Research in Psychoanalysis: Process, Development, Outcome edited by Theodore Shapiro and Robert Emde (Supplement to the Journal of the American Psychoanalytic Association, Volume  41). Published by  International Universities Press, Madison, Conn, 1993;  420 pp; £37.50.

At least some of us ought to read this book, and all of us may have cause to be grateful for its publication. It is a collection of papers presenting in shortened form some of the research now being done into the possibility of having verifiable ideas about what goes on in the course of an analysis. We need verifiable, testable ideas that would persuade

even Sanskritologists, musicologists and cosmologists to reach the same conclusions as we do using the same definitions and methods. At present this is an unattainable ideal, and this book is about the problems that arise in pursuit of this ideal and how some of the problems may be overcome. The definitions, methods and conclusions presented here are often cumbersome, but they may be the foundations of something much more elegant - indeed, every now and again we catch a glimpse of it.

I would advise the reader of this book not to start with the first section ‘Research on the Process’ although it lays the foundation for all serious work and gives the reader an idea of just how difficult it is to establish even the basic categories in terms of which to think of our task and the ground rules for research of this kind. Readers may feel more encouraged to persevere if they start with the middle section, ‘Outcome Studies’. The first paper in this section is an overview of the University of Columbia Records Project, which has been kept going in one form or another for over 40 years. Among the interesting findings is that, taking nearly 300 analysands, 8 out of 10 who are judged ‘analysable’ at the start of their analysis were deemed to have received ‘maximum benefit’ when they finished, but rather less than half of them were deemed to have ‘developed an analytic process’ (become psychologically minded) by then. The definition of what it is to ‘develop an analytic process’ is interesting; a patient judged to have developed it would have been rated high on the following criteria:

1. They brought psychological data into the treat