Skip to: Content
Skip to: Site Navigation
Skip to: Search


How many bomb mishaps?

December 26, 1980



A breif Reuters dispatch that came across our desk this week raises some disturbing questions about the Pentagon's handling of nuclear weapons, as well as the reliability of the "fail-safe" mechanisms used to ensure that such weapons are not accidentally detonated. Presumably there is no cause for alarm. On the other hand, we would feel remiss in not noting our concerns and urging Congress and the Pentagon to make a careful examination to determine whether existing safety procedures are adequate.

Skip to next paragraph

The dispatch, which triggered a flurry of inquiries from the news media, said that the Pentagon has hitherto underidentified the number of mishaps involving nuclear weapons. The Pentagon denies any deliberate misrepresentation. But the very issue of the numbers involved is one of the points that bothers us.

Before the Reuters article, the Pentagon's list of "broken arrows" (the code word for nuclear accidents) was relatively modest, ranging anywhere from 13 to 25 or so incidents. Whatever the accuracy of past counts, the Pentagon nowm concedes that there have been at least 32 serious incidents that it can publicly discuss. This at least three times the number given in the last official count in 1968. Some accidents cannot be identified, the Pentagon says, because of "international" considerations; in other words, because several "broken arrows" involved US weapons stored in or transported through other nations.

Private defense analysts believe the figure may be higher than the currently acknowledged 32 incidents. The Center for Defense Information, a liberal Washington-based "think tank," estimates that there were 95 accidents in just the period from 1950 through 1975.

Should not as much information as possible be made available to the public when a nuclear weapons mishap occurs? Are appropriate steps being taken to ensure that such accidents do not lead to detonation, with all its grave consequences? The weapons are under various commands. Tactical nuclear weapons , for example, come under the control of local area commanders. Is the Pentagon certain that safety procedures are uniformly carried out wherever weapons are located?

There is also question about the accuracy of the various mechanical devices used to prevent weapons from accidentally exploding. The Pentagon is justifiably loath to provide a complete accounting of all the switches used to detonate warheads. But is the report true that at least one mishap came periously close to detonation, as the Reuters article states? That involved a nuclear bomb that was jettisoned from a B-52 back in 1961, when five of six safety switches on the warhead were allegedly activated. That happened many years ago. But many US nuclear weapons are 25 to 30 years old, such as the Titan missiles. Are safety devices being constantly updated, based on the latest engineering technology?

The Pentagon points out that there has never been an accidental detonation of one of its weapons. It also maintains that its "personnel reliability program" for the mainte nance, security, and operational workers dealing with nuclear warheads is adequate At least two persons are always teamed for any task involving the weapons -- in part to monitor each other. Individuals are removed from work during periods of personal stress. Whenever weapons have to be transported, local and state police are called in for assistance.

Finally, the Pentagon argues, in a mishap no weapons has ever really even come close to detonation because there was never an "electronic interface" (triggering) of a warhead. In the case of a B-52, for example, the pilot, copilot and navigator must together go through a complicated start-up process to activate a weapon.

The Pentagon, as the Reuters dispatch notes, is currently in the process of updating its listing of nuclear accidents. There is also apparent controversy within the military establishment over defining exactly what constitutes a "broken arrow." In fairness to the military, the definition factor must not be overlooked, since it will determine the numbers count and could well lead to confusion about whether there is misrepresentation or just uncertainty about categorizing various types of incidents.

Nonetheless, the Defense Department has a clear obligation to provide whatever information it can about each "broken arrow" as it occurs -- not years later and only when reluctantly prodded into disclosure by a critical press report. Neither the Pentagon or Congress should become blase about safety procedures for nuclear weapons, a danger that arises when weapons have been around so long as to have gained a certain degree of "familiarity." To become indifferent to the potential risks would be an unpardonable betrayal of public trust.