IN 1968, THE “TREATY on the Non-Proliferation of Nuclear Weapons,” more commonly known as the Non-Proliferation Treaty (NPT), was negotiated into existence. The NPT was designed to restrict the number of countries that would have access to nuclear weapons to the five countries that had them at the time: the United States, the United Kingdom, the Soviet Union, France, and China. Not coincidentally, these were also the five permanent members of the United Nations Security Council. The long-term vision of the NPT was that signatories would commit (a) not to engage in proliferation activities, (b) to eventual disarmament among those who currently had nuclear weapons, (c) to supporting the peaceful use of nuclear technology for all signatories, and (d) to submit to inspections and safeguards by the International Atomic Energy Agency (IAEA) to ensure safety and compliance.
By the turn of the century, 190 countries had signed the NPT, with the only holdouts at the time being North Korea, Israel, Pakistan, and India.1 Those who refused to sign argued, as did many signatories, that given insufficient commitment to disarmament by the nuclear haves, the treaty was simply suppressing the sovereign and strategic rights of nuclear have-nots. In the years since the NPT went into effect, each of these four nonsignatories had, with varying degrees of success, developed its own nuclear weapons.
In July 2005, the United States and India set in motion what would become a three-year marathon of interrelated negotiations aimed at completing a “civilian nuclear agreement” between the two countries.2 The premise was relatively straightforward: India would agree to separate its military and civilian nuclear facilities and place the latter under IAEA safeguards in exchange for full civil (i.e., nonmilitary) nuclear cooperation (e.g., commerce) by the United States and the then-45-nation Nuclear Suppliers Group (NSG). India’s status as a nonsignatory to the NPT, however, made the negotiations difficult—some would have said inconceivable. To allow India to engage in civil nuclear commerce would, in the view of many, undermine American commitment to the NPT. If nonsignatories would be treated as well as signatories, what incentive would there be for anyone to sign? In contrast, the Bush administration and others among the NSG believed that despite being a nonsignatory of the NPT, and despite having developed its own weapons, India had not engaged in proliferation activities. Allowing it to participate in civilian nuclear commerce in exchange for some degree of IAEA inspections and safeguards would only promote continued responsible behavior and greater safety.
Reaching such an agreement was meant to be difficult. Negotiations would need to be coordinated and sequenced at many levels, across the globe. First, the United States would have to pass domestic legislation allowing it to engage with a nonsignatory of the NPT (accomplished via the Hyde Act in 2006). Then, the United States and India would have to negotiate a bilateral agreement (referred to as a 123 Agreement). Meanwhile, the IAEA would have to approve an agreement with India to place Indian civil nuclear facilities under IAEA safeguards, and the NSG would have to grant India an unprecedented waiver to allow it to have access to nuclear technology and fuel. Finally, the agreement between US and Indian diplomats would have to be approved by the US Congress and supported by India’s Parliament.
One of the more vexing problems that arose during these negotiations pertained to the consequences that would follow if India tested another nuclear weapon. In 1998, India had, to broad international condemnation, conducted five nuclear tests, resulting not only in sanctions by the United States and others, but also in retaliatory nuclear tests (for the first time) by Pakistan, a mere two weeks later. A year later, in 1999, due to Pakistani military incursions through the “Line of Control” separating India and Pakistan in Kashmir, the two had waged history’s first and only conventional war between known nuclear powers. With this terrifying backdrop in place, it is no surprise that support for the civil nuclear agreement among US lawmakers and many NSG nations was contingent upon guarantees that India would not test another nuclear weapon.3
Meanwhile, support in India was contingent upon exactly the opposite. There seemed no possibility that India’s Parliament would approve a deal if the agreement limited their perceived sovereign right to test nuclear weapons if and when they felt it was necessary. Indeed, this was the very reason India had not signed the NPT in the first place: a civil nuclear agreement that imposed NPT-type restrictions was entirely unacceptable. India had announced a voluntary moratorium on testing but was unwilling to make the moratorium binding.
How do you negotiate an agreement when the exact same issue is a deal breaker for both sides? How can you reconcile the interests of both parties when the minimum requirements of one (based on the logic of international security) are entirely unacceptable to the other (based on the logic of national sovereignty)?
In 2007, the United States and India negotiated their bilateral agreement; in 2008, the government of Prime Minister Singh survived a no-confidence vote in the Indian Parliament, the IAEA approved the safeguard agreement, and the 45-nation NSG granted its waiver. Later that same year, the US Congress approved the deal, and the two countries officially signed it on October 10, 2008.
How did this happen? Which side subordinated its demands and accepted the other side’s logic? Who made the courageous concession? It turns out no one did.
So did the 123 Agreement signed by the United States and India restrict nuclear tests? Did it stipulate the termination of nuclear commerce if India detonated a nuclear device? No one could say for sure.
On October 1, 2008, US Secretary of State Condoleezza Rice gave testimony to the Senate in which she declared, “Let me reassure you that an Indian test, as I have testified publicly, would result in most serious consequences. Existing US law would require automatic cutoff of cooperation, as well as a number of other sanctions, if India were to test.”4
On the other side, on October 3, 2008, Indian External Affairs Minister Pranab Mukherjee, when asked whether India had sacrificed the right to test, clarified that “we would not like to convert this voluntary moratorium into a treaty-bound obligation. That position has been maintained.”5
What is going on? What does the agreement actually say? If anyone ought to know, it would be Secretary Rice and Minister Mukherjee, the two people who signed the final agreement on October 10, 2008. The fact is that the 123 Agreement, and the network of agreements on which the US–India deal is structured, are deliberately vague. But the lack of precision is by design: this is the art of strategic ambiguity.
Strategic ambiguity is a risky tactic that can pay dividends when used at the appropriate time in the appropriate way. It is risky because it creates an agreement that can be interpreted differently by different parties—we will revisit this problem shortly. But multiple interpretations can also be valuable. This is because sometimes the problem isn’t that the two sides cannot live with each other’s demands, but that writing down or announcing explicitly what you’re willing to live with is too problematic.
In this case, negotiators in the United States and India understood that any agreement, no matter what language was used to write it down, would involve both sides knowing that if India tested another weapon, which of course it could technically do, the United States would be forced to terminate the agreement due to domestic and international pressure. It really did not matter what was or was not in the signed agreement. The practical reality was that there was no way to stop India from testing a nuclear weapon if it wished, and there was no way to stop the United States from pulling out of the agreement if India tested. Knowing this would be the American reaction was the best incentive for India not to test a weapon in the first place. In other words, incentives were aligned and everyone in the room agreed on everything—there was no misunderstanding. Yet, putting the deal in writing would be problematic. For weeks negotiators labored over how to draft language that would be acceptable to each side given its constraints. Any language akin to “if India tests a nuclear weapon …” was a nonstarter in India; the absence of such language was unacceptable in the United States. The eventual solution was an approach that runs contrary to the instincts of most lawyers: the agreement had to be sufficiently imprecise, allowing each side to interpret the agreement and present it to its constituency in the most favorable way.
To appreciate the role of strategic ambiguity in the negotiator’s tool kit—and to help draw some clear boundaries on when it is advisable—we need to first distinguish between different kinds of strategic relationships. There are those in which one or both parties have the incentive and ability to exploit the other—and will do so—unless a contract or treaty makes this impossible or extremely costly. In such a case, it is wise to have an agreement that clearly delineates the rights and responsibilities of each side and clarifies what behaviors are proscribed. Strategic ambiguity should be avoided in these instances. In contrast, there are relationships in which mutual interests are aligned enough so that the relationship is self-sustaining regardless of what you write down. Here, you have the flexibility to keep the contract incomplete or ambiguous if doing so helps you solve other problems (e.g., optics). In other words, strategic ambiguity should be limited to situations where other mechanisms are in place to enforce appropriate behavior from each party. India’s External Minister Pranab Mukherjee made it quite clear that the nuclear negotiations fit this criterion—that is, that the deal was inherently self-enforcing based on each side’s rights and interests—when he said, days before the signing of the agreement: “We have the right to test; others have a right to react.”6 This is not the kind of language that could go into the agreement but was perhaps suitable for a domestic audience. More generally, when the shadow of the audience looms large, people are sometimes willing to accept certain de facto limitations being imposed on them in a strategic relationship, but are unwilling to acknowledge or substantiate these in writing.
Strategic ambiguity can also be useful, ironically, when there is too little trust at the outset for the parties to reach a comprehensive agreement. Instead, parties may reach an initial deal that is incomplete or ambiguous, but which helps them stay engaged until trust is built. For example, in cross-cultural deal making, if a party fears that tying their hands with a long-term relationship is too risky, given the lack of a partner’s prior track record, they will avoid signing a deal that is too committal. But being explicitly noncommittal may also muddy the waters. For instance, consider a situation in which Company X is sourcing from a new manufacturer, Company Y, but is unwilling to commit to a multiyear contract or make any other commitments to offset Company Y’s investment in the deal. Company Y can live with the reality that, at the end of the day, if Y does not deliver a great product, X will look elsewhere. And yet, an agreement that is explicitly noncommittal, with numerous provisions that delineate all of the ways in which X is not responsible for the fate of Y, could send a very negative signal or force the parties to confront and get caught up haggling on issues that cannot be easily resolved. Introducing some degree of ambiguity regarding the nature of the relationship and the strength and length of commitment can provide both parties with the flexibility and freedom they need to overcome initial hesitations and pursue early-stage collaboration with the comfort of “no strings attached.” Even when both sides will benefit from behaving appropriately, some arrangements remain easy to understand and agree to in principle, but are difficult to write down with precision, especially at the outset of what might be a long-term, evolving relationship.
It is worth highlighting that in each of the situations we’ve considered, strategic ambiguity is not meant to substitute for a genuine and durable understanding between the two sides. If there are deep divisions on key issues of substance, strategic ambiguity not only will fail to help but may make matters worse. This is the topic we now visit in more detail, as we consider the potential misuse of framing to resolve conflict.