Although willing to answer the call, humanitarian organizations have been generally ill-equipped for what they have found: war zones where civilian populations are the intended victims, where access is difficult, where aid workers are in danger of being perceived as a threat or as a resource to be captured, and where their own physical safety is in doubt. Their standard operating procedures provided little guidance for how they might operate in places such as Bosnia, Rwanda, Afghanistan, and Iraq, forcing them to improve constantly. When, if ever, should they request armed protection and work with states? Would armed protection facilitate access or create the impression that aid workers were now one of the warring parties? Should they provide aid unconditionally? What if doing so means feeding the armies, militias, and killers who are responsible for and clearly benefit from terrorizing civilian populations? At what point should aid workers withdraw because the situation is too dangerous? Can aid really make a difference? The contemporary moment has proved so challenging that even stalwart defenders of humanitarianism concede that the moral necessity of humanitarian action is no longer self-evident. Aid workers, thus, should be forgiven if they seem almost nostalgic for the supposedly more straightforward emergencies of the past.