Broken Treaties: The Rise, Fall, and Future of Nuclear Arms Control Agreements

• Introduction
• Chapter 1 Why Arms Control Matters in a Fragmenting World
• Chapter 2 Deterrence and Restraint: The Architecture of Nuclear Order
• Chapter 3 From Atoms for Peace to the NPT: Origins and Grand Bargains
• Chapter 4 Inside the NPT: Safeguards, Peaceful Uses, and Disarmament Obligations
• Chapter 5 Compliance and Crisis: Iraq, Iran, and North Korea as Stress Tests
• Chapter 6 SALT I: The ABM Treaty and the Logic of Limits
• Chapter 7 SALT II: Negotiation, Non‑Ratification, and Enduring Legacies
• Chapter 8 The INF Treaty: Verification Breakthroughs and Eventual Collapse
• Chapter 9 START I: Deep Reductions and the Inspection State
• Chapter 10 START II and the Road Not Taken: MIRVs, Defense, and Doctrine
• Chapter 11 The Moscow Treaty (SORT): Bridging to a New Framework
• Chapter 12 New START: Counting Rules, Extensions, and Suspension Dilemmas
• Chapter 13 The Comprehensive Nuclear-Test-Ban: Science Diplomacy without Entry into Force
• Chapter 14 Test Moratoria: Politics, Signaling, and the Shadow of the CTBT
• Chapter 15 The Treaty on the Prohibition of Nuclear Weapons: Norms, Contestation, and Impact
• Chapter 16 Open Skies and Cooperative Transparency: The Rise and Retreat of Confidence-Building
• Chapter 17 Supply-Side Restraint: NSG, MTCR, Wassenaar, and the Australia Group
• Chapter 18 Verifying the Invisible: Satellites, Seismology, and Zero‑Knowledge Warhead Checks
• Chapter 19 Missile Defense, Prompt Strike, and the Offense–Defense Spiral
• Chapter 20 Cyber, AI, and Nuclear Command and Control: New Risks to Stability
• Chapter 21 Regional Nuclear Orders: Europe, South Asia, and Northeast Asia
• Chapter 22 Domestic Politics and Ratification: Legislatures, Courts, and Public Opinion
• Chapter 23 How Agreements Are Made: Negotiation Playbooks and Crisis Leverage
• Chapter 24 Pathways to Renewal: Risk Reduction, Transparency, and Incremental Caps
• Chapter 25 A Roadmap for Advocates and Policymakers: From Ceasefires to Codes and Treaties

Nuclear arms control has never been a synonym for trust; it is a method of managing distrust. From the earliest Cold War encounters to today’s multipolar competition, states have limited, counted, inspected, and sometimes dismantled the most destructive devices ever built—not because they became friends, but because they learned to make rivalry survivable. This book examines how that project rose, faltered, and can be repaired. It argues that durable arms control emerges when legal design aligns with political incentives and when verification technologies credibly measure what matters.

The story begins with the grand bargains that created the modern nuclear order: the Non‑Proliferation Treaty’s promise of peaceful access and eventual disarmament, the Strategic Arms Limitation and Reduction processes that capped and then cut strategic arsenals, and the global effort to end explosive testing through the Comprehensive Nuclear‑Test‑Ban. These achievements were not inevitable. They were forged amid crises, domestic backlash, and intense technical debate over what could be counted and how. Their successes teach that precision in definitions and transparency in implementation are as important as high‑level geopolitics.

Yet treaties do not fail only when parties cheat; they also falter when the world around them changes. Technological shifts—multiple warhead missiles, missile defense, precision conventional strike, cyber operations against command and control, and the algorithmic acceleration of decision cycles—can unsettle the cost‑benefit balance that once made restraint attractive. Political shocks—alliances strained, regional rivalries sharpened, legislatures polarized—can render ratification or continued compliance harder than withdrawal. Legal architecture must therefore be resilient enough to adapt to new systems, new actors, and new information.

This book treats arms control as a three‑legged stool: law, politics, and technology. Chapters on treaty design show how definitions, counting rules, and verification clauses structure incentives; case studies trace how leaders converted leverage into agreements; and technical chapters explore how inspectors, satellites, seismology, and emerging zero‑knowledge methods create credible verification without exposing secrets. Together, these perspectives explain both compliance and collapse across landmark agreements—from SALT and START to the test‑ban enterprise and newer normative projects.

Readers will also find that “success” is broader than ratified treaties. Moratoria, executive agreements, confidence‑building measures, and transparency regimes can stabilize relationships, reduce risks of miscalculation, and pave the way to formal constraints. Conversely, the demise of specific treaties can coexist with resilient practices and institutions that keep channels open and data flowing. Understanding this ecosystem helps advocates avoid all‑or‑nothing thinking and identify leverage points where incremental steps unlock larger bargains.

The final third of the book is forward‑looking. It offers practical playbooks for negotiators and legislators, proposes verification pilots that can be fielded quickly, and maps pathways for risk reduction among adversaries who may not be ready for sweeping treaties. These include reciprocal notifications, bounded no‑deployment zones, warhead‑level accounting trials, and codes of conduct for cyber operations touching nuclear command and control. Each proposal is built to be modular, legally sound, politically sellable, and technologically verifiable.

Broken Treaties is ultimately a guide for realists with ambition: people who accept that competition is durable but insist that catastrophe is not destiny. By learning why previous agreements worked or failed—and by aligning legal craft, political strategy, and technological credibility—we can rebuild a scaffolding of restraint sturdy enough for today’s rivalries and tomorrow’s innovations.

There is a simple test for whether nuclear arms control matters: ask if accidents, misread signals, or hair‑trigger politics could erase a city before lunch. If the answer is yes, then mechanisms that slow the clock, clarify intentions, and verify limits are not luxury goods; they are essential safety equipment. Arms control is not about friendship; it is about building guardrails for adversaries who cannot afford a head‑on collision.

The world is fragmenting, but danger is not distributed evenly. Power is diffusing among more nuclear‑armed states, while great‑power trust has thinned to a film. The post‑Cold War holiday from nuclear rivalry is over, and the calendar has returned to an era of long‑range missiles, dual‑use technologies, and leaders who meet only on screens. In that setting, restraint is not a moral preference; it is a practical instrument for keeping the peace.

Consider a few realities. Missiles that fly through space do not care whether the tension causing their launch was real or imagined. Satellite imagery reveals silos but not intentions. Algorithms in command‑and‑control systems can accelerate decisions while compressing human judgment. In such an environment, any tool that widens the margin for error earns its keep, whether it is a treaty, a hot line, or a mutually observed test pause.

Arms control can sound archaic, as if the subject were only silos and throw‑weights. It is not. It also encompasses the behavior of cyber teams probing nuclear systems, the movement of dual‑use materials, and the signals sent by military exercises near borders. The practice has expanded alongside the technology. Its aim remains the same: to keep competition from becoming catastrophe.

Many people think of arms control as a single grand project, but it is better described as a toolkit. Some tools are formal treaties with signatures and cameras on the ground; others are tacit norms and reciprocal notifications. Some are global, like the Non‑Proliferation Treaty; others are bilateral, like the caps on deployed warheads under New START. A functioning toolbox contains bolts, screws, and glue—each chosen for the joint that needs reinforcement.

The word “success” can mislead. A treaty that is negotiated, ratified, and observed for three decades before collapsing is not a failure; it is a prolonged success that ended. Conversely, a stalled negotiation that prevents a crisis or shapes expectations can be a quiet victory. Judging outcomes requires calibrating expectations to what the tools were designed to do at the time they were used.

There is also the question of who gets a seat at the table. The nuclear order emerged from a specific history—World War II, the early Cold War, decolonization—and carries that imprint. Newer states, new technologies, and new rules of the game have changed the audience. Arms control that works for five states will not necessarily work for nine, and that gap has tested old agreements and complicated new ones.

Law matters because it locks in what politics temporarily grants. A treaty transforms a unilateral gesture into a reciprocal obligation. It embeds verification, defines terms, and sets dispute‑resolution procedures. It is not magic—exit clauses exist and interpretations differ—but it changes the cost of reneging and the likelihood that violations will be seen, challenged, and corrected before they snowball.

Politics matters because leaders make choices under pressure. Domestic coalitions can reward or punish a signature. Legislatures can block ratification or fund workarounds. Populations may fear concessions or demand disarmament regardless of conditions. Effective arms control reads these currents and channels them into agreements that are politically durable, not merely theoretically elegant.

Technology matters because it determines what is knowable and what is contestable. If states cannot verify limits, they will not accept them. Advances in sensors, data analysis, and sampling can close gaps, while new systems—hypersonic glide vehicles, maneuvering warheads, or AI‑assisted targeting—can open them. The history of arms control is a back‑and‑forth between invention and limitation, like two children building and knocking over sandcastles.

The Cold War taught that constraints could coexist with deep hostility. The hot line after 1963, the Limited Test Ban Treaty, and the Anti‑Ballistic Missile Treaty were not born from affection; they were practical responses to the Cuban Missile Crisis and the realization that accidental war was too easy. They did not end the rivalry, but they turned an existential gamble into a manageable competition, with rules and channels where none had existed.

The 1990s offered a different lesson: momentum can be as dangerous as paralysis if it is untethered from verification. The euphoria that followed the Soviet collapse produced sweeping ambitions and, in some cases, sweeping assumptions. When political will waned and technical scrutiny tightened, agreements faltered, but the inspection cultures and data exchanges from that era still shape expectations today.

Looking around now, the toolbox looks battered and half‑empty. The Intermediate‑Range Nuclear Forces Treaty is gone. New START is suspended, though not yet terminated. The Comprehensive Test Ban Treaty is signed but not in force. Strategic stability talks stutter and restart. Supply‑side controls strain under the weight of new research ecosystems. Yet pockets of practice persist: notifications, scientific dialogues, and shared data that keep lines open and reduce miscalculation.

Some argue that in a fragmenting world, restraint is a luxury that the insecure cannot afford. Others claim that technology has outrun treaties, making them obsolete. A third view holds that arms control is an artifact of bipolarity, irrelevant in a multipolar age. None of these positions account for the stubborn facts: accidents happen, misperceptions are common, and the consequences of error are uniquely catastrophic.

Arms control’s relevance is not measured by the number of treaties in force alone. It is measured by whether the probability of nuclear use falls, whether crisis escalation speeds slow, and whether leaders retain a path back from the brink. By that metric, modest mechanisms—reciprocal launch notifications, data exchanges, hot lines—are as vital as grand bargains. They are the seat belts and airbags of international security.

The economics are often overlooked. Sustaining large arsenals is costly; modernization programs run into the hundreds of billions. Constraints can free resources for other priorities, from social programs to conventional readiness. Arms control is not charity; it is budgeting for stability, with the side effect of reducing both financial burden and the risk of a ruinous exchange.

There is a human dimension that is easy to miss. Nuclear operators, inspectors, and negotiators build professional relationships that survive political frost. The inspectors who swapped notes at a missile base in the 1990s may be the scientists comparing seismology readings today. These networks do not decide policy, but they shape the information on which decisions rest and the tone in which disputes are aired.

For advocates and policymakers, the temptation is to treat treaties as endpoints. A more useful frame is that treaties are crowns on a pyramid of practice. The base is data sharing, incident prevention, and crisis communication; the middle layer is risk‑reduction measures and moratoria; the peak is formal, legally binding constraints. Each layer supports the others, and skipping levels usually leads to collapse.

The risks are not static. New technologies—hypersonics, artificial intelligence, space‑based sensors, cyber capabilities—change the geometry of deterrence. They create new vulnerabilities and new temptations. Some enhance verification; others undermine it. Good arms control monitors these shifts and adjusts rules before they are permanently outrun, rather than waiting for a crisis to force a rethink.

Misunderstanding is a constant hazard. Terms like “strategic stability” or “first‑use” carry contested meanings. Definitions drive counting rules; counting rules drive compliance; compliance drives confidence. The more precise the language, the less room for fatal misinterpretation. Treaties are essentially legal code written for a world of imperfect trust, and like any code, it must be debugged before it runs.

Arms control also intersects with other regimes—export controls, non‑proliferation, missile defense, and space norms. None exist in isolation. A step that stabilizes one domain can destabilize another. Managing these cross‑effects requires a systems view, in which arms control is part of a wider security ecosystem rather than a stand‑alone project.

The question of fairness looms large. Smaller states often argue that nuclear powers have been slow to honor disarmament pledges, while nuclear powers point to regional threats that demand insurance. This gap in perception is not rhetorical fluff; it shapes participation in regimes and compliance with rules. Without mechanisms that address fairness—procedural and substantive—buy‑in erodes, and enforcement becomes selective.

The stakes are clearest in crises. When leaders have minutes to decide whether an inbound track is a missile or a glitch, and whether a signal is a bluff or a prelude, the margin for error shrinks. Established channels, shared procedures, and pre‑negotiated rules reduce the cognitive load at the worst moments. They do not guarantee good decisions, but they raise the odds that panic does not become policy.

A practical test for any new proposal is: does it buy time? If a measure widens the window for verification, consultation, or cancellation, it has value. If it adds friction without clarity, it risks being ignored. The best tools are those that slow the wrong actions while keeping the right ones available, a balance that requires engineering as much as diplomacy.

Another test is: does it survive leadership change? Personal chemistry between leaders can open doors, but institutions keep them from swinging shut with each election. Treaties that embed verification and transparency into routine practice are more durable than those that depend on the goodwill of a single administration. The rule of law beats the rule of personality, especially when nuclear weapons are involved.

The public often hears about arms control only when it fails. That visibility matters—democracies require consent—but it can create a distorted picture. There are thousands of hours of unglamorous work in sensor calibration, data reconciliation, and inspection routines that never make headlines. That quiet labor is what turns promises into credible reality.

There is also a temporal dimension. Nuclear effects last for generations; political incentives do not. Arms control aligns the time horizons of politics and consequences by making short‑term restraint beneficial and long‑term stability visible. It does so not by preaching, but by structuring choices so that the safe option is also the smart option, at least most of the time.

It is tempting to see arms control as an East‑West artifact. The reality is broader. Every region with nuclear weapons or nuclear ambitions faces its own version of the problem: how to manage fear without surrendering security. Lessons travel—verification methods from U.S.–Soviet practice inform other contexts—but local politics and history shape the menu of feasible solutions.

In a fragmenting world, the temptation is to abandon shared rules in favor of competitive autonomy. That path has costs: more redundancy, higher spending, and a sharper risk that a local crisis turns global. Arms control offers a counterweight by establishing minimal shared baselines—what not to do, what to announce, what to verify—that preserve autonomy while curbing the most dangerous behaviors.

The case for arms control does not rely on hope. It rests on experience. Decades of practice show that adversaries can agree to limits, verify them, and benefit from them without becoming friends. The guardrails work imperfectly, but they work better than none at all, which is why even a fragmenting world still finds reason to build and maintain them.

This chapter sets the stage for the rest of the book by sketching why the subject matters now. The chapters that follow unpack the institutions, the technologies, and the political dynamics that shaped past successes and failures. They also explore what might be built next, with the humility of people who know that tools are only as good as the hands that wield them.

Let the reader beware: the path from handshake to verified compliance is long, technical, and often boring. That is a feature, not a bug. Boring is what happens when systems work and nothing blows up. In a nuclear world, boring is underrated, and the craft of arms control is the art of keeping it that way.