The Maximalism Trap: How All-or-Nothing Diplomacy Undermines Nuclear Security

Nuclear negotiations rarely collapse because states have nothing to gain from an agreement. They collapse because the terms of agreement are framed in ways that make agreement impossible.

The pattern is visible in the Iran talks. Since the earlier days of negotiations Washington demanded the elimination of uranium enrichment, surrender of existing stockpiles, and limits on Iran’s missile program. Tehran has long treated those conditions as matters of sovereignty and deterrence. When negotiations require complete capitulation before cooperation can begin, the space for agreement disappears. What remains is coercion without constraint.

Our lab’s forthcoming systematic analysis of 43 arms control negotiations since 1945 points to a consistent finding: how states respond to uncertainty shapes outcomes more than the presence of uncertainty itself.

States facing deep uncertainty about an adversary’s capabilities and intentions tend toward one of two strategies. Some manage uncertainty incrementally — through information exchange, reciprocal steps, and transparency measures that reduce risk without requiring either side to surrender what matters most. Others seek to eliminate uncertainty entirely, insisting on comprehensive settlements that close every avenue of concern before agreement becomes possible. The first approach produces agreements. The second does not.

The 2015 Joint Comprehensive Plan Of Action (JCPOA), also known as the Iran Nuclear Deal, followed the first logic. It did not eliminate Iran’s nuclear capabilities. It constrained them. 

JCPOA capped enrichment levels, reduced stockpiles, limited centrifuge numbers, and gave international inspectors broad access to Iranian facilities in exchange for sanctions relief. Iran retained a nuclear program — but one that was monitored, restricted, and slower to move toward weaponization. The agreement reduced immediate risks while leaving the broader political rivalry unresolved. That was by design.

After the U.S. withdrawal from the JCPOA in 2018, the frame shifted. The question was no longer how to constrain Iran’s nuclear program, but whether Iran would accept a fundamental restructuring of its deterrence posture. The bargaining space collapsed.

The pattern is not unique to Iran. At the dawn of the nuclear age, the Baruch Plan proposed placing atomic energy under international control. Washington insisted on inspections before dismantling its nuclear monopoly; Moscow demanded disarmament first. Each required the other to concede entirely before cooperation could begin. Neither moved. The arms race followed.

The 2019 Hanoi summit offered a more recent version. Washington demanded sweeping denuclearization from North Korea. Pyongyang offered to dismantle the Yongbyon nuclear complex in exchange for sanctions relief — a limited but meaningful step. Neither side accepted the incremental option on the table. The talks ended without agreement.

The 1987 INF Treaty appears to stand apart: an agreement that eliminated an entire class of nuclear weapons and held for over three decades. But its success was not the product of maximalist demands. Concessions were symmetrical — both sides removed the same category of weapons that most directly threatened the other. The scope was narrow; verification was extensive. What looked like elimination was, really, a masterclass in uncertainty management.

The pattern across cases is consistent: nuclear weapons sit at the center of national security strategy and domestic politics. Governments rarely surrender capabilities that underpin their deterrence under external pressure. When negotiations demand comprehensive settlements before cooperation can begin, they remove the space that partial agreements require. They also destroy what often matters most: the incremental process through which adversaries build just enough confidence to take the next step.

What maximalism produces instead is a particular kind of stalemate — one that tends to harden over time. States that survive cycles of pressure learn to plan for resilience rather than compliance. They build redundancy, preserve technical expertise, and shorten timelines. Each cycle of coercion without settlement makes the next round of diplomacy harder.

Arms control has rarely advanced through grand bargains. The Limited Test Ban Treaty restricted nuclear testing without ending the arms race. SALT placed ceilings on strategic weapons without eliminating them. Incremental agreements allow rivals to cooperate without abandoning core interests. They create transparency, establish monitoring, and build the precedent upon which further steps become possible. Maximalist diplomacy works against those mechanisms — not just by failing to produce agreements, but by eroding the foundations on which future agreements could rest.

Today, the arms control architecture is under pressure from multiple directions. The last remaining bilateral treaty between the United States and Russia has expired. China is expanding its arsenal outside formal constraints. Regional nuclear crises are intensifying. In that environment, the temptation to demand sweeping solutions is understandable.

But the historical and empirical record points in the same direction: the conditions that make agreements most urgently needed — deep uncertainty, rapid military change, unstable political environments — are the same conditions that make maximalist demands most tempting and least likely to succeed.

Arms control has never worked by eliminating rivalry. It has worked by learning to manage it. That difference is not semantic. It is the precondition for stability.

This piece draws on research from The Arms Control Paradox (forthcoming from Stanford University Press, 2026) and a systematic analysis of 43 postwar arms control negotiations.