Warnings about the consequences of filling old bottles with new wine are legion. The bottles will split, the wine will run out, to the ruin of both. Only 20 years ago, such concerns were used in defence of a clear separation between the containing vessel of psychoanalysis and the cocktail of ethology and developmental psychology known as attachment theory. Nowadays, discussion focuses less on whether attachment theory should be assimilated into analytic therapies, than on how. No one has done more than Jeremy Holmes to bring about such acceptance and to demonstrate its potential. In summarising relevant developments for his readers, Holmes's continuing refusal to be theoretically or geographically insular serves him well.

Exploring in Security concentrates on clinical uses of attachment theory within individual psychotherapy. Like its predecessors, it rarely disappoints. Within a basic frame of relationship, meaning and change, Holmes outlines a number of therapeutic principles that are pragmatic but potent. He avoids facile adherence to a small set of recipes or solutions, illustrating how therapists’ success is likely to reflect their ability to act differently in different situations, as well as on their familiarity with a broad range of relevant theory. As he writes about these clinical lessons, what he offers is rich, complex and balanced in ways that soften conceptions of new or old. The book is not a manual and is likely to appeal especially to therapists of some experience willing to use the lens of attachment theory to ‘better understand what they intuitively do in the consulting room’. They are likely to find the book helpful in conveying this understanding to those they supervise and teach.

Two small reservations may be signs of our time. The blurb on the book's cover makes reference to evidence-based practice and a brief consideration of research into psychotherapy outcomes is included within. However, this section fails to get beyond an unnecessary stand-off between demonstration of effectiveness and ‘logic and theory’. Both are needed if promising outcomes in therapeutic trials are to be attributed to processes such as repair, mentalisation or reflective function that the book describes so well. A lack of analysis here means readers could confuse practices that are scientifically informed with ones that are strictly evidence-based. The other caveat is that copy-editing has been minimal and many kinds of typographical error will be found throughout. A book of this quality deserved surer production. Still, it will amply repay most therapists’ exploration.