As a result of the 1992 agenda for peace, of the events in Rwanda and Bosnia in the mid-1990s and of the 2000 “Brahimi report”, current peacebuilding missions generally have to assume the task of protection of civilians. Today, the number of people in need of tangible physical protection and security is at an all-time high. Peacebuilding missions and other international actors often respond with the creation of safe spaces that offer shelter and sanctuary inside the conflict zone. The list of possible examples includes the UN “Protection of Civilians-Sites” in South Sudan, certain “Internally Displaced Persons-Camps” in the DR Congo and the Central African Republic or consent-based demilitarized hospital zones of the ICRC. Yet, analytical engagement with the phenomenon is rare with few publications dealing with safe areas specifically and many approaches ignoring the relevance of space completely. This paper aims at preparing the phenomenon of safe spaces for theoretical analysis by developing precise definitions and establishing a distinct taxonomy. Based on a comprehensive overview of relevant empirical events between 1900 and today, different characteristics like size, consent, military/non-military character, success, underlying motives or the nature of the involved external actors will be identified and compared. The paper concludes with the development of ideal type definitions that cluster different kinds of empirical safe spaces. In doing so, the paper contributes to a better understanding of how peacebuilding missions deliver their task of protecting civilians and which short- and long-term consequences may result from this.
