The high rate of comorbidity between major depressive disorder and alcohol abuse is noted and the possible causes of the association are examined. Three reasons for the high degree of comorbidity are discarded--underlying dependent personality disorder and/or childhood deprivation, a common hereditary defect, and self-medication with alcohol for primary depression. Alternative explanations are favored, namely, the chance simultaneous occurrence and subsequent oversampling of two common disorders, and the fact that alcohol abuse is often the "horse" to the "cart" of major depressive disorder. The treatment implications of such a view are discussed.