Smart and Mann (in press) provide a brave and important analysis of some of the core issues and assumptions influencing the field of alcohol interventions: do treatment programs work? Which work best? Do they impact on alcohol problems at the aggregate, population level? In general, is it better to spend money on such programs targeted, as they are, towards the highest risk drinkers or on strategies directed at the entire population of drinkers? They make a strong case for taking seriously the argument that treatment programs should be incorporated into a comprehensive set of public health strategies to reduce population-level alcohol problems, while noting some of the difficult methodological issues that remain to be fully resolved in tackling the underlying research. While the evidence to date is reasonably encouraging, a higher level of sophistication needs to be applied.