Like a volcano bursting forth, September 11 vastly altered the terrain of U.S. foreign policy. Before that date, Central Asia was one of the most obscure places on earth to most Americans. In the weeks before September 11, Gary Condit was attracting several thousand inches of newsprint for every one devoted to Central Asia. In recent weeks, however, every mass-circulation news magazine and major American newspaper has featured photographs of U.S. soldiers in places such as Uzbekistan and Kyrgyzstan-countries that many Americans have considerable difficulty pronouncing or finding on a map.

As for the Washington foreign policy elite, only yesterday it was arguing that the United States had thrown itself too hastily into an activist Caspian energy policy. They pointed to the authoritarianism and corruption of governments there, and the hostility we were evoking from Central Asia's neighboring major powers. Since September 11, the United States has greatly increased aid-especially military aid, so far-to four of the five Central Asian republics and, far more dramatically, American military forces now use air bases in the region to carry out or support operations in Afghanistan. They appear to be digging in for the long haul. Central Asia also looks to become a major staging area for whatever nation-building exercises are undertaken in Afghanistan; if so, can the nomadic horde of consultants and NGOs-many of them no doubt from the United States-be far behind?

The United States has refrained from giving explicit security guarantees in exchange for access to Central Asian bases and staging areas, but the mere presence of U.S. forces under present circumstances suggests commitment. Such a suggestion abides, too, in the U.S.-Uzbekistan joint declaration of March 12 on a "Strategic Partnership and Cooperation Framework." That declaration reads, in part, that the United States "affirms that it would regard with grave concern any external threat to the security and territorial integrity of the Republic of Uzbekistan." Commitment is also implied in the amplified military-to-military relationships that normally and naturally accompany the provision of U.S. military aid.

Will this commitment wax or wane? Should it wax or wane? These are difficult questions, and the Bush Administration is now called upon to answer them. The key issue is the U.S. military presence; this is what most concerns the major powers that border Central Asia-Russia, China and Iran-and what symbolizes most vividly the expansion of U.S. global power to others worldwide. It is also the most important factor in creating U.S. political debts to local host governments.
Good arguments can be made both for and against a long-term U.S. military presence. On balance, however, the more prudent course would be to anticipate, and not to dread, several years' military basing in Central Asia. Military training relationships will make sense for decades. But we should acknowledge the potential dangers in such a course, the better to mitigate them to the extent possible. There are four such dangers.

First, none of the five Central Asian governments is a democracy, and most are plunged in an economic crisis comparable to the Great Depression. Polls suggest that these weaknesses threaten the legitimacy of several governments. Clearly, these governments will wish to use the U.S. need for access to their territory to slacken pressure on them with regard to political and economic reform. Worse, aid money provided to autocratic governments may exacerbate corruption, making better governance more difficult instead of less. They will also try to leverage their relationship with the United States in their regional rivalries with each other. And, of course, the United States risks being associated with unpopular regimes in the eyes of the peoples of these countries, and suffering when those regimes eventually fall.

Second, there are possible military costs as well as benefits from a protracted Central Asian deployment. A prolonged presence in Central Asia could expose U.S. forces to Islamist guerrilla attack. Moreover, the Pentagon is preoccupied by the strain on military effectiveness created by the increased dispersal of its forces. With the substantial contraction of the Services since the end of the Cold War, and a general unwillingness to provide a defense budget that is adequate for both existing forces and force modernization, this is a serious concern.

Third is the potential impact of a protracted Central Asian deployment on U.S. relationships with major powers, notably Russia, China and Iran. The sudden arrival of the American military next door-in Afghanistan, Uzbekistan and Kyrgyzstan-along with the upsurge of Palestinian terrorism and Israeli reaction, appears to have outweighed Iranian interest in detente with the United States. From an opportunity to escape isolation, America has become, in Iranian clerical eyes, the very country tightening an iron ring around Iran. Central Asian regimes will try to use the American presence to ward off burdensome pressures from Russia and China; we will therefore become enmeshed in a test of wills guaranteed to irritate one side or the other-maybe both. In the Russian case, a long-term American deployment would abrade the Russian sense of historical entitlement in Central Asia. The Russians-government elite and the public alike-may also claim a swindle: Just as the United States, they believe, reneged on a solemn promise not to expand NATO in the context of the agreement on German re-unification, so the Americans have taken advantage of Russian support for a temporary U.S. use of Central Asian facilities in the context of the war on terrorism to put Russia at a permanent geostrategic disadvantage. China worries that a protracted U.S. presence in Central Asia is part of an effort at encirclement and containment. There is a range of issues on which the United States seeks cooperation with Russia, China and Iran, and their displeasure over the U.S. presence in Central Asia could make securing that cooperation more difficult and expensive.

Finally, perhaps the most serious bundle of problems with staying involves not Russians or Chinese but ourselves. We are woefully ignorant about the area and, worse, our ignorance tends to be filled by wishful thinking. To substitute our daydreams, like the Middle East "peace process", for real knowledge of people and their cultures is one of the more unfortunate American traits.

It is worth reflecting, too, how scant our local knowledge is next to that of 19th-century Europe. Its entourage of entertaining misfits-knights-errant like "Chinese" Gordon of Khartoum; refugees from knighthood like Joseph Conrad; spiritual seekers like Sir Wilfred Blunt; monkish zealots like Pre de Foucault; secular ascetics like Doughty or Thesiger; connoisseurs of the exotic like the French naval officer and novelist Pierre Loti; sexual adventurers like Sir Richard Francis Burton and Lawrence of Arabia-kept the elites from which they had emerged and their imperial publics engaged with the intricacies of the world over which they projected power. Moreover, as an epoch of rapid change, "progress" was then making many tastes, styles and ways of life obsolete, above all among the old ruling classes. That imperial age had, through its romantic craving for the unfamiliar and its obsessive polymathy, niches for every cultural resource it possessed (and often squandered). By contrast, our comparable exotics-and there are some-do not please our foreign policy elite, people who are so overworked and so proud that their skills apply to every time and place that they disdain such esoterica. If so few of us really know the region, and if our political elite is disinterested in those who do, we court serious error. We have already made at least one: to think, as we apparently do, that Pashtuns will ever accept an Afghan government effectively dominated by Tajiks is fantasy. Overwhelming U.S. power can beat down such miscalculations for a while, but only for a while.

These four dangers in staying constitute real potential weaknesses. But leaving is even more dangerous. Like it or not, the United States finds itself in a situation where it must extend itself into the world in various places and ways if it is to meet the security challenges posed by apocalyptic terrorism joined to the proliferation of weapons of mass destruction. If the United States is going to succeed in the effort the President has put before the country, we must acknowledge that risks must be taken and prices must be paid. Whether done more in sorrow than in joy it does not matter; we must re-examine the scope not only of our foreign policy and defense policy, but of our entire contact with the world.

Why We Should Stay

The United States must stay in Central Asia, militarily and otherwise, for essentially three interconnected reasons. First, the nature and scope of international terrorism is broader than many appreciate. A critical part of the problem is that weak states abet terrorism, and Central Asia is home to several weak states.

Second, the United States stands to benefit enormously in the long run from the increased stability and the success of moderate Muslim societies and states. Central Asian Islam is for reasons of history and happenstance more moderate than most other types, and it is worth our effort to sustain that moderation.

Third, we need to stay in Central Asia in order to stabilize Afghanistan. The idea that it will be enough to smash the Taliban and give humanitarian aid to render the country essentially harmless to American security is wrong. It will be hard to stabilize that country, and it will not be accomplished anytime soon. Moreover, because of Afghanistan's geopolitical situation-bordering Iran, Pakistan, Uzbekistan, Turkmenistan, Tajikistan and China-it is a far more important country than even most American foreign policy experts tend to appreciate.

Islamist Terrorism and Weak States

The Bush Administration has waged war with inflexible resolution and inventiveness, but it has not cast much light on the real problem we face with international terrorism. Because certain bureaucracies were habituated to narrowly define the problem as singularly that of Al-Qaeda, the preconditions of Al-Qaeda's success have remained obscure. In fact, the problem of Islamist terrorism is fourfold.

The existence of small fanatical groups of full-time terrorists is only the first part of it. A second precondition is a climate of opinion in the Muslim world that tolerates preaching and recruitment for such groups. Third, such groups need sponsors, which Saudis, aided by a double-gaming Saudi state, were for Al-Qaeda and Pakistan was for the Taliban. To operate effectively, terrorists also need seams between state administrations where they can operate without their sponsors being identified or taking responsibility. So, fourth and finally, the terrorists require places to host them, such as Taliban Afghanistan, and before Afghanistan Sudan, and still to this day Syria and its Lebanese satrapy. In other words, terrorism needs failed or weak states to thrive.

This becomes clear if we review the prewar situation in Afghanistan. The Taliban sympathized with Al-Qaeda's interpretation of Islam, though only in general terms. In the course of the war, however, it became clear that the Taliban also needed Al-Qaeda: to supply highly motivated troops in the absence of a real Taliban army; to give money in the absence of a state apparatus that could raise or manage money; and to supply ignorant village mullahs with ideological guidance, as Al-Qaeda's role in the destruction of the Bamiyan Buddhas suggests. Put differently, the Taliban needed Al-Qaeda because they lacked a state in the modern sense, even though such an organization would be outlawed under normal circumstances. Al-Qaeda needed a place where state structures had been destroyed by civil conflict and left unredressed by international assistance.
Weak states in general pose serious problems for American foreign policy, but our realpolitik orientation on powerful states often obscures this fact. If we list the major headaches of U.S. foreign policy since the end of the Cold War, we find that more have come from weak states such as Bosnia and Haiti than from powerful states like China. September 11 has shown the need for a major re-focusing of U.S. foreign policy to deal with such neglected cases because weak states that happen also to be Muslim states are prime locales for terrorist refuges. It is no accident that Somalia, Puntland (an unrecognized government within the former Somalia) and Yemen became the foci of American concern after the initial victories in Afghanistan. Few have remarked, however, on how much these places resemble each other.

Of course, there are many patches of this planet that lack normal, functioning states in the modern sense. Aside from Somalia and Yemen, we may name the Democratic Republic of the Congo, Liberia, Sierra Leone, Georgia, Bosnia, Colombia, Iraq and others. It is no surprise that within these states are areas not recognized as states that fulfill the conditions for terrorism noted above-places such as the Republika Srpska and Herzeg-Bosna within Bosnia; Abkhazia, South Ossetia and Achara within Georgia; Chechnya within the Russian Federation; the Kurdish quasi-governments in northern Iraq; and FARC- (and paramilitary) controlled areas of Colombia.

Weak states that provide administration-free zones are open not only to fanatics but to criminals and drug lords, as well. Afghanistan is the source of the bulk of the world's heroin, and most of it passes through Central Asia. Moreover, the decay of government authority in Central Asia is already a reality in some states and a possibility in all of them, particularly in remote mountainous areas that have always been recalcitrant to central control. The potential for such Muslim areas to provide religious fanatics and havens and to tempt state sponsors-three of the elements of Islamist terrorism-is very high.

This is already an urgent problem. Tajikistan and Kyrgyzstan are on the verge of failed-state status. When the Islamic Movement of Uzbekistan (IMU), a terrorist organization made up of many different nationalities, invaded Kyrgyzstan a few years ago with a few hundred fighters, the Kyrgyz government was unable to dislodge them. They finally left after an enormous bribe. This disaster was the result of President Askar Akaev's failure to build a normal army, the historical precondition of a modern state. Shoring up the feeble statehood of several Central Asian states is an important anti-terrorist task for the United States, and nothing we can do to this end is as important as training combat-capable armed forces. We began doing this quietly after the IMU incursion, but the pace and scope of this aid has greatly increased since September 11. We need to see this process through to success. This will take time, and it will be facilitated immensely by a local U.S. military presence.

Finally on this point, weak states in the former Soviet Union are a particular problem for the United States. This is because we have treated the map that came into being at the end of 1991 as a permanent international settlement, like the negotiated settlements of 1815 and 1919. We regard those borders as permanent, but several states there, including Tajikistan, do not control all their nominal territory. That heightens the stakes of our general diplomatic posture of support for the consolidation of independent statehood in Central Asia along invariant territorial lines. Russian attempts to "re-integrate" the former Soviet Union, such as they are, run counter to our diplomatic design. We can affect Russian behavior, however, not only by negotiating with them but by changing the facts with which they work. Stronger states in Central Asia will diminish Russia's interest in a revisionist foreign policy. Thus, a policy aimed primarily at preventing and deterring terrorism can work at the same time as a bulwark against lingering imperial tendencies in Russian foreign policy.

The Value of Moderate Muslim Countries

Because the general atmosphere of Muslim countries provides an atmosphere conducive to anti-American terrorism, the United States has a great stake in the success of Muslim states with a moderate orientation. Muslims regard Islam as an umma-a single community-and the points of view that exist within it are important for the direction that the community as a whole will take. This is the reason that Israel has expended such effort to cultivate the Central Asian states. A similar effort should serve us as well.

We start with a major advantage. Traditional Islam in Central Asia never took the Wahhabi form that most often supplies the doctrinal element in contemporary Islamist terrorism. Attachment to formal as opposed to folk or Sufi Islam was never strong among the nomads, the ancestors of the Kazakhs, Kyrgyz and Turkmen, and the Soviet period made the ruling elites of every republic very secular in orientation. Central Asian elites understand and even exaggerate the danger of some forms of Islam to their new states. People are currently rediscovering Islam, some as cultural identity, some in traditional form, and a few as political extremists. But having suffered intra-ethnic violence and seen civil war in Tajikistan, nearly everyone is wary of anything that might spark fighting on a sectarian basis. Thus the Central Asian states, if they can become strong states, can provide significant support for our conception of international order: culturally diverse but tolerant and peaceful.

Closely related to such a conception is the more general American desire to help people, which is not only noble but self-interested. America and its friends account for some 78 percent of "gross world product." September 11 was a sign that the accumulation of American power has reached a point where almost all enemies will respond to us asymmetrically: that is, outside the traditional arena of world politics. America, like Rome at the end of the Second Punic War, seems to be turning into the kernel of a world empire. The only thing that can make a world empire acceptable, however, is some kind of beneficence. Policies such as ending the slave trade made Britain's position in the first half of the 19th century more legitimate. Beneficence-for us, assistance to those who want to be free, action for human rights and democracy, and tolerance-sometimes crosses other foreign policy aims, and it always runs the risk of hypocrisy. But it is still important. A policy to work intensively with the existing Central Asian regimes may be justified by national interest, but it requires as well an active concern for the long-term encouragement of positive trends in these societies.

Unless we act more seriously than we have in the recent past, however, our effort at improvement will indeed seem hypocritical. The United States, the most influential democratic country, routinely protests the lack of free elections in places like Turkmenistan, where the question of elections has no practical relevance. But we do nothing effective in cases where the question is current, where there is international leverage for positive change, and where we would pay only a small price for greater activism. Kyrgyzstan, for example, began as the most democratic country in Central Asia, but has strayed from the path. Enormous foreign debt, which cannot be repaid, dependence on foreign aid and international investment, and the threat of extreme Islamist groups give the democratic community enormous leverage in Bishkek. Major geopolitical interests are scarcely at issue. Nonetheless, democratic countries have passively observed the dimming of Kyrgyzstan's democratic prospects. When President Akaev's major rival, former Bishkek mayor Felix Kulov, was prevented by various shenanigans from actively contesting the 2000 presidential election, the only American responses were private ambassadorial remonstrances, a brief scolding from visiting Secretary of State Madeleine Albright, and a testy phrase or two in a State Department press briefing. There were no concrete consequences.

U.S. human rights policy needs to differentiate its strategies toward various countries, seeking objectives that are obtainable. Free elections are conceivable in Kyrgyzstan, and they might change the nature of the system. Free elections are impossible for years under the present Uzbek or Turkmen regimes, so we should not demand them. The registration of a human rights NGO, on the other hand, was achieved in Uzbekistan in the circumstances of the new U.S.-Uzbek relationship after September 11. Sometimes governments will accept some risks in domestic relations if their external security seems to be better guaranteed, and this is the effect of our arrival on all the Central Asian states. We should leverage our presence to achieve feasible change and thereby strive to reduce over time the problem of working with Central Asia's authoritarian regimes.

Of course, interference in the ways of Central Asian rulers may sour their zeal for military cooperation. We can reduce our risk in this respect by maintaining several military relationships in the region. That will enable us to shift from one base to another if need be, remaining relatively detached from tense Central Asian rivalries. Countering all the inconveniences of their relationships with America, too, will be an eagerness to have us as a counterweight to post-colonial Russia.

There is always a danger in overstaying a welcome, but in Central Asia today there is a danger of understaying a welcome, too. Just as it proved unwise to abandon Afghanistan after the withdrawal of the Red Army in 1989, so it would be a mistake to give up too soon trying to help Central Asian states become forces for moderation and stability in the Muslim world.

Stabilizing Afghanistan

Most important of all, it will prove extremely difficult, if not impossible, to stabilize Afghanistan without a military presence in the former Soviet republics to the north. Weak or failed states foster terrorism, especially if they are Muslim; if Afghanistan is prey to the lawlessness, misery and statelessness that now exist, then the Taliban, or another group like it, will rise again in a few years, attract groups such as Al-Qaeda and we will face September 11-or worse-all over again.

Our task in Afghanistan, which the Bush Administration understands well, is to restore the economy by a route other than renewed opium production, to build an Afghan state that can control Taliban remnants, and to educate Pashtun youth to a more centrist version of Islam. There has been scant recognition so far, however, on just how difficult this task will be.
To do any of this, America needs stability in Afghanistan, and no stability is possible without some approximate balance between the competing ethnic factions within the country and their foreign patrons. The interim government is "multiethnic", but bureaucratic titles mean little in a country that has been hacked apart by civil war for thirty years. In such a situation the "power ministries", to use the expressive post-Soviet jargon, are decisive. Mohammed Fahim and Yonus Qanuni, the Ministers of Defense and Interior, are the masters, along with their close ally, Foreign Minister Abdullah. The way they use their mastery is eloquently expressed by their choice of generals for the new "national" army: 36 of the initial 38 generals were Tajiks, most of them, like Fahim, Qanuni and Abdullah, of Panjshiri background.

Postwar Afghanistan is so far an endeavor to dominate fourteen or more discordant ethnic and religious groups not by the Northern Alliance, nor by Tajiks, nor even by Sunni Tajiks, but by Panjshiri Sunni Tajiks-the inhabitants of just one valley in a very big country. Such a project will inevitably invite the hostility of most other ethnic groups and factions. Moreover, because every Afghan group has a foreign patron, this audacious attempt throws the equilibrium among foreign states out of balance. During the years of Taliban rule, the Northern Alliance depended predominantly on Russia and Iran. Iran now plays a more independent game in western Afghanistan, but the Panjshiris still look to Russia.1

The Russian government has taken advantage of U.S. military victories in Afghanistan to pursue objectives there that are directly opposed to America's own policy and interests. With the Panjshiri Tajiks as their utensils, Moscow essays again the establishment of preponderant Russian influence over Afghan affairs. The Russian project in Afghanistan is staggering in its boldness. When there was a Soviet Union, something similar was tried with the instrument of Afghan communism, and then with an army that had overawed all of Eurasia. Now, shrunken Russia-with an economy contracted to less than South Korean dimensions, a weakened state and a demoralized army-tries again. It is hardly child's play. The mischief being done by Russia in Afghanistan should be considered when we weigh the utility-and the future-of our Central Asian military presence.

The ethnic group that has lost the most from the recent war is the Pashtuns. Pashtuns dominated the Taliban government and the Afghan monarchy from the 18th century, making it still less likely that Pashtuns will accept for long a settlement biased against them. The external patron of the Taliban Pashtuns was Pakistan, which has a huge stake in their fate because there are so many Pashtuns within Pakistan. The shipwreck of Pakistani interests in the war raises profound questions about the survival of General Musharraf, who made the brave decision to suddenly reverse a long Pakistani policy and help us. Musharraf has been able to control Islamic effervescence in Pakistan during the war, and thus to give us military bases. But given the war's outcome, the long tradition of Muslim extremism in the subcontinent, and Pakistan's huge social problems, how long can we rely on Pakistan for bases to operate in Afghanistan? As an insurance policy at the least, the utility of an alternative base for Afghan operations in Central Asia should be obvious.

The very surge of energy we Americans pull from within us to accomplish great tasks hides within it the temptation not to follow through, because it lifts us above everyday things. An entrepreneur who discovers a new way to make a fortune and builds it into a great corporation is likely, just because of that, to neglect the dreary slogging afterward: answering the mail, paying the bills, balancing the books. We can too easily "yield to. . . the relief of victory", as Churchill called it.2 When we had triumphantly routed the Iraqi army in 1991, we relaxed into these so human moods and left the work at the edge of completion. Having so recently had this experience and seen the results, we should know better than to repeat that lapse.