LATE ONE NIGHT in February 2013, I climbed into an unmarked U.S. government Gulfstream jet parked on the deserted tarmac at Andrews Air Force Base. Secretary Kerry’s parting words, delivered with his characteristic optimism and self-assurance, still rang in my head: “We’ve got the diplomatic opportunity of a lifetime.” I felt far more uncertain.
I spent much of that seventeen-hour flight to Oman reviewing briefing books, talking through strategy and tactics with our negotiating team, and trying to come to grips with the task before us. It had been thirty-five years since the United States and Iran had had sustained diplomatic contact. There was baggage on both sides, and massive mutual mistrust. The diplomatic stakes were high, with Iran’s nuclear program accelerating and military conflict between us an increasing possibility. The politics in both our capitals were explosive, with little room for diplomatic maneuver. International diplomacy had run aground, its thus far desultory exchanges missing a key ingredient—a direct discussion between the two principal protagonists, the United States and Iran.
For all the anxiety, it was also hard not to feel a sense of possibility. Here was a chance to do what diplomats spend their whole careers trying to do. Here was a chance to apply tough-minded diplomacy, backed up by the economic leverage of sanctions, the political leverage of an international consensus, and the military leverage of the potential use of force. And here was a chance to demonstrate the promise of American diplomacy after a decade of America at war.
IRAN HUNG OVER much of my career, a country synonymous in American foreign policy terms with troubles, threats, and blunders. Iran seemed a menacing and impenetrable presence, too big and dangerous to ignore, but too intransigent to engage. It was a minefield for diplomats, and nobody had a good map.
I took the Foreign Service entrance exam in November 1979, a few days after the seizure of our embassy in Tehran and the beginning of a hostage crisis that brought down a president. Iranian-backed terrorists twice bombed our embassy in Beirut, and killed more than two hundred Marines in another attack there. The Iran-Contra scandal nearly brought down a second president.
The sweeping success of Desert Storm in 1991 propelled American influence in the Middle East to its zenith. The Clinton administration worked hard to contain Iran, but also explored in the late 1990s a possible opening with the Khatami government. It never got very far. The post–9/11 landscape offered a similar opportunity, which we never seized. Instead, the U.S.-led overthrow of Iran’s bitter historical adversaries in Kabul and Baghdad, and the chaos that ensued, delivered Iran a strategic opening that it was only too pleased to exploit.
In late 2001, the U.S. intelligence community began to track two clandestine nuclear sites in Iran: a uranium enrichment plant at Natanz and a facility in Arak that could eventually produce weapons-grade plutonium. These efforts, undeclared to the IAEA, built on Iran’s overt civilian nuclear energy program, which began during the shah’s time—ironically, with the initial support of the United States.
The revelation of the covert sites in the summer of 2002 set off a diplomatic dance that continued for the next several years. The UN Security Council passed resolutions demanding that Iran suspend its enrichment work. Iran instead plowed stubbornly ahead. Given the unwillingness of the Bush administration to engage directly with Iran, our European allies (the United Kingdom, France, and Germany, or the “EU-3”) began a negotiation with the Iranians that showed fitful progress, as Tehran sought both to preserve its enrichment program and the long-term possibility of weaponization and at the same time avoid economic sanctions. Russia and China later joined the EU-3, which was eventually rebranded as the P5+1 (the five permanent members of the UN Security Council plus Germany).
This all boosted the market for international diplomatic acronyms, but didn’t make much of a dent in Iranian behavior. By the last year of the Bush administration in 2008, despite the imposition of several rounds of UN sanctions against Iran and growing international concern, the Iranians had accumulated half the amount of low-enriched uranium they would need to enrich further and make a single bomb. They were spinning more than four thousand primitive IR-1 centrifuges at Natanz, and were making halting progress toward more sophisticated models.
While the American intelligence community concluded famously in 2007 that the Iranian leadership had suspended its weapons work back in 2003, the fact that they were clearly determined to keep their options open in the face of mounting international pressure was deeply troubling. An unconstrained Iranian nuclear program or a regime clearly bent on a weapons program would add yet another layer of risk and fragility to an already unstable region. Our friends—the Gulf Arab states and, especially, Israel—had to take that threat seriously.
As the Bush administration grappled with the damage done by the Iraq War, some of its senior figures began to recognize that its stubborn insistence on not engaging directly in P5+1 diplomacy with Iran had become counterproductive. An early probe for direct talks in May 2003, orchestrated by the enterprising and well-intentioned Swiss ambassador in Tehran, was never pursued. I was traveling in the region when Tim Guldimann, about to complete his ambassadorial tour in Iran, came to Washington and met with my deputy, Jim Larocco, to present a short paper that he insisted had been drafted in cooperation with Iran’s ambassador in Paris, the nephew of Iranian foreign minister Kamal Kharazi. Guldimann said the whole effort had been sanctioned at high levels of the Iranian government. The document itself was intriguing, offering a wildly ambitious dialogue across the whole range of U.S.-Iranian differences. Jim and my other NEA colleagues pressed Guldimann hard on who exactly in Tehran had endorsed the paper, and how explicitly that was conveyed. Guldimann was too vague for Jim’s taste. The tangled history of ill-sourced messages and double-dealing cast a shadow on our deliberations.
We conveyed the document and an account of Jim’s conversation to Secretary Powell and Deputy Secretary Armitage—noting our doubts that it bore the stamp of the highest level of Iran’s leadership, but recommending that we test the proposition and reopen the contacts with Iran that had been suspended a year earlier. Powell and Armitage agreed. But in the heady immediate aftermath of the invasion of Iraq, there was little White House interest in talking to a charter member of the “axis of evil,” and a conviction that direct engagement would be a reward for bad behavior. For Vice President Cheney and hardliners in the administration, the calculus was clear: If the Iranians were worried about being next on the American hit parade after Saddam, it wasn’t a bad idea to let them stew a little.
Throughout the remaining two years of my time in NEA we continued to make the case for dialogue with Iran. I repeated the argument in my December 2004 transition memo to Secretary Rice. I also added a proposal—which was adopted—to restart a serious program of Persian-language training for a small cadre of American diplomats, and then station them in several posts on Iran’s periphery to develop expertise and prepare for an eventual resumption of contacts. The “Iran Watchers” initiative had as its inspiration what we had done more than seven decades before in preparing Russian-language specialists for eventual reopening of diplomatic relations with the Soviet Union.
By the time I returned to Washington from Moscow in the spring of 2008, the mood had begun to shift a little. Chastened by the postwar mess in Iraq, President Bush had replaced Don Rumsfeld with Bob Gates at the Pentagon. The vice president’s hawkish views were less dominant, and Rice was pressing on several fronts for more active American diplomacy.
In late May 2008, I sent Secretary Rice a long memo entitled “Regaining the Strategic Initiative on Iran.” I began by arguing that “our Iran policy is drifting dangerously between the current muddle of P5+1 diplomacy and more forceful options, with all of their huge downsides.” Our unwillingness to engage directly with Tehran was costing us more than the Iranians, and deprived us of valuable leverage. “The regime has constructed a narrative which portrays Iran as the victim of implacable American hostility,” I wrote, “increasingly gaining the diplomatic upper hand regionally and globally, with the American administration—not Iran—increasingly the isolated party. Reviving significant pressure against Iran’s nuclear program requires us to puncture that narrative.”1
I had two practical suggestions. First, it was long past time for the United States to join our European, Russian, and Chinese partners at the negotiating table. I had few illusions that the government of Mahmoud Ahmadinejad, let alone the deeply suspicious Supreme Leader, was ready to negotiate seriously. By not engaging, we were giving them an easy out—allowing them to hide behind the pretext that they couldn’t really be sure about P5+1 proposals, because the Americans weren’t there to back them up. Our physical presence would put us on the high ground, put the Iranians on the defensive, strengthen solidarity with our negotiating partners, and better position us to pivot to more sanctions if Tehran balked again.
My second idea would revive an initiative that had already been kicked around at lower levels in the administration. I suggested to Rice that we should propose quietly to the Iranians that we staff our interests section in Tehran with a few American diplomats, revising the arrangement that had been in effect since the assault on our embassy in Tehran in 1979, under which the Swiss represented our interests in Iran. We would reciprocate by allowing the Iranians to staff their interests section in Washington, managed by the Pakistani government, with a handful of Iranian diplomats. Like the argument for joining the P5+1 talks, the focus was on tactical advantages. I had little expectation that the Supreme Leader would actually agree to such a proposal. The last thing he wanted to see was a long line of Iranian visa applicants around a U.S. diplomatic facility in Tehran staffed by Americans; for the Ayatollah Khamenei, this would be the ultimate Trojan horse. The proposal, which would inevitably become public, would only further cement our grip on the high ground. I suggested that we pitch the idea to the Iranians through the Russians, who had good high-level channels in Tehran and whose support would be crucial if we had to go back to the UN Security Council for tougher sanctions.
I concluded with a broad argument, echoing the classic containment concept that Rice knew so well as a recovering Sovietologist. In dealing with a profoundly hostile adversary beset by its own serious internal contradictions, I said, “a successful strategy will require calculated risk-taking on our part…with the same combination of multiple pressure points, diplomatic coalition-building, wedge-driving among Iran and its uneasy partners, and selected contacts with the regime that animated much of Kennan’s concept.” Moreover, we should simultaneously explore “creatively subversive ways to accentuate the gap between the regime’s deeply conservative instincts and popular Iranian desire for normalization with the rest of the world, including the U.S.”2
Rice saw the possibilities immediately, and knew that we needed to inject some new American initiative into nuclear diplomacy with Iran. The proximate opportunity in the talks themselves was the presentation by Javier Solana, the de facto European Union foreign minister, of a renewed P5+1 proposal to Iran. The essence of his proposal was a freeze on Iranian nuclear activities, including enrichment, and a reciprocal freeze on new UN Security Council sanctions, which would allow space for negotiations on a comprehensive nuclear deal. Solana conveyed this plan in Tehran in June 2008, and the Iranians pledged to respond at a follow-up meeting in Geneva in July. Rice decided to seek the president’s approval for me to attend the Geneva meeting—and to also get his blessing on the interests section idea.
One morning in early July, I rode over with the secretary to one of her regular sessions with President Bush to make our pitch. Now nearing the end of his tenure, he looked a little grayer, but his decency and good humor were undiminished.
“Burnsie,” he said with a familiar smile as I walked in behind the secretary, “it’s good to have you back in Washington.” Vice President Cheney sat in an armchair next to the president, less visibly enthused about my homecoming. I joined Rice on a couch alongside Bush, and she quickly laid out our case. The president asked a couple questions about how the interests section proposal would work, and expressed skepticism about what impact joining the Geneva talks with the Iranians would have on their behavior, but saw the value of trying both. The vice president started to object, arguing that we shouldn’t reward the Iranians by appearing at a meeting. Bush cut him off. “Dick,” he said with a wave of his hand, “I’m okay with this, and I’ve made up my mind.” A lot had changed since the first term and the run-up to the Iraq War, I thought to myself. Diplomacy had its uses after all.
On July 19 in Geneva, amid massive media attention, I broke the taboo on direct American participation in the nuclear talks. I joined my P5+1 colleagues around an oblong table in a cramped meeting hall in the old city. I had been reminded by the secretary and Steve Hadley to keep my game face on and look appropriately sober while the cameras filmed the opening of the first session. They had also both suggested that it might be best to remain silent during the talks and simply witness the Iranian reply to Solana. The first point made sense. The second did not. If the purpose of joining the talks was to emphasize our seriousness and tag Iran as the diplomatic problem child, the silent treatment would backfire. Looking across the table directly at Saeed Jalili, the head of the Iranian delegation, I made a simple statement. I said that I hoped the Iranians understood the significance of the signal we were sending by joining the talks. We knew what was at stake on the nuclear issue; we were determined to prevent Iran from developing a bomb and to hold it to its international obligations; and we were firmly behind the P5+1 proposal. I emphasized that Iran had a rare opportunity before it; we could only hope that it would take advantage of it.
Jalili took careful notes, and smiled faintly throughout. I got lots of sidelong glances from him and his colleagues, who seemed to find the American presence unnerving. Jalili then embarked on nearly forty minutes of meandering philosophizing about Iran’s culture and history, and the constructive role it could play in the region. He could be stupefyingly opaque when he wanted to avoid straight answers, and this was certainly one of those occasions. He mentioned at one point that he still lectured part-time at Tehran University. I didn’t envy his students.
Jalili wound up his comments by handing over an Iranian “non-paper.” The English version was mistakenly headed “None Paper,” which turned out to be an apt description of its substance. Solana and the rest of us looked at it quickly, at which point my French colleague helpfully groaned and muttered, “Bullshit,” which caused Jalili to look somewhat startled—and me to lose my game face. Fortunately, the cameras were long gone.
In a quick note to Secretary Rice that evening, I reported that “five and a half hours with the Iranians today were a vivid reminder that we may not have been missing all that much over the years.” Nevertheless, our P5+1 colleagues were delighted that the United States was now visible and engaged. The Russians and Chinese seemed particularly impressed. However disappointing the Iranian response, we were back on the high ground.3
Neither joining the Geneva meeting nor the interests section initiative produced any substantive breakthroughs as the Bush administration came to an end. I joined Rice for a quiet meeting with Sergey Lavrov in Berlin later in July, and pitched the interests section idea. Lavrov agreed readily that Russia would convey it to Ali Akbar Velayati, the Supreme Leader’s foreign policy advisor. But then the war in Georgia intervened, the Russians lost interest in being the messenger, we lost interest in the Russians, and the idea never went any further. We had, however, laid some of the groundwork for Barack Obama’s much more active and imaginative approach to the Iranian nuclear dilemma.
AS HE MADE clear during his campaign for president, Obama sought a mandate to wind down America’s wars in the Middle East and to make diplomacy the tool of first resort for protecting American interests. He advocated direct, unconditional engagement with adversaries, embroiling him in an early disagreement with his hard-nosed rival in the Democratic primaries, Hillary Clinton. By the time he took office in January 2009, Iran loomed as the biggest test of both of those propositions—whether diplomacy backed up by economic and military leverage could produce results, and whether direct contacts with our toughest adversaries could pay off.
President Obama found an effective partner for his Iran diplomacy in Clinton. She was instinctively more cautious about engaging the Iranians, and more skeptical about the chances of ever reaching an agreement that would deny Tehran a bomb. She agreed, however, that direct engagement was both the best way to test Iranian seriousness and the best way to invest in the kind of wider international coalition that we’d need to generate more pressure on Iran if it failed those initial tests.
Three days after she was sworn in as secretary of state, I sent Clinton a memo entitled “A New Strategy Toward Iran.” I began by trying to encapsulate our fundamental purpose:
Recognizing that Iran is a significant regional player, our basic goal should be to seek a long-term basis for coexisting with Iranian influence while limiting Iranian excesses, to change Iran’s behavior but not its regime. That means, among other things, preventing Iran from achieving nuclear weapons capability; channeling its behavior so that it does not threaten our core interests in a stable, unitary Iraq and an Afghanistan that is not a platform for the export of violent extremism; and gradually reducing Iran’s capacity to threaten us and our friends through support for terrorist groups. We should also speak out consistently against human rights abuses in Iran.4
I argued for a comprehensive approach. As with China in the early 1970s, it made sense to employ careful and incremental tactics at the outset, but as part of a coherent long-term strategy. “We should set,” I said, “an early tone of respect and commitment to direct engagement, however severe our differences.” I added the obvious: “Dealing with Iran will require enormous patience, persistence and determination. Deeply conspiratorial and suspicious of American motives, and riven by factions especially eager to undermine one another in the run-up to Iran’s Presidential elections in June, the Iranian elite will be prone to false starts and deceit.” We shouldn’t underestimate the reality that, especially for the Supreme Leader and the hard men around him, animus toward the United States was the core organizing principle for the regime. But, I continued, “we should deal with the Iranian regime as a unitary actor, understanding that the Supreme Leader (not the President) is the highest authority. We have failed consistently in the past when we tried to play off one faction against another.”
I also emphasized that we shouldn’t lose sight of Iran’s vulnerabilities. “Iran is a formidable adversary…but it is not ten feet tall. Its economy is badly mismanaged, with rising rates of unemployment and inflation. It is vulnerable to the ongoing sharp decline in oil prices, and to its dependence on refined petroleum products. It has no real friends in the neighborhood, distrusted by the Arabs and the Turks, patronized by the Russians, and suspicious of the Afghans.” Finally, I stressed that “we need to be always conscious of the anxieties of our friends, as well as key domestic constituencies, as we proceed with Iran.” I warned that our Sunni Arab partners would be nervous that we were abandoning them for a new Persian love interest. The Israelis would be at least as worried, given the undeniable threat that Iran’s proxies and nuclear and missile programs posed for them. We’d have a big challenge managing Congress and its widespread aversion to serious engagement with Iran. And, I argued, “we must make sure that the Administration speaks with one voice, and avoids the divisions which beset the last Administration.”5
Convinced by the argument, Clinton brought discipline and skill to the task. President Obama was eager to begin, and he convened a series of meetings in early 2009 to hammer out a broad strategy, close to the one I had tried to lay out for the secretary. Obama’s inheritance on Iran was difficult. When he told the Iranians in his inaugural address on January 20 that “we will extend a hand if you are willing to unclench your fist,” Tehran had already stockpiled enough low-enriched uranium for a nuclear weapon. Its missile systems were advancing. And while we had no firm evidence of a revival of Iran’s earlier weaponization efforts, we could never be entirely sure.
In March, the president sent a videotaped Nowruz message to the Iranian people and, in a subtle effort to signal his lack of interest in forcing regime change, referred to the government by its formal name—the Islamic Republic. He committed the United States to “engagement that is honest and grounded in mutual respect.” The Iranian popular reaction was overwhelmingly positive. The regime, particularly the Supreme Leader, remained skeptical.6
In early May, the president sent a long secret letter to the Ayatollah Khamenei. The letter tried to thread a needle—the message needed to be clear, but written in a way that would not cause too much controversy if it was leaked. In the letter, Obama reinforced the broad points in the Nowruz message. He was direct about his unwavering determination to prevent Iran from acquiring a nuclear weapon, and his support for the P5+1 position that Iran was entitled to a peaceful civilian nuclear program. He also made clear that it was not the policy of his administration to pursue regime change, and indicated his readiness for direct dialogue. The Supreme Leader replied a few weeks later, trying to thread a similar needle. His message was rambling, but at least by the standards of revolutionary Iranian rhetoric not especially edgy or sharp. While it offered no explicit reply to the president’s offer of direct dialogue, we understood it nevertheless as a serious indication of his willingness to engage. President Obama responded quickly, in a short letter that proposed a discreet bilateral channel for talks, naming me and Puneet Talwar, a senior NSC staffer, as his emissaries.
All this halting momentum, modestly encouraging given the usual tribulations of dealing with Iran, came to an abrupt stop when the Iranian presidential elections in June turned into a bloodbath. The regime’s ballot stuffing and repression of a surprisingly potent Green Movement opposition led to violence in the streets, documented by cellphone-wielding Iranian citizens and broadcast around the world in dramatic fashion. The government cracked down with its customary brutality, with paramilitary militias beating demonstrators, thousands arrested, and dozens killed. The White House’s public response was initially tepid, less because of concern that it would jeopardize the fledgling effort at talks and more because the message from Green Movement leaders was not to suffocate them with an American embrace and allow the regime to paint them as U.S. stooges. In hindsight, we should have politely ignored those entreaties and been sharper in our public criticism from the start. Such criticism, which we eventually made quite strongly, was not only the right thing to do, it was also a useful reminder to the Iranian regime that we weren’t so desperate to get nuclear talks started that we’d turn a blind eye to threatening behavior, whether against Iran’s own citizens or our friends in the region.
As the summer of 2009 wore on, we continued to invest systematically in our P5+1 partners. Part of this had to do with an intriguing new idea that had emerged from IAEA director General Mohamed ElBaradei. Near the end of his tenure, ElBaradei still smarted from his frequent clashes with the Bush administration, but was anxious to help the new American administration get off on a more positive footing. The Iranians had sent a formal request to the IAEA in early summer, notifying them that the Tehran Research Reactor (TRR), which produced medical isotopes, had nearly exhausted the supply of 20 percent enriched uranium fuel plates that the Argentines had supplied in the 1990s. The implication seemed clear: Either ElBaradei would produce an alternative supplier, or the Iranians would produce the material themselves—and move closer to weapons-grade enrichment.
ElBaradei had the beginnings of a creative proposal. Why not call the Iranian bluff and supply the fuel plates, which posed no risk of being used for enrichment or weapons purposes? Bob Einhorn, my Baker-era Policy Planning colleague and now a senior advisor on nonproliferation issues at State, and Gary Samore, his counterpart at the NSC staff, took this one very interesting step further. Why not offer to supply the fuel plates for the TRR, but insist in return that the Iranians “pay” with about twelve hundred kilograms of 5 percent enriched uranium, roughly the amount that it would take to produce a batch of 20 percent fuel plates (and roughly the amount for one bomb’s worth of material) to replenish the original Argentine shipment? Subtracting it from the then Iranian stockpile of about sixteen hundred kilograms would leave only four hundred kilos, far less than what they would need if they wanted to try to break out toward a weapon. It would take the Iranians a year or so to get back to one bomb’s worth of material. That would provide time and space for serious negotiations about both the interim “freeze for freeze” proposal and a comprehensive solution.
Another priority that spring and summer was to strengthen cooperation with Russia on the Iran nuclear problem. The Georgia war in August 2008 had cratered U.S.-Russian relations, and the Obama administration had begun its effort to “reset” the relationship. Cooperation with Russia was the key to making the P5+1 effective. If the Iranians realized that they couldn’t separate Moscow and Washington, and that we and the Russians might actually work together on much tougher sanctions, there might be a chance of focusing minds in Tehran. President Obama’s conversation with Dmitry Medvedev in London in April 2009 was an excellent start. Secretary Clinton stayed in close touch with Foreign Minister Lavrov, and I had several long, quiet meetings with Deputy Foreign Minister Sergey Ryabkov, my counterpart and Russia’s representative in the P5+1 talks. Ryabkov and I discussed the TRR proposal, and began to outline a cooperative arrangement in which Russia might produce the fuel plates for the TRR and take the Iranian low-enriched material in return.
Events came to a head in September 2009. As the annual meeting of the UN General Assembly approached in New York, which would be followed shortly by a G-20 summit in Pittsburgh, U.S., British, and French intelligence uncovered damning evidence of a covert Iranian enrichment site, buried deep inside a mountain near Qom. What made the clandestine site especially alarming was its relatively modest scale; with a capacity of only about three thousand centrifuges, it was much too small to produce enriched uranium fuel for a civilian nuclear power plant, but big enough to produce material for one or two nuclear bombs a year. Apparently nervous that Western governments might be poised to expose them, the Iranians sent a brief, seemingly innocuous note to ElBaradei informing the IAEA (many months after they were obligated to) of vaguely described construction work near Qom, at a site they called Fordow.
ElBaradei walked into a previously scheduled meeting at the Waldorf Astoria hotel in New York with me, Samore, and Einhorn on the evening of September 20. A little jet-lagged, and unaware of what we already knew about Fordow, ElBaradei reached into his pocket and handed us the Iranian notification. As Gary, Bob, and I each took turns looking at it, struggling to seem nonchalant, we quickly realized that it referred to the covert enrichment facility at Qom. We now had a fair amount of leverage with the Iranians, and a powerful argument to use with the Russians. Medvedev was angered by the revelation, partly because the Russians had again been caught off guard, and partly because the Iranians had apparently deceived them too. When President Obama announced the breach we had uncovered a couple days later in Pittsburgh, it deepened the resolve of the P5+1 to push the Iranians hard at the meeting that had already been scheduled in Geneva on October 1, and left Tehran backpedaling.
Led again by Javier Solana, my P5+1 colleagues and I met with Saeed Jalili and the Iranian delegation at a chateau outside Geneva on a sunny day in early October. We spent a desultory three hours in the morning delivering familiar positions across the table. Impatient and concerned that we’d miss the moment, I took advantage of the break for lunch, walked up to Jalili, shook his hand, and said, “I think it would be useful if we sat down and talked.” He agreed, having presumably gotten advance permission from Tehran. And so began the highest-level conversation between the United States and Iran since 1979.
We walked over to a small side room and sat down around a polished round table with seats for four. Bob Einhorn joined me, and Jalili was accompanied by his deputy, Ali Bagheri. Puneet Talwar arrived a few minutes later and sat behind Bob. Jalili was more soft-spoken than in our prior encounter. There were no set pieces this time. This was the first bilateral talk we had ever had with the Iranians on the nuclear issue, and I didn’t want to waste it with a long preamble. I was also mindful that Jalili, with or without bombast, remained deeply suspicious of this whole interaction. He was a true believer in the Islamic Revolution, and he had come by his convictions through bitter experience. Wounded fighting the Iraqis in the 1980s, he had lost part of his right leg and walked with a distinct limp. Like many in his generation, he had learned the hard way in the trenches that Iran could trust no one and could only rely on itself.
I laid out carefully the TRR swap concept that ElBaradei had already previewed to the Iranians. Bob added some details, to make sure Jalili and Bagheri understood precisely what we were proposing. Jalili asked a few questions, but seemed to accept the core concept, and to appreciate how Iran would benefit from such a reciprocal arrangement. I also made clear, in a straightforward tone, that the consequences of rejecting the proposal, especially in light of the Qom revelation, were certain to be substantially tougher sanctions. Jalili seemed confident that Tehran would approve. “Our viewpoint,” he said, “is positive.” After we broke up, I asked Bob to go through the TRR proposal one more time with Jalili’s deputy. They produced a paragraph summarizing our understanding, which we agreed that Solana could make public. Our P5+1 partners were supportive, relieved that we finally seemed poised to make some headway.
While hopeful, I told Secretary Clinton on the phone later that afternoon that the chances were probably less than fifty-fifty that the deal would stick in Tehran. Unfortunately, my pessimism proved well founded. A follow-up meeting in Vienna, hosted by the IAEA later in October, collapsed when the Iranians tried to walk back key provisions, particularly the shipment of twelve hundred kilograms of material to Russia. That was the crucial confidence-building step. The irony was that President Ahmadinejad was the biggest booster of the TRR agreement in Tehran, anxious to improve his standing after the disastrous fixed election and show that he could “deliver” the Americans. I assumed that Jalili’s positive response in Geneva reflected Ahmadinejad’s eagerness, and perhaps also wider regime worries after the Qom revelation that they needed to find a way to ease tensions. The Iranian president’s political rivals, some of whom had been involved in the nuclear negotiations before and might otherwise have taken more supportive positions, didn’t want Ahmadinejad to get the credit for any breakthrough, however modest. Iranian politics are a brutal contact sport, and the TRR deal was one of its many casualties.
As we had warned Jalili, his rejection of the deal led us to pivot to greater pressure against Iran. Secretary Clinton played a particularly effective role in helping Susan Rice, our ambassador at the United Nations, cajole the members of the Security Council toward a much tougher sanctions resolution, finally passed as UN Security Council Resolution 1929 in early June 2010. The Iranians played their usual critical role in helping us to persuade key members of the council, announcing in February 2010, for example, that they were beginning to enrich to 20 percent, ostensibly for the TRR. Russia’s position was crucial; among the permanent, veto-wielding members of the council, we could count on strong support from Britain and France for more substantial sanctions, and China tended on the Iran issue at least to defer to Russia. Frustrated by the Iranians after the Qom disclosures and the failed TRR experiment, and increasingly confident in the possibilities of selective cooperation with the United States as the “reset” evolved, Medvedev eventually came around to support Resolution 1929.
An improvised effort in May by Brazil and Turkey to rescue the TRR proposal and stave off a new round of sanctions was too little, too late. In mid-May, Presidents Luiz Inácio Lula da Silva and Recep Tayyip Erdog˘an went to Tehran and announced with great fanfare that they had brokered a breakthrough. The problems with their vaguely worded declaration were manifold: Since the Iranians now had accumulated enough low-enriched uranium for two bombs, exporting half would still leave them with enough for a bomb, if they chose to enrich to weapons-grade; the arrangements for shipping the material out of Iran to be swapped for TRR fuel plates were unclear; and Iran had already started enriching to 20 percent, another new problem. The bigger issue was that we had put enormous effort into getting Russia and China on board for what became UN Security Council Resolution 1929, and it would have been foolish to turn back unless the Iranians had made a spectacular move. This wasn’t.
Passage of Resolution 1929, aimed in part at isolating Iran from the international financial system, was an enormous relief. I was at the high school graduation of our younger daughter, Sarah, in Georgetown on the day the vote took place in New York. Much to the consternation of the watch officers at the State Department Operations Center who had been connecting me with calls to P5+1 counterparts and a variety of American colleagues earlier that day, I happily turned my cellphone off for a few hours to enjoy Sarah’s moment.
Resolution 1929 provided a platform for additional U.S. sanctions against Iran, as well as significant new EU measures. The U.S. steps, adopted overwhelmingly by Congress two weeks later, were aimed in part at reducing international purchases of Iranian oil, the lifeblood of its crumbling economy. The EU followed in July with a stringent package of its own. Far more than any previous combination of sanctions, these took a serious toll on the Iranian economy. By the end of President Obama’s first term, the value of Iran’s currency and its oil exports had each declined by 50 percent.
Iranian nuclear advances, however, continued to move at a dangerous pace. By the end of 2012, it had a stockpile of nearly six bombs’ worth of 5 percent enriched material, and probably half a bomb’s worth of 20 percent material. It was spinning more and more centrifuges at its openly declared site at Natanz, installing centrifuge cascades at Fordow, experimenting with more advanced centrifuges, and continuing work on its heavy water plutonium-producing site at Arak. Its missile systems were increasing in range and sophistication.
The country most alarmed by these developments was Israel. Although appreciative of all the effort that had gone into stepping up sanctions, Prime Minister Netanyahu argued throughout the latter part of Obama’s first term that sanctions and diplomacy would be too slow to curb Iran’s nuclear ambitions, and that military action would likely be required. His clear preference was to press Obama toward U.S. military action, especially against the deeply embedded enrichment facility at Fordow. Obama was unconvinced by the logic or necessity of force at this stage, and deeply irritated by Netanyahu’s heavy-handed attempts to manipulate him in the run-up to the 2012 presidential elections. The Israeli leader’s efforts to badger and maneuver Obama into a more belligerent approach had the opposite effect, deepening and accelerating his commitment to finding a way short of war to stop the Iranians.
Even a sweepingly effective attack on the Iranian program, Obama believed, would only set back the Iranians by two or three years. They would undoubtedly regroup, take their program fully underground, and very likely make a decision to weaponize, with wide popular support in the aftermath of a unilateral U.S. or Israeli strike.
Obama and Clinton worked carefully to manage Netanyahu’s pressure and demonstrate U.S. determination to ensure by whatever means necessary that Iran would not acquire a nuclear weapon. National Security Advisor Tom Donilon deepened consultations with the Israelis on intelligence, as well as on sanctions and diplomatic strategy. We stepped up the transfer of sophisticated military systems to Israel, and accelerated our own plans for a new, fifteen-ton bomb that could penetrate Fordow. The United States and Israel reportedly jointly developed and deployed a malicious computer worm dubbed Stuxnet to sabotage, at least temporarily, the Iranian program. This campaign helped deflect Netanyahu’s push to bomb, but it was clear as President Obama began his second term that the drumbeat would get louder again if we couldn’t make diplomacy work. As Hillary Clinton would later describe it, “the table was set” for a renewed diplomatic push, with sanctions eating away at the Iranian economy and a Supreme Leader in Tehran nervous about a repetition of the unrest that had so unsettled his regime in the summer of 2009.7 What was still missing, however, was a direct channel with the Iranians.
THE STORY OF how the Omani back channel to Iran emerged seems, like so many things in diplomacy, a lot neater in retrospect than it did at the time. Sultan Qaboos, an engaging ruler of the old Arab school, had navigated complicated currents at home and in the region for more than four decades, and had maintained a good rapport with the Supreme Leader in Tehran. Eager to play the intermediary, and nervous about the dangers of conflict so close to home, Oman sent the new U.S. administration a series of low-key overtures about its readiness to establish a channel to Iran.
The principal messenger for the sultan was Salem Ismaily, a clever, urbane, persistent, and resourceful advisor, who, in the ambiguous way in which Middle East elites often function, moved easily between the worlds of officialdom and private business, and was often used as a trusted fixer and negotiator. Salem was supremely confident of his ability to set up a reliable channel to Tehran, although sometimes murky about who exactly he was dealing with on the Iranian side. Given the checkered history of American-Iranian contacts, we were always skeptical of new initiatives, which often turned out to be overenthusiastic at best, and duplicitous at worst.
But Salem’s steady and upbeat insistence that he could deliver, backed up by long-standing trust in Qaboos, set the Omani overtures apart. What solidified my confidence in Salem and his relationships in Tehran was his role in securing the release of three young American hikers who had strayed into Iran along the border with Iraqi Kurdistan in the summer of 2009. They were arrested and thrown into the dismal confines of Evin Prison in downtown Tehran, where American embassy hostages had been held many years before. The hikers faced deep uncertainty. We tried through a variety of channels to secure their release, with no luck—until Salem got involved. Using his contacts in Tehran and the sultan’s reputation and resources, he managed over the next two years to negotiate the release of all three Americans.
In October 2011, shortly after the hikers had returned home, Secretary Clinton met with Qaboos in Muscat, and concluded that the Omani channel was our best bet. President Obama spoke a couple times by phone with the sultan, and was similarly impressed, especially by Qaboos’s conviction that he could deliver contacts with Iranians fully authorized by the Supreme Leader. I shared the view that the Omanis offered a promising opening, although I always wondered whether their relative success with the Iranians was a matter of their influence and ingenuity, or perhaps simply that they were a convenient vehicle for the Iranian regime when it decided to unburden itself of problems (like the hikers) or test channels with some plausible deniability. It also always appealed to the Iranians to sow dissension among the Gulf Arabs and use the Omanis to irritate the Saudis.
Even more energetic in his promotion of the Oman channel was Senator John Kerry, chairman of the Senate Foreign Relations Committee. Kerry had long been persuaded that the United States had to come to terms with Iran’s nuclear progress and engage directly. Coordinating with Clinton and Donilon, he had a series of meetings and phone calls with the sultan and Ismaily in late 2011 and 2012, which made him a passionate advocate of the Oman channel. He made clear his commitment to exploring a dialogue with the Iranians, and his interest in playing a personal role.
At around the same time, Salem came to us with a new proposal, which he and the sultan were certain bore the approval of the Supreme Leader. He suggested a direct U.S.-Iranian meeting in Muscat, quietly facilitated by the Omanis. He asserted that the Iranians would be prepared to address any issue, but wanted to focus in particular on the nuclear problem. He was uncertain who would lead the Iranian team, but thought it might be Ali Velayati. After some debate in Washington, we decided to suggest a preliminary, preparatory meeting at a lower level. We had been burned so many times in the past few decades that caution seemed wise.
Jake Sullivan, still serving as Secretary Clinton’s Policy Planning director, and Puneet Talwar were natural choices to represent the United States at this exploratory session. Puneet had joined me for innumerable P5+1 rounds, and was a key player in the TRR initiative. Jake was Hillary Clinton’s closest policy advisor. He had her full confidence, and the president’s trust.
In early July, Jake was off on yet another overseas trip with the secretary when I called him and asked if he could break off for a couple days in Oman. He didn’t hesitate, and made his way from Paris to Muscat, where, hosted by Salem, he and Puneet spent a long and not particularly encouraging day with a mid-level Iranian delegation. Jake reported that the Iranians had been almost entirely in “receive mode,” and seemed intent on securing some kind of substantive down payment for any future talks. They were particularly focused on the thorny issue of their “right” to enrichment—something that the Treaty on the Non-Proliferation of Nuclear Weapons did not explicitly convey, and that their continuing violation of successive IAEA and UNSC resolutions did little to promote. With his usual candor, an unusually effective mix of Minnesota politeness and East Coast hardheadedness, Jake made clear that the issue was what the Iranians would do to satisfy powerful international concerns, not the other way around.
Over the next few months, as the American elections approached in November, both sides regrouped. We made clear, through Salem, that we were ready for further meetings but weren’t in any rush. Sanctions pressure was building, and we wanted the Iranian government to feel the pain.
Following President Obama’s second inauguration, John Kerry succeeded Hillary Clinton as secretary of state. On the same day his nomination was announced, Kerry asked me to stay on as deputy secretary. I had planned to retire, but I was glad to accept.
It was obvious that Secretary Kerry, like the president, wanted to make Iran negotiations a priority for the second term. Sensing an opening, Salem conveyed renewed interest from the Iranians in meeting again, this time at a higher level, with a deputy foreign minister heading their delegation. We eventually settled on March 1, 2013, in Oman. I would lead the American team, with Jake as my alter ego. He had already become my closest collaborator in the Obama administration—the best of his generation of foreign policy thinkers and practitioners, strategically creative as well as tactically adept. We’d be joined by Puneet and Bob Einhorn; Jim Timbie, whose encyclopedic knowledge of nuclear issues and four-plus decades of negotiating experience made him a quiet national treasure; Richard Nephew, a specialist on sanctions; and Norm Roule, the senior advisor on Iran in our intelligence community.
The president convened several meetings in February to review our approach. In all my three decades in government, this was—along with the bin Laden raid in 2011—the most tightly held effort. The White House Situation Room, usually crowded with cabinet officials and backbenchers, was unusually spare. Only a handful of people in the White House and the State Department knew of the secret talks, and we went to great lengths to preserve their discretion. Meetings on this issue didn’t go on our public calendars, or bore innocuous titles; documents related to the talks were kept only on the ultra-secure White House communications systems.
By this point, the president’s grasp of the Iranian nuclear issue and his policy sense were both well developed. His expectations were realistic; he knew the odds that the Iranians would be willing or able to accept sharp limitations on their nuclear program were low. He also knew the dangers of being played by Tehran, as had happened too many times before. Yet he was determined to test the proposition, and secure in the investment he and Secretary Clinton had made in the first term on sanctions and international solidarity. “We’re as well positioned to negotiate as we’ve ever been,” Obama said in one session in the Situation Room. “We’ve set this up right. Now we’ll see if we can make this work.”
President Obama laid out the framework for our effort crisply. First, our direct channel wasn’t a substitute for the wider P5+1 channel, but a pragmatic complement. Agreements on the nuclear issue could only come through that broader group, since it was the source of much of the international pressure that was starting to weigh on Tehran.
Second, we would have to keep the bilateral channel secret. The Iranians, facing enormous internal pressures, and filled with suspicion of American motives, warned that any premature disclosure would make it impossible to continue. Secrecy would help prevent opponents in both capitals from smothering the initiative in its crib—but it would carry future costs, feeding stab-in-the-back criticisms from some of our closest partners, particularly the Israelis, Saudis, and Emiratis. We knew that secret talks would be hard to sustain. In the age of omnipresent information technology and never-ending media scrutiny, it was unlikely that we could avoid disclosure for long. Oman was a relatively quiet and off-the-beaten-path place, but it was also a fishbowl in its own way, under the intermittent scrutiny of a variety of intelligence services.
We also knew that transitioning from a direct channel back into multilateral talks would be complicated and awkward. Some of the best foreign diplomats with whom I had ever worked had been or remained part of the P5+1. I didn’t much enjoy the idea of keeping our efforts from them. But we would never have gotten as far as we later did, as relatively fast as we did, if we had been trying to negotiate with the Iranians in the glare of international publicity, and solely in the inevitably more cumbersome P5+1 process.
The president stressed a third point. We would focus the back-channel talks on the nuclear issue, which was the most pressing and explosive of our many problems with Iran. It was the concern around which we had united the international community and built such powerful pressure, and the one on which the Iranians seemed prepared to engage. Moreover, our Gulf Arab partners were adamant that we not widen the aperture of the nuclear talks and address Iran’s non-nuclear transgressions, unless they were in the room. The purpose of the secret bilateral talks would be to test Iranian seriousness on the nuclear issue, and jump-start the broader P5+1 process.
This was a fairly transactional and unsentimental view of the nuclear negotiations, without any grand illusions of overnight transformations in Iranian behavior or U.S.-Iranian relations. I was a short-term pessimist about the prospect for such changes, given the cold-blooded nature of that regime, its resilience and practiced capacity to repress, and the opportunities before it to meddle in a troubled Arab world. We knew we’d have to embed any progress on the nuclear issue in a wider strategy to push back against threatening Iranian behavior in the region, and preserve leverage and non-nuclear sanctions to draw on.
The president’s fourth bit of guidance cut right to the core of the transactional challenge. We would indicate to the Iranians, carefully, that if they were prepared to accept tight, long-term constraints on their nuclear program, with heavily intrusive verification and monitoring arrangements, we would be prepared to explore the possibility of a limited domestic enrichment program as part of a comprehensive agreement. There had been considerable internal back-and-forth on this issue, beneath the president’s level, less over whether to play this card than when. I thought it was best to do it at the outset, as a sign of our seriousness and a test of theirs, putting the burden squarely on the Iranians to show that they would accept tough constraints, and make clear in practical terms that they wouldn’t be able to break out to a bomb. Tom Donilon was a little uneasy about that tactic. He wanted to see more tangible evidence of Iranian seriousness before playing a card that—however carefully framed the proposition—would be hard to put back in the deck.
The president was convinced that we’d never get an agreement with the Iranians without some limited form of domestic enrichment. They had the knowledge to enrich, and there was no way you could bomb, sanction, or wish that away. Maybe we could have gotten to a zero enrichment outcome a decade earlier, when they were spinning a few dozen centrifuges. That was extremely unlikely to happen in 2013, with the Iranians operating some nineteen thousand centrifuges, and with broad popular support across the country for enrichment as part of a civilian program. The president wanted to cut to the chase in the back-channel talks, and that made sense to me. He approved the caveated formula we suggested, but placed heavy emphasis on the verb “explore.” This was not a promise or a guarantee.
Finally, the president reminded all of us that the chances of success were “well under fifty-fifty.” We’d keep all our other options open. We’d keep developing the “bunker buster” bomb we’d need to strike Fordow. The Pentagon would “set the theater” and demonstrate through regular deployments and thorough preparations that our military was prepared to act. We’d keep up other efforts to slow and obstruct the Iranian program. And we’d be ready to pivot again from a failed negotiating effort to even stronger sanctions. If direct talks went nowhere, it would be harder to blame the United States, and easier to build more pressure against a recalcitrant Iran.
At the end of our last meeting, the president shook hands with Jake and me and said simply, “Good luck.” Obama was staking a lot on this uncertain enterprise, and we were determined to do all we could to make it work.
WE WOULD SOON get used to long flights to Oman in unmarked planes with blank passenger manifests, but I was too restless on that first trip at the end of February 2013 to sleep, wondering what lay ahead. Salem met us on arrival at a military airfield in Muscat, upbeat as always, and confirmed that the Iranian delegation, led by Deputy Foreign Minister Ali Asghar Khaji, had arrived just before us.
We drove about thirty minutes outside Muscat, to a secluded military officers’ club on the Arabian Sea. Its four walls became the boundaries of our little universe for the next few days. Our team spent that first evening reviewing our approach to the start of talks the next morning and hashing out roles and responsibilities. Our expectations were modest, but after many months of internal debate and planning, we were just glad to finally get started.
The next morning was typically hot and humid. I went out on an early-morning stroll, but the sauna-like conditions did little to cure me of my jet lag. Salem, the chief of the royal court, and the head of Omani intelligence greeted both delegations as we walked into the meeting room, which offered a panoramic view of the sea. I shook hands with Khaji and his colleagues, who included Reza Zabib, the chief of the Iranian Foreign Ministry’s North America division; Davoud Mohammadnia, from the Ministry of Internal Security, or MOIS; representatives of Iran’s atomic energy agency; a capable Iranian interpreter; and what I always assumed were a number of listening devices to record our conversations. We sat on opposite sides of a long table, too weighted down by history to enjoy the view or the moment. The Omanis, clustered around the head of the table, offered a few brief words of welcome and then departed. There was then an awkward pause.
I broke the silence by asking Khaji if he wanted to speak first. In what was an early indication that the Iranians were mostly on a reconnaissance mission, he deferred to me. I then made a brief presentation, along the lines the president had approved in the Situation Room a few days before. I tried to strike a respectful but candid tone. This was an important moment for both of us, a rare chance to talk directly and privately. We had no illusions about how hard this would be. While it was not our purpose to point fingers or lecture, there were profound mutual suspicions, and a long record of Iranian defiance of its international obligations that had provoked widespread concern. There were too many unanswered questions; too many obvious disconnects between the requirements of a realistic civilian nuclear energy program and the pace and lack of transparency of Iran’s efforts; and too much disregard for the clear requirements of a series of UN Security Council resolutions. There was a serious and growing risk that Iran’s domestic enrichment capacity could be quickly and covertly converted to produce weapons-grade, highly enriched uranium. That concern cut right to the core of what was at stake and the dilemma we faced.
The nuclear issue was not just a dispute between Iran and the United States, but between Iran and the P5+1, and the broader international community. I repeated the president’s message in his earlier letters to the Supreme Leader—it was not the policy of the United States to seek regime change in Iran, but we were absolutely determined to ensure that Iran did not acquire a nuclear weapon. Any hope for a diplomatic resolution would require Iran to understand the depth of international concerns and act upon them. I repeated bluntly that failure to take advantage of the uncertain window before us would certainly increase the costs to Iran, and the risk of military conflict.
The key to any diplomatic progress would be Iran’s willingness to take concrete, substantial measures to give the rest of us confidence that a peaceful program could not be converted into a weapons program. If Iran were ready to do that, I continued, we would be willing to explore whether and how a domestic enrichment program could be pursued in Iran, as part of a comprehensive settlement of the nuclear issue. That settlement would require many difficult, long-term Iranian commitments, and intense verification and monitoring provisions. Should we eventually reach a satisfactory agreement, the United States would be prepared to call for an end to all United Nations and unilateral sanctions against the Iranian nuclear program. Such a process would likely have to unfold in phases, with early practical steps to build confidence, leading to a comprehensive agreement.
There was enormous skepticism in both our capitals, I concluded. I shared much of that skepticism, but if the Supreme Leader’s fatwa against nuclear weapons was serious, then it shouldn’t be impossible to find a diplomatic path forward and prove the skeptics wrong. We were certainly prepared to try.
As I went through this presentation, Khaji and his colleagues listened and took copious notes. There were a few cold stares, and some head-shaking, but no interruptions. The Iranians in Oman were a welcome change and stark contrast to the doctrinaire, obstructionist Jalili-led delegation in the P5+1 talks. They were professionals, mostly career diplomats, and it wasn’t hard to sense a shift in style and seriousness of engagement.
When I was done, Khaji took the floor. His tone was measured, even when he recited a long and predictable list of grievances about American policy. He spoke sharply about the unfairness of UN Security Council resolutions, the assassinations of Iranian nuclear scientists, and the U.S. public emphasis that “all options are on the table.” He objected to American references over the years to the use of “carrots and sticks” against Iran. Raising his voice, he exclaimed, “Iranians are not donkeys!”
Khaji had little of substance to offer, although he stressed that he “wanted to look to the future.” The Iranian delegation had clearly taken note of my heavily caveated comment about domestic enrichment, but they were looking for (and probably expecting) more. Khaji asserted that Iran would defend its “right” to the whole nuclear fuel cycle, including enrichment, “at any cost.” We went back and forth over this argument, emphasizing our conviction that no such explicit “right” was granted by the NPT. The problem that Iran’s defiant behavior had created was simply that there were serious and growing international doubts about whether it wanted only a civilian program, or might pursue a military one. The onus was on Iran to disprove those doubts. Constant reassertion of imaginary rights was not going to get anywhere.
In the first of several one-on-one conversations, Khaji and I sat privately after the opening plenary meeting. He was approachable but guarded, and acknowledged that he was encouraged that we were finally talking to each other directly. He appealed almost plaintively for acceptance of Iran’s right to enrich, implying that it would be hard to engage without it. At one point he pulled out a bulky file of papers, which appeared to be Omani notes from alleged conversations with various Americans, including members of Congress, acknowledging Iran’s right to enrich. I explained that in our political system, members of Congress did not speak for the president. I repeated our position, said I was sure those conversations were taken out of context or were well-meaning Omani garbles, and stressed that we needed to focus on what was practical if we were going to get this process off the ground.
We spent the next couple days covering essentially the same terrain over and over. I took evening walks with Khaji around the officers’ club compound, and Jake and my other colleagues had similar chats with their counterparts. There was a powerful cognitive dissonance at the heart of our discussions at this stage, which never entirely disappeared throughout the secret talks. The Iranians would maintain, in a tone of wounded pride, that the nuclear problem was all a big misunderstanding, that they had done nothing wrong, never had explored steps toward a bomb, and had acted within their international rights. Sanctions were unjust and should be lifted. We responded firmly that the Iranians would never get anywhere in nuclear negotiations if they didn’t realize that they had a gaping credibility problem—not just with the United States but with the wider international community.
Near the end of our final session on March 3, Jake and I told the Iranians bluntly that there wasn’t much point in continuing the secret talks if they weren’t going to think much more seriously about the tangible steps they would need to take. The Iranians were wildly unrealistic in their expectations; they weren’t in the same ballpark, or even playing the same sport, as an increasingly determined international community.
Zabib made an impassioned plea as we wrapped up the last meeting. He recounted a trip to New York years before, and a large sign he had encountered at JFK airport on his arrival. Torturing his English syntax a bit, he recalled that it read “Think the Big.” “That’s what you Americans must do,” he said. “Think the big. If you do, everything will become better.” I smiled at Jake. Neither of us could think of a succinct way to explain that the issue here was not the elasticity of our thought process, but the seriousness of Iran’s commitment. So we simply urged them to think about everything we had discussed over the previous three days in Oman, and to consider the choices that lay before them.
It was an incongruous conclusion to an incongruous first round. We reported back to Washington that we were “miles apart on substance.” Khaji was an able diplomat, but not empowered. None of that was unexpected, after so many years of not talking to one another directly. The atmospherics were significantly better than the more sterile P5+1 process had been, with at least the possibility of less polemical and more practical conversations, and maybe more room for creativity. Zabib’s plea to “think the big” did not exactly fill us with confidence, but it was at least a start.
We retraced our seventeen-hour journey and arrived back in Washington on March 4, our secret still intact. The next day, with Secretary Kerry out of the country, the president asked to see Jake and me in the Oval Office so we could report directly on our discussions in Oman. The president was sitting in his usual chair in front of the fireplace, having just finished his regular morning intelligence briefing. He listened attentively. We went through our impressions, careful not to oversell what was just a first step on a long road. “I never expected immediate progress,” the president said. “This may or may not work. But it’s the right thing to do. Let’s just hope we can keep it quiet, and keep it going.”
WE HAD LEFT that Omani beach compound still not convinced that our secret initiative had a future. Events over the rest of the spring were not reassuring. The P5+1 met with Jalili and the Iranians in Almaty in early April, but made no headway. We had made clear through Salem Ismaily and the Omanis that we were ready to resume the back-channel talks, but the Iranians were consumed by the run-up to their presidential elections in June. There was no point in running after them; we had no idea who would succeed Ahmadinejad, and it made sense to wait and see.
The unexpected election of Hassan Rouhani in June 2013 created a modest new sense of possibility. A former lead nuclear negotiator and a wily survivor in the unsentimental world of revolutionary Iranian politics, Rouhani saw the toll that international sanctions (as well as Ahmadinejad’s erratic populist mismanagement) had taken on the country’s economy. He managed to persuade the Supreme Leader that Iran needed to explore a more serious nuclear negotiation and consider some real compromises, or face a resurgence of the internal political unrest that had jarred the regime in the summer of 2009. In hindsight, it was useful to have launched the secret channel while Ahmadinejad was still president and before Rouhani was elected. Had we waited until after the election, it might have appeared to the ever-suspicious Khamenei that we were fixated on Rouhani and neglecting the ultimate decision-maker. It also cost Rouhani far less political capital to push for direct talks with the Americans when the more hawkish Ahmadinejad government had already crossed that Rubicon.
Rouhani had a resourceful partner in his new foreign minister, Javad Zarif, who emerged as the new face of Iranian diplomacy. With a doctorate from the University of Denver, where he (like Condi Rice) had studied under Madeleine Albright’s father, Zarif had served for many years as Iran’s permanent representative to the United Nations in New York. He was a formidable diplomat. He knew how to navigate his own treacherous political system and squeeze the most out of his instructions—and he knew how to use his talents and sympathetic image, as well as a sometimes frustrating gift for melodrama, to cajole and maneuver the rest of us.
President Obama’s short congratulatory message to Rouhani received a rapid and positive reply. Rouhani was inaugurated on August 4, and two days later announced publicly his readiness to resume the P5+1 talks. The Iranians also told the Omanis that they wanted to restart the back-channel process, and Salem came away from conversations in Tehran that summer convinced of newfound seriousness. We agreed to meet again at the beach compound in early September.
In preparation, we took stock of our approach. Our long-term challenge remained the same: to cut off the pathways that the Iranians might use to develop a nuclear weapon. By the time of Rouhani’s inauguration, they had accumulated a substantial stockpile of enriched uranium. Meanwhile, they were continuing construction of their heavy water plant at Arak, moving steadily to create a potential plutonium pathway to a bomb. A covert effort remained our biggest concern.
We had substantial negotiating assets. UNSC sanctions, as well as U.S. and EU measures, were having a major impact. Markets for Iran’s oil exports were rapidly contracting, and Tehran was starved for hard currency, with over $100 billion in oil revenues frozen and inaccessible in overseas banks.
We were determined to get the most out of that leverage. A two-stage process still made the most sense, given the mistrust between us, and the urgency of freezing their nuclear progress to give us time to try to negotiate a comprehensive deal. We’d seek in an initial phase to stop the advance of the Iranian nuclear program, across all of its fronts, in return for no further nuclear sanctions. In addition, we’d try to roll back key aspects of their program, especially their 20 percent enrichment effort, in return for limited sanctions relief. We’d also seek initially to apply the most intrusive inspection procedures possible, across the entire supply chain, from uranium mines and mills to centrifuge production and storage. That would create a solid precedent for longer-term verification provisions in a comprehensive agreement.
The president convened a session shortly before we returned to Oman to give us our final guidance. Vice President Biden, Secretary Kerry, and Susan Rice, who had recently succeeded Tom Donilon as national security advisor, were there too, along with a tiny circle of officials who knew of the back channel. All the principals by this point knew the details of the nuclear issue and our approach. The president’s grasp of arcane technical details was impressive, as was John Kerry’s, who was eager to dive in himself at the right time.
As the president concluded the meeting, he motioned Jake and me over as he walked out the door of the Situation Room. “You guys know what needs to get done,” he said. “I trust you. So don’t screw it up.” He smiled slightly, but we couldn’t decide whether to feel buoyed by his confidence, anxious about his warning, or some of both. Those are the moments when you’d almost prefer the comfortable straitjacket of fourteen pages of single-spaced instructions.
The Iranian delegation this time was led by two deputy foreign ministers, Majid Takht-Ravanchi and Abbas Araghchi. In Zarif’s mold, both had done graduate degrees in the West, Ravanchi at the University of Kansas and Araghchi at the University of Kent in the United Kingdom. Both were tough Iranian patriots, skeptical of doing business with Americans and dogged in arguing their positions. We discovered that they could also be creative problem-solvers. In the hours and hours of conversations that followed, we exhausted the whole range of emotions, searching for practical solutions, occasionally pounding the table or walking out of the room, and sometimes even finding a little bit of humor in our shared predicament. While trust was always in short supply between Iranians and Americans, I developed considerable professional respect for Ravanchi and Araghchi, although I doubted that expressing that publicly would be career-enhancing for either of them.
Ravanchi and Araghchi were professional diplomats, not ideologues, but they were no less committed to Iran, no less proud of the revolution, no less determined to show that they could hold their own in the diplomatic arena. They were often guarded about the difficulties they faced at home, although they would sometimes confide that they had a Supreme Leader who was just waiting to say “I told you so” and prove that the Americans could not be trusted and that Obama was just as bent on regime change as Bush.
From the start, the atmosphere in the September round was much different, and more encouraging, than what we had experienced in March. We conducted the negotiations in English, without translation, which made for much easier and more informal discussions. Ravanchi and Araghchi were comfortable in their interactions with us, always careful to stay within the bounds of their instructions, but uninhibited in going back and forth in more formal plenary sessions or smaller conversations. They agreed that first day that a two-phase approach was best. We talked about the broad outline of a comprehensive deal, but agreed quickly that we’d get stuck if we tried at this early stage to get much beyond general principles. We spent most of our time on what it would take to put together a six-month interim accord. They had thoroughly digested our exchanges with Khaji. While they would regularly come back at the issue of a “right” to enrichment, they understood our position and didn’t belabor the point, at least at this stage.
The pattern for our meetings in the back channel soon took shape. We would begin with a five-on-five plenary session, and then break for separate conversations, sometimes one-on-one and often Araghchi and Ravanchi with Jake and me, while our colleagues would get into more detail on the limitations and verification measures we had in mind, and the sanctions relief at the front of the Iranians’ mind.
Our biggest challenge was countering Iranian expectations of the magnitude of sanctions relief that we could offer for the interim phase. We believed that the best tool we had for providing limited relief was the frozen oil revenue that was gradually building up in foreign banks, at a rate of roughly $18 billion every six months. Metering out a fraction of those funds would allow us to preserve the overall architecture of sanctions, and keep all the leverage we’d later need for comprehensive talks. Not surprisingly, the Iranians had a vastly different definition of what “limited” relief meant. They insisted stubbornly that all $18 billion should be released in return for their acceptance of six-month limitations on their program. We indicated early on that we could consider no more than about $4 billion.
Another problem was Iranian concern about how much they could count on a commitment by the U.S. administration not to enact new sanctions for six months, given the role of Congress. It was not an unreasonable worry. We explained at length how our system worked, and why we believed an administration commitment would hold up, assuming that Iran limited its program substantially along the lines we were discussing. But the Iranians were never entirely reassured by the formulas we offered, and the truth is that we weren’t either, given the uncertain state of American politics. “The best thing we can do,” I told Araghchi, “is make a solid agreement, and then live up to it scrupulously.” Those words would ring hollow a few years later.
While it was clear after this session that we still had a tough slog ahead of us, we could now see that a first-step understanding was possible. With relatively modest sanctions relief, we could freeze their program, and roll it back in some important respects. Rouhani and Zarif seemed to want to show their own critics at home that they could produce early progress and begin to ease sanctions pressure. We reported all this to the White House and Secretary Kerry, conscious of the value of underpromising and overdelivering. We also stressed the Iranians’ concern that we keep this channel secret. In our side conversations, both Ravanchi and Araghchi had worried that premature disclosure would torpedo the talks at home.
We agreed with the Iranians to meet next in New York later in September, when the annual UN General Assembly session would provide good cover for Ravanchi and Araghchi to come with Zarif, as well as Rouhani, who was making his first visit to the UN as Iran’s president. We had four rounds of talks over nearly two weeks in New York, and made considerable progress. Jake and I put an initial draft text on the table in our first discussion with Ravanchi and Araghchi on the evening of September 18, in a room at the Waldorf Astoria. We were acutely aware of the danger of negotiating with ourselves as time wore on, but it quickly became apparent that we could much more effectively drive the process by taking the pen. We also realized that the only way we could really be sure that we were making progress was to put notional understandings on paper. Araghchi in particular bristled at our continuing insistence that we couldn’t provide more than a small fraction of frozen Iranian oil revenue during the initial six-month period, and at our emphasis on mothballing nuclear infrastructure to reassure us that frozen centrifuges could not simply be reactivated. We kept at it, and over the next week painstakingly removed brackets around contested language and agreed on significant portions of the interim agreement.
As productive as the New York rounds were, they had their moments of minor drama too. Just as the Iranians were walking down the hall to our meeting room at the Waldorf, Jake and I noticed that hanging on the wall across from the room was a large framed photo of the shah visiting the Waldorf in the 1970s. We tried to take it down quickly, but it was firmly attached to the wall. Ravanchi and Araghchi didn’t seem to notice as we hurriedly ushered them through the door. The last thing we wanted was to offend our counterparts or inspire a forty-minute recitation about America’s support for the ancien régime.
We conducted several more rounds later in the week across town, away from the hustle and bustle of the United Nations meetings, at a hotel on Manhattan’s West Side. The Iranian delegation had no trouble moving in and out of the hotel quietly. With its kaleidoscope of humanity, Manhattan was one place where five guys with white shirts buttoned all the way up and no ties could blend in easily.
Our progress in the direct bilateral talks set the stage for Secretary Kerry’s first encounter with Zarif, on the margins of a P5+1 ministerial meeting at the UN on September 26. In their thirty-minute tête-à-tête, Kerry and Zarif reviewed the encouraging results of the back channel so far, and agreed that we should keep at it. It was the first half hour of what would be endless hours of face-to-face meetings, texts, and telephone calls between them; their relationship and drive was at the heart of everything that was later achieved.
Meanwhile, Jake was trying to explore the possibility of a meeting between Rouhani and Obama. The initial signals we received from Zarif and the Iranian negotiators were positive, but Rouhani and his political advisors got more concerned about the potential backlash in Tehran the more they considered the idea. Rouhani had already made quite a splash at the UN, sounding decidedly unlike Ahmadinejad as he acknowledged the Holocaust, and working with Zarif in a flurry of meetings and interviews to put a much different face on Iranian diplomacy. Once the Iranians began to press us to agree to preconditions for even a brief pull-aside encounter, invoking the familiar plea for some recognition of a “right” to enrich, it was apparent that the effort to engineer a meeting was not worth it.
Somewhat to our surprise, the Iranians came back to us with a proposal for a phone call between the two presidents on Rouhani’s last day in New York. There were no preconditions. The call was connected as Rouhani was in his car on the way to the airport. The brief silence during the connection felt like a lifetime. Not surprisingly, Jake was anxious—about whether he had been given the right number, or whether this was all a setup in which some radio host in Canada would pop up on the line in an elaborate prank. Finally, the call connected. Obama and Rouhani had a cordial fifteen-minute conversation. Obama congratulated Rouhani on his election, and stressed that they now had an historic opportunity to resolve the nuclear issue. Mindful that a variety of foreign intelligence services might be listening, Obama made only an oblique reference to the secret bilateral talks. Rouhani responded in the same constructive tone, closing with a somewhat surreal “Have a nice day” in English. The glimmer of possibility was steadily getting brighter.
We met twice more in October in the familiar confines of the Omani beach club. Sanctions relief remained a source of great irritation to the Iranians. We wouldn’t budge from our position of roughly $4 billion in relief over six months, far short of the $18 billion that the Iranians sought. We still had sharp differences over restrictions on the Arak heavy water facility, as well as over the continuing Iranian insistence on language about their “right” to enrich. Trying to anticipate some of the main lines of concern expressed by critics of the P5+1 process, we pressed relatively late in the game for a freeze on new centrifuge production, not just their installation. That set Araghchi off. “What are you going to demand next?” he asked, with an air of deep exasperation.
We talked at length with the Iranians about how best to handle the resumption of P5+1 meetings in Geneva in mid-October, in the middle of our extended back-channel talks that month. We both recognized that we were approaching the point where we would merge the two processes, but we had made surprising strides in the secret bilateral talks, and thought it was worth seeing how far we could get by the end of October. Araghchi suggested that Iran make a general presentation to the P5+1, laying out the broad contours of a two-phase approach, including interim and comprehensive agreements. The Iranians left out the details to which we had tentatively agreed, as well as the areas of continuing disagreement, but it was useful to introduce the framework. By the time we completed an intensive two-day back-channel session on October 26–27, we had a draft text that still had five or six contested passages, but that was beginning to resemble a solid step forward after so many years of tension on the nuclear issue. In a meeting with Sultan Qaboos just before we flew home from Muscat, I again expressed our appreciation for everything he had done to make this channel work. “We’re getting close,” I assured him.
THE PRESIDENT WAS pleased with the progress we had made, but intent in his lawyerly way on “buttoning it down tight.” We were genuinely surprised that we had come such a long way in such a short time. We were equally surprised that the back channel had stayed secret through eight rounds. We realized that that would not last much longer, with at least two journalists already beginning to put some of the pieces together.
A new P5+1 round under the leadership of EU high representative Cathy Ashton was scheduled to begin on November 7 in Geneva. With Wendy Sherman, my exceptional successor as undersecretary for political affairs and as head of our P5+1 delegation, joining us for the late October back-channel talks in Oman, we had told the Iranians that we would inform our multilateral partners of our direct bilateral meetings before the November session. They were a little nervous, but understood that the time had come. We scheduled another back-channel round on November 5–7 to see if we could remove another bracket or two in the draft text, before turning it over to our P5+1 partners for their consideration as the basis for rapid completion of an interim agreement.
Wendy had the unenviable task of briefing our P5+1 colleagues on the back-channel effort. The debate about when to tell our closest allies about the secret talks had been extensive. I was torn, having spent years as undersecretary working with my P5+1 counterparts, and understanding the very real concerns of the Israelis and our Gulf Arab partners, but also acutely aware of the risks of leaks and premature public disclosures. The White House preferred, in any case, to hold off as long as we could in the fall of 2013, but by the end of October there was no longer any good reason to wait.
Starting with Ashton, who knew as well as anyone that bilateral U.S.-Iran talks were essential, Wendy laid out the quiet effort we had been making, and the main areas of agreement and disagreement with the Iranians. Some of our partners were not entirely surprised. The British government, for example, had excellent contacts in Oman, and was generally aware of our progress. The president had also taken Prime Minister Netanyahu into his confidence at the end of September, in a one-on-one conversation at the White House. Netanyahu was not surprised either, since the Israelis had their own sources in the region, but he was decidedly less understanding than the Brits. He saw our back channel as a betrayal.
On November 5, we met with the Iranians at the Mandarin Hotel in Geneva, on the other side of town from the InterContinental, where the P5+1 session would take place a couple days later. The president and Secretary Kerry told us to make a final push to improve the draft text, which now bore the suitably anodyne title “draft joint working document.” We made a little more progress on defining a “pause” on Arak, but still had bracketed language there. We had made quite a bit of headway on specifying the elements of a freeze on enrichment at Natanz and Fordow, and on conversion and dilution of Iran’s existing stockpile of 20 percent enriched material. We were close to an understanding on sanctions relief in return, at roughly the $4 billion figure over six months that we had set out at the start of the back-channel talks. We also settled on an unprecedented set of verification and monitoring measures that would serve as a solid foundation for much more detailed arrangements in an eventual comprehensive agreement. The draft text we produced with Ravanchi and Araghchi still had three or four difficult brackets to resolve in its four and a half single-spaced pages.
It was probably inevitable that the handover to the P5+1 would have its awkward moments. Some of our European colleagues were impressed by our progress, but not happy about being kept in the dark. Ashton did a superb job of focusing the group on the opportunity the draft text offered. With Zarif already in Geneva to take charge of the Iranian team, John Kerry flew in on November 8. French foreign minister Laurent Fabius was close on his heels, bringing both considerable Gallic ego, a bit bruised over the back channel, and some solid ideas on how to tighten language, especially on Arak. Sergey Lavrov and the other ministers flew in too.
The next couple of days had lots of drama, some contrived and some reflecting real frustration, emotion, and exhaustion. Pressures were building in Washington for another round of sanctions, and Zarif faced his own share of domestic suspicions and second-guessing. Some of our P5+1 partners were still smarting over the back channel. Ashton and Kerry skillfully defused most of the tensions within the P5+1, and we developed a revised text that the group supported. It built on the back-channel draft, filling in new proposed language in some of the bracketed areas, and adding a few new sentences.
Zarif was not thrilled to see this updated text on November 9. The Iranians knew that the bilateral draft we had been working on for months still had brackets with unresolved differences over language. They also knew that it would have to be reviewed and accepted by the rest of the P5+1, who would undoubtedly want to put their own stamp on it. As Zarif reminded Kerry, he faced a tough audience in Tehran, and any shifts in language, however minor, were troublesome. Like other accomplished diplomats, Zarif was also a gifted thespian, and his head-in-hands expressions of gloom and duplicity unsettled some of the other ministers.
After a long day and night of discussions, the ministers agreed to consult in their capitals and convene another, hopefully final, round of talks in Geneva on November 22. The back channel had still not become public, and we worried that their revelation would complicate completion of an interim agreement. Jake and I arrived back in Geneva on November 20 to help bridge the final gaps with the Iranians. Coordinating closely with Ashton, and joined by Sherman, we met with Araghchi and Ravanchi on the twenty-first. We further narrowed our differences. The Iranians seemed more relaxed about preambular language on enrichment, in which we had carefully separated the words “right” and “enrichment,” using the first to refer explicitly to NPT language on the widely acknowledged right of members in good standing to peaceful nuclear energy, and the second in the much more conditional sense of an Iranian demand that might be applied if mutually agreed, long-term limitations on its program were developed. We made some headway on Arak, as Araghchi and Ravanchi grudgingly accepted French edits to more tightly define a cessation of construction activity at the site.
We also pinned down an excellent set of verification measures, including 24/7 surveillance arrangements at Natanz and Fordow, and access to each step along the nuclear supply chain. On sanctions relief, we wound our way toward the formula that Kerry and Zarif eventually agreed upon. Its core was $4.2 billion in unfrozen Iranian oil revenue, metered out in six monthly installments. It had a few additional provisions, notably a relaxation of sanctions on the auto industry, whose main beneficiary was French automaker Renault. Zarif recounted to us with a mischievous glint in his eye that Fabius had spent most of their bilateral sessions in Geneva on this issue, not on Arak and the other questions on which he had been so voluble in public.
Throughout November 22 and 23, John Kerry was his usual relentless self, nudging Zarif toward the finish line, working with Ashton to manage the P5+1, and staying in close touch with the president by secure phone. Jake and I came over to the InterContinental Hotel for the final push, using service elevators and stairwells to get up to the secretary’s suite. Our cloak-and-dagger seemed a little silly at this stage, but it had become habitual over the past eight months, and we figured it was worth it if we could keep the back channel under wraps until an interim agreement, now termed the “Joint Plan of Action,” could be reached.
We could sense that the Iranians were in a hurry to finish the deal, before their own politics became an even bigger impediment. We didn’t think we needed to concede anything further on sanctions, and were confident that we had succeeded in preserving most of our leverage for the much more complex task of negotiating a comprehensive accord. Borrowing a famous Mel Gibson line from the movie Braveheart, as he urged his Scottish compatriots to stand firm in the face of charging English cavalry—and with a little of the giddiness that comes from high stakes and little sleep—Jake and I kept repeating to each other “Hold, hold, hold” as the Iranians kept probing for concessions.
By 2 A.M. on November 24, we were nearly there. The ministers were straining one another’s patience by this point, and I met with Ravanchi to iron out the last bits of language. Tired and relieved, we quietly congratulated one another. Ashton mobilized all the ministers for a signing ceremony at 4 A.M. Araghchi called me thirty minutes before the ceremony to say that he had “just two or three more changes to make” in the text. The Iranians were never entirely satisfied until they had overreached on nearly every issue and tested every last ounce of flexibility. I laughed politely. “It’s a little late for that,” I said. “We’re done.”
The Joint Plan of Action (JPOA) was a modest, temporary, and practical step. Iran froze its nuclear program for an initial six months, and rolled it back in key respects, especially in disposing of its existing stockpile of 20 percent enriched uranium. It accepted intrusive monitoring arrangements. In return it got limited sanctions relief, and a commitment not to increase sanctions for six months.
The JPOA aroused more than a little controversy. Prime Minister Netanyahu said publicly on November 24 that it was “the deal of the century” for Iran. I told Ravanchi that hyperbolic statements like that should help his selling job at home, and he smiled with some satisfaction. Then the congressional critics joined in, predicting that the Iranians would cheat and the whole edifice of sanctions that we had so painstakingly put together over the years would collapse, long before a comprehensive deal could be negotiated. None of that turned out to be true. The JPOA was a solid agreement, in many ways better for us than for the Iranians, who still faced huge economic pressure. It offered us and the Iranians an opportunity to show that we could actually each live up to our sides of a fair bargain, and it gave the president and Secretary Kerry the time and space to negotiate a final agreement.
NEWS OF OUR back channel broke a few hours after the signing of the JPOA, helping to explain how the P5+1 and Iran had concluded the interim deal so quickly. Spent after this long effort and more than three decades in the Foreign Service, I intended to retire at the end of 2013. I had promised John Kerry when he asked me to remain as deputy secretary that I’d stay on for his first year. In the end, encouraged by him and the president, and admiring them both immensely, I would keep at it for an additional year, until late 2014.
I was especially touched when President Obama invited me to lunch at the White House to reinforce the case for continuing at State. He was an adroit closer. We sat in his small private dining room just off the Oval Office, with tall windows looking out onto the Rose Garden. Over a relaxed conversation, we covered everything from our daughters and the current NBA season to the Iran negotiations and the state of the State Department. “I don’t want to play on your Irish Catholic guilt,” he said, “but I consider you to be the ultimate professional, and it would mean a lot to me if you would stay for another year.” I noted that he was doing a pretty good job on the Irish Catholic guilt part—and that he had had me at the lunch invitation.
With the back channel now history, Jake and I played a supporting and episodic role in the negotiations for a comprehensive nuclear agreement that consumed 2014 and the first half of 2015. In all those hours and days of secret talks, we had built some rapport with Araghchi, Ravanchi, and their other colleagues, as well as with Zarif. While the Iranians knew that the road to a comprehensive deal went through the P5+1, it was also clear that what were now overt and frequent U.S.-Iran contacts were the core of the effort. Even the distinctly unsentimental Iranians could get a little nostalgic sometimes about the seemingly simpler days of the back-channel talks in 2013.
Secretary Kerry threw himself into the comprehensive process, and he and Zarif were its prime movers. Wendy was tireless, and a deft leader of a vastly expanded negotiating team, including Timbie and Roule and terrific experts from Treasury, Energy, and other departments. Energy Secretary Ernie Moniz’s nuclear expertise and creativity helped to bridge gaps with his Iranian counterpart, Ali Salehi, a fellow MIT alum. I joined our team a few times in the cramped confines of the Palais Coburg hotel in Vienna, where both the slow rhythm of multilateral negotiations and the buffet menu became very familiar. At Kerry’s request, I saw Zarif privately a couple times in the second half of 2014. Before marathon talks in Lausanne in the spring of 2015, I met quietly in Geneva with Araghchi and Ravanchi. With congressional impatience and appetite for new sanctions growing, and the Iranians backtracking on key issues, I told them bluntly, “We have come so far, but maybe we should start thinking about a world without an agreement.” That helped get their attention.
Kerry’s talks with Zarif and Ashton in Lausanne in late March and early April 2015 were the longest continuous negotiation that a secretary of state had engaged in since Camp David in 1978. A framework for a Joint Comprehensive Plan of Action (JCPOA) was announced on April 2, and the final deal emerged in July. In return for the gradual lifting of sanctions, Iran made a permanent commitment never to develop a nuclear weapon, and accepted substantial, long-term limitations on its civilian nuclear program. Ninety-eight percent of Iran’s stockpile of enriched material was removed, and so were nearly two-thirds of its centrifuges. The deal also cut off Iran’s other potential pathways to a bomb, eliminating the heavy water reactor core at Arak and the capacity to produce weapons-grade plutonium. Extensive verification and monitoring measures were put in place, some of them permanent. For the next decade, at least, Iran’s “breakout time”—the time it would theoretically take to enrich enough weapons-grade uranium for a bomb—was extended from the two or three months frozen in the JPOA to at least one year. We had achieved our objective, and we diverted a potential path to war.
IT WAS HARD to imagine when we embarked on that first secret flight to Oman in early 2013 that diplomacy could resolve the Iranian nuclear issue, at that time the most combustible challenge on the international landscape. The even longer history of grievance and suspicion in America’s relations with Iran was another massive obstacle. The politics in Tehran and Washington were corrosive, offering little room for maneuver or incentive for risk-taking. The nuclear problem itself was maddeningly complicated and opaque. There was little reason to think that we could overcome any one of those obstacles, let alone all of them.
Neither the JPOA nor the JCPOA were perfect agreements. In a perfect world, there would be no nuclear enrichment in Iran, and its existing enrichment facilities would have been dismantled. But we don’t live in a perfect world, and perfect is rarely on the diplomatic menu. We couldn’t neatly erase by military or diplomatic means Iran’s basic know-how about enrichment. What we could do was to sharply constrain it over a long duration, monitor it with unprecedented intrusiveness, and prevent its leadership from building a bomb.
For all its trade-offs and imperfections, this was a classic illustration of how diplomacy can work. We set out at the beginning of the Obama administration, building on tentative steps taken at the end of the Bush administration, to test Iranian seriousness directly and invest in a wider coalition, and to build a stronger sanctions program. Our willingness to engage in direct talks and think creatively was a critical ingredient. It put the Iranians on the defensive, removed a pretext for their inaction, and solidified our coalition. When Tehran proved unwilling or incapable, it gave us the opportunity to build substantial economic leverage. Always lurking just over the horizon was the reality of American military power, backing up our determination to ensure that, by one means or another, Iran would not develop a nuclear weapon.
When our leverage had reached a kind of critical mass, we had to use it or risk losing it. Sanctions had so much impact on Iran because they were international, and widely, if often grudgingly, supported. Once Rouhani and Zarif took office and portrayed Iran in a more pragmatic and sympathetic light, it was time to put diplomacy to a rigorous test. Framing the issue as a question of whether Iran could accept sufficiently tough, long-term constraints in return for sanctions lifting and the possibility of limited domestic enrichment was key. There would have been no agreement without sharp constraints and strong monitoring—but there would also have been no agreement if we had insisted on zero enrichment. As Araghchi once put it to us, a civilian nuclear program, including enrichment, was “our source of national pride, our moon shot.”
In the first few years after completion of the JCPOA, contrary to the prediction of its opponents that Iran would cheat, the IAEA and the U.S. intelligence community repeatedly affirmed Iranian compliance. Iran’s economy did not become a juggernaut as a result of sanctions relief. The agreement deprived the regime of the argument that outside pressure—not chronic mismanagement, corruption, and misallocation of resources—was the source of the grim economic circumstance of most Iranians. Widespread protests in the summer of 2017 demonstrated that the clerical leadership was not sitting comfortably in Tehran. Much as the Supreme Leader feared during the nuclear negotiations, the deal had exposed the regime’s vulnerabilities, not erased them.
Meanwhile, Iran continued to export instability across the Middle East, exploiting and accelerating chaos in Syria and Yemen, its forces and proxies locked in a bitter regional competition with Saudi Arabia and other Sunni Arab states. President Obama had always understood that the nuclear agreement would have to be embedded in a wider strategy for reassuring our friends and partners, who were unnerved by the prospect that dialogue with Tehran might someday temper our support for them. The nuclear deal explicitly reserved the option for the United States and its partners to take measures against the Iranian government for non-nuclear transgressions; but it was still tempting for critics to caricature the administration’s approach as constraining Iran’s nuclear ambitions but enabling its regional troublemaking.
Donald Trump came into office with visceral contempt for the JCPOA, which he called “the worst deal ever.” He was dismissive of its practical merits in limiting Iran’s nuclear program, and of the whole notion that there was value in the classic diplomacy of building coalitions and hammering out negotiated solutions, with all the give-and-take they required. His was a much more unilateralist impulse, aimed not so much at a better deal with the Iranians as at squeezing them so hard that they’d either capitulate or implode. Despite the entreaties of other P5+1 players, and despite zero evidence of Iranian noncompliance, Trump pulled the United States out of the JCPOA in May 2018.
I was surprised only that he had taken so long to withdraw, given the vehemence of his views. It was nevertheless a dispiriting moment, after years of effort to produce an agreement in which I continued to believe firmly. I wondered what we might have done differently to better insulate the deal. Perhaps we could have pressured the Iranians longer through the interim accord, the JPOA, and extracted more concessions from Tehran—on the duration of certain enrichment restrictions, for example. But the reality was that politics in both the United States and Iran were tortured and impatient, and it was always a lot harder than it looked from the outside to hold the P5+1 together, especially after serious rifts began to emerge over other problems, like Ukraine or the South China Sea.
We could have done a better job, both before and after the comprehensive nuclear agreement was reached, of confronting the wider challenge of Iran in the Middle East. A willingness to take more risks against the Assad regime after the Syrian civil war began in 2011 would have sent a strong signal to Iran, and cushioned the disquieting effect of the nuclear deal for the Saudis and our other traditional friends. Some of their angst, however, was simply unavoidable. They were deeply worried by the tumult of the Arab Spring, and the prospect of an eventual regional order in which Iran couldn’t be denied a place. But we could have done more to show that the nuclear agreement was the start, not the end, of a tough-minded policy toward Iran.
It certainly would have helped shield the JCPOA from Trump’s decision to withdraw if we had been able to anchor it better politically at home. It would have been harder to undo as a formal treaty than as an executive agreement. In a deeply polarized Washington, however, the two-thirds affirmative vote in the Senate required for a treaty was virtually impossible. The fact that public opinion polls showed 60 percent of Americans were opposed to withdrawing from the nuclear agreement was not a sturdy enough defense.
Trump’s abrogation was another reminder of how much easier it is to tear down diplomacy than to build it. Pulling out of the nuclear deal alienated allies who had joined us in the effort for many years. Reimposition of U.S. sanctions in the face of opposition from partners further damaged a tool of policy already suffering from abuse, driving other countries to lessen reliance on the dollar and the U.S. financial system. It also betrayed an obsession with Iran that exaggerated its strategic weight and undermined larger priorities like rebuilding alliances or managing great power rivals.
Trump’s demolition of the Iran deal was a further blow to our own credibility, to international confidence that we could keep our end of a bargain. “Credibility” can be an overused term in Washington, a town sometimes too prone to badger presidents into using force to prop up our currency and influence around the world. But it matters in American diplomacy, especially at a post-primacy moment when our ability to mobilize others around common concerns is becoming more crucial. With its echoes of the muscular unilateralism on the road to the Iraq War in 2003 and the seductive appeal of remaking regional order through American power, the decision to abandon the JCPOA signaled anew a dangerous dismissiveness toward diplomacy. It was exactly the kind of risky, cocky, ill-considered bet that had shredded our influence before, and could easily do so again.