A plea by a professional ""counterterrorist"" for Western governments simply to murder all terrorists. Rivers argues that taking terrorists into custody encourages their comrades to seek their release through further incidents and that leniency merely enables terrorists to strike again. Thus, they must be killed, either in swift attacks during incidents or by specially trained units that would track down terrorists wherever they are. Sometimes this wandering narrative is fascinating, for example, when Rivers discusses how he would have handled the assault on the Move headquarters rowhouse in Philadelphia last year or the attempt to free the American hostages in Iran in 1980. (He would have used much smaller numbers of people and quietly.) He also wonders about the advisability of the US crossing Libya's ""Line of Death"" in the Gulf of Sidra: ""One has to confront the fundamental question: does the sinking of attack gunboats or the blasting of antiaircraft facilities have any lasting effect on the war against the terrorists?"" But on larger issues, his historical sense is virtually nil. He reduces the Spanish Civil War to ""fascists fighting communists for control of Iberia."" The Church Committee's curbs on CIA covert actions (which included assassination attempts on foreign officials and efforts to depose legitimate governments) are called a ""slaughter."" Also Rivers makes little distinction between terrorists. He lumps together the Baader. Meinhof Gang (which had little public support) with the IRA (which for better or worse represents not a negligible portion of Irish Catholics in Northern Ireland). Moreover, he shows little concern for the victims of the kind of terror that the US participates in or winks at. He never brings up death squads in El Salvador or Guatemala (with easily 100,000 victims in the past six years). Napalm in Vietnam goes unmentioned, as do other notorious incidents. While Rivers offers interesting nuts-and-bolts information and a logical recommendation for elimination of terrorists, he frames the question in an unsatisfactorily limited way.