This study attempts at evaluating the Turkish National Amnesty Law of 2003. The Law being active for 6 months period, granted amnesty for the non-violent terrorists and granted pardon for the violent terrorist in case they cooperate with the law enforcement. The law aims to decrease terrorism. The monthly data is obtained for the number of armed and unarmed incidents, and number of deaths and injuries, for three time periods: 10 month before the intervention, 6 month during the intervention, and 10 month after the intervention; for the 4 major and the other terrorist organizations in Turkey. Hence, there are four dependent variables, three time series and five cross sections. After applying the Generalized Least Square in cross sectional multiple time series analysis, the study found that the policy has no impact on any of the dependent variables. There are at least three explanations of the findings: First, the terrorist organizations may escalate their activities to intimidate their associates from being the beneficiaries of the policy. Second, the beneficiaries have been un-proportionally are the associates of the terror organizations not active today. Third, the post intervention period may not be enough to assess the policy.