The challenge presented by a rising China is the principal issue
facing American foreign policy. This is not always obvious to most
Americans or even to many of our leaders. Since the end of the Cold
War, defense policy has been absorbed in second-order problems of
deterring or defeating mid-level powers such as Iraq, North Korea and
Serbia, and in third-order problems of peacekeeping and humanitarian
intervention. Over the long term, however, the first priority of a
serious foreign policy is to handle challenges from discontented,
nuclear-armed, major powers.

It is hardly inevitable that China will be a threat to American
interests, but the United States is much more likely to go to war
with China than it is with any other major power. Other current or
emerging great powers either are aligned with the United States (NATO
countries and Japan), are struggling against crippling decline
(Russia), or, while having a tense diplomatic relationship with
Washington, have no plausible occasion for war with America (India).
China, by contrast, is a rising power with high expectations,
unresolved grievances and an undemocratic government.

Debate about whether and how China might threaten U.S. security
interests has often been simplistically polarized. Views range from
alarmist to complacent: from those who see China emerging as a hefty
and dangerous superpower, to those who believe the country's
prospects are vastly overrated; and from those who see its economic
growth as an engine for building threatening military capabilities,
to those who see that growth as a welcome force for political
liberalization and international cooperation.

Most strategic debate about China still focuses on a few simple
questions. With respect to capabilities, these revolve mainly around
whether the Chinese armed forces will develop to the point that they
rival U.S. military power, and whether the economic surge--with its
implications for military transformation--will continue indefinitely
or stall. With regard to intentions, China watchers want to know how
thoroughly and how soon the country will integrate into a global
economy that allegedly constrains conflict; whether Beijing will
adopt aggressive aims as its power grows; and whether political
liberalization will occur as its wealth grows. Concern also zeroes in
on whether the People's Republic of China (PRC) has the ability to
take Taiwan by force.

These are relevant questions at the most basic level, but they are
the wrong ones to generate progress in a mature debate. The most
worrisome possibilities are those that lie beyond the answers to
these questions, and between alarmist and complacent viewpoints. The
truth is that China can pose a grave problem even if it does not
become a military power on the American model, does not intend to
commit aggression, integrates into a global economy, and liberalizes
politically. Similarly, the United States could face a dangerous
conflict over Taiwan even if it turns out that Beijing lacks the
capacity to conquer the island.

Will China's Military Power Rival America's?

There is little disagreement that the People's Liberation Army (PLA),
a generic designation for all the Chinese armed forces, remains a
threadbare force, well below Western standards. Pockets of excellence
notwithstanding, most personnel are poorly educated and trained.
Weapons systems are old, and even those acquired most recently are
inferior to those in Western arsenals. Many units spend a good deal
of time in non-military activities; staffs do not practice complex,
large-scale operations; exercises and training regimens are limited;
and equipment is not well maintained. Even according to the highest
estimates, defense spending per soldier is low by First World
standards, indicating the dominance of quantity over quality in the
Chinese forces.

The main disagreement among Western analysts of China's military is
about whether the PLA is poised to move out of its unimpressive
condition and into a new era of modernity, efficiency and
competitiveness, as anticipated economic reform and growth translate
into military improvement. Arguments to that effect are supported by
a number of PLA writings about a prospective Chinese "revolution in
military affairs" (RMA). The current backwardness of the PLA reflects
its low priority in the country's modernization efforts since the
1970s, and the diversion of the energy of the military into business
activities (which are now supposed to be curtailed). In this view,
the military potential of the PLA will be liberated when the
political leadership decides to give it significantly more of the
resources generated by economic development.

For China to develop a military on the model of the United States
would be a tremendous stretch. The main issue is not whether Beijing
will have high defense budgets or access to cutting-edge technology.
A rich China might well be able to acquire most types of advanced
weaponry. Deeply ingrained habits of threat assessment in the U.S.
defense community encourage focusing on these factors. Unfortunately,
however, basic "bean counts" of manpower and units and the quality of
weapons platforms are poor measures of truly modern military
capability. More fundamental to that assessment is whether the PLA
establishment is capable of using whatever increase in resources it
might receive to build the complex supporting infrastructure
necessary to make Chinese forces competitive in combat. The PLA's
current mediocrity--like that of many armed forces in the world--may
be rooted in a history, ideology and culture that are incompatible
with the patterns of organization and social interaction necessary to
rival the best First World militaries. This does not mean that the
PLA can never break out of this box. It does mean that it will not be
easy to do, and will not occur quickly.

Modern military effectiveness has become more a matter of quality
than of quantity, and less a matter of pure firepower than of the
capacity to coordinate complex systems. The essence of the American
RMA lies in the interweaving of capacities in organization, doctrine,
training, maintenance, support systems, weaponry and the overall
level of professionalism. These factors are harder to measure, but
they are what make it feasible to assimilate and apply
state-of-the-art weapons effectively. Those capacities require high
levels of education throughout a military force, a culture of
initiative and innovation, and an orientation toward operating
through skill networks as much as through traditional command and
obedience hierarchies. Few militaries have developed these
capabilities. Indeed, experience in the Persian Gulf and Kosovo
indicate that the United States is in a class by itself in these
respects. If the PLA has the resources to integrate complexity and a
willingness to delegate authority to overcome its mediocrity, it is a
well-kept secret. As yet it is not obvious that PLA effectiveness is
likely to be closer to that of the American military than to that of,
say, Iraq.

If the PLA remains second-rate, should the world breathe a sigh of
relief? Not entirely. First, American military power is not the only
relevant standard of comparison. Other armed forces in Asia that the
PLA could come up against are much closer to the Chinese standard
than to the American. (This is true even of Taiwan's technologically
sophisticated military, whose long isolation has eroded its quality.)
Second, the United States has global interests and often finds itself
distracted or pinned down in other regions. Third, the Chinese do not
need to match U.S. capabilities to cope with them. Rather than trying
to match an American revolution in military affairs, they might do
better to develop a counter-revolution by devising asymmetrical
strategic options on various parts of the technological spectrum that
can circumvent U.S. advantages.

One such example could be "cyberwar" attacks on the complex network
of information systems that stitches American military superiority
together. Another could be the use of new weapons like land attack
cruise missiles or lower tech weapons such as naval mines to impede
American access to the region. Still another could be the
modification of China's no first-use policy on nuclear weapons,
making an exception for repelling an invasion of Chinese territory.
Although it is almost unimaginable that China would use nuclear
weapons in an effort to gain political concessions from Taiwan, it
might threaten their use to deter U.S. military action on behalf of
the island. As Disarmament Ambassador Sha Zukang said in 1996, "As
far as Taiwan is concerned it is a province of China. . . . So the
policy of no-first-use does not apply." Even though the Foreign
Ministry subsequently repudiated the statement, nothing makes a
future adaptation of doctrine in that direction unthinkable.

Pundits on defense policy commonly observe that China lacks power
projection capabilities--the ability to send and sustain combat
forces far from home. By U.S. standards this is true. Talk about
obtaining aircraft carriers has produced nothing deployable, the navy
and air forces lack the requisite assets for "lift" (transporting and
supplying large units for operations abroad), and Chinese forces have
negligible logistical capacity as we know it. For those worried about
facing a Chinese force on the American model, these are all good
grounds for optimism. Indeed, there is general agreement that Beijing
lacks the capability at present to invade Taiwan and that it could
take decades to overcome the obstacles. But China does not need to
match American standards to reshape the Asian strategic environment.
This becomes especially clear when one considers where power
projection will be an issue.

Taiwan is both the most dangerous and most likely instance of Chinese
power projection (more on which below), but it is not the only one.
China has conceivable points of conflict in several places that would
not require its forces to cross large bodies of water, and where it
would not be facing opposition as potent as Taiwan's military.
Although less likely than conflict over Taiwan, an imbroglio in Korea
would be scarcely less dangerous if the Pyongyang regime were to
collapse and South Korean or American forces were to move into the
vacuum without Beijing's agreement. Far too little attention has been
focused on the odds of miscalculation in a confused situation of this
sort. The PLA does not have the American army's logistical capacity,
but even a half century ago it managed to project a force of hundreds
of thousands of men deep into Korea.

While the Chinese navy is weak, some of its neighbors' navies are
weaker still. Two of these neighbors, Vietnam and the Philippines,
have outstanding sovereignty disputes with China and have not fared
well in naval skirmishes in the past three decades. We also cannot
rule out the possibility of a land attack. The PLA did poorly in its
invasion of Vietnam over twenty years ago, but the Vietnamese army is
now less than half the size it was then, and the Vietnamese economic
base is far more inferior to China's than it was in 1978. Logistical
limitations would hamper, but not preclude, PRC action in Mongolia,
or in the Russian Far East, if that region were to fall out of
Moscow's effective control. Granted, conflict over these places is
improbable. The problem is that the same could have been said of most
wars before they happened.

Will China Become the World's Leading Economy?

Military potential grows out of economic capacity. China's economy,
like its military, is neither to be envied nor denigrated. In recent
years it has been the fastest growing major economy in the world.
Until strains became evident in the late 1990s, it was common to
project high growth rates straight into the future and to see China's
GNP surpassing America's early in the twenty-first century. But
China's economy faces daunting challenges. And even if it solves many
of its problems, the central government may not have sufficient
control over the fruits of growth to use them for military coercion.

Even if China achieved the highest GNP in the world, low per capita
wealth would persist, limiting disposable income thatcould be
reallocated to the military. In any case, the fantastic economic
surge of the last two decades has sputtered. Although China's economy
seems to be recovering, it has just completed three years of
deflation, and this despite dangerous levels of government deficit
spending on infrastructure projects. Such deficits, and the high
ratio of unrecoverable bank loans to GNP, risk exposing the economy
to the same dangers experienced in Southeast Asia and Korea in 1997.

Straight-line projections in the negative direction would, of course,
be as naive as the excessive expectations of the 1990s. One common
and rather condescending mistake made by some foreign analysts of
China is to offer a long list of problems and then forget to mention
that Chinese elites are smart enough to do something about them.
Under the economic leadership of Premier Zhu Rongji, the Chinese
Communist Party has tried to reform the financial sector, force the
military to abandon much of its commercial empire (including its
lucrative smuggling operations), and meet the stringent demands of
China's accession to the World Trade Organization (WTO).

Still, most of the methods available to spur growth involve
politically risky measures, such as reform of the financial sector
and state-owned enterprises, that threaten to leave tens of millions
of additional urban residents jobless. These would join the ranks of
the floating population of more than 100 million already in search of
work. Similar problems affect the military, as cutbacks in the
overall size of the PLA (needed for modernization and
professionalization) create yet more disgruntled citizens. To
maintain comfortable lifestyles, especially for officers, the
military became dependent on business activities and large-scale
smuggling of goods otherwise subject to high tariffs. If able
officers and soldiers are to stay in uniform, then, the military
units need to be subsidized at a high rate in order to implement
successfully the plan to get the PLA out of commercial activities and
back into the barracks.

All in all, there are no easy ways out for Beijing. Even if all
growth-spurring measures work and the Communist Party maintains
stability during the transition, the process of doing so will be
expensive--especially for a central government that takes in a far
smaller percentage of GNP than do Western governments. As China grows
through capitalist reforms, Beijing must scramble to find new ways to
tax private wealth. However inefficient they may have been, large
state-owned enterprises provided a large portion of government
revenues. The soft loans that officials forced government-owned banks
to give to these enterprises functioned as an indirect tax on
families, which, before Chinese membership of the WTO, had no
alternatives to the state banks. If the Party is successful in
demolishing both the monopoly of state-owned banks and the sturdy
safety net for state-owned enterprises, it will need to raise money
elsewhere for the government operating budget. That budget, in turn,
will be strained by increases in welfare spending needed to maintain
social stability--pensions for the unemployed and retired, and
compensations to the military for its lost sources of legal and
illegal revenue.

What is left over will be available to invest in education (critical
to creating sustained growth in high-tech areas) and in military
modernization. However, even if China's economy and high-tech sectors
were to grow rapidly, it is not clear that Beijing will be able to
channel resources effectively to create armed forces that could rival
those of the United States, even in the region. For one thing, it is
doubtful that the state will be able to channel young technical
talent into defense research and development. China's turn to Russia
and Israel for military technology is worrisome, but if it continues
it will say more about China's inability to close the technological
gap with the West than anything else.

Will China be Pacified by Globalization?

To pessimists steeped in realpolitik, a rich China will necessarily be
a threat, because economic power can be translated into military
power and power generates ambition. To optimists impressed with the
revolutionary implications of globalization, however, a more powerful
China will not be a threat because it will have too much to lose from
disrupting international trade and investment. The latter view is
more common in the West than the former, which seems to many to reek
of old thinking.

The notion that a web of commercial ties discourages war, however, is
itself quite old, if not exactly venerable. It was popularized by
Henry Thomas Buckle in the 1850s, by Norman Angell just a year before
World War I erupted, and again in the 1970s, when interdependence was
said to have reduced the utility of force. Most recently, at a White
House rally for permanent normal trade relations, Al Gore quoted his
father as saying, "When goods do not cross borders, armies will."

The argument this time around is that the proposition is finally true
because the nature of interdependence has changed in a crucial way. A
century ago it was characterized by vertical trade between imperial
centers and colonies, trade in final products between wealthier
nations, and portfolio investments. Today there is much more direct
investment and transnational production of goods, which fosters "a
growing interpenetration of economies, in the sense that one economy
owns part of another." With the PRC, Taiwan, Japan and the United
States all owning pieces of each other, how can they fight without
destroying their own property? As the Chinese elite make more and
more money from bilateral investments with Taiwan, simple greed will
prevent huffing and puffing from crossing the line to war. If
development thoroughly enmeshes the PRC in the globalized economy,
therefore, peace will follow.

There are at least two problems with this latest version of
interdependence theory. One is that both sides in a political dispute
have a stake in not overturning profitable economic integration. The
PRC might not want to kill the golden goose, but neither would Taiwan
or the United States. Why, then, should Beijing be any more anxious
to back down in a crisis than Taipei or Washington? Mutual dependence
makes a political conflict a game of chicken, in which each side
expects the other to bow to the stakes, and in which collision may
result rather than concession.

The second problem is that there is little reason to assume that
sober economic interest will necessarily override national honor in a
crisis. A tough stand by Beijing may be viewed from the inside as
essential for regime survival, even if it is not seen by detached
observers as being in China's "national interest." In an imbroglio
over Taiwan, which capitals will feel the strongest emotional
inhibitions against backing down? Beijing and Taipei both have a
greater material, moral and historical stake in the outcome than does
the United States.

The global economy does indeed change logical incentives to
compromise in political conflict, but not to the degree that it makes
Beijing likely to be more "reasonable" than anyone else. Economic
globalization does not eliminate the high priority that nations place
on their political identity and integrity. Drawing China into the web
of global interdependence may do more to encourage peace than war,
but it cannot guarantee that the pursuit of heartfelt political
interests will be blocked by a fear of economic consequences.

Will China Become Aggressive?

Whether China has aggressive motives is what most policymakers want
to know about Beijing's strategic intentions. Optimists say the
answer is no, because the PRC is ideologically anti-imperialist and
seeks only respect as a status quo great power. Pessimists say the
answer is yes, because a seething set of Chinese grudges and
territorial ambitions are on hold only for a lack of confidence in
capability, or simply because all great powers tend to become
aggressive when they get the chance.

But such focusing on the unlikely odds of deliberate aggression
diverts attention from possibilities that are both much more likely
and almost as dangerous. Most countries viewed as aggressors by their
adversaries view their own behavior as defensive and legitimate.
Whether Beijing is a tiger in waiting, about to set out deliberately
on a predatory rampage, is not the most relevant question. No
evidence suggests that Chinese leaders have an interest in naked
conquest of the sort practiced by Genghis Khan, Napoleon Bonaparte or
Adolf Hitler. The model more often invoked by pessimists is Kaiser
Wilhelm's Germany. Like Germany a century ago, China is a
late-blooming great power emerging into a world already ordered
strategically by earlier arrivals; a continental power surrounded by
other powers who are collectively stronger but individually weaker
(with the exception of the United States and, perhaps, Japan); a
bustling country with great expectations, dissatisfied with its place
in the international pecking order, if only with regard to
international prestige and respect. The quest for a rightful "place
in the sun" will, it is argued, inevitably foster growing friction
with Japan, Russia, India or the United States.

Optimists do not have a hard time brushing off this analogy to a
state of a different culture on a different continent at a different
time, a long-gone era when imperialism was the norm for civilized
international behavior. Their benign view, that economic development
and trade will inevitably make China fat and happy, uninterested in
throwing its weight around, strikes them as common sense. It could
turn out to be true. It is more an article of faith, however, than a
prediction grounded in historical experience. The United States, for
example, is quite interested in gaining the goodies from
globalization, yet on the world stage it sometimes throws its weight
around with righteous abandon.

Indeed, the most disturbing analogy for China's future behavior may
not be Germany but the United States. If China acts with the same
degree of caution and responsibility in its region in this century as
the United States did in its neighborhood in the past century, Asia
is in for big trouble. Washington intervened frequently in Mexico,
Central America and the Caribbean for reasons most Americans consider
legitimate, defensive, altruistic and humane. The United States and
its allies in Asia, however, would see comparable Chinese regional
policing as a mortal threat. Even if China does not throw its weight
around, the fact that there are others who can respond to the growth
of Chinese power sets up the possibility of a classic spiral of tense
actions and reactions. China faces alliances involving the United
States, Japan, Australia and South Korea, and potential alliances in
Southeast Asia.

Would Chinese Liberalization Guarantee Peace?

Many assume that as long as democratization accompanies the growth of
Chinese power, China will not necessarily pose a security challenge.
This would hold true even if China proves able to maintain high
levels of economic and technological growth, a healthy degree of
government accumulation of the increasing national wealth, and,
thereby, military modernization. This theory of the democratic
peace--that developed democracies virtually never fight one
another--is currently the most influential political science theory
among American foreign policy elites.

Even if we accept the democratic peace theory at face value, there
are several problems with applying it to China. First, as Fareed
Zakaria has noted, the theory really applies only to liberal
democracies on the Western model, ones with restraints on government
action and guarantees of minority rights. Democrat-ization in China
could just as conceivably turn in an illiberal direction, on the
model of post-Tito Yugoslavia, Iran or other unpleasant examples of
violent activism.

Second, the democratic peace theory does not apply clearly to civil
war. Democracies must recognize each other as democracies for the
theory to apply. They also have to view each other as legitimate,
independent and sovereign states. No matter how many Americans and Taiwanese
believe that Taiwan is or should be a sovereign state, this view is widely
rejected on the mainland (and is not a premise of past
or current U.S. policy).

Third, while liberal democracy is pacific, the process of becoming a
democracy can be violent and destabilizing. This is particularly true
of democratizing states that lack developed civil societies,
independent news media, healthy outlets for popular grievances, and a
marketplace for ideas where countervailing views can be debated. This
gives elites incentives to manipulate populist or nationalist themes
and to adopt tough international policies as an electoral strategy.

The Chinese Communist Party has behaved like many authoritarian
regimes, but with much more success. It has systematically prevented
the rise of both an independent press and a civil society. Although
the foreign press has penetrated China, domestic political
publications are still strictly circumscribed by the state. As the
recent crackdown on the Falun Gong demonstrates, the Party is afraid
of any group that organizes for any purpose without state sanction,
regardless of how apolitical it appears to be.

The Chinese government's concerns about its legitimacy are not mere
expressions of paranoia. The intensity of criticism of the leadership
that one hears privately in places ranging from taxi cabs to
government offices is astonishing. Awareness of its unpopularity
gives the government in Beijing incentives to use nationalism as a
replacement for the now hollow shell of communist ideology. But the
Party is also aware that nationalism is not an inert tool to be
pulled out of a kit and manipulated at the whim of the government. It
is double-edged. Volatile and potentially uncontrollable, especially
on emotional issues such as relations with Japan and Taiwan,
nationalism is powerful enough to prop up a communist party in a
capitalist society, but it could also severely damage the party if it
were turned against the state. Officials in Beijing are aware that
nationalism was a major force in the Communist Party's overthrow of
the Kuomintang, as it was in the 1911-12 revolution that overthrew
the Qing dynasty. This is probably why, during the row with Japan
over the Senkaku Islands in the summer of 1996, the Party actively
prevented students and workers from marching in protest to the
Japanese embassy.

According to one prominent Chinese foreign affairs expert, since the
May 1999 NATO bombing of the embassy in Belgrade, the authorities in
Beijing have been very concerned about an increasing trend in public
opinion that views the leaders as soft in responding to international
humiliations, and sees them even as "traitors" (maiguo zei)
interested only in business prospects. He and others argue that the
protests outside the American embassy in Beijing were not so much
instigated by the Communist Party, as many in the West assumed, but
rather were managed, controlled and ultimately suppressed by the
Party. One retired Chinese military officer touched on a similar
theme during the 1996 Senkaku crisis. He maintained that, if the
Party allowed the people to protest unhindered, the first day they
would be protesting against Japan, the next day against the lack of
response by the government, and on the third day against the
government itself.

In the early phases of democratization, China should be ripe for
jingoism. Hypernationalism could be exploited to mobilize popular
support and to deflect criticism of the state, especially given the
existence of irredentist claims, and the danger of ethnic and
regional "splittism" on the mainland and in Taiwan. If
democratization were to occur in the current context of weak
institutions, political leaders and opposition parties would have
incentives to appeal to nationalism in ways that could destabilize
the region. In fact, this is a favorite, though perhaps cynical,
argument offered to foreigners by communist opponents of multiparty
democracy.

There are, however, plausible scenarios in which Chinese
democratization might reduce international conflict. Democratization
could make the mainland more imaginative with regard to the
frameworks for unification offered to Taiwan, thus making meaningful
cross-strait political dialogue more likely. Political liberalization
might also make the prospect of eventual accommodation with the
mainland more palatable to Taiwan and discourage the island's
diplomatic adventurism. But that possibility is not necessarily more
likely than its opposite: a renewed belief by Taiwan that the island
has its own national identity, that unification with the mainland
under any circumstances is unacceptable, and that democratization on
the mainland is a threat to Taiwanese goals. Taiwan may see a closing
window of opportunity, with the hopes of gaining true independence
reduced as the West's sympathy for Taiwan's opposition to unification
hardens.

If Americans view mainland democratization as genuine, such fears on
Taiwan are well placed. Would the United States risk war to prevent a
democratic Taiwan from becoming part of a larger democratic China?
The main reasons a president could currently use to mobilize
Americans around action in defense of Taiwan's democracy would simply
disappear. Many observers view democracy as Taiwan's biggest security
asset, because it increases U.S. support. They are right. But it is
not just democracy that provides Taiwan's security; it is the current
contrast between Taiwan's democracy and the mainland's
authoritarianism.

Can the Mainland Conquer Taiwan?

THE POSSIBILITY of war with China over Taiwan is arguably the most dangerous threat that U.S. security policy faces in the coming decade. No other flashpoint is more likely to bring the United States into combat with a major power, and no other contingency compels Washington to respond with such ambiguous commitment. U.S. policy regarding the defense of Taiwan is uncertain, and thus so is the understanding in Beijing and Taipei-and in Washington-over how strongly the United States might react in different circumstances. Because Taiwan is more independent than either Washington or Beijing might prefer, neither great power can fully control developments that might ignite a crisis. This is a classic recipe for surprise, miscalculation and uncontrolled escalation.

Traditional questions about Chinese intentions and capabilities miss the mark in analyzing the likelihood of war and the probable course war would take. A PRC attack on a Taiwan following the pursuit of formal independence from the Chinese nation would be viewed (quite sincerely) in Beijing as purely defensive, to preserve generally recognized territorial sovereignty. Many outside China would view the attack as a sign of belligerence. But military activity against an independence-minded Taiwan might have little relevance to Beijing's behavior on other issues, even for other sovereignty disputes such as those over the Senkaku or Spratly Islands.

The niceties of balance of power calculations could prove relatively unimportant in determining whether China would use force over Taiwan and whether it would do so effectively. U.S. efforts to create a stable balance across the Taiwan Strait might deter the use of force under certain circumstances, but certainly not all. Moreover, such efforts would miss the major point of cross-strait strategic interaction. China's military strategy in a conflict over Taiwan would likely be to punish and coerce rather than to control, tasks for which its military may be able to use force to great effect. The PLA's ability to mount a Normandy-style assault on the island is not the toughest question. Geography (the water barrier), together with U.S. supplies, would provide powerful means to Taiwan for blocking such an invasion, even without direct U.S. combat involvement.

A greater challenge would be a blockade by the PRC, which has a large number of submarines and mines. Taiwan's proximity to the mainland and its dependence on international trade and investment enhance the potential effect of blockades--or coercive campaigns involving ballistic and cruise missiles--even if the military impact would be modest. The PRC might thus be able to damage severely the island's economy regardless of the number of F-16s, AWACS aircraft and theater missile defense batteries the island can bring to bear. Moreover, to break a blockade by sweeping the seas would likely require a direct attack on Chinese vessels. If Chinese forces had not already targeted U.S. ships by that point, it would be up to Washington to decide to fire the first shots against a nuclear-armed country that was attempting to regain limited control of what it believes is its own territory.

Some think that the United States should give Taiwan military assistance or defend it directly even in the extreme case that it openly declares legal independence. Many assume that the United States could deter an attack from the mainland and that, if worse came to worst, the United States would prevail in a war should deterrence fail. These assumptions, unfortunately, are suspect. Before being deterred, Beijing would have to weigh the costs of inaction against action. The perceived cost of inaction against Taiwanese independence is very high. No leader can count on survival if labeled the next Li Hongzhang, the diplomat who ceded Taiwan to Japan in the 1895 Treaty of Shimonoseki.

Many in Beijing believe that the United States lacks the national will to pursue a war against China to save Taiwan. If the prospect of casualties did not deter the United States from intervening, the reasoning goes, even low levels of casualties would frighten it into early withdrawal. Following this logic, China need not defeat the U.S. military in wartime or close the gap in military power in peacetime. Rather, the strategic requirement is much lower: to put a number of American soldiers, sailors and airmen at risk.

It is dangerous that so many Chinese seem to subscribe to this "Somalia analogy." Washington would probably not be deterred by fears of casualties if it decided that Taiwan was being bullied without serious provocation, any more than it was deterred from attacking Iraq in 1991 by high pre-war casualty estimates. Nor is it likely that, once the United States had made the momentous decision to gear up for combat against a power like China, it would quit easily after suffering a small number of casualties.

Thinking over the long term, however, it is hard to imagine how the United States could "win" a war to preserve Taiwan's independence against a resolute China. Too many analyses inside the Beltway stop at the operational level of analysis, assuming that tactical victories answer the strategic question. Sinking the Chinese navy and defeating an invasion attempt against the island would not be the end of the story. Unless the U.S. Air Force were to mount a massive and sustained assault against mainland targets, the PRC would maintain the capability to disrupt commerce, squeeze Taiwan, and keep U.S. personnel at risk. As one American naval officer put it, as a nation much larger than Iraq or Yugoslavia, "China is a cruise missile sponge." This will be doubly true once China builds more road-mobile, solid-fuel missiles and learns better ways to hide its military assets.

Moreover, strikes against the mainland would involve huge risks. Recall that for three years, while Chinese forces were killing U.S. soldiers in Korea, the Truman administration refrained from carrying combat to the mainland for fear of a wider war--and this at a time when China had no nuclear weapons and its Soviet allies had fewer than China now has. China maintains the capacity to strike the U.S. homeland with nuclear missiles, and to strike U.S. bases in the region with both conventional and nuclear missiles. China has or is feverishly obtaining increasingly sophisticated systems, including Russian SA-10 air defense batteries, stealth detection technologies, anti-ship missiles, land attack cruise missiles, accurate ballistic missiles, and new submarines. Any of these could give the United States and its regional allies pause before widening a campaign against the Chinese mainland.

And if the issue is a PRC blockade of Taiwan, who will bear the onus of starting a war between China and the United States? If a conventional engagement leaves U.S. naval forces in control of the Taiwan Strait, can anyone be confident that Beijing would not dream of using a nuclear weapon against the Seventh Fleet? And then what? Such a scenario of nuclear escalation seems fancifully alarmist to many in the post-Cold War era. But is it any more so than such concerns ever were when defense planning focused on crises with the Soviet Union? Is this an experiment a U.S. commander-in-chief should run?

Even if we dismiss entirely any nuclear danger, there are still considerable problems for the United States. If Chinese conventional capabilities do not deter American escalation, and if Chinese forces prove relatively ineffective against U.S. weaponry, a broader question remains: How long would the United States be willing to continue a war of attrition against a country of more than a billion people? How long would it be willing even to camp multiple aircraft carrier battle groups and minesweepers off the Chinese coast? What would American allies such as Japan--where crucial U.S. bases are located--do? Taiwan will always be just 100 miles from mainland China, and Chinese nationalism is extremely unlikely to wither under American bombardment. Indeed, it would probably harden. What, then, is the endgame? A negotiated deal that rejected Taiwanese independence but protected de facto autonomy would be tempting in this situation. In this limited but very real sense, the effort to defend an independent Taiwan wou ld have failed and China would be victorious, since occupation and direct control of Taiwan is not Beijing's stated strategic goal.

The point here is not that a U.S. commitment to defend Taiwan is meaningless. It could be a major factor in deterring more activist impulses in Beijing among those who might want to do more than just prevent a declaration of independence, or those who view current trends in cross-strait relations with much more pessimism than their colleagues. But we should not overestimate the U.S. ability to deter and defeat China in all circumstances.

Reasons for Pessimism

OPTIMISTS on the China challenge are often guilty of contradictory arguments. On the one hand, they argue that China will only become a dangerous enemy if the United States treats it like one. At the same time, they attempt to demonstrate why China will not be able to develop the military capacity to pose any appreciable threat to us for a very long time. How can China be both hopelessly weak and potentially dangerous?

There are ways to square this circle by asking the right questions. China can pose security challenges to the United States even if it is unlikely to narrow the gap in military power. This is true because of geography; because of America's reliance on alliances to project power; and because of China's capacity to harm U.S. forces, U.S. regional allies, and the American homeland, even while losing a war in the technical, military sense.

Optimists are correct to focus on Chinese intentions and the potentially pacifying influences that the United States and other international actors can have on China. But they often assume too much about the positive effects of globalization, interdependence and political liberalization, because they underestimate the role of nationalist emotion and the possibility of misperceptions and inadvertence in war. They also forget that interdependence is a two-way street that restrains not only the Chinese, but China's potential adversaries as well.

In addressing the China challenge, the United States needs to think hard about three related questions: first, how to avoid crises and war through prudent, coercive diplomacy; second, how to manage crises and fight a war if the avoidance effort fails; third, how to end crises and terminate war at costs acceptable to the United States and its allies.

In terms of coercive diplomacy, the United States needs to balance deterrence and reassurance, recognizing that, on the one hand, deterrence is complicated by the increasing capabilities of the PLA, and that, on the other hand, if certain core interests of China are disregarded by the United States, even a relatively weak China might resort to force. The essence of deterrence and war planning lies in considering military responses-- including crisis management and diplomacy--to a Chinese attack. It might be advantageous to have the capacity to sink a Chinese destroyer pre-emptively before it launched its missiles, but an American admiral might never receive authorization to do so in a real crisis. Similarly, it might be beneficial to have the capacity to strike theater missile sites pre-emptively, but such action might cause unwanted escalation and alienate needed allies. This is one reason that effective local missile defenses are a good idea, even though they may be much less effective and much more expens ive than "shooting the archer."

The United States also needs to consider for how long and at what cost it is willing to fight for certain goals. For how long would the United States be willing to fight to keep Taipei from an agreement with Beijing in which Taipei maintains a very high degree of military and political autonomy, but is nonetheless, in terms of abstract sovereignty, part of China? (In answering this question, remember that the United States made similar concessions to Belgrade regarding Kosovo after defeating Yugoslavia militarily.) It is better to ask such questions about the end of a war before the unpredictable forces of emotion regarding sunk costs and reputation take hold after a war breaks out.

China's growing power causes so many headaches largely because its strategic implications are not fully clear. But before one laments the rise of Chinese power, one should consider an even more uncertain alternative: Chinese weakness and collapse. Nothing ordains that China's march to great power status cannot be derailed. Severe economic dislocation and political fragmentation could throw the country into disorder, and the central government could prove too crippled to use external adventures to rally support and maintain unity. Hard-bitten realists should hesitate before hoping for such developments, however. The last time China was weak and disunified--in the era of warlordism and revolution in the first half of the twentieth century--it was a disaster, not only for China, but also for international peace and stability.

Richard K. Betts is director of the Institute of War and Peace Studies at Columbia University, and a senior fellow at the Council on Foreign Relations.

Thomas J. Christensen is associate professor of political science and a member of the Security Studies Program at the Massachusetts Institute of Technology.