be scary if one walks alone at this late night.” My son replied, “Yeah, right. But I feel okay, in the car, with you, Dad.” He went on to talk about different look of trees and artificial structures in the park from daytime that evoked the look of ghosts or monsters. He looked very amused. It struck me in that I did not do attempt to soothe my son or to elicit his fantasy about scenery. His thought was very spontaneous. I suppose what enabled him to express a scary fantasy in amusing way was his feeling secure at the very moment, with me, in the car. While having an interesting chat about ghosts, we arrived home and I continued to ponder on the subject.

At last, I realized that I am attempting to describe what is fundamentally extremely difficult because the ‘holding’ of a parent represents the environment of years in which the verbal language of the child is not fully developed. In other words, it mostly belongs to a preverbal era. So, I feel what ‘holding’ is, but I have a difficulty to put it into words when I venture to describe it. As an analyst, however, the professional tries to translate experiences that were not yet represented in words, I tried to continue to answer the question: “What holding is like in a clinical sense and how it could be applied in daily practice?”

Review of the Definition of ‘Holding’

Before looking into my answer, let me briefly go over about Dr. Winnicott who coined the term, ‘holding.’ Donald Woods Winnicott was born in 1896 in England and died in 1971. He studied medicine and then pediatrics. He had two personal analyses, first with James Strachey for 10 years and second shorter one with Joan Riviere. His influential supervisor was Melanie Klein and he considered himself lucky to have such a generous and pioneering teacher as Klein (Winnicott 1962). But he chose to remain in the middle group after ‘controversial discussions’ between Melanie Klein and Anna Freud because of some theoretic disagreement between Klein and himself. Unlike Klein, who put much emphasis on internal object world, Winnicott comparatively put more emphasis on early external mothering environment, which enabled him to develop many very important unique psychoanalytic ideas such as transitional objects, the ‘good enough,’ rather than, ideal mother, holding environment and so on. His originality stemmed from direct observations of numerous children, mothers and their interactions in his practice and postulation of his own theory of understanding childhood state of mind, either healthy or pathologic (Winnicott 1945). He is also known for treating large number of so-called ‘sicker’ patients in his practice and his approach has been helpful to treat them.

Although defining ‘holding’ or ‘holding environment’ in an abstract concept would be distant from practicing such things clinically, it is always useful to grasp the orientation through theoretical review.

According to Moore and Fine (1990), ‘holding’ is defined as below:

“A maternal provision that organizes a facilitative environment that the dependent infant needs. Holding refers to the natural skill and constancy of care of the good enough mother. Through this holding the infant experiences an omnipotence that Winnicott regards as an essential and ordinary feature of a healthy child’s development. It provides sufficient security that after a while the infant is able to tolerate the inevitable failures of empathy that result in rage and terror when the holding is lost” (Moore and Fine 1990, p.205-206).

In the same book wrote about the ‘good enough mother’ as, “Such a mother… offers at the right time instead of imposing her own timing and needs. Then, when infant must face frustration, aggression, and loss, she also provides support within a setting of ongoing basic empathy and holding. … so that the infant has a gratifying human context for a subjective sense of his or her own being, expression, and creativity” (Moore and Fine 1990, p.206).

Elusiveness is partly due to Winnicott’s writing style. He wrote lots of papers but, not in a theoretically systemized way. It gave readers the chance to experience powerful intuitive response (Ogden 2001). But his writing could not be fitted into a clear theory (Modell 1985). Another reason that the concept of ‘holding’ is so elusive is that Winnicott’s analogy of the treatment situation as a mother-infant relationship has been so powerful to give some of us an illusion to understand what he meant. Although we seem to know what good enough mother is but, in fact we do have different versions of it since we had experienced different mothers and different relationships. So must have had been Dr. Winnicott. As Winnicott (1962, p.169) wrote, “analysts are not alike,” we probably have different clinical notions of ‘holding.’

‘Holding’ has Many Facets

Now I want to break down ‘holding’ into several facets for the sake of illustrating my own understanding. Among others I want to investigate following three domains: Setting, Boundary and Analysts’ attitude.

Setting means the agreed prearrangement of the treatment frame such as the couch (or the chair), the regular frequency of appointments, verbal communication (preferred mode of communication), the exclusive attention to how the mind works (method), payment (cancel policy), roles of each party and so forth. It defines time and space of the treatment and gives protagonists inside of the setting unique respective roles. Often, in a classical Winnicottian sense, the analytic setting represents a holding environment itself (Abram 1996). I basically agree on the notion of Abram, but it blurs the essence of clinical usefulness of holing environment, so I deliberately choose the narrower sense of setting in this writing.

Boundary represents what is possible or impossible in the present therapeutic relationship. I think the boundary is the inevitable creation of the setting. The setting necessarily limits the patient’s and the analyst’s freedom to act in whatever way they would act in social life outside treatment. The setting forms the boundary of each participant and the treatment itself. Anything that the patient brings in, whether a demand or expectation, if goes beyond the boundary, is, in a sense, unrealistic and could be viewed as transference phenomenon (Giovacchini 1987). As in the anecdote with my son, the moving car created the boundary between the secure inside and scary outside. A passenger’s seat and a driver’s seat, represents the boundary between my son and me just like role of a patient and an analyst in the treatment. In other words, my son did not expect me to give up driving a car out of fear or expect the car’s metal frame to be vaporized into the air. If either happened, it would render my son exposed and insecure and would prevent him from enjoying fantasizing about ghosts and monsters. So, boundary may act in paradoxical ways in analysis in that it gives participants safety to play around, and it also restrain their freedom to be all-free at the same time.

Lastly, the attitude of the analyst, the focus of this presentation. Even though pre-determined setting and its accompanied boundary are well understood by each participant, I think the attitude of the analyst plays the pivotal role to establish a clinically helpful holding environment. By attitude, I particularly mean the analyst’s capacity to think himself as one comprising component of the holding environment. In this stance the analyst works in three realities simultaneously: 1) in analyst’s reality, 2) in patient’s reality, and 3) in the middle of realities of the patient and the analyst as a comprising component of the holding environment. I would say that the analyst holds the temporary small universe for the patient’s ego to grow strong and realistic enough to break out of it eventually. In that universe any product from the patient should be allowed to enter while the producing process is influenced by the analyst in a facilitating and non-traumatic way.

The clinical usefulness of ‘holding’ comes from an appropriate maintaining of the holding on the part of the analyst with relations to a patient who is regressed into or arrested in early states of mind. This replicates temporarily the atmosphere of early infant-mother ‘good enough’ relationship, which safely accepts the fantasies of the patient. And, at the same time, it softly and steadily signals the patient that this space is not permanent one but just a way point to shared outer reality.

One result of misuse of the concept of ‘holding’ could be to be pulled in too deeply, and this causes the loss of actual holding function of the analyst as in countertransference enactments with ‘sicker’ patents. We know well that we should not venture to be their replacement ‘good mothers’ or ‘fathers.’ But their notorious transference-countertransference pull to enact a certain role, responsive to such a primitive state of relationship is well known (Sandler 1976). Sometimes the analyst unwittingly supplies their needs for them to be cured and rationalizes this violation.

‘Holding’ in a Clinical Sense

To summarize, the clinical meaning or aim of ‘holding environment,’ hence, is to facilitate fantasy formation of the patient and simultaneously to help development of reality sense. In other words, to facilitate the mentalizing function or to help the patient to attain the symbolic level of mental functioning. Then ‘holding’ provides the safe haven to prevent the destruction of therapeutic opportunities from rupturing the relationship because of frustrated primitive wishes, most notably omnipotent or primitive narcissistic one. I think my recapitulation as to the clinical meaning of ‘holding’ could lead to another important contribution of Winnicott, ‘transitional object’ or ‘transitional phenomena’ (Winnicott 1953), but they are beyond the scope of this paper. My purpose here, however, is to stress that ‘holding’ is an inextricable element of our treatment that facilitates the above-mentioned psychoanalytic processes that are deemed to be essential requisites of any successful analysis no matter what theoretical school is followed.

As I stated about the importance of analyst’s attitude towards the concept of ‘holding,’ if the analyst misunderstands the clinical implication of the concept of ‘holding’ due to one’s theoretical orientation or one’s character style, the treatment would result in failure, no change or damaging the patient. I will now introduce the vignette from another article (Giovacchini 1987) first to show how far the treatment goes astray if an empathic, responsive, and engaging analyst fails to recognize his own unconscious omnipotent wishes and rationalizes this failure as holding. Here the analyst is drawn too deeply into the patient’s inner reality, loses his own identity and distorts outer reality. And then I will present my own vignette in which I experienced a similar pull and enacted it. I also could not recognize my own primitive omnipotent wishes until supervisor helped me to see them. I lost holding functioning while I was drawn to the patient’s reality and lost my own.

Vignette 1

“The therapist of early forties was treating the young woman in her middle twenties. He asked for the supervision from Dr. Giovacchini because of her erratic demands and sometimes violent behavior. At first, he anticipated a ‘quiet’ analysis because the patient was shy, meek, and undemanding. He was, however, unpleasantly surprised when it turned out to be otherwise.

At first, he was aware of fatherly feelings toward the patient. The analyst became sufficiently concerned that he insisted on hospitalization. He became more and more concerned about her self-destructive behaviors such as taking a dangerous combination of drugs, roaming the streets at night, having unsafe sexual relationships and so on. He suggested calling him at a prearranged time every evening in addition to her daily sessions. This brought about temporary reassurance, but telephone calls pervaded his personal life severely but without his recognition of annoyance. He formulated this situation as reaching rock-bottom regression which they both had to survive, and he was doing everything for that survival. He hoped developmental growth might be stimulated after experiencing and surviving such a regression, which was a postulation Alexander (1956) suggested many years ago.

The telephone calls and her outer daily life reached at the point beyond his patience and tolerance. He tried to put a stop to it, but she became furious. The patient felt betrayed and abandoned and it had reached an intensity that made it impossible to continue treatment on an outpatient basis. The day the analyst said this to her, she would not leave the room after the end of the session and invaded the time of the next patient. He literally carried her out of the consultation room. He felt embarrassed and humiliated. The patient returned several hours later and destroyed the waiting room furniture. That was the point that he decided to seek consultation. He cancelled all future sessions and promised her he would call back and further discuss the situation about the future of the treatment. She was frightened of implicit threat to terminate therapy and agreed on his wishes. She also asked for a consultation for herself, and he was glad to refer her to a colleague” (Giovacchini 1987, p.156-158).

Vignette 2

The patient is in her early thirties who has been in analysis for three years. She at first consulted me for medically unexplained somatic symptom such as sustained muscle tension in the lower abdomen area. That symptom arose under the stressful relationship with her bossy and domineering roommate at her university dormitory. She could not tolerate criticism from her roommate nor assert herself for her own sake. She became more helpless and dependent than before. Even before the symptom development she was sexually naïve and was attracted to and dependent on powerful and assertive female figures around her. After her symptom developed, she became easily upset and quarrelsome when she was treated unfairly. But this part of the dynamic change was hidden under her colorful somatic symptoms until the analytic process facilitated the emergence of her unconscious infantile wishes. I recognized them by observing transference demands and by monitoring my countertransference responses. For example, the former were complaints about the method of ‘free association,’ the slow nature of the treatment, the frequency of the sessions, and by demanding responses of approval from me after her associations, etc. My countertransference was manifested by mild discomfort about her complaints. I considered them as early signs of a maternal transference where I was her controlling mother. This was interpreted and enabled her to become comfortable with the analytic processes.

As the analysis evolved her immaturity became more obvious: she suppressed enormous resentment and anger she felt by maintaining a meek, polite, and submissive appearance. She denied the responsibility that she had to understand and accept that her angry feeling is inappropriate.

She said, “Why am I responsible for that anger? Mom made me angry. She injected that feeling into me. That anger is not mine so it’s Mom who are responsible for abolishing it.”

The analyst became more concerned about the tension arose between the patient and me regarding the usefulness of ‘not directly gratifying’ all of patient’s needs. I started to wonder about the best way of helping the patient. After one and half year of analysis, the following interaction between the patient and me and between my supervisor and me occurred.

I thought that the analysis was going well because it enabled me to interpret regressive and projective defenses of the patients, but I did not expect this to result in her becoming extremely sensitive to small differences of opinions with her mother and/or her mother’s insensible responses to her emotional state. She felt ignored and became furious at such moments. Her angry outbursts were justified by her as necessary if she were to separate from her or if she were to change her mother’s attitude into more empathic way.

I became concerned that the unintended consequences of the analysis required a change in technique to stop her from damaging her relationship with her mother unnecessarily. I wanted to provide direct advice by disapproving of her inappropriate anger. It was a ‘blind spot’ of mine because I did not realize I was taking her mother’s side against my patient. My supervisor interpreted this identification with the patient’s mother that I was acting out. This enabled me to interpret my patient’s underlying omnipotent transference wish to have everything she desired from her mother. If I remained in the role of her mother, I could not work through the regressive pull from the patient and my own wish to force her to stop being inappropriately angry.


As we have seen in both vignettes, the concept of ‘holding environment’-our benevolent effort to establish an empathic tie with the patient can be transformed into the loss of the holding if we try to “create a holding environment in reality” (Giovacchini 1987, p.160). Instead, the analyst limits it to the relationship between inner reality of patient and outer reality. If the analyst does not permit the part of himself to form the holding environment, the patient will externalize inner psychic reality onto the outer world and will not have opportunities to observe one’s authentic materials. This is another example of the failure of the holding which is not like above mentioned unconditioned unrealistic acceptance. I was in a such a rigid state when my worries about being pulled into the patient’s needs caused me to lose technical neutrality. As I was deeply pulled into, my comments lost reference to my sense of reality and mind was operating in the frame of reference of the infantile world of the patient. After I recognized that I was also captured by my own infantile wishes, I was able to escape from it and was able to return to my usual comfortable position.

I will now turn to the drawbacks of ‘create a holding environment in reality.’ It fails because the patient’s regression is too deep, making her unable to observe the sessions herself. Hence regression pervades into the daily life of the patient. Her infantile wishes for dyadic relationship grow and seek immediate gratification, which lead the patient into substituting anger for reality testing. Therefore, the patient resists accepting the multi-dimensionality of reality and is reduced to dichotomous perception and thinking. In turn this causes primitive defenses such as splitting, projective identification to prevail. Such processes make the ego of the patient weaker and weaker. Hence the afflicted patient is easily overwhelmed by primitive impulses that are released by the expression of powerful affects. This completes the vicious cycle.

Lastly, I want to briefly comment about the symbolic meaning of the analyst as the interpreter of infantile omnipotent wishes, who shows the patient the existence of the different being, differentiated from the patient. The interpreting analyst not only facilitates the capacity to tell fantasy from the real, but also paves the way to form the reference frame of experiencing from dichotomous, concrete thinking into multi-dimensional, stereoscopic, and symbolic thinking. The analysist can have many opportunities to help the patient to confront (re-confront) and master (re-master) oedipal themes and to decrease the future likelihood of pathologic regression to an infantile state of the mind.

Conflicts of Interest

The author has no potential conflicts of interest to disclose.

  1. Abram J. The language of Winnicott. Perthshire: Clunie Press;1996. p.162-181.
  2. Alexander F. Two forms of regression and their therapeutic implications. Psychoanal Q 1956;25:178-196.
    Pubmed CrossRef
  3. Giovacchini PL. Treatment, holding environment and transitional space. Mod Psychoanal 1987;12:151-162.
  4. Modell AH. The works of Winnicott and the evolution of his thought. J Am Psychoanal Assoc 1985;33(Supplement):113-137.
    Pubmed CrossRef
  5. Moore BE, Fine BD. Psychoanalytic terms and concepts. New Haven and London: Yale University Press;1990. p.204-209.
  6. Ogden TH. Reading Winnicott. Psychoanal Q 2001;70:299-323.
    Pubmed CrossRef
  7. Sandler J. Countertransference and role-responsiveness. Int Rev Psychoanal 1976;3:43-47.
  8. Winnicott DW. Primitive emotional development. Int J Psychoanal 1945; 26:137-143.
  9. Winnicott DW. Transitional objects and transitional phenomena. In Collected Papers. New York: Basic Books;1975. p.229-242. (Original work published 1953)
  10. Winnicott DW. A personal view of the Kleinian contribution. In The Maturational Processes and the Facilitating Environment. New York: International Universities Press, Inc.;1965. p.173. (Original work published 1962)

April 2022, 33 (2)
Full Text(PDF) Free

Social Network Service

Cited By Articles
  • CrossRef (0)

Author ORCID Information