There is a popular belief that the experimental administration of alcohol to individuals who have chronic, severe alcohol problems (‘alcoholics’) is inherently dangerous or unethical. This creates an environment in which researchers who desire to conduct a study involving the administration of alcohol to persons with severe alcohol problems must defend the relative safety and reasonableness of this practice when, in fact, scientific justification for not using this important methodologic technique in alcohol research is lacking. The primary purpose of this manuscript is to present and discuss the safety, ethical, and practical considerations of research involving administration of alcohol to subjects who have had difficulty refraining from harmful alcohol use in the natural setting. The authors also describe a study in which they monitored the short-term effects of administering 0.4/0.6 g/kg alcohol to 16 recently abstinent subjects who had chronic, severe alcohol problems. This study revealed no evidence that the administration of beverage alcohol in the experimental setting to such individuals causes an uncontrollable desire for more alcohol, precipitates immediate relapse, or creates any behavioral problems. The data also suggested that the knowledge gained from the effects of alcohol ingestion in the experimental setting might help many subjects to understand more completely their addiction or drinking behaviour. It is concluded that there is no overriding reason why alcohol cannot, with due precaution, be safely and ethically administered in the experimental setting to human subjects who suffer from alcohol problems.