On July 16, 1945, at 5:29 AM in the Jornada del Muerto desert of New Mexico, the Trinity test detonated the first nuclear weapon in history. The fireball rose to ten thousand feet. The shock wave knocked observers off their feet at distances of miles. The sand beneath the tower vitrified into a new mineral — trinitite — green and glassy from the heat. Robert Oppenheimer, watching the fireball rise from a bunker, recalled a line from the Bhagavad Gita: "Now I am become Death, the destroyer of worlds." Kenneth Bainbridge, the test director, turned to him afterward and said: "Now we're all sons of bitches."
Three weeks later, the same technology destroyed Hiroshima. A single bomb, carried by a single aircraft, killed an estimated 70,000-80,000 people instantly. By the end of 1945, the death toll had risen to between 90,000 and 140,000. Not a military campaign. Not a sustained bombing operation. One bomb, one city, one morning. Three days after Hiroshima, a second bomb destroyed Nagasaki. Japan surrendered.
The nuclear age had begun, and humanity had acquired something it had never possessed before: the practical ability to destroy its own civilization. Not metaphorically — literally. The combined nuclear arsenals of the United States and Russia contain enough destructive capacity to kill billions of people directly, and the fires they would ignite might trigger global agricultural collapse that could kill billions more. We have built this capability, live under its shadow, and have narrowly avoided using it multiple times since 1945. Understanding nuclear weapons — how they work, how they became part of geopolitical strategy, and what risks they represent — is not an optional topic for the curious. It is a prerequisite for understanding the world.
"The question is not whether nuclear weapons will be used. The question is how to prevent their use and ultimately to eliminate them." — Robert McNamara, The Fog of War (2003)
Key Definitions
Fission: The splitting of a heavy atomic nucleus (typically uranium-235 or plutonium-239) by a neutron, releasing energy and more neutrons in a chain reaction.
Fusion: The combining of light atomic nuclei (hydrogen isotopes deuterium and tritium) at extreme temperatures and pressures, releasing even more energy per unit mass than fission.
Critical mass: The minimum amount of fissile material needed in a given geometry and configuration to sustain a self-amplifying chain reaction.
Chain reaction: The self-sustaining process in which each nuclear fission event releases neutrons that trigger additional fission events, releasing more neutrons, exponentially amplifying energy release.
Little Boy (Hiroshima): The first nuclear weapon used in combat, a gun-type uranium-235 fission bomb with a yield of approximately 15 kilotons, dropped on Hiroshima on August 6, 1945.
Fat Man (Nagasaki): The second nuclear weapon used in combat, an implosion-type plutonium-239 fission bomb with a yield of approximately 21 kilotons, dropped on Nagasaki on August 9, 1945.
Thermonuclear weapon (hydrogen bomb): A two-stage weapon in which a fission primary ("trigger") creates the conditions to ignite a fusion secondary, releasing yields in the megaton range — approximately 1,000 times more powerful than early fission weapons.
Yield: The energy released by a nuclear detonation, measured in kilotons (kt, thousands of tons of TNT equivalent) or megatons (mt, millions of tons of TNT equivalent).
Mutual assured destruction (MAD): The strategic doctrine that nuclear war between two well-armed states is deterred because each side can absorb a first strike and still retaliate with unacceptable damage.
First strike / second strike capability: First-strike capability is the ability to destroy the adversary's retaliatory forces before they can be used; second-strike capability is the ability to retaliate even after absorbing a first strike.
Nuclear triad: The three delivery platforms — land-based ICBMs, submarine-launched ballistic missiles, and strategic bombers — providing redundancy and survivability for nuclear forces.
Deterrence theory: The strategic doctrine that the threat of nuclear retaliation can prevent nuclear attack by making it suicidal for a rational adversary.
Non-Proliferation Treaty (NPT): The 1968 treaty establishing the international nonproliferation regime, with 191 state parties; recognizes five nuclear weapon states and commits them to disarmament while prohibiting other signatories from acquiring weapons.
Nuclear winter: The hypothesis that a large-scale nuclear exchange would inject sufficient soot into the upper atmosphere to cause global temperature declines and agricultural collapse.
The Physics: Why Nuclear Weapons Are Different
Conventional explosives work by rapid chemical reactions — breaking and forming molecular bonds — that release energy at a rate of roughly 3-7 megajoules per kilogram of explosive. Nuclear weapons work by reactions in atomic nuclei — reactions governed by the strong nuclear force, which is approximately 1 million times more powerful than chemical bonds at nuclear distances. The result is energy densities that are simply incommensurable with conventional explosives.
When uranium-235 or plutonium-239 absorbs a neutron, the resulting nucleus becomes unstable and splits into smaller fragments — fission products — plus two or three more neutrons plus approximately 200 megaelectronvolts of energy, released primarily as kinetic energy of the fragments and gamma radiation. Those 200 MeV corresponds to roughly 3.2 × 10^-11 joules per fission event. That sounds tiny, but the number of atoms in even a kilogram of uranium is approximately 2.5 × 10^24 — and if all of them fissioned (which does not happen in practice), the energy release would be approximately 8 × 10^13 joules — equivalent to nearly 20 kilotons of TNT.
The challenge is not releasing the energy but releasing it fast enough and in a compact enough configuration to produce an explosion rather than a slow fizzle. The chain reaction must go supercritical: each fission event must, on average, produce more than one new fission event. This requires that free neutrons not escape the material faster than they cause new fissions — hence the concept of critical mass.
For uranium-235, the "bare sphere" critical mass is approximately 52 kilograms. For plutonium-239, it is approximately 10 kilograms. These masses can be reduced substantially by using a neutron reflector (a surrounding material that bounces escaping neutrons back into the fissile material) or by compressing the fissile material to higher-than-normal density (as the implosion design does).
Thermonuclear weapons, first tested by the United States in 1952 (Ivy Mike, approximately 10.4 megatons) and the Soviet Union in 1955 (RDS-37, approximately 1.6 megatons), work by using a fission bomb as a trigger. The intense X-ray radiation from the fission primary compresses and heats a secondary assembly of fusion fuel — typically lithium deuteride — to the conditions needed for fusion. The fusion reaction releases neutrons that can trigger additional fission in a surrounding shell of uranium-238 (which is not fissile but can be "fast-fissioned" by fusion neutrons), further amplifying the yield. There is no theoretical upper limit to thermonuclear weapon yield: the Soviet "Tsar Bomba" test of 1961 achieved approximately 57 megatons.
Nuclear Weapon Designs and Historical Yields
| Weapon / Test | Country | Year | Type | Yield | Notes |
|---|---|---|---|---|---|
| Little Boy (Hiroshima) | United States | 1945 | Gun-type fission (U-235) | ~15 kt | First combat use; no test needed |
| Fat Man (Nagasaki) | United States | 1945 | Implosion fission (Pu-239) | ~21 kt | Second combat use |
| Ivy Mike | United States | 1952 | Thermonuclear (hydrogen bomb) | ~10.4 Mt | First H-bomb; liquid deuterium fuel |
| Castle Bravo | United States | 1954 | Thermonuclear (dry) | ~15 Mt | Largest US test; unexpected 2.5x yield |
| Tsar Bomba | Soviet Union | 1961 | Three-stage thermonuclear | ~57 Mt | Largest weapon ever detonated |
| Typical modern warhead | US / Russia | 2020s | Two-stage thermonuclear | 100-500 kt | ICBM and SLBM deployed warheads |
kt = kilotons (thousands of tons TNT equivalent); Mt = megatons (millions of tons). The Hiroshima bomb destroyed a city; modern strategic warheads are 7 to 40 times more powerful.
The Manhattan Project: Building the Bomb
The Manhattan Project was the most expensive and scientifically demanding weapons program in history to that point. Its origins lay in the awareness, spreading through the physics community by 1939, that nuclear fission had been achieved (Hahn and Strassmann in Berlin, confirmed by Frisch and Meitner in Copenhagen) and that a weapon based on chain reaction was theoretically possible.
The Einstein-Szilard letter of August 1939 — signed by Albert Einstein and drafted by Leo Szilard — warned President Roosevelt that Germany might be pursuing such a weapon and urged American development. The letter was prescient about German interest and premature about German progress; the German nuclear program was eventually determined to be far less advanced than feared, partly because key physicists were excluded on racial grounds and partly due to fundamental technical misjudgments.
Enrico Fermi's Chicago Pile-1, constructed in a converted squash court under the stands of Stagg Field at the University of Chicago, achieved the first controlled sustained nuclear chain reaction on December 2, 1942. This demonstration that a chain reaction could be controlled — could be started and stopped — was the proof of concept that justified the industrial-scale program.
Los Alamos, the secret laboratory in the New Mexico mountains where the weapons were designed, was directed by J. Robert Oppenheimer, a theoretical physicist who proved extraordinary both as a scientific mind and as a manager of competing, brilliant, and difficult personalities. The laboratory employed, at its peak, several thousand people, including perhaps the most concentrated collection of physics talent in history: Fermi, Hans Bethe, Niels Bohr (under the name "Nicholas Baker"), Richard Feynman, Edward Teller, John von Neumann, and dozens of others.
The $2 billion project — an almost incomprehensible sum in 1945, equivalent to roughly $30 billion today — built and operated three major production sites simultaneously: the Y-12 electromagnetic separation plant at Oak Ridge, Tennessee (for uranium enrichment); the K-25 gaseous diffusion plant, also at Oak Ridge (for further enrichment); and the Hanford Engineer Works in Washington State (for plutonium production). The coordination of these efforts across multiple sites with tight security and under wartime pressure remains an organizational as well as scientific achievement.
Trinity, the first test, used a plutonium implosion design (Fat Man). The uranium gun design (Little Boy) was considered too reliable to need testing. Both designs had been confirmed by theory, calculation, and experiment; only the implosion design, with its precisely synchronized detonation of explosive lenses, required actual proof.
Hiroshima and Nagasaki: What the Weapons Did
At 8:15 AM on August 6, 1945, Little Boy detonated approximately 600 meters above the center of Hiroshima. The fireball reached temperatures of several million degrees Celsius. The thermal pulse ignited fires across a radius of roughly 3 kilometers, which combined into a firestorm that consumed the wooden city. The blast wave destroyed most structures within a 1.6-kilometer radius of the hypocenter and caused severe damage to a radius of 3 kilometers. The prompt radiation — gamma rays and neutrons released in the first minute — was lethal for those within about 1 kilometer who were not shielded by intervening material.
The immediate death toll was approximately 70,000-80,000. Many more died over the following weeks and months from radiation sickness, burns, and injuries exacerbated by the collapse of Hiroshima's medical infrastructure — most doctors and nurses were among the dead. By the end of 1945, the total death toll was estimated at between 90,000 and 140,000. The population of Hiroshima at the time was approximately 350,000; roughly half were dead or severely injured within months.
Three days later, Fat Man detonated over Nagasaki. The Nagasaki bomb was more powerful (21 vs 15 kilotons) but killed fewer people — approximately 60,000-80,000 by year's end — partly because Nagasaki's geography (it is surrounded by hills) confined the blast and fire effects, and partly because Hiroshima had given the remaining urban populations in Japan some warning.
The Radiation Effects Research Foundation (RERF), established jointly by the US and Japan in 1975 and continuing to the present, has followed approximately 120,000 survivors of the two bombings — about 94,000 hibakusha and 26,000 controls — through regular medical examinations. Their data is the primary source of human evidence on radiation-induced cancer. Among survivors exposed to significant doses, solid cancer mortality was approximately 10 percent higher than the control group; leukemia rates were significantly elevated, peaking in the late 1940s and early 1950s. Children exposed in utero showed elevated rates of developmental problems and intellectual disability at high doses. The RERF's estimates of radiation-cancer dose-response relationships underlie the international radiation protection standards that govern nuclear medicine, nuclear power, and occupational radiation limits.
Nuclear Deterrence: The Strategy of Mutual Hostage-Taking
Bernard Brodie wrote in "The Absolute Weapon" (1946): "Thus far the chief purpose of our military establishment has been to win wars. From now on its chief purpose must be to avert them. It can have almost no other useful purpose." This insight — that nuclear weapons changed the function of military force from winning to preventing — was the seed of nuclear deterrence theory.
The theory's full development came with the thermonuclear age and the massive buildup of the 1950s. The core logic of Mutual Assured Destruction is a Nash equilibrium in a game-theoretic sense: each side maintains forces sufficient to absorb a first strike and still retaliate with unacceptable damage. Neither side has an incentive to strike first, because striking first does not prevent devastating retaliation. The equilibrium is stable as long as each side believes the other's retaliatory threat is credible — as long as each side believes the other would actually carry through on the promise to destroy their cities.
The problem, as Thomas Schelling analyzed in "Arms and Influence" (1966), is making the threat credible. Why would a rational leader, after absorbing a nuclear first strike that destroyed their cities, retaliate with nuclear weapons and trigger a second retaliatory strike on whatever survived? Pure revenge? The logic is shaky, and adversaries might calculate that deterrence would "fail" after the fact — that a disarmed or devastated state would rationally accept surrender rather than risk further destruction.
Schelling's solution was to understand deterrence as a game of commitment: reducing the decision-maker's freedom to back down in a crisis makes the threat credible. Automaticity helps: if the retaliatory launch is triggered automatically by detection of incoming missiles, there is no decision to back down from. Delegation helps: if the authority to launch is delegated to submarine commanders who would launch in the absence of "all-clear" signals, the retaliatory threat is decoupled from central decision-making. The phrase "leaving something to chance" — the nuclear threat is credible precisely because it might be carried out irrationally — became a cornerstone of nuclear strategy.
Herman Kahn's "On Thermonuclear War" (1960) was widely criticized for its clinical analysis of nuclear scenarios but was genuinely influential in strategic thinking. Kahn insisted that nuclear deterrence required thinking through scenarios of failure: what if deterrence broke down? What would a nuclear war look like? Could it be "managed"? His concept of an "escalation ladder" — different levels of nuclear threat, each designed to signal resolve without immediately triggering all-out war — influenced NATO strategy for decades.
The Nuclear Triad and Its Logic
The strategic logic of the nuclear triad flows directly from deterrence theory's requirement for secure second-strike capability.
Land-based ICBMs are in fixed, known locations — Minuteman III silos are scattered across the American Midwest, in hardened concrete shelters whose coordinates are well-known. They are accurate and can be launched quickly, but they are vulnerable: in a coordinated first strike, an adversary might attempt to destroy them before they can be used. Their deterrent value comes partly from this vulnerability: the adversary must launch near-simultaneously against hundreds of silos, a technically demanding and risky operation, and if even some survive, they can retaliate.
Ballistic missile submarines are the most survivable delivery system. The US operates 14 Ohio-class submarines, each carrying 20 Trident II D5 ballistic missiles; each missile can carry up to 8 independently targeted warheads. The submarines operate at undisclosed locations in the world's oceans; tracking them is extraordinarily difficult and becomes harder as technology improves. Even if all land-based ICBMs and bomber bases were destroyed in a first strike, surviving submarines would retain the ability to devastate the attacker. The existence of this capability is the clearest reason why a nuclear first strike is irrational.
Strategic bombers add flexibility: unlike ballistic missiles, which cannot be recalled once launched, bombers can be sent out and called back. This allows a nuclear power to visibly signal resolve in a crisis — putting bombers on alert, extending patrol routes — without the irreversibility of a missile launch. B-2 Spirit stealth bombers, capable of penetrating advanced air defense systems, add a complicating factor to adversary planning.
The United States and Russia each maintain roughly 5,000 nuclear warheads in various states of readiness (deployed, in storage, or awaiting dismantlement). China is undergoing a significant modernization and expansion of its nuclear forces, with estimates of its arsenal varying widely; US intelligence assessments from 2023 suggested China may be expanding toward 1,500 warheads by 2035. Other nuclear states — the UK, France, India, Pakistan, Israel, North Korea — maintain smaller arsenals with different strategic rationales.
Arms Control: Constraining the Weapons
The recognition that nuclear arsenals could continue growing indefinitely, and that growth was destabilizing, drove successive arms control efforts.
The Limited Test Ban Treaty (1963) prohibited nuclear tests in the atmosphere, underwater, and in space, following years of concern about radioactive fallout from atmospheric tests — strontium-90 found in children's teeth, radioactive rain, nuclear ash on Pacific atolls. It did not limit weapons numbers but eliminated the most visible and contaminating tests.
SALT I (1972) and SALT II (1979) capped offensive strategic weapons — ICBMs, SLBMs, bombers — for the first time, freezing the rough numeric parity the two superpowers had reached. SALT II was never ratified by the US Senate after the Soviet invasion of Afghanistan but was observed informally for years.
START I (1991), signed as the Soviet Union dissolved, made the first actual reductions: both sides cut to 6,000 deployed warheads. New START (2010), signed by Obama and Medvedev, reduced each side to 1,550 deployed strategic warheads — a massive reduction from Cold War peaks of tens of thousands on each side.
The INF Treaty (1987) eliminated an entire class of nuclear weapons — ground-based ballistic and cruise missiles with ranges between 500 and 5,500 kilometers. It was withdrawn by the Trump administration in 2019, citing Russian violations (the development of the SSC-8 missile system); Russia then resumed development of previously prohibited systems. The collapse of the INF Treaty is widely regarded as a significant setback for arms control, potentially triggering a new generation of intermediate-range nuclear deployment in Europe and Asia.
New START expired in 2026. Arms control negotiations between the United States and Russia have been paralyzed by the war in Ukraine and the broader deterioration of relations. The architecture of strategic stability that was built across five decades of difficult negotiations is substantially weakened.
Proliferation: Who Has Nuclear Weapons and Why
Scott Sagan, in his 1996/97 debate with Kenneth Waltz in "The Spread of Nuclear Weapons," identified three models for why states pursue or forgo nuclear weapons: the security model (states build weapons when they face threatening adversaries and lack sufficient conventional deterrence), the domestic politics model (nuclear programs can serve the interests of particular bureaucratic and military factions regardless of strategic logic), and the norms model (nuclear weapons carry symbolic value as markers of great-power status, and conversely, renouncing them signals peaceful intent).
All three models explain real cases. India's 1974 test followed China's 1964 test and the 1971 war with Pakistan; Pakistan's 1998 test was explicitly a response to India's. North Korea's program, culminating in the 2006 first test and subsequent tests of progressively more capable weapons and delivery systems, reflects security concerns (the memory of the Korean War, the observation of what happened to Iraq and Libya after they gave up WMD programs) and the domestic legitimation needs of the Kim dynasty. South Africa's six weapons, built secretly and dismantled as apartheid ended, reflected both security concerns (Cold War proxy conflicts on the southern African periphery) and the new government's decision that a nuclear arsenal was inconsistent with its post-apartheid identity.
Israel's nuclear opacity — it has never confirmed or denied possessing nuclear weapons — is a deliberate policy. Israel is widely believed to possess approximately 90 warheads; its opacity serves a specific strategic function, deterring adversaries without triggering a regional nuclear arms race or pressure to sign the NPT.
The NPT's fundamental tension is between its nonproliferation and disarmament pillars. Non-nuclear weapon states accepted restrictions on their own programs on the basis of a commitment by the five recognized nuclear weapon states to work toward disarmament. That commitment has not been honored: the nuclear weapon states have modernized and in some cases expanded their arsenals, while insisting on the non-nuclear states' obligations. This asymmetry has eroded the treaty's legitimacy among developing countries and makes future negotiations more difficult.
The Existential Risk: Near-Misses and Future Dangers
The Bulletin of Atomic Scientists' Doomsday Clock is a symbolic device for communicating nuclear risk to a general audience. In January 2023, the Bulletin's science and security board moved it to 90 seconds to midnight — the closest to midnight in its 75-year history, closer even than the peaks of the Cold War. The rationale included the war in Ukraine, the breakdown of arms control, Russian nuclear rhetoric, the stalled Non-Proliferation Review Conference, and advances in hypersonic weapons and artificial intelligence.
The historical record of near-misses — situations where nuclear war could plausibly have begun through accident, miscalculation, or technical failure — is more extensive than is generally acknowledged. The Cuban Missile Crisis saw multiple moments where local commanders had both the authority and inclination to use nuclear weapons, including a Soviet submarine commander preparing to launch a nuclear torpedo before being overruled. The Petrov incident of 1983 — when Stanislav Petrov correctly identified a Soviet early-warning alarm as a false positive and did not report it up the chain, possibly preventing a retaliatory launch — has become emblematic of the role of individual judgment (and luck) in avoiding nuclear war.
Toby Ord's "The Precipice" (2020) estimates the probability of nuclear catastrophe within the next century at approximately 1 percent — small in any given year but significant over a century, and with consequences that could permanently truncate civilization's potential. The estimate is deeply uncertain, but the uncertainty itself is significant: we do not have the data to confidently bound the risk of events that have happened rarely and that have a catastrophic lower tail.
Current risk factors include: the continuing deployment of weapons on launch-on-warning postures (missiles that can be launched within minutes of detecting an incoming attack, before confirmation can be obtained); the development of hypersonic glide vehicles that reduce warning times for defending states; the expansion of Chinese nuclear forces introducing new actors into the strategic stability calculation; the North Korean program and its integration into Korean peninsula crisis scenarios; and the potential for cyberattacks on nuclear command, control, and communication systems to either disable them or spoof false alerts.
The fundamental tension of the nuclear age has not been resolved: the weapons that provide deterrence are also the weapons that represent existential risk. Managing this tension — through arms control, through technical safeguards, through crisis communication channels, through the cultivation of strategic stability — is among the most consequential ongoing tasks in international relations.
Related Articles
References
- Rhodes, Richard. The Making of the Atomic Bomb. Simon and Schuster, 1986.
- Brodie, Bernard, ed. The Absolute Weapon: Atomic Power and World Order. Harcourt Brace, 1946.
- Schelling, Thomas C. Arms and Influence. Yale University Press, 1966.
- Kahn, Herman. On Thermonuclear War. Princeton University Press, 1960.
- Sagan, Scott D. "Why Do States Build Nuclear Weapons? Three Models in Search of a Bomb." International Security 21(3): 54-86, 1996/97.
- Radiation Effects Research Foundation (RERF). Life Span Study and Adult Health Study reports. Available at: rerf.or.jp.
- Sagan, Carl, Richard Turco, Owen Toon, Thomas Ackerman, and James Pollack. "Nuclear Winter: Global Consequences of Multiple Nuclear Explosions." Science 222(4630): 1283-1292, 1983.
- Robock, Alan, and Brian Toon. "Let's End the Peril of a Nuclear Winter." Scientific American, 2012.
- Ord, Toby. The Precipice: Existential Risk and the Future of Humanity. Hachette Books, 2020.
- Bulletin of Atomic Scientists. Doomsday Clock Statement, January 2023. Available at: thebulletin.org.
- Tannenwald, Nina. The Nuclear Taboo: The United States and the Non-Use of Nuclear Weapons Since 1945. Cambridge University Press, 2007.
- Cirincione, Joseph. Bomb Scare: The History and Future of Nuclear Weapons. Columbia University Press, 2007.
Frequently Asked Questions
How does a nuclear weapon work? What is the difference between fission and fusion?
Nuclear weapons derive their energy from one of two fundamental nuclear reactions: fission (splitting heavy atomic nuclei) or fusion (combining light atomic nuclei). In both cases, the reaction converts a small amount of mass directly into energy, as described by Einstein's equation E=mc^2. Because the speed of light squared is an enormous number, even tiny mass conversions release catastrophic amounts of energy. In a fission weapon, the nuclei of heavy atoms — specifically uranium-235 or plutonium-239 — are split by neutrons. When a nucleus splits, it releases more neutrons, which can split more nuclei, which release more neutrons, producing a chain reaction. The key to making a weapon rather than just a slow reactor is achieving a 'supercritical' mass: enough fissile material in a compact enough configuration that the chain reaction accelerates rather than dissipating, releasing most of the energy in a tiny fraction of a second. Little Boy, the bomb dropped on Hiroshima, used a gun-type design in which a subcritical mass of uranium-235 was fired down a barrel into another subcritical mass, combining them into a supercritical whole. Fat Man, dropped on Nagasaki, used an implosion design: conventional explosives surrounding a subcritical sphere of plutonium-239 were detonated simultaneously, compressing the sphere to supercritical density. In a fusion weapon — a thermonuclear or hydrogen bomb — a fission bomb is used as a trigger to achieve the temperatures and pressures needed to fuse hydrogen isotopes (deuterium and tritium). The fusion reaction releases even more energy per unit mass than fission, and crucially, it is not limited by the critical mass constraint — you can make fusion weapons arbitrarily powerful by adding more fusion fuel.
How destructive were Hiroshima and Nagasaki, and what were the long-term health effects?
The immediate destruction of Hiroshima and Nagasaki was on a scale that had no precedent in conventional warfare. Little Boy, which detonated over Hiroshima on August 6, 1945, had a yield of approximately 15 kilotons (equivalent to 15,000 tons of TNT). Fat Man, detonated over Nagasaki on August 9, 1945, had a yield of approximately 21 kilotons. The immediate effects were three-fold. The blast wave from the Hiroshima bomb destroyed structures within roughly 1.6 kilometers of the hypocenter and caused severe damage to a much larger radius. The thermal pulse — a fireball reaching tens of millions of degrees Celsius at its center — caused fires across Hiroshima that combined into a firestorm. Ionizing radiation (gamma rays and neutrons) was lethal within approximately 1 kilometer of the hypocenter for those exposed without shielding. Estimated immediate deaths were approximately 70,000-80,000 in Hiroshima and 40,000 in Nagasaki; by December 1945, the toll had risen to approximately 90,000-140,000 in Hiroshima and 60,000-80,000 in Nagasaki, as radiation sickness and burn injuries killed survivors of the initial blast. The Radiation Effects Research Foundation (RERF), established jointly by the United States and Japan to study the long-term effects on hibakusha (survivors), has followed approximately 120,000 survivors for decades. The cohort shows significantly elevated rates of leukemia (peaking in the 1950s) and solid cancers (emerging from the 1960s onward), with risk increasing with radiation dose received. The RERF data remain the primary source of human data on radiation-cancer relationships and inform current radiation safety standards worldwide.
What is nuclear deterrence theory, and does it work?
Nuclear deterrence theory holds that the threat of nuclear retaliation can prevent nuclear attack by making the costs of attack so catastrophically high that no rational adversary would initiate one. Bernard Brodie, writing in 1946 in the days immediately after Hiroshima, was the first to articulate this in systematic terms: nuclear weapons had fundamentally changed the purpose of military power from winning wars to preventing them. The full development of deterrence theory came with the nuclear buildup of the 1950s and 1960s. Herman Kahn's 'On Thermonuclear War' (1960), controversial but analytically influential, worked through the logic of nuclear conflict systematically. Thomas Schelling's 'Arms and Influence' (1966) examined the game-theoretic structure of deterrent threats: a threat is credible only if the threatening party would actually carry it out, which means making the threat believable in advance requires either automating the response or reducing the decision-maker's ability to back down. Robert McNamara, as Secretary of Defense, operationalized deterrence through the concept of assured destruction: maintaining the ability to destroy a specified fraction of the adversary's population and industry even after absorbing a first strike. The empirical record on whether deterrence 'works' is difficult to assess because we are counting non-events: the nuclear attacks that did not happen. The Cold War ended without nuclear use, which some analysts attribute to deterrence and others to luck. Near-misses — the 1962 Cuban Missile Crisis, the 1983 Able Archer exercise (which Soviet intelligence interpreted as possible cover for a real attack), the 1983 Petrov incident — suggest that the stability of deterrence is significantly less than its theoretical logic implies.
What is the nuclear triad, and why does it matter?
The nuclear triad refers to the three platforms through which nuclear weapons are delivered: land-based intercontinental ballistic missiles (ICBMs), submarine-launched ballistic missiles (SLBMs), and strategic bombers carrying nuclear gravity bombs or cruise missiles. Each leg of the triad provides distinct survivability and strategic advantages, and the logic of deterrence theory holds that having all three significantly reduces the risk that a first strike could destroy an adversary's retaliatory capability. Land-based ICBMs are highly accurate — they can be targeted precisely — but they are in fixed, known locations, making them vulnerable to a coordinated counterforce first strike. Their deterrent value lies partly in this vulnerability: an adversary who might contemplate a first strike must successfully destroy all ICBMs simultaneously, which is technically extremely demanding. Submarine-launched ballistic missiles are the most survivable leg of the triad. Ballistic missile submarines (SSBNs) operate continuously at sea in undisclosed locations; they cannot be targeted without first detecting and tracking them, which modern submarines are designed to prevent. They provide assured second-strike capability: even if land-based ICBMs and bomber bases were destroyed, submarines could still retaliate. Strategic bombers are the most flexible leg: they can be launched and then recalled, can carry conventional as well as nuclear payloads, and can be placed on ground alert to signal resolve in a crisis. Their disadvantage is speed: unlike ballistic missiles, which reach targets in minutes, bombers take hours. The United States and Russia each maintain all three legs; other nuclear states typically have fewer. The 2010 New START treaty limited the US and Russia to 1,550 deployed warheads each on these platforms.
What is the Non-Proliferation Treaty, and has it worked?
The Nuclear Non-Proliferation Treaty (NPT), which entered into force in 1970, is the cornerstone of the international nuclear nonproliferation regime. It has 191 state parties, making it one of the most widely ratified arms control agreements in history. The NPT rests on three pillars: non-nuclear weapon states agree not to acquire nuclear weapons; the five recognized nuclear weapon states (the United States, Russia, the United Kingdom, France, and China) agree to work toward disarmament; and all parties have the right to peaceful nuclear energy with IAEA safeguards. The empirical record on nonproliferation is better than the pessimism of the 1960s predicted: US government analysts in that era expected twenty-five or more nuclear-armed states by 1975. Instead, only nine states currently possess nuclear weapons, and several states that had active programs — Argentina, Brazil, South Africa (which built six weapons and then dismantled them), Belarus, Kazakhstan, Ukraine — have either stopped or reversed their programs. However, the three states that have acquired weapons outside the NPT — India (first test 1974), Pakistan (first test 1998), and North Korea (first test 2006) — and Israel's undeclared arsenal present ongoing challenges to the regime. The nuclear weapons states' failure to fulfill disarmament obligations has weakened the treaty's legitimacy among non-nuclear states. And the NPT faces a fundamental structural problem: it permits the enrichment of uranium and the reprocessing of plutonium for civilian purposes, activities that bring states within weeks of weapons capability. Iran's program has operated in this legal grey zone for years.
What is nuclear winter, and how concerned should we be about it?
Nuclear winter is the hypothesis that a large-scale nuclear exchange would cause sufficient smoke and soot to be lofted into the upper atmosphere — from the fires ignited by nuclear weapons in cities and forests — to block sunlight globally for months to years, causing agricultural failures and potentially mass starvation across the globe, including in countries not directly targeted. The hypothesis was developed in a landmark 1983 study by Sagan, Turco, Toon, Ackerman, and Pollack (the TTAPS study, named after their initials), published in Science. Their initial estimates were dramatic: a major nuclear exchange could cause a decade of subfreezing global temperatures. Subsequent research has refined the estimates. A 2019 study by Robock, Toon, and colleagues in Science Advances found that even a regional nuclear war — their model used a hypothetical India-Pakistan exchange involving 100 Hiroshima-scale weapons, a small fraction of current arsenals — would inject approximately 5 teragrams of soot into the upper atmosphere, causing a global average temperature decline of 1.5 degrees Celsius for several years and disrupting monsoon patterns that provide water and food for billions of people. A large-scale US-Russia nuclear exchange, involving hundreds to thousands of weapons, could inject 150 teragrams of soot and cause temperature declines of 8 degrees Celsius or more — sufficient to cause global agricultural collapse. The nuclear winter hypothesis remains somewhat controversial in its specific parameters, but the broad mechanism — soot injection causing cooling — is considered credible by most atmospheric scientists. It represents the ultimate 'indirect effect' of nuclear weapons: their most dangerous consequences might fall on populations with no strategic relevance to the conflict that triggers them.
How close have we come to accidental nuclear war, and what are the main risks today?
The historical record of nuclear near-misses is more alarming than is generally appreciated. The Cuban Missile Crisis in October 1962 is the best-known case. Less well-known is that during the crisis, a US Navy destroyer was depth-charging a Soviet submarine whose crew, unable to communicate with Moscow for days, had no way of knowing whether nuclear war had started. The submarine's captain prepared to launch a nuclear torpedo; only the objection of one officer — Vasili Arkhipov — prevented it. In 1983, Soviet early-warning satellites detected what appeared to be five incoming US ICBMs. The duty officer, Lieutenant Colonel Stanislav Petrov, was supposed to report the alarm to his superiors. Instead, reasoning that a real US first strike would involve hundreds of missiles, he classified it as a false alarm — correctly; it was sunlight reflecting off clouds in an unusual way. In 1995, a Norwegian weather rocket triggered Russian early-warning systems, and President Yeltsin had his nuclear briefcase activated. He had ten minutes to decide whether to launch. The rocket was identified as benign with minutes to spare. Current risks come from several sources: the collapse of the intermediate-range nuclear forces treaty in 2019 and the potential expiration of New START; the development of hypersonic glide vehicles that travel too fast for current warning systems; the potential for artificial intelligence-enabled cyber attacks on nuclear command and control; and North Korea's growing arsenal and the risk of miscalculation in a crisis. The Bulletin of Atomic Scientists set its Doomsday Clock to 90 seconds to midnight in January 2023 — the closest it has been since its founding in 1947.