The beast in the bottle

The Natural History of Alcoholism Revisited is a revised and updated version of a text first produced in 1983. That was, and still is, regarded as a classic and certainly broke new ground during the 1980s. The importance of the original text derived from the fact that it traced the development of alcohol problems among a community sample originally recruited as adolescents and followed up for 35 years. The new text provides an update based on developments over the past 15 years; and its importance again derives from the fact that almost all the alcohol abusers identified in the first version have been followed up for an additional 15 years to make 50 years in all. It goes without saying that 50-year follow-up studies are few and far between.

The form of the new book is somewhat unusual. When encountering a revised version of a previous work, the reader perhaps expects to find a version that has been extensively reworked with new material interwoven seamlessly into the text. George Vaillant's most recent opus, however, deviates from this model. The basic text remains the same. Instead of producing a new version, Vaillant has chosen to introduce a number of separate sections, each of which is subtitled by the word "Revisited". These sections are simply added on at the end of existing sections with no attempt at integration. A consequence is that people already familiar with the 1983 version will encounter a number of deja vu experiences. Does the new text take the reader beyond the conclusions of the 1983 editions? The answer is both yes and no.

Vaillant's 50-year follow-up now stands as a milestone within the addiction literature. In that sense it is required reading. However, a problem with a 50-year follow-up study is that in order to ensure comparability between data sets over time, conceptual frameworks that were in place at the start of the study still necessarily persist, and the importation of new viewpoints is somewhat more problematic. Consequently, although by no means ignoring the importance of social, environmental and learning factors, Vaillant's new text on "alcoholism" still adopts a predominantly "disease" orientation. There is thus a difference in ideology between Vaillant's text and other recent texts such as Liberating Solutions to Alcohol Problems (1995). While most people would agree that there are important medical and genetic factors involved in alcohol problems, many would perhaps give greater prominence to psychological, learning and social factors.

On the other hand, a number of Vaillant's conclusions require endorsement without qualification: the development of alcoholism requires no underlying psychopathology; alcohol problems should not be viewed simply as symptoms of inability to cope or underlying distress. Vaillant also revisits the evidence concerning safe limits of alcohol consumption and possible cardiac protective effects that have been identified in a number of recent studies.

One would have to conclude that Vaillant's text should be on the shelves of any serious-minded worker concerned with alcohol problems. The data are beautifully presented and described and the conclusions eminently reasonable. One would wish, however, to supplement Vaillant's views with some of the more radical psychological, ethnographic and possibly social constructionist views of addiction in order to gain a full overview, even if ultimately one chose to reject certain perspectives. This is perhaps best summarised on page 347 where Vaillant makes reference to the work of Thomas Szatz (1972). Szatz, he reports, would have us believe that alcoholism is a mythical beast. Unfortunately, argues Vaillant, sometimes mythical beasts have real horns. It was never my impression, however, that Szatz meant that mythical beasts were in fact hornless, but that they sometimes had an impressive array of horns and were famous for the amount of damage they could inflict.

John B. Davies is professor of psychology, University of Strathclyde.