Keeping America's Military Edge
Journal Article, Foreign Affairs, volume 80, issue 1, pages 90-105
Author: Ashton B. Carter, Former Co-Director, Preventive Defense Project, Harvard & Stanford Universities
January/February 2001, Volume 80, Number 1
Keeping America's Military Edge
By Ashton B. Carter
THE (NOT SO) GREAT DEBATE
Last fall's presidential campaign rarely ventured into discussions of national defense. When it did, these all-too-brief excursions mostly concerned setting priorities among the almost numberless tasks that should be taken up by the world's leading power. That meant answering several questions: Will China and Russia pose future threats, or can they be cooperatively integrated into the international system? Is preparing to fight two major wars in Korea and the Persian Gulf still the appropriate organizing principle for U.S. forces and budgets? When and how should the United States participate in peacekeeping and conflict prevention?
Such debates focused on the ends of U.S. foreign and defense policy. Although important, they left unanswered an equally fundamental question: whether the United States has the means to implement the policies it decides on. And this lapse was unfortunate, for there is mounting evidence that the necessary means may in fact be wanting. The U.S. national security establishment is becoming deficient, not so much in deciding what to do, as in the ability to get it done. As a result, the new president and his administration may well find their ability to translate policy decisions into results sharply limited. Reversing this trend will depend on the management of means.
At one point, the campaign did touch briefly on means -- namely, when the two sides debated the "readiness" of the U.S. military. George W. Bush charged that U.S. forces were not ready for their current missions. Al Gore replied that the Pentagon was spending a larger fraction of its budget on training and exercises -- the two factors that contribute most directly to readiness -- than ever before. But whatever the merits of their arguments, both sides overlooked a key issue. The "readiness" about which they were arguing was that of today's forces to meet today's threats. A more fruitful debate would have focused on a longer-term version of readiness and a wider view of the defense establishment.
The near-term proficiency of the U.S. military is, in fact, unrivaled. The United States today retains a commanding military edge over all other nations. This advantage is the result not only of large budgets, but also of important managerial and organizational changes made over the years. Conscription was replaced by voluntary service in 1973. Military advice to the president and joint command over military operations were strengthened by the Goldwater-Nichols Defense Reorganization Act of 1986. And then there was the fundamental decision made by Cold War-era Pentagon managers to rely on a distinctive American asset -- innovative technology -- to overcome the Warsaw Pact's superiority in numbers. This reliance on a technological edge, dubbed the "offset strategy" by former Secretary of Defense William Perry, was essential to ensuring America's overall military edge.
That edge persists today. The long-term readiness of the national security establishment to face tomorrow's threats, however, remains in question. Rather than debate the shape of today's forces, therefore, Bush and Gore could more usefully have proposed the managerial changes necessary to ensure that the U.S. military keeps its advantage years from now in the face of the globalization, commercialization, and information revolutions that are transforming the world. Unless the next president acts to address these changes by transforming the way the U.S. defense establishment does business, the nation's military dominance may, in time, begin to waver.
TROUBLE IN THE BACK ROOM
One need not look far to find signs that the next president must start paying attention to his role as a manager of means, not just a definer of ends. The fact is that America's combat forces -- which, since the end of the Cold War, have won victory in the Persian Gulf, peace in the Balkans, and respect from friend and foe alike -- are much better at their jobs than the institutions that support and complement them. The point of the spear may be sharp and hard, but much of the rest of the national security establishment is broken. The underpinnings of quality performance in government functions are eroding. Top-flight people refuse to serve at all levels of government, from high political posts to the civilian and uniformed services, because the conditions of public service are often demeaning and frustrating. Good people already in government are leaving, and those who remain often feel that their potential for creative leadership is stifled. What former Deputy Secretary of Defense John Hamre called the "back room" of defense -- systems for auditing and accounting, attracting and rewarding public servants, contracting for weapons and services, controlling exports, and security classification and background checks -- all show signs of bureaucratic decline and apply an accumulation of rules rather than logic to their assigned missions. In the past decade, military bases and depots have not been reduced nearly as much as the armed forces themselves, resulting in a bottom-heavy structure with a large "tail-to-tooth" ratio and billions of wasted defense dollars. Department of Defense (DOD) acquisition personnel are still burdened with the Federal Acquisition Regulations, a guide as thick as a big-city telephone book. Forces meant to fight jointly are still equipped, sometimes incompatibly, by the separate services and defense agencies.
Meanwhile, outside the armed forces, the research and industrial base on which America's high-tech edge relies is being transformed by the forces of globalization and commercialization. But the DOD persists in many old habits regarding research and development (R&D) and weapons procurement. As a consequence, the U.S. military is not fully exploiting or even staying abreast of the information revolution. It is scarcely even in the game when it comes to biotechnology, which could have an even more profound impact on human conflict than has information technology. As for the defense industry itself, on which the technological edge ultimately depends, it is having a hard time raising capital and is losing many of its talented engineers and managers. Finally, transatlantic defense-industry cooperation, important for efficiency and NATO cohesion, remains an unsolved problem for all allied governments.
President Eisenhower once said that the right system may not guarantee success, but the wrong system guarantees failure. Why, then, does Washington pay so little attention to the organization and management of the national security system? The reasons are several. Most senior officials are appointed for their policy acumen, not their management skills. Most regard their term in office as too short and precarious to spend time making structural improvements that will benefit their successors. Government organization and management -- closing unneeded bases, for example -- is usually the stuff of small politics, a reminder of former House Speaker Thomas (Tip) O'Neill's maxim that "all politics are local." Overcoming inertia, ingrained habits, and entrenched interests and bureaucracies is sometimes more difficult than winning a spirited national debate on a major policy issue. Reforms of any depth require close cooperation between Congress and the executive branch, which is never easy to achieve.
Despite the difficulties, however, now is the time to heed Eisenhower's maxim. The transition to a new administration provides an opportunity to undertake needed changes in the national security system -- an opportunity that comes only every four or eight years. Early in a presidential transition, civilian jobs are not yet filled with officials who, once appointed, might resist change in their functions. The new administration has not yet started making do with the dysfunctions it inherited. Congress and the voters expect new measures. The moment is therefore ripe to address the management problems in defense.
The necessary changes must meet three basic needs. First, the national security establishment is being asked to accomplish new missions, ranging from peacekeeping to dealing with terrorism committed with weapons of mass destruction (WMD). It needs to figure out how to develop an edge in these new areas. Second, where the United States already has an edge -- in joint operations, high-tech military systems, and quality people in uniform -- it must keep it, despite dramatic changes occurring in the market and in technology. Third, in areas where DOD management has long underperformed, such as base closures and reforming the acquisition system, a true "revolution in business affairs" is overdue.
Today, some of the most critical security missions -- counterterrorism, combating WMD proliferation, homeland defense (including protection against computer network attacks and biological weapons), information warfare, peacekeeping, civil reconstruction, and conflict prevention (or "preventive defense") -- are accomplished in an ad-hoc fashion by unwieldy combinations of departments and agencies designed a half-century ago for a different world. Too many of these missions are institutionally "homeless": nowhere are the authority, resources, and accountability brought together in sharp managerial focus. Although it is widely agreed that the United States needs the means to accomplish these homeless missions (even if debate continues over exactly when and where it should perform them), the U.S. government is not well structured for these jobs.
Many of the homeless missions, although apparently unrelated, have a key element in common: they cut across different cabinet departments and require the coordinated action of several agencies. This is hardly surprising, since the problems they address do not respect neat distinctions between foreign and domestic threats, military and economic remedies, or states of war and states of peace. Such distinctions are outdated and merely reflect what the world was like in the era after World War II, when the American national security establishment was founded. It is due to such distinctions that no one agency is automatically in charge of these missions or responsible for developing and directing the capabilities they require.
One example of a homeless mission is preparing for "catastrophic terrorism" -- that is, attacks on the United States not made during a formal war but nonetheless on a warlike scale. These assaults may involve nuclear, chemical, or biological weapons, cyber-attacks on critical infrastructures, or assaults on government institutions. Such an incident could simultaneously be described in different ways: as an attack, a crime, or a disaster. This means that the Departments of Defense, Justice, and others, as well as the Federal Emergency Management Agency (FEMA), could all lay claim to being in charge of the federal government's role in prevention and response. The gravity of these threats and the danger that responses might fall through the cracks in Washington's bureaucracies have already led to calls for a new cabinet department of homeland defense, for the appointment of a "catastrophic terrorism czar" in the White House, for turning such missions over to the Pentagon (as was done with air and missile defense), and for legislation to add more members to the National Security Council (NSC).
The Clinton administration wisely avoided embracing any of these suggestions and relied instead on the coordinating power of the NSC. The council duly negotiated a series of presidential directives with the relevant cabinet departments, spelling out who was in charge in various circumstances: the FBI in most domestic cases, for example, and the State Department in most overseas situations. Other agencies, like the CIA and the DOD, were assigned support roles.
Most of the agencies assigned central roles in the NSC policy, however, have neither the capability nor the resources to carry out their jobs, which thus remain unfunded mandates. This means that although policy may be clear and coordinated in these cases, there is no coherent interagency program for actually accomplishing the homeless missions. To properly fight catastrophic terrorism requires a multi-year, multi-agency, presidentially approved and enforced plan to develop the capabilities (such as surveillance and detection, public health, and forensic measures) necessary to protect the country, while avoiding waste and safeguarding civil liberties. Without a proper interagency program, mere coordination will produce a "come as you are" party, where agencies can offer only their current resources. These current capabilities, even taken together, are not adequate for meeting the threat of catastrophic terrorism.
The nub of the problem is not the NSC's membership, but its practices, which are ill-suited to address homeless missions. In the years since the Eisenhower administration, the NSC has become a highly effective mechanism for coordinating policy -- but not actual, working programs. This is evident in the council's staff, who have the highest foreign policy expertise but almost no experience in programming, budgeting, or the management of technical programs or large organizations. Meanwhile, the Office of Management and Budget (OMB), which shares coordinating responsibility with the NSC, has much less clout over the programs of security agencies than it does over domestic agencies.
To address these shortcomings, the new administration should charge the NSC (working with the OMB) to develop multi-year interagency budget plans for each of the homeless missions. This task will require staff with program as well as policy expertise -- perhaps working for a new deputy national security adviser appointed for this purpose. The new staff would be confined to planning and programming and would not have the kind of operational role that led to the Iran-contra scandal of the Reagan-era NSC. Meanwhile, the active involvement of the vice president in coordinating cabinet members has been helpful in past efforts of this sort and would be valuable here. And the congressional committees and subcommittees overseeing separate agencies should coordinate their responses to the president's budgets for each of the homeless missions.
JOINT AND SEVERAL LIABILITIES
Joint war-fighting, unlike the homeless missions, is a traditional concept and an area where the United States has established an unparalleled edge over other nations. This advantage owes in part to the Goldwater-Nichols Act of 1986, which was intended to ensure that U.S. forces fight together, rather than in separate Army, Navy, and Air Force campaigns. The act gave unified commanders-in-chief (CINCS) clear authority for joint operations. But it assigned the jobs of organizing, training, and equipping the forces to the separate armed forces. So although operations are now "joint," equipment and personnel are still acquired severally.
The Goldwater-Nichols reforms have been a great success by almost any measure. But they did not address the question of how several separate acquisition systems can produce truly joint forces. One way to address that now -- an option that should be rejected -- would be to create a truly joint acquisition process at the expense of the services' own authority. This move would have the theoretical advantage of giving the power to configure and buy joint forces to their ultimate users -- the war-fighting CINCS. However, in practice, this option would weaken the services -- the sort of proud, living institutions of which there are far too few in government. Joint acquisition would undermine the services' proven ability to provide the best land, naval, air, and amphibious forces in the world. In addition, transferring responsibility for requirements, budgets, or acquisition to the joint CINCS would divert their attention from their principal tasks: namely, to command operations, plan for regional contingencies, and maintain U.S. military relations with friends and allies. The CINCS have no staff specialized in acquisition. Giving them that new responsibility would therefore weaken, not strengthen, the acquisition system.
A second option is to maintain the current arrangement. Indeed, this system may seem the best balance between the demonstrated expertise of the services and the need for jointness. But in fact, maintaining the status quo would result in increasing imbalance, due to the changes being wrought in warfare by information technology. According to the logic of the so-called revolution in military affairs, military capability improves dramatically when battlefield systems of tanks, ships, and airplanes are linked into a "system of systems" through an information network that allows all U.S. forces to fully coordinate their operations and to know the precise location and disposition of enemy and friendly forces. Exploiting the potential of such networked warfare obviously requires compatibility among the several armed services. Even more important, the network that connects them is an inherently joint system that falls under none of their traditional acquisition responsibilities.
It is clear, therefore, that a strengthened mechanism is needed to develop and acquire military systems that are truly joint. A third, middle-ground option offers the best chance for sustaining the expertise of the individual services while giving appropriate voice to joint considerations in acquisitions. Rather than involving all the CINCS in the process, a single CINC -- the Joint Forces Command CINC -- should be given the power to inject joint thinking into the acquisition process on behalf of the chairman of the Joint Chiefs of Staff and the other CINCS. This is, in fact, the avenue supposedly already being pursued by the DOD. But the Joint Forces Command has not yet been given the tools for the job. To make the process work will require four additional steps. First, the Joint Forces CINC should prepare for the chairman of the Joint Chiefs a broad road map, updated annually, for developing truly joint systems. Next, the Joint Forces Command should be given the personnel and resources necessary to take on its new acquisition responsibilities. These should include some direct authority over budgets for purchasing inherently joint systems (such as the battlefield networks described above). The person chosen to head the Joint Forces Command should be a senior officer with experience as a CINC, service chief, or vice chief. Finally, he or she should become a member of the DOD's key decision-making bodies on acquisition matters: the Joint Requirements Oversight Council and the Defense Resources Board.
The United States' "offset strategy" -- the Cold War notion of using superior technology to counter opponents' advantages in numbers, geography, or willingness to risk casualties -- has become key to Washington's way of waging war. The fruits of the offset strategy were dramatically demonstrated during Operation Desert Storm -- a war no one expected -- when reconnaissance satellites, stealth aircraft, precision weapons, and other unique technologies made short work of Iraq's forces, which resembled a miniature Warsaw Pact military. In light of such success, the United States is determined to keep its technological edge and apply it to post-Cold War security missions.
The offset strategy rests on the DOD's ability to draw on a strong technology base and on a defense industry that uses such technology in new defense systems. But current trends in high-tech industry -- the hallmarks of the "dot-com" era -- are changing the field so fundamentally that once-successful DOD practices can no longer guarantee a future technological edge. Consider the numbers: according to the National Science Foundation, in 1980, the West (as it was then defined) spent a total of about $240 billion (in today's dollars) on scientific research and development. The U.S. DOD contributed about $40 billion, or one-sixth of the total. Last year, by contrast, the corresponding total for R&D spending reached $360 billion. But DOD funding accounted for only one-twelfth of the total -- half of its 1980 share.
Meanwhile, the circumstances that the DOD grew accustomed to in the Cold War, when it first formulated the offset strategy, are becoming inverted. During the Cold War, technologies critical for defense originated in a "defense technology base" -- that is, in laboratories and companies that were mostly American, and for which the American military was the main customer. These days, militarily critical technologies are increasingly developed by global companies for which the U.S. military is a small market. DOD-funded research pioneered the microchip, massive parallel processing, the Internet, software engineering techniques, and other technologies that are now being exploited and further developed by the well-financed private sector. In the past, America's military advantage was conferred by its possession of technology, unique to the defense sector, that the country's opponents could not get. No longer. In the future, the technological edge that the offset strategy requires will be maintained only if the DOD gets much better and faster than its opponents at absorbing commercial technology into defense systems. After all, America's potential enemies will have access to much of the same technology.
Learning to integrate commercial technology faster than potential opponents can will require the Pentagon to increase its R&D spending, as much to keep abreast of developing technology as to make its own contributions. It will also require the DOD to employ experts familiar with commercial technologies and practices -- and not just specialized niches, such as stealth technology, that are unique to defense. Washington must also start using commercial buying practices and systems in defense procurement. If the DOD sticks to its idiosyncratic buying methods and cumbersome contracting procedures, it will always remain a generation behind commercial practice.
Above all, maintaining America's technological edge will require a healthy and innovative defense industry. Unfortunately, the industry is heading in just the opposite direction. Defense companies must justify themselves to investors by the same standards of profit and cash flow as commercial firms. But Wall Street regards defense firms as part of the "old economy," and thus market forces are drawing capital away from them. The market capitalization of all the major defense firms put together is today about half that of Wal-Mart's and just a quarter of Microsoft's. The list of U.S. corporations that have left the defense market reads like a Who's Who of industrial America: IBM, Texas Instruments, Ford, Chrysler, Westinghouse, and others. Meanwhile, "new economy" companies are wholly absorbed in the pursuit of rapidly growing commercial markets and have shown little interest in the sluggish defense market. Finally, the most worrisome trend is the flow of talented people leaving the defense industry for other fields, which reward executives with generous stock options and offer engineers the satisfaction of knowing they are on the cutting edge.
For the DOD to maintain the offset strategy in this climate, it must align its own practices with the market forces acting on both defense companies and on the commercial businesses that generate key technology. Policies should not prop up weak defense companies and accentuate their isolation from the industrial mainstream. Instead, they should work with, not against, market forces. The U.S. government's acquisition and contracting policies should reward the defense industry when it follows sound business practices in pursuit of innovation and efficiency, such as sharing savings from cost-cutting and consolidations, and allowing higher profits when industry delivers value in terms of cost, schedule, and performance. Today's procurement practices usually provide precisely the opposite incentives. Although the Pentagon cannot reverse the 1990s trend that concentrated the industry into a handful of large, purely defense contractors, it can and should encourage second-tier and third-tier companies that serve commercial markets to remain in the defense market. The DOD should also encourage robust transatlantic defense industry links, which will provide classic free-trade efficiencies and reinforce NATO solidarity.
A successful offset strategy cannot be based on a defense industry that is isolated from globalized commerce and innovation -- a rust-belt defense industry in a dot-com economy. Although the crisis has not yet reached such proportions, it may well do so unless Washington soon makes substantial changes in its Cold War practices.
PUTTING PEOPLE FIRST
In addition to its technological edge, the U.S. military already enjoys an advantage in the quality of its uniformed personnel. Here, too, keeping that edge will require adapting to the new environment -- changing labor markets, demographics, and the types of employees needed. America's armed services, for example, traditionally target recruiting at high school graduates who choose not to go on to college. But the fraction of people who make such a choice is now shrinking -- today it is only one in three. This is forcing military recruiters to draw from a smaller and smaller pool. To change this trend and retain access to high-quality recruits, the Pentagon needs to somehow make military service compatible with college ambitions. The changes needed may be fundamental. For example, the traditional career paths for officers and enlisted troops may be inappropriate when an increasing fraction of both pursue college educations. More paths should be opened for promising enlisted members to become officers, which would increase their opportunities for advancement. The growing number of two-career families also poses challenges to traditional concepts of military life. Meanwhile, the military needs to do a better job competing with today's booming economy for employees trained in information technology.
All these problems are well known to private businesses, which regard keeping personnel policies and costs under thoughtful control to be one of management's most important responsibilities. But military compensation policy is often set in political bidding wars that usually result in across-the-board increases, rather than targeted adjustments that reward individuals for their contributions and that emphasize the DOD's changing personnel needs. Military compensation needs to start being managed as a system, not a political football.
The same thinking should be applied to quality of life for military families. Here the DOD too often thinks in terms of merely increasing the provision of amenities (such as housing or grocery stores). This is an anachronistic vestige of the nineteenth-century military practice of providing everything a garrisoned soldier on the prairie needed. Today, at least in most locations, civilian markets are more than adequate for the needs of military families. Improving troops' quality of life should therefore focus on assisting military families to use the resources that already exist. Often this means simply giving them money to address their own needs, rather than acting as if the government knew what they needed better than they did.
The DOD's civilian personnel system needs even more fundamental reform. Unlike the uniformed system, the civilian system has not had the edge in quality for some time. This system is out of touch with the labor market and the changing needs of the DOD. Worse, it stifles professional development and innovation in its workforce. To remedy this, the secretary of defense should request legislative authority to remove the DOD's civilian workforce from the civil service and set up a new system directly responsible to the secretary -- as several other agencies have already done. The new system would have more flexible pay and hiring rules, portable pensions, and other provisions that allow people to enter, leave, and reenter government service. The civilian system should tie compensation to performance and attach employees' grades to their persons rather than to the jobs they happen to occupy at the moment. And it should provide for professional training and rotation to other agencies so that employees can get a feel for how similar issues are viewed in other departments. Such a system would be the opposite of the current one in almost every respect, but would vastly improve the quality of the one-third of the Pentagon's workforce that does not wear a uniform.
REVOLTING BUSINESS AFFAIRS
Although the United States has long held a substantial advantage in the core area of military operations, it has long lacked such an edge in the management of support functions. The acquisition system may need to be reformed, but this need is even more urgent in most of the DOD's other business affairs: accounting and auditing, security classification and background checks, export controls administration, utilization of bases and depots, logistics, and a host of other noncore functions. Indeed, former Deputy Secretary of Defense John White declared in 1995 that the DOD's performance in such areas was so far below standard practice in the private sector that a "revolution in business affairs" was needed.
One example of the attempt to achieve this revolution was the DOD's effort to outsource functions that are not inherently governmental in nature and that could be performed more efficiently by the private sector. There is no reason why DOD employees need to carry out nonmilitary tasks such as finance and accounting, teaching trade skills to soldiers, or managing long-distance communications and noncombat logistics. Holding large competitions between governmental and private suppliers of such functions would allow Washington to identify which jobs should be transferred to the private sector.
Unfortunately, Congress and the OMB have put in place numerous rules and procedures that thwart the competitive process and protect government jobs. These should be overturned. In 1994, Secretary Perry took a simple but far-reaching step in the analogous case of "military specifications" -- special DOD requirements imposed on items it buys that are often unnecessary and expensive. Perry decreed that program managers who had previously been required to obtain a waiver to avoid using military specifications would now need a waiver to apply such specifications. The new secretary of defense should, in a similar fashion, turn the presumption against outsourcing on its head by declaring the private sector to be the preferred supplier of noncore goods and services. The secretary should seek legislative authority to loosen the many strictures that currently exist, and should conduct a competitive program to fully realize the available efficiencies.
As the Cold War ended, then Secretary of Defense Dick Cheney, a former member of Congress, realized that it would be much easier to reduce the overall size of the military force than to reduce the number of bases and depots scattered about the country, many of them located in the home districts of influential representatives. Congress had in fact already erected by statute an intimidating gauntlet of studies, findings, public hearings, and other steps the DOD would need to take before closing a base. Cheney nevertheless published a list of unneeded bases and began running that gauntlet. And once Congress realized how much time and energy this process would consume and saw that Cheney was determined to stay the course, it decided to change the rules, allowing Washington to get the job done with less political hazard for all concerned. The so-called Base Realignment and Closure (BRAC) procedure achieved three rounds of closures. But in 1995, Congress accused the Clinton administration of manipulating the BRAC process for partisan advantage and, weary from three rounds, refused to reauthorize the legislation. The previous, cumbersome base closing law took effect again. This standoff has now lasted for five years. Meanwhile, money pours into bases that serve no national security purpose.
The new secretary of defense should take a page from Cheney's playbook. The secretary should publish a new list of unneeded bases and announce the intention to close them under the extant law -- namely, by the cumbersome procedure. Congress, faced with this unpalatable prospect, will likely back down and reinstate some version of the BRAC process. Billions of dollars will be saved.
The new administration and Congress will take office with a strong inheritance: America's commanding military edge. But this advantage is not a birthright or a fact of nature. It will need to be sustained in the face of dramatic changes in the makeup of U.S. society, the globalization and commercialization of technology and the defense industry, and the nature of modern conflict. America's military edge must be assured in new missions such as homeland defense and peacekeeping, where U.S. power is not yet fully developed. And the national security establishment as a whole must boost its managerial efficiency, dynamism, and the motivation of its workforce. In an area as important as national defense, it is simply unacceptable for government to lag so far behind the private sector. Without superpower competition or a sense of imminent danger to galvanize it, however, the U.S. defense establishment has drifted toward complacency, toward just getting by, without making the kinds of difficult choices that ensure real superiority.
The new administration will be judged by the results it obtains, not just by the policies it propounds. The president and his defense team would do well, therefore, to cast an eye over the means at hand to pursue their policy ends -- the people and organizations of the national security agencies. Four years is long enough for the managerial problems of the defense establishment to take their toll. Four years is also long enough for changes made to take form -- as a sustained military edge that this president could then pass on to the next one.
Ashton B. Carter is Professor of Science and International Affairs at Harvard University's Kennedy School of Government, and served as Assistant Secretary of Defense for International Security Policy from 1993 to 1996. This article draws on Keeping the Edge: Managing Defense for the Future, published by MIT Press, which the author edited with John P. White.
The contents of Foreign Affairs are protected by copyright.
© 2005 Council on Foreign Relations, Inc., all rights reserved.
Please see the pdf file below for the full text of this article:
Full text of this publication is available at:
For Academic Citation: