The cliché that people fall in love with their therapist is well established in the popular imagination and often treated with some amusement. In reality, what some clients are experiencing is a potentially harmful side effect of psychotherapy and one that should be taken far more seriously by the profession.

When a client falls in love with a therapist it is likely to be ‘transference’: the predisposition we all have to transfer onto people in the present experiences and related emotions and unmet longings associated with people from our past. In the initial stages of therapy, such transferences are usually idealising, because clients tend to project onto their therapists the qualities they longed for from their early carers, and so experience them in a particularly positive way. This can help establish the therapy and the initial intensity usually fades once the ‘honeymoon’ period of the therapy is over. However, for a small but significant number of people, the experience is very different: the idealisation intensifies rather than fades, and the client becomes increasingly consumed with and dependent on thoughts about the therapist. This can be immensely disruptive to the client and to their family, and can lead to adverse consequences, as the client’s autonomy and capacity to think rationally are typically compromised.

‘I was like a rabbit caught within blinding headlights. I was uneasy, but in a childlike way also excited by being so special to a person such as him. I was very confused, experiencing feelings that I had not felt before. I can only describe it as like being caught up in an emotional earthquake.’1

This article is based on the accounts of people who have had this experience of adverse idealising transference (AIT) and who have contacted the Clinic for Boundaries Studies (CfBS), an organisation that helps people who feel they have been harmed by their experience of psychotherapy or other professional relationships. Several hundred people contact the CfBS each year, and a substantial number report lasting harm as a result of AIT. The phenomenon is often discussed within the discourse on erotic transference but my emphasis is on the idealising rather than the erotic aspect; not all clients experience erotic feelings and, even when they do, these feelings are almost always infantile and sensual rather than adult and sexual. What is desired is not man but mother.2 This is also why any consequent sexual contact with the therapist is likely to be experienced as confusing and exploitative.

The CfBS working definition of AIT is ‘a transference reaction that impacts on a person, so that over a sustained period their ability to function in their usual way is adversely impaired’. The effects can last over a long period of time, as evidenced by the large number of people who contact the CfBS about an idealising transference that began decades ago and is still unresolved. It is also evidenced by the many accounts of transference harm that can be found on the internet (see and in the client literature.3 Members of the public who have experienced this believe that therapists are not sufficiently knowledgeable about the phenomenon, or are unaware that their actions can both cause and exacerbate the problem. Yet the professional literature does not appear to associate idealising transferences with serious and lasting harm, which is why I want to share what CfBS has learnt about AIT.

The propensity to develop intense feelings for a therapist has been known about since the earliest days of psychoanalysis and Anna O’s famous obsession with Freud’s colleague, Breuer. Freud4 used the analogy of a chemist handling highly explosive materials to describe the potentially catastrophic consequences when such feelings develop. Freud’s focus, however, was on the therapist’s experience of the phenomenon, rather than on the adverse effects on the client. He emphasised the erotic aspect, and believed that tenacious manifestations of the transference were a resistance to treatment and an attempt to seek cure through a new relationship. For this reason, he was clear that such transferences should be analysed and not reciprocated.

This view was further developed by Klein and her followers, who proposed that destructive and aggressive feelings are also present and should be interpreted, particularly in relation to envy and difficulties in tolerating the independent existence of the analyst.5 From the 1970s, following Kohut’s emphasis on the idealising rather than erotic aspect of the transference,6 a more positive conceptualisation of the phenomenon began to emerge. Kohut focused on the facilitative aspects of the transference and insisted that it should not be interpreted but left to take its course until firmly established. This view was controversial and attracted criticism, particularly from contemporary Kleinians. In subsequent years some therapists went further, suggesting that erotic desire in therapy should be facilitated.7 Our experience is that this encourages AIT to develop. Client accounts and insightful blogs on the internet vividly describe the harm that can result. A particularly extreme example of a toxic mix of cult-like idealisation and sexualisation in group therapy has recently been described by the mental health blogger Phil Dore.8

Although the literature greatly underplays the role of the therapist in AIT, it is important to state that there is a type of AIT that develops independent of the therapist, and quickly becomes very negative. This is known variously as malign, malignant, regressive or psychotic transference, and was elaborated in particular by Little,9 and more recently by Hedges.10 It refers to a situation in which people with no history of psychosis become regressed and develop intense, delusional ideas about the therapist’s actions in the therapy. Hedges believes that this is likely to happen just as a successful therapeutic alliance is forming, because the person’s fear overcomes their desire for connection.10

Effects of AIT

People who contact the CfBS about an experience of AIT emphasise in particular their feeling that they are disempowered. They often compare the experience to that of a powerful mood-enhancing drug, a religious experience or an addiction. They often use words such as hypnotic, enchanting, magical and sublime, and describe striking imaginary scenarios to illustrate the primacy of the therapist’s position in their life. For example, one person described being haunted by the image of a lifeboat with only one space, because she knew she would give it to her therapist and not her much-loved children.

Although people often describe the transference as making them feel ‘alive’, they also describe profound confusion, distress and shame. Retrospective accounts also often describe feeling in thrall to the therapist and considerable disruption to social and family life, as the person’s interest in other previously important relationships diminishes.

‘I cannot over-emphasize the devastating effect all this had on my husband and children. I think they could not recognise the person they had known – a family-orientated wife and mother. It was as if an alien had invaded my being and I was speaking and behaving in ways that were just not me. It is difficult after these years to understand the intensity of my feelings for him and the total subjugation of my will to his.’1

Some of the most common feelings and beliefs that clients describe when AIT is developing are:

  • believing that a ‘real’ relationship with the therapist would result in deep contentment
  • feeling that other aspects of life are diminishing in importance, including relationships with friends, a partner or children
  • feeling that the problems that brought the person into therapy in the first place are no longer important
  • feeling panic or depression at the thought of the therapy ending.

‘My feelings for Marion intensified. During the 166 and a half hours a week when I was not with her, I thought about her constantly. The rest of my life was dwarfed into insignificance… “Relationship” was no longer an adequate word to describe what bound us together. In my mind, I was transported into another world where I existed in a state of rhapsodic communication with Marion. We did nothing, we said nothing, we just were.’11

A transference of this kind clearly affects a person’s judgment and interferes with their autonomy, leaving them vulnerable to sexual, emotional and financial exploitation. It also masks the problems that brought the person into therapy, and so masquerades as a cure. A client may spend thousands of pounds on therapy, only to discover that their presenting problems have not been addressed.

‘A magic trick had been performed on me: in just a few hours of sitting alone in a room with Paul, a large part of my mind had effectively been taken over, leaving me with little left to expend on my work, social life and other parts of normal life.’12

Therapist characteristics and AIT

From our discussions with clients seeking our help, we have noted particular therapist characteristics that appear to be associated with AIT. These characteristics sit along a continuum but fall broadly into five overlapping categories. They begin, at the most severe end, with the psychopath who becomes a therapist. This is the ‘unscrupulous therapist’ described in the Foster report,13 who sets out to use transference to create dependency and then intentionally exploits the client for emotional, financial or sexual advantage, for years or even decades. Clients of such therapists often describe being drawn into cold, humiliating sexual activity, and/or financial and emotional exploitation, where they are coerced into making self-defeating choices. These therapists frequently exploit their knowledge of the client’s developmental vulnerabilities in order to exert maximum power and control.

Then there are the opportunist therapists, who may not set out to exploit the transference but cannot resist doing so when it emerges. They reap the emotional, financial and/or sexual rewards, and often convince themselves that the client’s feelings are ‘real’ and that sexual exploitation is ‘an affair’. They typically have poor professional boundaries, operate from a narcissistic position and often have relationship problems themselves, so the client becomes a source of comfort and validation. Clients of therapists in this category frequently describe getting into role-reversal situations with them.

The third category comprises therapists who offer love in the belief that they can compensate for their client’s history of poor parenting. Clients often respond with appreciation and idealisation, which encourages the therapist to continue practising in this way. If the client is predisposed to developing AIT, they are likely to find it difficult to tolerate the constraints of a time-limited love relationship. If they then act out their frustration, it is not uncommon for therapists to feel justified in terminating the therapy, without being aware of the part they have played. Some therapists in this category do not set out to offer love but respond to the client’s demand that they prove that they care and find themselves breaching boundaries if the client’s demands then escalate and cannot be satisfied.

‘When I met Karen I was struck by her warmth and confidence. She said she was an expert on my condition, that my life would change. I felt elated, as if I’d been blessed and chosen. We had a special bond, she looked out for me like nobody had done before… She said she would always be there for me, and when I questioned “how” she hugged me, fixed her eyes on mine and said, “Trust me.” This didn’t help. I needed to know how and this irritated her. I began to be silent during sessions and Karen said I was trying to sabotage the therapy and didn’t want to get better. She told me that her other clients improved because they trusted her. At the next session, when I saw her previous client leave, I experienced a sudden, visceral feeling of rage. I was drenched in fizzing emotion and couldn’t think. I went into the room, picked up a glass and smashed it. I held it to my throat… That was the last time I saw Karen.’14

The fourth group of therapists refuses to engage with the transference. They may do little or nothing to encourage the idealisation; when it emerges they ignore it, or treat it in a pejorative or disapproving manner. These therapists may feel incompetent, irritated or ashamed that this situation has arisen, and this produces shame and confusion in the client. The client then conceals the idealising feelings and they flourish in silence, until the adverse aspect becomes apparent because the feelings can no longer be hidden. Sometimes clients of these therapists simply leave the therapy and then find it impossible to resolve the transference.

The fifth and final category, at the far end of the continuum, is therapists who act appropriately but find that the client is predisposed to developing a regressive transference. This is likely to become apparent just when the therapist feels the therapy is going well, and it frequently involves delusional ideas about the therapist’s actions and intentions. It may be impossible for the therapist to resolve the situation because the client’s beliefs are so tenacious. Therapists in this group may find themselves the subject of a complaint, because the client truly believes they have acted inappropriately.10 This may also happen because the therapist cannot deal with the strain and ends the therapy without the agreed notice period.

Non-therapist factors in AIT

Clearly there will also be factors that contribute to AIT but are outside the therapist’s control. These include the client’s early developmental experience. Kohut6 theorised that the necessary conditions arise if a mother is unable to attune to the particular needs of her baby, and the baby is unable to internalise the mother. Little9 similarly associated the phenomenon with the infant’s poor experience of mothering. Blum15 made particular mention of children who have experienced sexualised parenting, and Hedges10 proposed that the absence of nurturing in infancy leaves some people with an insatiable desire for the mother they never had. This accords with the experience of people who contact the CfBS, who usually describe experiences of inadequate parenting, rather than overt abuse. Since not all people who have had inadequate parenting are predisposed to AIT, it seems likely that constitutional factors such as genetic makeup and the neurobiology of the brain also play a part.

The therapeutic setting may also contribute to AIT. Low lighting, a calm comfortable room, prolonged eye contact and finding themselves the focus of another’s intense interest may be a unique experience for the client and may be unconsciously associated with a promise of love and nurture. Because the ‘love’ is one way, it tends to mimic a maternal relationship rather than a mutual romantic relationship, which makes the experience all the more unique.

Gender is also a factor: AIT affects female clients more than male clients, and appears to occur more frequently between male therapists and female clients, although we at CfBS know of instances where the client and therapist are both female. When the client is male, the therapist involved is usually also male. We have occasionally come across AIT with male clients and female therapists, but the therapist has always been much older than the client. We have never come across AIT between an older male client and a younger female therapist, although this gender/age combination is perhaps also uncommon.

What helps to reduce the risk of AIT?

Risk of harm could be reduced if therapists routinely assessed their clients’ vulnerability to AIT, especially the more regressive forms of AIT, at the beginning of the therapy. We have noticed that the following three traits are strongly associated with regressive AITs:

  • the client has a history of dependent/idealised relationships, especially with health professionals
  • they are primarily seeking care, not insight into their problems
  • they hold unrealistic views about what therapy can provide.

We have worked successfully with clients who have a history of dependent/idealised relationships but do not have the other two traits. Where all three traits are present, our experience is that the client will be so strongly predisposed to regressive AIT that serious consideration should be given as to the appropriateness of one-to-one therapy. Consideration should also be given to the gender of the therapist. While the literature suggests that intense transferences are not gender specific, we have found that clients are often predisposed to develop AIT with therapists of a specific gender, and that they want to see a therapist of that gender.

The risk of AIT can also be reduced by responding appropriately when clients bring up transference concerns, as AIT is much more likely to occur if the first indications are ignored. Most clients tell us that they attempted to discuss their concerns at an early stage in the therapy, but did not get a helpful response. Instead of being alerted to the potential for harm, therapists often act in ways that escalate the problem. This includes disclosing their own feelings of attraction for the client, assuring the client that the feelings will pass and asserting that the feelings really are a reflection of the therapist’s qualities.

Crucially, we have observed that therapists whose clients develop AIT are unlikely to discuss transference, and are dismissive or hostile when clients suggest it. Clients also describe therapists becoming irritated, defensive and rejecting in response to discussion about the adverse effects on the client’s life (see the box for how the risk of AIT may be reduced).

Informed consent and AIT

Clients who develop AIT tell us they wish they had been warned about the possibility before the therapy began. They often point out that a drug with the same adverse potential would only be prescribed with informed consent. Most feel that, if they had been informed of the risk beforehand, the experience would have been less confusing and traumatic. This is discussed in some published client accounts of adverse experiences of therapy. For example, Simpson12 writes that she would have considered descriptions of idealising transference to be ‘far fetched’ if anyone had tried to warn her about it, but states: ‘If I had been warned, and decided to ignore the warning, I think I would have felt less cheated.’ Other therapy ex-clients, writing on the internet, describe feeling ‘furious’ and ‘tricked’ because they were not warned about this ( Many describe how they tried to research the phenomenon but felt frustrated by the absence of information.

Informed consent has not been embraced by the counselling and psychotherapy profession.16 This is perhaps because there has been almost no discussion of the risk of harm from transference. There may also be a fear that clients will be discouraged from engaging in psychotherapy or made anxious by raising the issue. These are, however, all factors that other healthcare workers negotiate successfully. For example, it would be unethical for a surgeon to recommend an operation without first discussing the potential adverse effects.

In our experience, AIT interferes with clients’ capacity for rational thought, making them vulnerable to both dependency and exploitation. As such, AIT is a potentially serious side effect of psychotherapy. The absence of discussion in the professional literature about this type of harm is concerning because there is much a therapist can do to discourage an idealising transference from becoming adverse. If therapists don’t know about it, however, there is a clear risk that they may unwittingly encourage it. What we hear from people who have developed AIT is that therapists should have a greater awareness of what might encourage it, and that clients should be informed about the risk before they embark on psychotherapy.

With thanks to the clients of the Clinic for Boundaries Studies who have given permission for their experience to be used.

Dawn Devereux was Director of Public Support at the Clinic for Boundaries Studies and is currently on sabbatical. She has a special interest in helping to resolve problematic situations in therapy.


1. Matheson M. Broken boundaries. London: Witness; 2008.
2. Stoller RJ. Perversion: the erotic form of hatred. London: Karnac; 1975.
3. Bates Y. Shouldn’t I be feeling better by now? Client views of therapy. London: Palgrave Macmillan; 2006.
4. Freud S. Observations on transference love. London: Penguin Books; 1915.
5. Kernberg OF. Further contributions to the treatment of narcissistic personalities. International Journal of Psychoanalysis 1974; 55: 215–240.
6. Kohut H. The analysis of the self. London: The University of Chicago Press; 1971.
7. Mann D. Erotic transference and countertransference: clinical practice in psychotherapy. London: Routledge; 1999.
8. Dore P. The strange family of Derek Gale. [Blog.] Unsafe Spaces 2016; 2 March. (accessed 1 March, 2016).
9. Little M. The delusional transference (transference psychosis). International Journal of Psychoanalysis 1958; 38: 134–138.
10. Hedges L. In search of the lost mother of infancy. New York: Jason Aronson; 1994.
11. Alexander R. Folie à deux. London: Free Association Books Ltd; 1995.
12. Simpson N. Untwining the transference. In: Bates Y. Shouldn’t I be feeling better by now? London: Palgrave Macmillan; 2006.
13. Foster J. Enquiry into the practice and effects of scientology. London: Her Majesty’s Stationery Office; 1971.
14. Nash P. Two years of hell. New York: Survivors’ Forum; 2002.
15. Blum HB. The concept of the eroticized transference. Journal of the American Association 1973; 21: 61–76.
16. Nutt D, Sharp M. Uncritical positive regard? Issues in the efficacy and safety of psychotherapy. Journal of Psychopharmacology 2008; 22: 3–6.