Throughout this summer and fall, barely mentioned in America's presidential campaign, Iran moved steadily closer to a showdown with the United States (and other countries) over its nuclear plans.
In June the International Atomic Energy Agency said that Iran had not been forthcoming about the extent of its nuclear programs. In July, Iran indicated that it would not ratify a protocol of the Nuclear Non-Proliferation Treaty giving inspectors greater liberty within its borders. In August the Iranian Defense Minister warned that if Iran suspected a foreign power—specifically the United States or Israel—of preparing to strike its emerging nuclear facilities, it might launch a pre-emptive strike of its own, of which one target could be the U.S. forces next door in Iraq. In September, Iran announced that it was preparing thirty-seven tons of uranium for enrichment, supposedly for power plants, and it took an even tougher line against the IAEA. In October it announced that it had missiles capable of hitting targets 1,250 miles away—as far as southeastern Europe to the west and India to the east. Also, an Iranian Foreign Ministry spokesman rejected a proposal by Senator John Kerry that if the United States promised to supply all the nuclear fuel Iran needed for peaceful power-generating purposes, Iran would stop developing enrichment facilities (which could also help it build weapons). Meanwhile, the government of Israel kept sending subtle and not-so-subtle warnings that if Iran went too far with its plans, Israel would act first to protect itself, as it had in 1981 by bombing the Iraqi nuclear facility at Osirak.
Preoccupied as they were with Iraq (and with refighting Vietnam), the presidential candidates did not spend much time on Iran. But after the election the winner will have no choice. The decisions that a President will have to make about Iran are like those that involve Iraq—but harder. A regime at odds with the United States, and suspected of encouraging Islamic terrorists, is believed to be developing very destructive weapons. In Iran's case, however, the governmental hostility to the United States is longer-standing (the United States implicitly backed Saddam Hussein during the Iran-Iraq War of the 1980s), the ties to terrorist groups are clearer, and the evidence of an ongoing nuclear-weapons program is stronger. Iran is bigger, more powerful, and richer than Iraq, and it enjoys more international legitimacy than Iraq ever did under Saddam Hussein. The motives and goals of Iran's mullah government have been even harder for U.S. intelligence agencies to understand and predict than Saddam Hussein's were. And Iran is deeply involved in America's ongoing predicament in Iraq. Shiites in Iran maintain close cultural and financial contacts with Iraqi Shiite communities on the other side of the nearly 1,000-mile border between the countries. So far Iraq's Shiites have generally been less resistant to the U.S. occupation than its Sunnis. Most American experts believe that if it wanted to, Iran could incite Iraqi Shiites to join the insurgency in far greater numbers.
As a preview of the problems Iran will pose for the next American President, and of the ways in which that President might respond, The Atlantic conducted a war game this fall, simulating preparations for a U.S. assault on Iran.
"War game" is a catchall term used by the military to cover a wide range of exercises. Some games run for weeks and involve real troops maneuvering across oceans or terrain against others playing the role of the enemy force. Some are computerized simulations of aerial, maritime, or land warfare. Others are purely talking-and-thinking processes, in which a group of people in a room try to work out the best solution to a hypothetical crisis. Sometimes participants are told to stay "in role"—to say and do only what a Secretary of State or an Army brigade commander or an enemy strategist would most likely say and do in a given situation. Other times they are told to express their own personal views. What the exercises have in common is the attempt to simulate many aspects of conflict—operational, strategic, diplomatic, emotional, and psychological—without the cost, carnage, and irreversibility of real war. The point of a war game is to learn from simulated mistakes in order to avoid making them if conflict actually occurs.
Our exercise was stripped down to the essentials. It took place in one room, it ran for three hours, and it dealt strictly with how an American President might respond, militarily or otherwise, to Iran's rapid progress toward developing nuclear weapons. It wasn't meant to explore every twist or repercussion of past U.S. actions and future U.S. approaches to Iran. Reports of that nature are proliferating more rapidly than weapons.
Rather, we were looking for what Sam Gardiner, a retired Air Force colonel, has called the "clarifying effect" of intense immersion in simulated decision-making. Such simulations are Gardiner's specialty. For more than two decades he has conducted war games at the National War College and many other military institutions. Starting in 1989, two years before the Gulf War and fourteen years before Operation Iraqi Freedom, he created and ran at least fifty exercises involving an attack on Iraq. The light-force strategy that General Tommy Franks used to take Baghdad last year first surfaced in a war game Gardiner designed in the 1980s. In 2002, as the real invasion of Iraq drew near, Gardiner worked as a private citizen to develop nonclassified simulations of the situation that would follow the fall of Baghdad. These had little effect on U.S. policy, but proved to be prescient about the main challenges in restoring order to Iraq.
Gardiner told me that the war games he has run as a military instructor frequently accomplish as much as several standard lectures or panel discussions do in helping participants think through the implications of their decisions and beliefs. For our purposes he designed an exercise to force attention on the three or four main issues the next President will have to face about Iran, without purporting to answer all the questions the exercise raised.
The scenario he set was an imagined meeting of the "Principals Committee"—that is, the most senior national-security officials of the next Administration. The meeting would occur as soon as either Administration was ready to deal with Iran, but after a November meeting of the IAEA. In the real world the IAEA is in fact meeting in November, and has set a deadline for Iran to satisfy its demands by the time of the meeting. For the purposes of the simulation Iran is assumed to have defied the deadline. That is a safe bet in the real world as well.
And so our group of principals gathered, to provide their best judgment to the President. Each of them had direct experience in making similar decisions. In the role of CIA director was David Kay, who after the Gulf War went to Iraq as the chief nuclear-weapons inspector for the IAEA and the United Nations Special Commission (UNSCOM), and went back in June of 2003 to lead the search for weapons of mass destruction. Kay resigned that post in January of this year, after concluding that there had been no weapons stockpiles at the time of the war.
Playing Secretary of State were Kenneth Pollack, of the Brookings Institution, and Reuel Marc Gerecht, of the American Enterprise Institute. Although neither is active in partisan politics (nor is anyone else who served on the panel), the views they expressed about Iran in our discussion were fairly distinct, with Gerecht playing a more Republican role in the discussions, and Pollack a more Democratic one. (This was the war game's one attempt to allow for different outcomes in the election.)
Both Pollack and Gerecht are veterans of the CIA. Pollack was a CIA Iran-Iraq analyst for seven years, and later served as the National Security Council's director for Persian Gulf affairs during the last two years of the Clinton Administration. In 2002 his book The Threatening Storm: The Case for Invading Iraq was highly influential in warning about the long-term weapons threat posed by Saddam Hussein. (Last January, in this magazine, Pollack examined how pre-war intelligence had gone wrong.) His book about U.S.-Iranian tensions, The Persian Puzzle, has just been published. Gerecht worked for nine years in the CIA's Directorate of Operations, where he recruited agents in the Middle East. In 1997, under the pseudonym Edward Shirley, he published Know Thine Enemy: A Spy's Journey Into Revolutionary Iran, which described a clandestine trip. He has written frequently about Iran, Afghanistan, and the craft of intelligence for this and other publications.
The simulated White House chief of staff was Kenneth Bacon, the chief Pentagon spokesman during much of the Clinton Administration, who is now the head of Refugees International. Before the invasion Bacon was closely involved in preparing for postwar humanitarian needs in Iraq.
Finally, the Secretary of Defense was Michael Mazarr, a professor of national-security strategy at the National War College, who has written about preventing nuclear proliferation in Iran, among other countries, and has collaborated with Gardiner on previous war games.
This war game was loose about requiring players to stay "in role." Sometimes the participants expressed their institutions' views; other times they stepped out of role and spoke for themselves. Gardiner usually sat at the conference table with the five others and served as National Security Adviser, pushing his panel to resolve their disagreements and decide on recommendations for the President. Occasionally he stepped into other roles at a briefing podium. For instance, as the general in charge of Central Command (centcom)—the equivalent of Tommy Franks before the Iraq War and John Abizaid now—he explained detailed military plans.
Over the years Gardiner has concluded that role-playing exercises usually work best if the participants feel they are onstage, being observed; this makes them take everything more seriously and try harder to perform. So the exercise was videotaped, and several people were invited to watch and comment on it. One was Graham Allison, of Harvard University's Kennedy School of Government, a leading scholar of presidential decision-making, who served as a Pentagon official in the first Clinton Administration, specializing in nuclear-arms control. His Essence of Decision, a study of how the Kennedy Administration handled the 1962 Cuban Missile Crisis, is the classic work in its field; his latest book, which includes a discussion of Iran, is Nuclear Terrorism: The Ultimate Preventable Catastrophe. Two other observers were active-duty officers: Marine Corps Colonel Thomas X. Hammes, who has specialized in counterinsurgency and whose book about dealing with Iran (and many other challenges), The Sling and the Stone, was published this summer; and Army Major Donald Vandergriff, whose most recent book, about reforming the internal culture of the Army, is The Path to Victory (2002). The fourth observer was Herbert Striner, formerly of the Brookings Institution, who as a young analyst at an Army think-tank, Operations Research Organization, led a team devising limited-war plans for Iran—back in the 1950s. Striner's team developed scenarios for one other regional war as well: in French Indochina, later known as Vietnam.
Promptly at nine o'clock one Friday morning in September, Gardiner called his group of advisers to order. In his role as National Security Adviser he said that over the next three hours they needed to agree on options and recommendations to send to the President in the face of Iran's latest refusal to meet demands and the latest evidence of its progress toward nuclear weaponry. Gardiner had already decided what questions not to ask. One was whether the United States could tolerate Iran's emergence as a nuclear power. That is, should Iran be likened to Saddam Hussein's Iraq, in whose possession nuclear weapons would pose an unacceptable threat, or to Pakistan, India, or even North Korea, whose nuclear ambitions the United States regrets but has decided to live with for now? If that discussion were to begin, it would leave time for nothing else.
Gardiner also chose to avoid posing directly the main question the game was supposed to illuminate: whether and when the United States should seriously consider military action against Iran. If he started with that question, Gardiner said, any experienced group of officials would tell him to first be sure he had exhausted the diplomatic options. So in order to force discussion about what, exactly, a military "solution" would mean, Gardiner structured the game to determine how the panel assessed evidence of the threat from Iran; whether it was willing to recommend steps that would keep the option of military action open, and what that action might look like; and how it would make the case for a potential military strike to an audience in the United States and around the world.
Before the game began, Gardiner emphasized one other point about his approach, the importance of which would become clear when the discussions were over. He had taken pains to make the material he would present as accurate, realistic, and true to standard national-security practice as possible. None of it was classified, but all of it reflected the most plausible current nonclassified information he could obtain. The detailed plans for an assault on Iran had also been carefully devised. They reflected the present state of Pentagon thinking about the importance of technology, information networks, and Special Forces operations. Afterward participants who had sat through real briefings of this sort said that Gardiner's version was authentic.
His commitment to realism extended to presenting all his information in a series of PowerPoint slides, on which U.S. military planners are so dependent that it is hard to imagine how Dwight Eisenhower pulled off D-Day without them. PowerPoint's imperfections as a deliberative tool are well known. Its formulaic outline structure can overemphasize some ideas or options and conceal others, and the amateurish graphic presentation of data often impedes understanding. But any simulation of a modern military exercise would be unconvincing without it. Gardiner's presentation used PowerPoint for its explanatory function and as a spine for discussion, its best use; several of the slides have been reproduced for this article.
In his first trip to the podium Gardiner introduced himself as the director of central intelligence. (That was David Kay's role too, but during this phase he just sat and listened.) His assignment was to explain what U.S. intelligence knew and didn't know about Iran's progress toward nuclear weapons, and what it thought about possible impediments to that progress—notably Israel's potential to launch a pre-emptive attack on Iran's nuclear sites.
"As DCI, I've got to talk about uncertainty," Gardiner began—the way future intelligence officers presumably will after the Iraq-WMD experience, when George Tenet, as CIA director, claimed that the case for Iraq's having weapons was a "slam-dunk." "It's an important part of this problem. The [intelligence] community believes that Iran could have a nuclear weapon in three years." He let that sink in and then added ominously, "Unless they have something we don't know about, or unless someone has given them or sold them something we don't know about"—or unless, on top of these "known unknowns," some "unknown unknowns" were speeding the pace of Iran's program.
One response to imperfect data about an adversary is to assume the worst and prepare for it, so that any other outcome is a happy surprise. That was the recommendation of Reuel Gerecht, playing the conservative Secretary of State. "We should assume Iran will move as fast as possible," he said several times. "It would be negligent of any American strategic planners to assume a slower pace." But that was not necessarily what the DCI was driving at in underscoring the limits of outside knowledge about Iran. Mainly he meant to emphasize a complication the United States would face in making its decisions. Given Iran's clear intent to build a bomb, and given the progress it has already made, sometime in the next two or three years it will cross a series of "red lines," after which the program will be much harder for outsiders to stop. Gardiner illustrated with a slide (figure 1).
Iran will cross one of the red lines when it produces enough enriched uranium for a bomb, and another when it has weapons in enough places that it would be impossible to remove them in one strike. "Here's the intelligence dilemma," Gardiner said. "We are facing a future in which this is probably Iran's primary national priority. And we have these red lines in front of us, and we"—meaning the intelligence agencies—"won't be able to tell you when they cross them." Hazy knowledge about Iran's nuclear progress doesn't dictate assuming the worst, Gardiner said. But it does mean that time is not on America's side. At some point, relatively soon, Iran will have an arsenal that no outsiders can destroy, and America will not know in advance when that point has arrived.
Then the threat assessment moved to two wild-card factors: Iran's current involvement in Iraq, and Israel's potential involvement with Iran. Both complicate and constrain the options open to the United States, Gardiner said. Iran's influence on the Shiite areas of Iraq is broad, deep, and obviously based on a vastly greater knowledge of the people and customs than the United States can bring to bear. So far Iran has seemed to share America's interest in calming the Shiite areas, rather than have them erupt on its border. But if it needs a way to make trouble for the United States, one is at hand.
As for Israel, no one can be sure what it will do if threatened. Yet from the U.S. perspective, it looks as if a successful pre-emptive raid might be impossible—or at least so risky as to give the most determined Israeli planners pause. Partly this is because of the same lack of knowledge that handicaps the United States. When Menachem Begin dispatched Israeli fighter planes to destroy Iraq's Osirak plant, he knew there was only one target, and that if it was eliminated, Iraq's nuclear program would be set back for many years. In our scenario as in real life, the Americans thought Ariel Sharon and his successors could not be sure how many important targets were in Iran, or exactly where all of them were, or whether Israel could destroy enough of them to make the raid worth the international outrage and the likely counterattack. Plus, operationally it would be hard.
But for the purposes of our scenario, Israel kept up its threats to take unilateral action. It was time again for PowerPoint. Figure 2 shows the known targets that might be involved in some way in Iran's nuclear program. And figure 3 shows the route Israeli warplanes would have to take to get to them. Osirak, near Baghdad, was by comparison practically next door, and the Israeli planes made the round trip without refueling. To get to Iran, Israeli planes would have to fly over Saudi Arabia and Jordan, probably a casus belli in itself given current political conditions; or over Turkey, also a problem; or over American-controlled Iraq, which would require (and signal) U.S. approval of the mission.
With this the DCI left the podium—and Sam Gardiner, now sitting at the table as National Security Adviser, asked what initial assessments the principals made of the Iranian threat.
On one point there was concord. Despite Gardiner's emphasis on the tentative nature of the intelligence, the principals said it was sufficient to demonstrate the gravity of the threat. David Kay, a real-life nuclear inspector who was now the DCI at the table, said that comparisons with Iraq were important—and underscored how difficult the Iranian problem would be. "It needs to be emphasized," he said, "that the bases for conclusions about Iran are different, and we think stronger than they were with regard to Iraq." He explained that international inspectors withdrew from Iraq in 1998, so outsiders had suspicions rather than hard knowledge about what was happening. In Iran inspectors had been present throughout, and had seen evidence of the "clandestine and very difficult-to-penetrate nature of the program," which "leaves no doubt that it is designed for a nuclear-weapons program." What is worse, he said, "this is a lot more dangerous than the Iraqi program, in that the Iranians have proven, demonstrated connections with very vicious international terrorist regimes who have shown their willingness to use any weapons they acquire" against the United States and its allies. Others spoke in the same vein.
The real debate concerned Israel. The less America worried about reaction from Europe and the Muslim world, the more likely it was to encourage or condone Israeli action, in the hope that Israel could solve the problem on its own. The more it worried about long-term relations with the Arab world, the more determined it would be to discourage the Israelis from acting.
Most of the principals thought the Israelis were bluffing, and that their real goal was to put pressure on the United States to act. "It's hard to fault them for making this threat," said Pollack, as the Democratic Secretary of State, "because in the absence of Israeli pressure how seriously would the United States be considering this option? Based on my discussions with the Israelis, I think they know they don't have the technical expertise to deal with this problem. I think they know they just don't have the planes to get there—setting aside every other problem."
"They might be able to get there—the problem would be getting home," retorted Gerecht, who had the most positive view on the usefulness of an Israeli strike.
Bacon, as White House chief of staff, said, "Unless they can take out every single Iranian missile, they know they will get a relatively swift counterattack, perhaps with chemical weapons. So the threat they want to eliminate won't be eliminated." Both he and Pollack recommended that the Administration ask the Israelis to pipe down.
"There are two things we've got to remember with regard to the Israelis," Kay said. "First of all, if we tell them anything, they are certain to play it back through their network that we are 'bringing pressure to bear' on them. That has been a traditional Israeli response. It's the nature of a free democracy that they will do that. The second thing we've got to be careful of and might talk to the Israelis about: our intelligence estimate that we have three years to operate could change if the Iranians thought the Israelis might pre-empt sooner. We'd like to have that full three years, if not more. So when we're talking with the Israelis, toning down their rhetoric can be described as a means of dealing with the threat."
Woven in and out of this discussion was a parallel consideration of Iraq: whether, and how, Iran might undermine America's interests there or target its troops. Pollack said this was of great concern. "We have an enormous commitment to Iraq, and we can't afford to allow Iraq to fail," he said. "One of the interesting things that I'm going to ask the CentCom commander when we hear his presentation is, Can he maintain even the current level of security in Iraq, which of course is absolutely dismal, and still have the troops available for anything in Iran?" As it happened, the question never came up in just this form in the stage of the game that featured a simulated centcom commander. But Pollack's concern about the strain on U.S. military resources was shared by the other panelists. "The second side of the problem," Pollack continued, "is that one of the things we have going for us in Iraq, if I can use that term, is that the Iranians really have not made a major effort to thwart us … If they wanted to make our lives rough in Iraq, they could make Iraq hell." Provoking Iran in any way, therefore, could mean even fewer troops to handle Iraq—and even worse problems for them to deal with.
Kay agreed. "They may decide that a bloody defeat for the United States, even if it means chaos in Iraq, is something they actually would prefer. Iranians are a terribly strategic political culture … They might well accelerate their destabilization operation, in the belief that their best reply to us is to ensure that we have to go to helicopters and evacuate the Green Zone."
More views were heard—Gerecht commented, for example, on the impossibility of knowing the real intentions of the Iranian government—before Gardiner called a halt to this first phase of the exercise. He asked for a vote on one specific recommendation to the President: Should the United States encourage or discourage Israel in its threat to strike? The Secretary of Defense, the DCI, the White House chief of staff, and Secretary of State Pollack urged strong pressure on Israel to back off. "The threat of Israeli military action both harms us and harms our ability to get others to take courses of action that might indeed affect the Iranians," Kay said. "Every time a European hears that the Israelis are planning an Osirak-type action, it makes it harder to get their cooperation." Secretary of State Gerecht thought a successful attack was probably beyond Israel's technical capability, but that the United States should not publicly criticize or disagree with its best ally in the Middle East.
Sam Gardiner took the podium again. Now he was four-star General Gardiner, commander of CentCom. The President wanted to understand the options he actually had for a military approach to Iran. The general and his staff had prepared plans for three escalating levels of involvement: a punitive raid against key Revolutionary Guard units, to retaliate for Iranian actions elsewhere, most likely in Iraq; a pre-emptive air strike on possible nuclear facilities; and a "regime change" operation, involving the forcible removal of the mullahs' government in Tehran. Either of the first two could be done on its own, but the third would require the first two as preparatory steps. In the real world the second option—a pre-emptive air strike against Iranian nuclear sites—is the one most often discussed. Gardiner said that in his briefing as war-game leader he would present versions of all three plans based as closely as possible on current military thinking. He would then ask the principals to recommend not that an attack be launched but that the President authorize the preparatory steps to make all three possible.
The first option was straightforward and, according to Gardiner, low-risk. The United States knew where the Revolutionary Guard units were, and it knew how to attack them. "We will use Stealth airplanes, U.S.-based B-2 bombers, and cruise missiles to attack," Gardiner said. "We could do this in one night." These strikes on military units would not in themselves do anything about Iran's nuclear program. Gardiner mentioned them because they would be a necessary first step in laying the groundwork for the ultimate scenario of forced regime change, and because they would offer the United States a "measured" retaliatory option if Iran were proved to be encouraging disorder in Iraq.
The pre-emptive air strike was the same one that had been deemed too demanding for the Israelis. The general's staff had identified 300 "aim points" in Iran. Some 125 of them were sites thought to be involved in producing nuclear, chemical, or biological weapons. The rest were part of Iran's air-defense or command system. "I call this a low-risk option also," Gardiner said, speaking for CentCom. "I'm not doing that as political risk—that's your job. I mean it's a low-risk military option." Gardiner said this plan would start with an attack on air-defense sites and would take five days in all.
Then there was option No. 3. Gardiner called this plan "moderate risk," but said the best judgment of the military was that it would succeed. To explain it he spent thirty minutes presenting the very sorts of slides most likely to impress civilians: those with sweeping arrows indicating the rapid movement of men across terrain. (When the exercise was over, I told David Kay that an observer who had not often seen such charts remarked on how "cool" they looked. "Yes, and the longer you've been around, the more you learn to be skeptical of the 'cool' factor in PowerPoint," Kay said. "I don't think the President had seen many charts like that before," he added, referring to President Bush as he reviewed war plans for Iraq.)
The overall plan of attack was this: a "deception" effort from the south, to distract Iranian troops; a main-force assault across the long border with Iraq; airborne and Special Forces attacks from Afghanistan and Azerbaijan; and cruise missiles from ships at sea. Gardiner presented more-detailed possibilities for the deployment. A relatively light assault, like the one on Afghanistan, is depicted in figure 4. A "heavier" assault would involve more troops and machines attacking across two main fronts (figure 5).
In all their variety, these and other regime-change plans he described had two factors in common. One is that they minimized "stability" efforts—everything that would happen after the capital fell. "We want to take out of this operation what has caused us problems in Iraq," Gardiner of CentCom said, referring to the postwar morass. "The idea is to give the President an option that he can execute that will involve about twenty days of buildup that will probably not be seen by the world. Thirty days of operation to regime change and taking down the nuclear system, and little or no stability operations. Our objective is to be on the outskirts of Tehran in about two weeks. The notion is we will not have a Battle of Tehran; we don't want to do that. We want to have a battle around the city. We want to bring our combat power to the vicinity of Tehran and use Special Operations to take the targets inside the capital. We have no intention of getting bogged down in stability operations in Iran afterwards. Go in quickly, change the regime, find a replacement, and get out quickly after having destroyed—rendered inoperative—the nuclear facilities." How could the military dare suggest such a plan, after the disastrous consequences of ignoring "stability" responsibilities in Iraq? Even now, Gardiner said after the war game, the military sees post-conflict operations as peripheral to its duties. If these jobs need to be done, someone else must take responsibility for them.
The other common factor was the need for troops, machinery, and weapons to be nearby and ready to move. Positioning troops would not be that big a problem. When one unit was replacing another in Iraq, for a while both units would be in place, and the attack could happen then. But getting enough machinery into place was more complicated, because airfields in nearby Georgia and Azerbaijan are too small to handle a large flow of military cargo planes (figure 6).
As centcom commander, Gardiner cautioned that any of the measures against Iran would carry strategic risks. The two major dangers were that Iran would use its influence to inflame anti-American violence in Iraq, and that it would use its leverage to jack up oil prices, hurting America's economy and the world's. In this sense option No. 2—the pre-emptive air raid—would pose as much risk as the full assault, he said. In either case the Iranian regime would conclude that America was bent on its destruction, and it would have no reason to hold back on any tool of retaliation it could find. "The region is like a mobile," he said. "Once an element is set in motion, it is impossible to say where the whole thing will come to rest." But the President had asked for a full range of military options, and unless his closest advisers were willing to go to him empty-handed, they needed to approve the steps that would keep all the possibilities alive. That meant authorizing the Department of Defense to begin expanding airfields, mainly in Azerbaijan, and to dedicate $700 million to that purpose. (As it happens, this is the same amount Tommy Franks requested in July of 2002, to keep open the possibility of war in Iraq.) "This is not about executing the plan," Gardiner of centcom said. "We're preparing options for the President; the whole issue of execution is separate. We need some money to build facilities."
Gardiner remained at the podium to answer questions as the CentCom commander, and the discussion began. The panelists skipped immediately to the regime-change option, and about it there was unanimity: the plan had been modeled carefully on the real assault on Iraq, and all five advisers were appalled by it.
"You need to take this back to Tampa," David Kay said, to open the discussion. Tampa, of course, is the headquarters for CentCom units operating in Iraq and Afghanistan. "Or put it someplace else I'd suggest, but we're in public." What was remarkable about the briefing, he said, was all the charts that were not there. "What were the countermoves?" he asked. "The military countermoves—not the political ones you offloaded to my Secretaries of State but the obvious military countermoves that the Iranians have? A very easy military counter is to raise the cost of your military operation inside Iraq. Are you prepared to do that?"
The deeper problem, Kay said, lay with the request for money to "keep options open." "That, quite frankly, is a bunch of bullshit," he said. "Approval of the further planning process forecloses a number of options immediately. I would love to see a strategic communications plan that would allow us to continue diplomatic and other options immediately with our European allies when this leaks; inevitably this will leak."
The next twenty minutes of discussion was to the same effect. Who, exactly, would succeed the mullahs in command? How on earth would U.S. troops get out as quickly as they had come in? "Speaking as the President's chief of staff, I think you are doing the President an enormous disservice," Kenneth Bacon said. "One, it will leak. Two, it will be politically and diplomatically disastrous when it leaks … I think your invasion plan is a dangerous plan even to have on the table in the position of being leaked … I would throw it in Tampa Bay and hope the sharks would eat it."
"This is a paranoid regime," Kenneth Pollack said of Iran. "Even if the development of the Caucasus airfields … even if it weren't about them, they would assume it was about them. So that in and of itself will likely provoke a response. The Iranians are not inert targets! If they started to think we were moving in the direction of a military move against them, they would start fighting us right away."
Michael Mazarr, as Secretary of Defense, said he did not want the authority that was on offer to his department. "Tell the President my personal judgment would be the only circumstances in which we could possibly consider launching any significant operation in Iran would be the most extreme provocation, the most imminent threat," he said.
Even the hardest-liner, Reuel Gerecht, was critical. "I would agree that our problems with the Islamic republic will not be over until the regime is changed," he said. If the United States could launch a genuine surprise attack—suddenly, from aircraft carriers, rather than after a months-long buildup of surrounding airfields—he would look at it favorably. But on practical grounds, he said, "I would vote against the regime-change options displayed here."
Further unhappy back-and-forth ensued, with the CentCom commander defending the importance of keeping all options open, and the principals warning of trouble when news of the plan got out. When Gardiner called an end to this segment, there was little objection to the most modest of the military proposals—being ready, if need be, for a punitive strike on the Revolutionary Guards. The participants touched only briefly on the Osirak-style strike during the war game, but afterward most of them expressed doubt about its feasibility. The United States simply knew too little about which nuclear projects were under way and where they could be destroyed with confidence. If it launched an attack and removed some unknown proportion of the facilities, the United States might retard Iran's progress by an unknown number of months or years—at the cost of inviting all-out Iranian retaliation. "Pre-emption is only a tactic that puts off the nuclear development," Gardiner said after the exercise. "It cannot make it go away. Since our intelligence is so limited, we won't even know what we achieved after an attack. If we set it back a year, what do we do a year later? A pre-emptive strike would carry low military risk but high strategic risk."
During the war game the regime-change plan got five nays. But it was clear to all that several other big issues lay on the table, unresolved. How could the President effectively negotiate with the Iranians if his own advisers concluded that he had no good military option to use as a threat? How could the world's most powerful and sophisticated military lack the ability to take an opponent by surprise? How could leaders of that military imagine, after Iraq, that they could ever again propose a "quick in-and-out" battle plan? Why was it so hard to develop plans that allowed for the possibility that an adversary would be clever and ruthless? Why was it so hard for the United States to predict the actions and vulnerabilities of a regime it had opposed for twenty-five years?
At noon the war game ended. As a simulation it had produced recommendations that the President send a go-slow signal to the Israelis and that he not authorize any work on airfields in Central Asia. His advisers recommended that he not even be shown Centcom's plans for invading Iran.
The three hours of this exercise were obviously not enough time for the panel of advisers to decide on all aspects of a new policy toward Iran. But the intended purpose of the exercise was to highlight the real options a real President might consider. What did it reveal? Gardiner called for a wrap-up from participants and observers immediately after the event. From their comments, plus interviews with the participants in the following week, three big themes emerged: the exercise demonstrated something about Iraq, something about the way governments make decisions, and something about Iran.
Iraq was a foreground topic throughout the game, since it was where a threatened Iran might most easily retaliate. It was even more powerful in its background role. Every aspect of discussion about Iran was colored by knowledge of how similar decisions had played out in Iraq. What the United States knew and didn't know about secret weapons projects. What could go wrong with its military plans. How much difficulty it might face in even a medium-size country. "Compared with Iraq, Iran has three times the population, four times the land area, and five times the problems," Kenneth Pollack said during the war game. A similar calculation could be heard in almost every discussion among the principals, including those who had strongly supported the war in Iraq. This was most obvious in the dismissal of the full-scale regime-change plan—which, Gardiner emphasized, was a reflection of real-life military thinking, not a straw man. "I have been working on these options for almost eighteen months," he said later. "I tried them in class with my military students. They were the best I could do. I was looking for a concept that would limit our involvement in stability operations. We just don't have the forces to do that in Iran. The two lesser concepts"—punitive raids on the Revolutionary Guard and pre-emptive air strikes—"were really quite good from a military perspective." And of course the sweeping third concept, in the very similar form of Tommy Franks's plan, had been approved by a real President without the cautionary example of Iraq to learn from.
Exactly what learning from Iraq will mean is important but impossible to say. "Iraq" could become shorthand for a comprehensive disaster—one of intention, execution, and effect. "Usually we don't make the same mistakes immediately," Graham Allison said. "We make different mistakes." In an attempt to avoid "another Iraq," in Iran or elsewhere, a different Administration would no doubt make new mistakes. If George Bush is re-elected, the lessons of Iraq in his second term will depend crucially on who is there to heed them. All second-term Presidents have the same problem, "which is that the top guys are tired out and leave—or tired out and stay," Kay said. "You get the second-best and the second-brightest, it's really true." "There will be new people, and even the old ones will behave differently," Gardiner said. "The CIA will not make unequivocal statements. There will be more effort by everyone to question plans." But Kay said that the signal traits of the George W. Bush Administration—a small group of key decision-makers, no fundamental challenge of prevailing views—would most likely persist. "I have come to the conclusion that it is a function of the way the President thinks, operates, declares his policy ahead of time," Kay said. "It is inherent in the nature of George Bush, and therefore inherent in the system."
What went wrong in Iraq, according to our participants, can in almost all cases be traced back to the way the Administration made decisions. "Most people with detailed knowledge of Iraq, from the CIA to the State Department to the Brits, thought it was a crazy quilt held together in an artificial state," Allison said. Because no such people were involved in the decision to go to war, the Administration expected a much easier reception than it met—with ruinous consequences. There was no strong institutional system for reconciling differences between the Pentagon, the State Department, the CIA, and other institutions, and the person who theoretically might have done this, Condoleezza Rice, was weak. "If you don't have a deliberate process in which the National Security Adviser is playing a strong role, clarifying contrary views, and hammering out points of difference, you have the situation you did," Allison said. "There was no analytic memo that all the parties looked at that said, 'Here's how we see the shape of this problem; here is the logic that leads to targeting Iraq rather than North Korea.'"
"Process" sounds dull, and even worse is "government decision-making," but these topics provoked the most impassioned comments from panelists and observers when they were interviewed after the war game. All were alarmed about the way governments now make life-and-death decisions; this was, after Iraq, the second big message of the exercise.
"Companies deciding which kind of toothpaste to market have much more rigorous, established decision-making processes to refer to than the most senior officials of the U.S. government deciding whether or not to go to war," Michael Mazarr said. "On average, the national-security apparatus of the United States makes decisions far less rigorously than it ought to, and is capable of. The Bush Administration is more instinctual, more small-group-driven, less concerned about being sure they have covered every assumption, than other recent Administrations, particularly that of George H. W. Bush. But the problem is bigger than one Administration or set of decision-makers."
Gardiner pointed out how rare it is for political leaders to ask, "And what comes after that? And then?" Thomas Hammes, the Marine expert in counterinsurgency, said that presentations by military planners feed this weakness in their civilian superiors, by assuming that the adversary will cooperate. "We never 'red-celled' the enemy in this exercise" (that is, let him have the first move), Hammes said after the Iran war game. "What if they try to pre-empt us? What if we threaten them, and the next day we find mines in Baltimore Harbor and the Golden Gate, with a warning that there will be more? Do we want to start this game?" Such a failure of imagination—which Hammes said is common in military-run war games—has a profound effect, because it leads to war plans like the ones from Gardiner's CentCom, or from Tommy Franks, which in turn lull Presidents into false confidence. "There is no such thing as a quick, clean war," he said. "War will always take you in directions different from what you intended. The only guy in recent history who started a war and got what he intended was Bismarck," who achieved the unification of Germany after several European wars.
Gardiner pointed out that none of the principals had even bothered to ask whether Congress would play a part in the decision to go to war. "This game was consistent with a pattern I have been seeing in games for the past ten years," he said. "It is not the fault of the military, but they have learned to move faster than democracy was meant to move."
And what did the exercise show about Iran? In the week after the war game I interviewed the partici- pants about the views they had expressed "in role" and about their personal recommendations for the next President's approach. From these conversations, and from the participants' other writings and statements about Iran, the following themes emerged.
About Iran's intentions there is no disagreement. Iran is trying to develop nuclear weapons, and unless its policy is changed by the incentives it is offered or the warnings it receives, it will succeed.
About America's military options there is almost as clear a view. In circumstances of all-out war the United States could mount an invasion of Iran if it had to. If sufficiently provoked—by evidence that Iran was involved in a terrorist incident, for example, or that it was fomenting violence in Iraq—the United States could probably be effective with a punitive bomb-and-missile attack on Revolutionary Guard units.
But for the purposes most likely to interest the next American President—that is, as a tool to slow or stop Iran's progress toward nuclear weaponry—the available military options are likely to fail in the long term. A full-scale "regime change" operation has both obvious and hidden risks. The obvious ones are that the United States lacks enough manpower and equipment to take on Iran while still tied down in Iraq, and that domestic and international objections would be enormous. The most important hidden problem, exposed in the war-game discussions, was that a full assault would require such drawn-out preparations that the Iranian government would know months in advance what was coming. Its leaders would have every incentive to strike pre-emptively in their own defense. Unlike Saddam Hussein's Iraq, a threatened Iran would have many ways to harm America and its interests. Apart from cross-border disruptions in Iraq, it might form an outright alliance with al-Qaeda to support major new attacks within the United States. It could work with other oil producers to punish America economically. It could, as Hammes warned, apply the logic of "asymmetric," or "fourth-generation," warfare, in which a superficially weak adversary avoids a direct challenge to U.S. military power and instead strikes the most vulnerable points in American civilian society, as al-Qaeda did on 9/11. If it thought that the U.S. goal was to install a wholly new regime rather than to change the current regime's behavior, it would have no incentive for restraint.
What about a pre-emptive strike of our own, like the Osirak raid? The problem is that Iran's nuclear program is now much more advanced than Iraq's was at the time of the raid. Already the U.S. government has no way of knowing exactly how many sites Iran has, or how many it would be able to destroy, or how much time it would buy in doing so. Worse, it would have no way of predicting the long-term strategic impact of such a strike. A strike might delay by three years Iran's attainment of its goal—but at the cost of further embittering the regime and its people. Iran's intentions when it did get the bomb would be all the more hostile.
Here the United States faces what the military refers to as a "branches and sequels" decision—that is, an assessment of best and second-best outcomes. It would prefer that Iran never obtain nuclear weapons. But if Iran does, America would like Iran to see itself more or less as India does—as a regional power whose nuclear status symbolizes its strength relative to regional rivals, but whose very attainment of this position makes it more committed to defending the status quo. The United States would prefer, of course, that Iran not reach a new level of power with a vendetta against America. One of our panelists thought that a strike would help the United States, simply by buying time. The rest disagreed. Iran would rebuild after a strike, and from that point on it would be much more reluctant to be talked or bargained out of pursuing its goals—and it would have far more reason, once armed, to use nuclear weapons to America's detriment.
Most of our panelists felt that the case against a U.S. strike was all the more powerful against an Israeli strike. With its much smaller air force and much more limited freedom to use airspace, Israel would probably do even less "helpful" damage to Iranian sites. The hostile reaction—against both Israel and the United States—would be potentially more lethal to both Israel and its strongest backer.
A realistic awareness of these constraints will put the next President in an awkward position. In the end, according to our panelists, he should understand that he cannot prudently order an attack on Iran. But his chances of negotiating his way out of the situation will be greater if the Iranians don't know that. He will have to brandish the threat of a possible attack while offering the incentive of economic and diplomatic favors should Iran abandon its plans. "If you say there is no acceptable military option, then you end any possibility that there will be a non-nuclear Iran," David Kay said after the war game. "If the Iranians believe they will not suffer any harm, they will go right ahead." Hammes agreed: "The threat is always an important part of the negotiating process. But you want to fool the enemy, not fool yourself. You can't delude yourself into thinking you can do something you can't." Is it therefore irresponsible to say in public, as our participants did and we do here, that the United States has no military solution to the Iran problem? Hammes said no. Iran could not be sure that an American President, seeing what he considered to be clear provocation, would not strike. "You can never assume that just because a government knows something is unviable, it won't go ahead and do it. The Iraqis knew it was not viable to invade Iran, but they still did it. History shows that countries make very serious mistakes."
So this is how the war game turned out: with a finding that the next American President must, through bluff and patience, change the actions of a government whose motives he does not understand well, and over which his influence is limited. "After all this effort, I am left with two simple sentences for policymakers," Sam Gardiner said of his exercise. "You have no military solution for the issues of Iran. And you have to make diplomacy work."