How Wars Are Fought 279ity to kill hundreds of thousands without discrimination. Many in the U.S. military,
for example, considered atomic weapons simply to be more eco nom ical extensions of
conventional bombs. But these first steps into the nuclear age— the first and last time
nuclear weapons were deliberately used against states in war— had already hinted at a
key prob lem related to their use: the long- lasting effects of radiation. During the Cold
War, both the United States and the Soviet Union constructed larger and more lethal
weapons, and developed more accurate delivery systems, ballistic missiles, and cruise
missiles, each capable of killing the earth’s population many times over. Thermonuclear
weapons led to the possibility that combatants could not limit the destruction of a
nuclear exchange to a target only— nuclear weapons were now hundreds of times more
power ful than those dropped on Hiroshima. A nuclear conflict might rapidly escalate
into an exchange that could extinguish life on earth, either by radiation from fallout or
by altering the climate in a “nuclear winter.” This mutual assured destruction (or MAD)
led the major antagonists to shelve plans to fight using nuclear weapons. Instead, they
fought through proxies, using more conventional weapons (see Chapter 2).
The fact that nuclear weapons have never been employed in war since their use
against Japan has prompted two impor tant debates about the po liti cal effects of nuclear
weapons. First, did nuclear deterrence prevent a third world war and therefore justify
the risk and expense the Soviet Union, the United States, Britain, China, and France
sustained through their development and deployment of nuclear weapons during the
Cold War? Second, if nuclear deterrence causes peace—if the very destructiveness of
nuclear weapons makes rational decision makers unlikely to use them or initiate a war
that could escalate to their use— could the spread of nuclear weapons to other countries,
called nuclear proliferation, cause peace? Scott Sagan and Kenneth Waltz debated
these issues in the 1980s. They renewed the debate in the beginning of the twenty- first
century after India and Pakistan— fierce rivals— had each acquired nuclear capability.
Waltz argues that “more may be better,” that under certain circumstances (namely, a
rational government and a secure retaliatory capacity), the proliferation of nuclear weap-
ons implies an expanding zone of deterrence and a lower risk of interstate war. Sagan
strongly disagrees, arguing that the proliferation of nuclear weapons is more likely to
lead to a failure in deterrence or an accidental war.^15 Sagan argues that the conditions
Waltz cites for nuclear peace- causing are rare, and certainly not pres ent in South Asia.
This debate over the threat the possession of nuclear weapons poses has gained a
new salience as the technology to build nuclear weapons has proliferated. The tangled
network of the Pakistani official A. Q. Khan, who provided ele ments of nuclear tech-
nology from Eu rope to Pakistan and then North Korea, has led many to reexamine
the stabilizing effect of proliferation. More crucially, nuclear theorists have ques-
tioned whether a nuclear- capable Iran would make war in the region more or less
likely. If Waltz is right, so long as Iran has a rational government and a number of
weapons secure from a preemptive first strike, the risk of major conventional war in