This study seeks to compare the short term and long-term efficacy of two widely different treatment approaches in widespread use in clinical settings for treating patients with PTSD. We will compare treatment responses to fluoxetine (which acts directly on biological systems), with a psychological treatment, Eye Movement Desensitization and Reprocessing (EMDR), and a pill placebo control group. This study seeks to clarify 1) the differential treatment effects of these different treatment modalities, 2) whether symptom improvement is accompanied by changes in pathophysiology, 3) the long term effectiveness of these treatments. In recent years a variety of treatment approaches have been shown to be effective in the treatment of PTSD. These include exposure therapies (PE), stress inoculation training, Eye Movement Desensitization and Reprocessing (EMDR) and psychopharmacological treatment with serotonin re-uptake blockers. While PE has been compared with SIT and a study is currently underway comparing cognitive-behavioral treatment with EMDR, no study has as yet compared the relative merits of pharmacotherapy alone versus an exposure treatment. While it is commonly held that, in order to recover, people with PTSD need to """"""""process"""""""" their traumatic memories, the effectiveness of some treatments that do not involve the processing of traumatic memories, such as SIT or pharmacotherapy, raises the question whether these therapies might be as effective as therapies that focus on addressing the memories of the trauma. In clinical practice, many patients with PTSD appear to be effectively treated with pharmacological agents alone, without trauma-focused therapy. One hundred and forty subjects with PTSD will be recruited into the study. They will be randomly assigned to one of three conditions: 1) a manualized treatment which focuses on """"""""processing"""""""" traumatic memories (EMDR), 2) a double blind psychopharmacological treatment (fluoxetine) and 3) a placebo control group. After 8 weeks of active treatment, subjects will be evaluated, cease treatment, and assessed again after 8 weeks and at 6 months in order to evaluate the long term effects. Trained raters will remain blind to the subjects' treatment condition throughout the study. We will assess treatment outcome with a multi-modal psychological and biological assessment battery: 1) standard psychological tests for PTSD (CAPS), 2) neuroendocrine function (cortisol), 3) psychophysiological response to traumatic scripts (pre-post changes in heart rate and skin conductance), 4) comorbid problems associated with PTSD, 5) social and occupational functioning. Treatment adherence will monitored throughout the study.