Anthony Winston reviews and describes current perspectives on the aetiology of borderline personality disorder, covering areas such as the role of traumatic factors, attachment theory and self-psychology. The author links these to the psychopathology of the borderline patient who experiences difficulties such as regulation of affect, impulse control and cognitive capacity to both think and reflect. The psychological function of splitting is considered along with other defence mechanisms as having adaptive potential for the borderline patient. Winston describes in a helpful way the links between human development, psychodynamics and psychopathology as he reveals some of the more recent contributions to the field of BPD. At the beginning of his article, he emphasises the problems encountered when attempting to treat or manage patients with BPD. He centres these difficulties on the counter-transference of the clinician, pointing out how such feelings can “all too easily be transformed into therapeutic nihilism”. His article can easily be read as an attempt to counter such a state of mind and he appears to try to generate a more hopeful if not optimistic outlook. In doing so, I fear that he goes beyond his own account of the research evidence for the effectiveness of treatment. Such enthusiasm or optimism can readily be understood as a reactive polar opposite to nihilism or pessimism. I would like to consider further Winston's statements germane to this aspect of his article and raise for further discussion some other ideas which extend beyond the narrower confines of aetiology and psychological therapy.

When describing the particular psychological therapies – namely dialectic behaviour therapy (DBT), psychoanalytic psychotherapy, interpersonal therapies, cognitive–analytic therapy (CAT) and schema-focused cognitive therapy – Winston acknowledges that in each case the evidence for their efficiency is still lacking. For example, with CAT, he states that “though promising it has yet to be evaluated adequately in clinical practice”. He describes schema-focused cognitive therapy as “another novel but untested approach”. He also describes a brief approach developed by Klerman within interpersonal therapy, for which “a small pilot study has been carried out but the results have yet to be published”. Psychoanalytic psychotherapy, he argues, has never been subject to formal evaluation. I am not sure that this view is strictly correct in light of the work of the Menniger Group in Topeka (Reference WallersteinWallerstein, 1986) – it depends on what is understood by ‘formal evaluation’. Regarding DBT, he discusses the work of Linehan and colleagues comparing this approach with a “treatment as usual” approach for self-harming behaviour, and then states that “despite this essential negative finding, DBT has attracted considerable interest”. Winston then proceeds to point out methodological criticisms of Lineham's studies. In this way, he appears realistically to appraise the different psychotherapeutic approaches to treating BPD, but when he turns to the future he again seems to believe that the evidence suggests that patients with BPD tend to improve, stabilise or mature over longer periods of time. It is not clear from his article what this evidence is, or whether he is referring to a natural phenomenon (Reference StoneStone, 1993). He admits that there is “urgent need for more research into the outcome of different forms of treatment” and acknowledges that the different therapies have yet to be properly assessed. Yet, in spite of the lack of evidence, he seems to be optimistic in his outlook. Winston does admit that the jury is still out in considering the case for short-term therapies; he poses the possibility of some form of unspecified early intervention with at-risk families in order to help prevent the development of BPD. Given that he has previously stated an estimated prevalence of BPD in the community of 2%, the nature of any proposed interventions would surely need to be on a grand scale, in the context of much wider socio-political initiatives.

Winston addresses the thorny issue of where specialised therapies might fit within the total mental health service; he regards it as probable that most borderline patients will continue to be the responsibility of general psychiatrists. This matter is open to ongoing debate (Reference Cawthra and GibbCawthra & Gibb, 1998). I concur with Winston that the place of such specialised therapies within generic mental health services needs to be determined. Psychoanalytic understanding of mental mechanisms including denial, displacement, projection and transference has enormously assisted the psychiatrist in appreciating how and why the borderline patient can have such a powerful and disturbing impact on the clinical team, which may become divided in ways that mirror the inner world of the patient (Reference MainMain, 1989). Here, our understanding of the borderline patient can help with clinical management in a broad sense. The development and promotion of treatment by specialised individual therapies or within therapeutic communities, as described by Winston, needs careful consideration before more resources are committed to them. Kernberg (1984) – often quoted in this field – warns against the frequently encountered scenario whereby an enthusiastic group establishes a therapeutic community model in a sector of the hospital (service), so forming an ‘ideal society’, which generates gratification, excitement, hope and perhaps a messianic spirit in both staff and patients, to be followed later by bitter disappointment because of the “lack of understanding” and apparent rejection of this ideal society by the hospital (service) within which it has developed. The incipient danger of idealisation as a state of mind affecting not only borderline patients but also psychotherapists and psychiatrists needs to be addressed before we can have a sustainable conviction that by receiving treatment, patients with BPD can be significantly changed. Many clinicians, especially general psychiatrists, remain unconvinced about the existing arguments for therapeutic optimism. Given the endeavours of those working in the field, coupled with the spirit of collaboration between the different schools of thought, there are grounds to be more hopeful that in future we will make progress in helping patients with BPD. Furthermore, we may be more able to determine which individuals within this diagnostic category could have therapeutic strategies better tailored to meet their individual needs (Reference Horwitz, Gabbard and AllenHorwitz et al, 1996). One factor that seems to be regarded as crucial by some in determining treatability is that of patient motivation – a most enigmatic quality (Higgitt & Fonagy, 1992).

In commenting on Winston's paper, it would be somewhat unfair and wrong to suggest that his overall message sounds like that of a Jackanorian Utopian ideal in a post-modernist world. Equally, it would be inaccurate to regard my own comments as simply those of a Jeremiah or doubting Thomas. To do either would be a distortion of the truth and probably a form of splitting, but it might contribute to further refection on this most difficult but important topic.