Deterrence and Security in the 21st Century: China, Britain, France, and the Enduring Legacy of the Nuclear Revolution. By Avery Goldstein. Stanford, CA: Stanford University Press, 2000. 356p. $49.50

Clausewitz once observed that in war everything is simple, but the simplest things are very complicated. This seems to apply doubly so to nuclear deterrence. The principle is very simple: A potential enemy is persuaded not to do anything rash by the prospect of devastating retaliation. But it soon gets complicated. What difference does it make if the idea is not only to protect the homeland but also allies? As potential enemies acquire their own means of devastating retaliation, issues of preemption arise, and this requires close attention to the details of force structure. How varied, overwhelming, and surprising need the attacker be, especially if there is little interest in preemption? How much need the defender disperse, conceal, or protect forces, or develop antimissile defenses, just in case the other side is contemplating preemption? What happens if both sides are contemplating preemption at the same time? Can understandings, tacit or negotiated, between potential enemies ease the dangers of a crisis getting out of hand?