Historically, a major controversy between practitioners and scientists in the alcohol-use treatment field has revolved around treatment models. The majority of practitioners have strongly supported the Minnesota or traditional model, with many championing it as uniquely effective. The key assumptions of the Minnesota model are that all alcoholics are alike; alcoholism treatment is more effective when delivered in residential settings; all alcoholism treatment should be intensive and abstinence-oriented; and the 12-step approach is more effective than alternative approaches in treating alcoholism problems. Scientists have criticized the Minnesota model and have often advocated its replacement with a science-based behavioral treatment model. Practitioners' and scientists' allegiance to contrasting treatment models has been a major obstacle to effective technology transfer. In the first part of this article, key issues in this debate are reviewed, and the extent to which current empirical evidence supports the challenges raised by the research community is considered. In the second part of this paper, some alternative paradigms are suggested for conducting alcoholism treatment research in clinical practice settings. These paradigms allow for a more collaborative partnership between researchers and clinicians. Specific examples of research that has used these paradigms are provided, and suggestions are offered for how these paradigms can usefully augment the current focus on clinical trials research and lead to improvements in the quality of care. 69 references