The aim of this paper was to study, separately and jointly, the adequacy of the A1 and A2 criteria of posttraumatic stress disorder (PTSD) as proposed in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR). Four hundred thirty-two college students answered a questionnaire of traumatic events and a PTSD symptom scale. When comparing groups meeting or not A2 –but no groups meeting or not A1 or groups meeting or not A1+A2–, significant differences of moderate degree were found in posttraumatic symptomatology, and significant differences were also found in the conditional risk of PTSD, although the effect size was low. These data show that PTSD can occur with equal probability in response to stressful but not traumatic events, and cast doubt on the usefulness of criteria A1 and A1+A2. We suggest abolishing criterion A1 and broadening criterion A2, but considering it a risk factor rather than a diagnostic requirement.