Familial risk and environmental stress promote the development of alcohol dependence. We investigated whether a positive family history of alcoholism affects the neuroendocrine response to a standardized laboratory stress test in healthy subjects without alcohol use disorders. Twenty-four high-risk subjects with a paternal history of alcoholism (PHA) and 16 family history negative (FHN) controls were evaluated. Psychosocial stress was induced by having subjects deliver a 5-min speech and mental arithmetics in front of an audience on separate days, after drinking either placebo or ethanol (0.6 g/kg) in a randomized sequence. Adrenocorticotropin (ACTH) was measured in 10 plasma samples covering up to 75 min after the stress test. Plasma arginine vasopressin (AVP) was determined before the stressor, at the time of maximum ACTH secretion, and at 75 min after stress onset. The stress test induced a phasic increase in ACTH secretion. At the time of maximum ACTH, AVP was significantly increased in relation to baseline. Compared to placebo, alcohol administration significantly attenuated maximum ACTH concentration in PHA but not FHN subjects, and decreased AVP measured in the same samples in PHA but not FHN subjects. We conclude that activation of the hypothalamic-pituitary-adrenal system by psychosocial stress is accompanied by an increase in peripheral plasma AVP levels. Secretion of both ACTH and AVP suggest that alcohol attenuates the stress response selectively in PHA but not FHN subjects. This might imply some short-term positive alcohol effect in sons of alcoholics, but also constitute a mechanism by which their risk to develop alcohol use disorders is increased.