Ninety-four consecutive responders to treatment alcoholics, 39 with and 55 without psychiatric comorbidity, were compared, regarding affective temperaments, according to the formulation of Akiskal and Mallya, with 50 healthy volunteers displaying the same social characteristics and belonging to the same environment. No differences were observed between alcoholics and controls on the hyperthymic scale. Significant discrepancies were measured on the depressive, cyclothymic, and irritable scales, where alcoholics scored higher, regardless of the presence or absence of dual diagnosis. In a multivariate discriminant analysis, mainly cyclothymic, but also depressive traits to a lesser degree, make it possible to distinguish between alcoholics and controls, but not between alcoholics with and without a dual diagnosis. The present study shows the main ways in which alcoholics can be distinguished from controls in terms of cyclothymic traits, with a depressive component, and why these characteristics are unrelated to the presence of dual diagnosis.