Integration of Theory

Dear Editor

I would like to offer a response to the two papers in the Winter issue (BJP 15(2), 1998) by Stephen Mitchell and Susie Orbach.

The increasing number of calls for the ‘integration’ of theories must, I think, indicate a desire both for more clarity and for less conflict between the various theoryespousing groups. I do not know how theories ‘integrate’ (and I rather suspect they don't), but it would be good if we could think more about what we are saying and see what are the overlaps, disagreements and so on. I would like to make an attempt, in the hope that I may also provoke response from readers.

Stephen Mitchell's paper is one of a number in recent years advocating the inclusion (or some similar term) of attachment theory in psychotherapeutic work, e.g. Migone and Lotti (1998). Although I work analytically, as does Stephen Mitchell, I have no problem with attachment theory, which seems to me to have illuminated human behaviour and to have established its findings by way of the careful evaluation of empirical evidence. Furthermore, its findings have, on the whole, confirmed the findings of analytic work which should be a cause for satisfaction to analytic folk. I am thinking here not only of work specifically labelled attachment theory but also of the work of experimentalists like Daniel Stern, who may not place themselves in this camp.

My question is: as an analytical practitioner, what do I do with this work? Attachment theory, being experimentally based, has an underpinning of evidence not readily available to theory derived from the consulting room. It is theory about a population and moreover a theory which is probabilistic, i.e. statistically based. It makes statements which aim to be general or universal in nature. How are these to be placed within psychoanalytic practice?

To take Mitchell's work with Connie, Connie's separation is depicted as a hole, and Mitchell devotes his efforts to understanding the nature of this hole, what is (not) in the hole,