There was a time, some decades ago, when the psychoanalytic establishment derided John Bowlby's attachment theory as insufficiently deep. It was regarded as too concerned with surface interpersonal relationships at the expense of the map or template of intrapsychic relationships between internal objects, and too behavioural in understanding motivation at the expense of the unconscious and drives. Attachment theory, in turn, denigrated psychoanalysis, particularly Kleinian, for its solipsistic avoidance of the impact of the external world and its lack of scientific and biological rigour.

It must be a source of considerable satisfaction to attachment theorists that the tide has turned. They are being welcomed in from the cold as psychoanalysis reconstructs itself, seemingly in an attempt to reinvigorate its empirical credentials, and as it starts to develop an interest in having a proper dialogue with neuroscience, cognitive theory and developmental psychology. Attachment theory has the potential to provide a rational framework within which integration can take place: integration not only between the types of enquiry concerned but also between modalities of psychotherapy increasingly well versed in mutual respect but still without bridges of theoretical contact. Jeremy Holmes's contribution to this process is considerable and he is coming close to breaking new ground through the clarity he brings to the integrative project. Nowhere is this more in evidence than in his new book.

Sadly, for all attachment theory's accessibility as a framework, there is an unfortunate gravitational pull towards a reduction of the complex to just a handful of core constructs, the four types of attachment: secure, avoidant, ambivalent and incoherent. Furthermore, translation from one modality or school of psychotherapy to another, substituting one set of terms for another, does not always extend understanding itself, and palls at times.

Holmes's book really takes off when it considers psychosocial intergenerational transmission, with particular emphasis on links between handling-style in infancy, the subsequent development of narrative style, which represents the individual's relationship to him- or herself, and adult attachment patterns. Peter Fonagy's work on the development of a child's ‘theory of mind’ and ‘reflexive function’ joins forces with ideas such as autobiographical competence and nodal memories to produce a relational theory with narrative as its core organiser. I particularly liked Holmes's reworking of Winnicott's concept of mirroring. This emerging narrative theory is as pertinent to the domain of cognitive therapy as it is to psychoanalytic psychotherapy, explaining and justifying why the dialogue of therapy is an instrument of change.

There is much else in addition to this, for instance a fascinating chapter on money and psychotherapy, and a description of brief attachment-based therapy, making this thoroughly recommendable.