BATTLE ROYALE
If the attacks of September 11, 2001, changed everything, then it is also true that some things changed more quickly than others. Within a year and a half, the United States fought one war and prepared for another, and Congress passed two sweeping bills, the U.S.A. Patriot Act and the Homeland Security Act, to address the war on terrorism's domestic front. But it took more than three years, a best-selling report, and a bruising fight on Capitol Hill for Congress to pass a bill addressing the issue of intelligence failure. That legislation, the Intelligence Reform and Terrorism Prevention Act, was signed by President George W. Bush in December 2004 and hailed by many intelligence reform advocates as a major victory. Its core innovations included the creation of a new director of national intelligence (DNI) and the charter of a national counterterrorism center, two recommendations of the 9/11 Commission that sought to remedy insufficient information sharing within the long-fractured intelligence community and establish more coherence in setting and meeting priorities.
This transformation owed much to the 9/11 Commission and the victims' families who backed it. Its gripping report concluded with a set of proposals for overhauling not just the spy agencies but congressional oversight as well. Rather than disbanding after issuing its report, the commissioners kept the heat on Congress and the White House through the summer and autumn of 2004. The momentum of the report, combined with a unique convergence of political forces, paved the way for a set of reforms that had previously lacked the traction to succeed in getting implemented.
This achievement is all the more remarkable given that between the end of the Cold War and the September 11 attacks, no fewer than six independent commissions and three government reviews tackled intelligence reform -- all to no avail. One long-standing concern was that the CIA director also served as the titular head of the 14 other agencies in the intelligence community, most of which are in the Pentagon, even though he effectively lacked control over their budgets or personnel. This "dual hatting" meant that no one was really in charge of the intelligence community. Other experts warned that spy agencies had not successfully reconfigured themselves to meet the amorphous challenges of the post-Cold War world, terrorism chief among them.
The September 11 attacks threw these issues into bold relief while also highlighting the need for better human intelligence and tighter cooperation among agencies -- especially between foreign intelligence collection and domestic counterterrorism efforts. Congress finished its own September 11 inquiry in 2003 with a bipartisan report that called for, among other things, an end to the CIA director's dual hatting through the creation of a more powerful DNI. Yet once again, efforts responding to these recommendations languished in the Republican-controlled Congress, stymied by quiet resistance from the White House, the Pentagon, and the Pentagon's allies on Capitol Hill. Since the Defense Department's intelligence agencies get the biggest piece of the pie -- around 80 percent of the total intelligence budget -- there was deeply ingrained opposition to the idea of a strong DNI who could divert resources out of the Defense Department into other intelligence agencies, including the CIA. (The total budget remains classified but is thought to come to around $42 billion.) And with wars under way in Afghanistan and Iraq, this bureaucratic concern eventually meshed with a strategic argument: any attempt that might interfere with the Pentagon's intelligence budget would harm U.S. troops on the ground.
Against this backdrop, the passage of a sweeping intelligence reform bill was a political feat of the highest order. The first-ever DNI, former Ambassador John Negroponte, was confirmed in May and has since set up shop and introduced several reforms in such products as the Presidential Daily Brief and the National Intelligence Estimates. Another panel, Bush's Commission on the Intelligence Capabilities of the United States Regarding Weapons of Mass Destruction, issued its own report last March, which included recommendations for fine-tuning the intelligence bill's provisions. Most of those were implemented through executive order in June.
But the battle over the bill and its aftermath have underscored a more pessimistic reality: the resilience of turf protection and bureaucratic self-preservation in Washington. The bitter, drawn-out congressional fight preceding the bill's passage made clear just how ambivalent the political will for reform was, even in the aftermath of the successful report by the 9/11 Commission. Just as problematic was Congress' failure to reorganize its committees and beef up oversight of the spy agencies when it had the chance, which left it poorly positioned to follow through on this legislation and oversee Negroponte's attempts to carve out his new powers. Negroponte has drawn praise for his diplomatic skills, but concern is growing that the former diplomat is better suited to forging smooth relations with the powers that be than making hard decisions on structural reform. And only those hard decisions can address the ongoing problems with interagency cooperation and information sharing. As with any legislation, the success of the intelligence bill depends largely on its implementation. But in this case, the political momentum to build on initial gains is running out of steam just at the critical point.
WHERE YOU STAND IS WHERE YOU SIT
The congressional battle over intelligence reform was distinctive in several respects. Rather than Republicans pitted against Democrats, a bipartisan group of senators squared off against House Republicans. The catalyst was a report by an outside commission, not the advocacy of the White House or a particular lawmaker. What was typical, however, was the fierceness of the turf battle that ensued, both in public and in secret conference talks.
The 9/11 Commission's bipartisanship was a defining part of its strategy. Two well-regarded moderates, Republican Thomas Kean and Democrat Lee Hamilton, headed the panel's ideologically diverse ten-member team, which was split between the two parties. The report's narrative provided exhaustive detail about intelligence failures in both the Clinton and Bush administrations without openly pinning blame on specific individuals. And after the report's release, the commissioners focused their efforts on getting Congress to act on its proposals rather than on retrospective finger-pointing.
This strategy yielded impressive results in terms of both public opinion and traction in the Senate. Polls showed that the commission had a high standing among Democrats and Republicans alike -- with approval ratings above 60 percent -- and its report sold phenomenally well. In the GOP-controlled Senate, meanwhile, the centrist leaders of the Homeland Security and Governmental Affairs Committee, Susan Collins (R-Maine) and Joseph Lieberman (D-Conn.), crafted a robust intelligence overhaul bill within two and a half months of the report's July 2004 publication. The bill hewed closely to the commission's proposals, establishing a DNI with strong budget powers, a national counterterrorism center, and a unified and declassified intelligence appropriation. After bringing onboard powerful skeptics such as Senate Armed Services Committee Chair John Warner (R-Va.), the Collins-Lieberman bill swept through the Senate with a vote of 96-2.
But a far more cautious reaction came from the White House. It was a month after the report's release before Bush first issued a vague statement on its proposals, expressing support for a DNI with "strong budget authority" but rejecting a declassified budget. In contrast to John Kerry, he avoided the subject of intelligence reform throughout the 2004 election. Both Democrats and Republicans close to the bill's negotiations say that the White House did come around in August 2004 to accepting the political need for an intelligence reform bill but that it was a decision wrought by pragmatism rather than conviction. "The White House saw the writing on the wall, so they decided to make lemonade out of lemons," is how Senate Intelligence Committee Chair Pat Roberts (R-Kans.) put it.
If the White House's reaction was tepid, the response among most House Republicans ranged from skeptical to hostile. This opposition did not arise in reaction to the carefully neutral 9/11 Commission report itself but as a response to the commission's confrontations with the administration and its GOP allies on the Hill over the course of its investigation. The White House and most congressional Republicans initially opposed the panel's creation in 2002. After they had started their probe, the commissioners had to bargain hard for limited access to the highly classified Presidential Daily Brief, for getting then National Security Adviser Condoleezza Rice to testify under oath, and for more funding to continue working past their initial deadline. Perhaps most dramatic was an unsuccessful attempt by some congressional Republicans in the spring of 2004 to make Democratic Commissioner and former Deputy Attorney General Jamie Gorelick step down and testify on her alleged role in building the "wall" between intelligence and law enforcement in the mid-1990s -- a part of the investigation from which she had already recused herself and a charge that the report itself largely debunked.
The core issue for some House Republicans was the commission's basic composition and its mandate to include the Bush White House in its probe. Most were furious over the tough questioning that the Democratic commissioners forced on Bush administration officials such as Rice, not to mention the blistering public testimony by former counterterrorism official Richard Clarke. As one GOP source close to the House leadership put it, "Our guys in the rank and file just didn't like the feel of the commission. The process was tainted by its Democratic composition, which caused a lot of uneasiness. [Commissioner Timothy] Roemer had this close association with the September 11 families and House Democrats. Clarke also really pissed people off."
Some Republicans went so far as to charge that the commission endangered national security. House Majority Leader Tom DeLay (R-Tex.) warned in April 2004 that "the politicization of the commission undermines the war effort and endangers our troops." Others made more personal attacks. Representative Jack Kingston (R-Ga.) blasted the panel as "a reunion of political has-beens who haven't had face time since Seinfeld was a weekly show."
So from the start, with the exception of a small group of moderate northeasterners, House Republicans indicated that they would take their own time, and go their own way, in crafting a bill. Many Republican House members immediately warned that the Senate legislation would undermine the war effort. Led by House Armed Services Committee Chair Duncan Hunter (R-Calif.) and House Judiciary Committee Chair James Sensenbrenner (R-Wis.), and flanked by the House Republican leadership, this group quashed a bipartisan effort to pass a House bill similar to the Collins-Lieberman version in the Senate. That move paved the way for the House to push through a GOP-written bill in October that watered down the DNI's authorities and included a slew of provisions on immigration and law enforcement that were neither in the 9/11 Commission report nor in the Senate bill.
Thus began the bruising conference negotiations to reconcile the House and Senate intelligence bills. Although a majority of all conferees -- Republican and Democratic senators as well as House Democrats -- backed the Collins-Lieberman version, the intelligence bill needed the support of a majority of conferees from each chamber to pass. House GOP intransigence, accordingly, was a critical sticking point. Although House Republican negotiators eventually dropped their provisions on immigration and law enforcement, wrangling over the DNI's authority to spend, hire, and fire -- the three pillars of power in Washington -- stymied conference talks through the fall, past the November election, and into a lame-duck session.
At that point, House Republicans got more wind in their sails. The election confirmed that the White House and both chambers would remain in Republican hands -- and with a more conservative majority at that. Meanwhile, the leaders of the 9/11 Commission began making comments that appeared to tilt more favorably toward the House bill. Both developments put the Senate negotiators on more precarious footing. As one Senate aide close to the talks put it, "The commissioners misread what was going on and didn't pressure the House Republicans to move. The fundamental error was to think that you could get the House Republicans to move if you said enough nice things about the bill. But instead that gave the House an excuse not to act, because they could say, 'The commission thinks our bill is good, too.'"
Thus emboldened, Hunter and his allies stalled talks once again after the election, again citing the needs of U.S. troops in battle -- even though the Senate negotiators had already made concessions. One critical compromise was to back off from the 9/11 Commission's proposal to declassify the intelligence budget and give it a separate, unified appropriation rather than keep the traditional practice of parceling out the intelligence budget in various spending bills and hiding it in classified, "black" accounts. Nor did it placate House Republicans that the Senate bill never even gave the DNI budgetary control over tactical and joint military intelligence, the kind of real-time information that soldiers need on the ground. But those provisions were not enough to shake Hunter and his allies. Even House Speaker Dennis Hastert (R-Ill.) underestimated the solidity of his caucus' position. After the speaker announced shortly before Thanksgiving that a compromise had been reached, Hunter refused to endorse the deal, forcing negotiators back to the drawing board.
It was to the surprise of many, then, when negotiators reached a deal in early December. Conventional wisdom at the time credited a last-gasp offer by Collins and Lieberman: new language that guaranteed the president would issue guidelines to make sure that the DNI would not "abrogate the statutory responsibilities" of other departments. In fact, that language did not substantially change the legal authorities of the DNI, as they had already been spelled out in the bill. Rather, shifting political dynamics brought about the deal. As sources close to the talks on both sides saw it, the White House would have been embarrassed by the prospect of a failed bill after Bush's electoral triumph, while Senate negotiators were all the more willing to acquiesce before the more conservative 109th Congress took over in January. Hence, the compromise language gave all sides what they needed, if not what they wanted: House Republicans got face-saving language, and everyone else got a DNI bill. But the tortuous negotiations and the bill's near death highlighted the fragility of the political will behind this progress. The report and the efforts of the 9/11 Commission succeeded in moving Congress and the White House, but only so far. In the end, House Republicans succeeded in redrafting the bill so that it came much closer to meeting their terms. Intelligence reform had reached its high-water mark and would go no further.
HE WHO HAS THE GOLD MAKES THE RULES
Among the bill's most important provisions is the one granting the DNI significant powers over money and people. Although the legislation does not grant the DNI his own unified intelligence budget, it gives him authority on paper to develop and execute around 80 percent of intelligence spending, including the budgets of the CIA, the Defense Intelligence Agency, and the intelligence shops of the FBI and the Departments of State, Homeland Security, Energy, and the Treasury. That authority also extends to the biggest part of the intelligence budget, namely, the money for the Pentagon's national collection agencies: the National Security Agency (which handles communications intercepts), the National Reconnaissance Office (satellites), and the National Geospatial-Intelligence Agency (photography and maps). Budgetary execution is a particularly important power -- and one that the CIA director never had except over his own agency -- because it lets the DNI control the pace of funding once appropriated funds flow from the Treasury Department to the various agencies. In practical terms, it means that once Congress has written the checks, the DNI can direct money to those programs he thinks need the money the soonest.
The bill also grants the DNI the right to transfer unilaterally a capped amount of money and personnel among agencies, the power to name a deputy DNI with four deputies of his own, and the authority to concur in the nominations of the chiefs of the ten agencies under his authority. The DNI's office absorbed the CIA's National Intelligence Council (the think tank that coordinates the National Intelligence Estimates) and the new National Counterterrorism Center, or NCTC (which had previously been parked at the CIA as well, where it was called the Terrorist Threat Integration Center). But the CIA also came out well despite the shakeup. During the restructuring that followed the bill's passage, the CIA won the exclusive authority to manage all overseas human intelligence collection, including that of the Pentagon and the FBI. In addition, the CIA's Counterterrorism Center, not the DNI or the NCTC, remains in charge of disrupting terror plots abroad.
The streamlining of authority under the purview of the DNI was not matched, however, in the legislative branch. Pointing out that oversight of the budget for the Department of Homeland Security was split among 88 different committees and subcommittees, the 9/11 Commission report recommended the creation of strong homeland security committees in both houses of Congress and the consolidation of authorizing and appropriating powers for the intelligence committees, which write the intelligence authorization bills but effectively lack full control over the actual budget. The latter move would have also facilitated the implementation of the commission's proposal for a single, unified intelligence appropriation and ended the practice of chopping up the intelligence budget into secret accounts. Later on, some lawmakers also floated the idea of creating a subcommittee for intelligence on the House and Senate Appropriations Committees, another way to streamline the intelligence budget process.
Congress had a chance to make progress on these proposals while negotiators were hammering out the details of the reform bill, but the reorganization attempt fizzled immediately. Virtually none of the 9/11 Commission's recommendations for Congress itself have been adopted. Appropriators refused to cede any power to the intelligence committees or to entertain the idea of creating an intelligence appropriations subcommittee, which would have likely diminished the power of one subcommittee on appropriations (Defense) and eliminated another (Military Construction). The House, which has a weak homeland security committee, failed to beef up that panel's authorization powers, while the Senate, which had had no homeland security committee at all, decided to give only very modest jurisdiction over homeland security to Collins and Lieberman's committee, limiting its purview to functions outside of transportation security, immigration, and the Coast Guard -- the three biggest items in the Department of Homeland Security's budget.
The failures of congressional reorganization received little press coverage, but the implications are serious. First, the continuation of the status quo keeps intact the unusual degree of control over the intelligence budget by appropriators (who write the checks in the end) rather than authorizers (who write the bills outlining their respective budgets), even though only the latter have a full staff of experts at their disposal. Second, the continuation of the status quo makes it more difficult for Negroponte and his successors to count on a unified budget process, since it forces the DNI to deal with multiple actors rather than a few individuals. That splintering of jurisdiction, in turn, will make any future attempts at intelligence reform more difficult to push through Congress. Third and last, the decision to keep the intelligence budget split up among separate, classified accounts means that appropriators themselves cannot transfer money among their respective subcommittees -- say, from Defense over to the State Department's Bureau of Intelligence and Research or to the FBI to aid its counterterrorism efforts -- even if they decide that such a transfer reflects an important priority. Coordination of the intelligence budget, in short, remains as unwieldy as it was before.
A Republican member of the 9/11 Commission, John Lehman, put the problem this way: "Let's say Negroponte decides to make some fundamental changes. He can do all he wants, but the intelligence authorization bill will still have to go through several committees and get tangled up in a very complicated process with appropriations. There have to be at least two enlightened players for any rational progress to be made. But there's no one person the DNI can go to on the Hill." One main reason why this conundrum remains is related to the reform bill's compromise to keep the total amount of the intelli-gence budget secret -- a tactically necessary concession by the Senate because both the White House and the House Republicans rejected the idea of budget declassification from the start. The latter argued that terrorist groups and other enemies would gain too much knowledge about U.S. intelligence through the disclosure of its budget, whereas the Senate negotiators and the 9/11 Commission dismissed that threat as an unjustified excuse to keep large sums of money out of the view of congressional oversight and out of the public eye. The then interim CIA director, John McLaughlin, had even given a yellow light for declassification during a congressional hearing. But the proposal was a nonstarter, and the Senate compromise effectively killed with one stone the push for a unified budget for the DNI and a more unified intelligence budget process in Congress. The creation of an intelligence appropriations subcommittee or an intelligence committee with appropriating authorities would have led to a unified intelligence budget, and a unified budget is effectively a declassified one. (With a unified intelligence budget, it takes only simple arithmetic to figure out the total once the president submits his annual budget request. Conversely, chopping up the intelligence budget keeps it secret by allowing smaller slices of the pie to be hidden as classified accounts in large spending bills.)
The perpetuation of secrecy has other implications as well. It means that lawmakers have great difficulty bringing up any intelligence-related amendments to the House or Senate floor once a bill is out of committee, because doing so could violate congressional rules barring the disclosure of classified information. More broadly, it also means that Congress and the American public cannot have an open debate about intelligence priorities. And by keeping the intelligence committees relatively weak -- no power of the purse, no way to amend spending bills, no appropriation subcommittee to coordinate the budget -- Congress left those panels poorly equipped to act as a check on abuses or failings within the intelligence agencies. Accordingly, the prospects for stronger congressional oversight of intelligence remain bleak. Kean, chair of the 9/11 Commission, put it this way: "Only Congress can hold the intelligence agencies accountable, and that's why it has to function effectively. ... Negroponte might be doing the job with the support of the president, but we have a problem finding out whether he can do it at all since this is behind closed doors. No one else can fix this but Congress, and if Congress can't fix it, then there's no one to make sure Negroponte can do his job."
LEARNING TO SHARE
One of the critical failures highlighted by the 9/11 Commission report was the disastrous lack of communication among various intelligence agencies -- especially between the FBI and the CIA. The many missed opportunities are by now well known. In 2000, the CIA did not alert other agencies about the only two September 11 hijackers who were known to have al Qaeda ties before they arrived in the United States, and later the FBI failed to devote sufficient time and agents to pursue them. Investigators in the case of the October 2000 bombing of the U.S.S. Cole did not receive critical information from the CIA and the FBI, and in August 2001, the FBI did not grasp the significance of Zacarias Moussaoui's flight training, which was revealed after he was arrested and which would have led them to press him for information about his al Qaeda links. Whether capitalizing on these opportunities would have cracked the September 11 plot will be debated for years, but they all stand as examples of how the system should not work.
There is widespread agreement that coordination and information sharing have improved since the attacks. Both the creation in 2002 of the Terrorist Threat Integration Center and its subsequent morphing into the NCTC last year through the DNI bill and related executive orders have helped coordinate intelligence analysis across agencies. The NCTC now runs daily meetings with representatives from the CIA, the FBI, the Defense Intelligence Agency, the National Security Agency, and the Departments of State and Homeland Security. It also meets daily to discuss the drafting of analytic products such as the president's daily threat report.
But thanks to the years of entrenched independence of the individual agencies and a long tradition of information hoarding, major shortcomings remain. Although the NCTC is generally regarded as an improvement over the status quo, the status quo is a poor benchmark for comparison. NCTC analysts have access to 26 information networks across the intelligence agencies (housed in three computers on each analyst's desk), but there is no one database that unites these networks, nor is it possible to conduct a search across all 26. It was only in the autumn of 2004, three years after the attacks, that analysts first began to be able to conduct a search across two of the networks. NCTC analysts can share information with each other person to person, but if they wish to share information with intelligence colleagues outside of the NCTC, they must get the permission of the originating agency. (For example, an FBI analyst at the center wishing to exchange intelligence with a CIA analyst outside of the NCTC may do so, but only with official permission from the FBI.) Although that may sound like a reasonable safeguard, "there is still a lack of understanding on what needs to get out," according to one senior intelligence official.
Meanwhile, Negroponte has a dizzying "to do" list that includes responsibilities that extend beyond those he has related to the NCTC. Intelligence sharing between special operations units on the ground in places such as Iraq and Afghanistan and the office of the DNI is only in an embryonic state. Conversely, state and local authorities now receive a stream of terrorist-related information from federal agencies, but much of it is redundant and poorly processed. There is still no single office or individual in the intelligence community to mandate information sharing across all agencies. The overhaul of the FBI -- whose counterterrorism, counterintelligence, and intelligence authorities are now supposed to come under the umbrella of the DNI as one fused unit -- poses further questions. The office of the DNI is still negotiating with the FBI and Congress on just how robust Negroponte's powers over the FBI's national security functions will be. The recent collapse of the bureau's unfinished $170 million Virtual Case File computer system serves as a stark example of its inability to move forward into the information age. Many critics also charge that the fiercely independent bureau has not adapted its case-based law enforcement culture to the very different needs of intelligence collection or put enough resources into cultivating intelligence analysts, an area long seen as a backwater. Most of the fewer than two dozen FBI analysts at the NCTC are drawn from the younger, less experienced ranks, due to a paucity of seasoned analysts.
Finally, concern is growing that Negroponte's office is simply another layer of bureaucracy over all agencies rather than a force that can push through necessary structural changes to streamline the intelligence community and foster more accountability. In May, Negroponte's deputy, former chief of the National Security Agency Michael Hayden, tapped four officials for collection, analysis, intelligence community management, and customer relations ("customer" meaning anyone who uses intelligence, from the president to Congress). But these officials have amorphous responsibilities covering a broad range of services rather than responsibility over a specific agency, mission, or branch of intelligence, and they all hold lower executive-branch ranks than the intelligence chiefs they are supposedly overseeing. Accordingly, some critics worry, the office of the DNI can neither ensure greater accountability nor use the muscle of its deputies to encourage more "jointness" among the intelligence agencies. As 9/11 Commissioner Lehman said, "This model doesn't give anybody line authority over anything. You just have meetings and studies that focus on the process rather than the product."
Negroponte has been in office only since May, but diplomacy, rather than boldness, has marked his critical first few months. Negroponte has significant power on paper, but he has spent much energy on projects that do little to further a bureaucratic transformation. His office has commendably revamped plans for the Presidential Daily Brief (which Negroponte or an assistant, rather than the CIA director, delivers to the president), the president's daily threat report, and the National Intelligence Estimates, so that more nuance and input from all agencies are incorporated. But if the spirit of the intelligence overhaul effort was to create a CEO for the intelligence community, not just a new top briefer for the president, that transformation has yet to happen. As Admiral William Studeman of the WMD commission argued, "The PDB is not a good use of Negroponte's time. It should take less than one percent. It's an important thing relative to an important customer but he needs to be focusing on the rest of the community." Another senior intelligence official expressed the problem more bluntly: "Only Negroponte needs to understand what the intelligence business is, and it's not just about the PDB. It's about looking at organizational issues. People at the top are trying to play nice with each other, but nobody is willing to make tough calls on what the CIA and FBI can and cannot do. The DNI is putting together staffs to compensate for structural deficiencies rather than addressing the deficiencies themselves."
BUSINESS AS USUAL
The fact that an intelligence shakeup occurred at all remains a singular accomplishment. A unique convergence of forces -- the resounding success of the 9/11 Commission, growing media attention on intelligence shortcomings, shifting dynamics in Congress and the White House -- allowed for some progress despite long-entrenched opposition. But even broad support in the Senate, overwhelming public opinion, and a best-selling report were not enough to move Congress nearly as far as overhaul proponents had hoped. The story thus serves as a textbook case of how the opaque ways of Washington's bureaucratic warfare undermine sound policy. Insiders used both official and unofficial levers of power to keep as much of their turf intact as they could. As with so many bills, a compromise resulted.
Most worrying is that this compromise, already laden with caveats, could become a standalone accomplishment rather than part of a long-term transformation. A DNI is in place and the NCTC is up and running, but the risk of complacency is rising. Last September, the disbanded 9/11 Commission began issuing a series of toughly worded statements warning of the incompleteness of the effort, but this time around the panel did not shake Congress into action. All the while, the challenges to the U.S. intelligence agencies are growing, from North Korea to Iran to the insurgency in Iraq. And Hurricane Katrina tragically underscored the poor state of the nation's emergency response infrastructure. Indeed, the rise and fall of the reform effort brings to mind Dean Acheson's famous quip. "Washington," he noted, "is like a self-sealing tank on a military aircraft. When a bullet passes through, it closes up." The bullet of intelligence reform may have hit its mark, but the tank remains intact.
You are reading a free article.
Subscribe to Foreign Affairs to get unlimited access.
- Paywall-free reading of new articles and a century of archives
- Unlock access to iOS/Android apps to save editions for offline reading
- Six issues a year in print, online, and audio editions