Command and Control

Dropping a nuclear weapon was never a good idea: According to a study released by the Armed Forces Special Weapons Project in 1958, “Extreme shocks can cause failure of one or more of the presently used safety devices and warhead components, which could contribute to a full-scale nuclear detonation, particularly if the X-unit is already charged.” See “A Study on Evaluation of Warhead Safing Devices,” Headquarters Field Command, Armed Forces Special Weapons Project, FC/03580460, March 31, 1958, (SECRET/RESTRICTED DATA/declassified), p. 18.

when the Genie was armed, it didn’t need a firing signal: See “Vulnerability Program Summary: Joint DOD-AEC Weapon Vulnerability Program,” Armed Forces Special Weapons Project, FC/010 May 1958 (SECRET/RESTRICTED DATA/declassified), p. 44.

a B-29 bomber prepared to take off from Fairfield-Suisun: For the story of the plane crash and its aftermath, see Jim Houk, “The Travis Crash Exhibit,” Travis Air Museum News, vol. XVII, no. 3 (1999), pp. 1, 5–11; John L. Frisbee, “The Greater Mark of Valor,” Air Force Magazine, February 1986; and the accident report reproduced in Maggelet and Oskins, Broken Arrow, pp. 65–77.

“a long training mission”: Quoted in “Bomb-Laden B-29 Hits Trailer Camp; 17 Killed, 60 Hurt,” New York Times, August 7, 1950.

an American B-47 bomber took off from Lakenheath: I first learned about this accident from a document obtained by the National Security Archive: “B-47 Wreckage at Lakenheath Air Base,” Cable, T-5262, July 22, 1956 (SECRET/declassified). The accident report is reproduced in Maggelet and Oskins, Broken Arrow, pp. 85–87.

“The B-47 tore apart the igloo”: “B-47 Wreckage at Lakenheath Air Base.”

“Some day there will be an accidental explosion”: Morgenstern made the assertion in 1959. Quoted in Joel Larus, Nuclear Weapons Safety and the Common Defense (Columbus, OH: Ohio State University, 1967), p. 17–18.

“Maintaining a nuclear capability”: “A Survey of Nuclear Weapon Safety Problems,” p. 14.

“Acceptable Military Risks from Accidental Detonation”: Although I did not obtain the Army study, its conclusions are explored in “Acceptable Premature Probabilities for Nuclear Weapons,” Headquarters Field Command, Armed Forces Special Weapons Project, FC/10570136, October 1, 1957 (SECRET/RESTRICTRED DATA/declassified).

the acceptable probability of a hydrogen bomb… should be 1 in 100,000: See ibid., p. 4.

The acceptable risk of an atomic bomb… set at 1 in 125: See ibid. p. 4

the “psychological impact of a nuclear detonation”: Ibid.

“there will likely be a tendency to blame”: Ibid.

Human error had been excluded as a possible cause: Ibid., p. 6.

“The unpredictable behavior of human beings”: Ibid.

the odds of a hydrogen bomb exploding… should be one in ten million: Ibid., p. 13.

odds of a hydrogen bomb detonating by accident, every decade, would be one in five: For a nuclear weapon with a yield greater than 10 kilotons, removed from stockpile storage, the study proposed an accidental detonation rate of 1 in 50,000 over the course of ten years. Putting 10,000 of those weapons into “handling, maintenance, assembly and test operations,” therefore, lowered the odds of an accidental detonation to 1 in 5 every decade. See Ibid., p. 14.

the odds of an atomic bomb detonating by accident… would be about 100 percent: For a nuclear weapon with a yield lower than 10 kilotons, removed from stockpile storage, the study proposed an accidental detonation rate of 1 in 10,000 per weapon over the course of ten years. If the United States possessed 10,000 of such weapons, at least one of them would most likely detonate by accident within that period. See ibid., p. 14.

During a fire, the high explosives of a weapon might burn: See “Factors Affecting the Vulnerability of Atomic Weapons to Fire, Full Scale Test Report No. 2,” Armour Research Foundation of Illinois Institute of Technology, for Air Force Special Weapons Center, February 1958 (SECRET/RESTRICTED DATA/declassified), and “Vulnerability Program Summary,” pp. 10–20, 58–60.

The time factor for the Genie was three minutes: Cited in “Vulnerability Program Summary,” p. 59.

Carl Carlson, a young physicist at Sandia, came to believe: A short biographical sketch of Carlson — who advocated passionately on behalf of nuclear weapon safety, resigned from Sandia in frustration at one point, and later took his own life — can be found in Stevens, “Origins and Evolution of S2C at Sandia,” p. 236.

“the real key”: “A Survey of Nuclear Weapon Safety Problems,” p. 28.

the T-249 control box made it easy to arm a weapon: See ibid., pp. 21–27.

“a weapon which requires only the receipt of intelligence: Ibid., p. 51.

“always/never”: Peter Douglas Feaver succinctly explains and defines the “always/never problem” of controlling nuclear weapons in his book, Guarding the Guardians, pp. 12–20, 28–32.

“a higher degree of nuclear safing”: Quoted in “A Survey of Nuclear Weapon Safety Problems,” p. 13.

“Such safing,” Quarles instructed: Quoted in ibid.

The Optimum Mix