Skip to main content

Non-Explosive Nuclear Testing

James Schlesinger, former secretary of defense, secretary of energy and chairman of the Atomic Energy Commission (AEC), is mistaken in arguing that ratifying the Comprehensive Test Ban Treaty–thus ending all nuclear testingmeans that “[o]ver the decades, the erosion of confidence [in the– reliability of U.S. nuclear weapons] would inevitably be substantial” and therefore “[p]eriodic testing remains desirable” (“Nukes: Test Them or Lose Them,” editorial page, Nov. 19).

From 1945 to 1992, the United States conducted 1,030 nuclear tests (not including 24 joint U.S.-British tests). Of these, 83.5% (860) were related to developing new nuclear weapons, 9.5% (98) concerned the effects of nuclear explosions on military hardware, 3.3% (34) were conducted to minimize the risk of accidental detonation, and 2.6% (27) were exploded as part of the Plowshare program to explore civilian uses for nuclear explosives (for example, excavation and mining). The remaining 1% (11 tests) were designed to help detect and locate underground nuclear explosions and analyze the risk of detonation in transportation accidents. In fact, since 1970 only 12 out of 408 tests were conducted to identify or assess the correction of defects in deployed weapons.

Since the AEC initiated its Stockpile Evaluation Program in 1958, more than 13,800 weapons of 45 different types have been disassembled, inspected, and (non-explosively) tested. Only about half of the 800 or so resulting findings have been deemed “actionable,” requiring a modification to either the warhead or the production process which made it or lowering the warhead’s presumed yield or reliability. Most of these findings occurred within the first few years of a warhead’s entry into the stockpile (the newest U.S. warheads were built in 1990). As weapons age, fewer modifications have been required. Moreover, nearly all age-related problems concern non-nuclear components which can be fully assessed and corrected without resorting to nuclear tests. Between fiscal 1980 and 1992, non-explosive testing of weapons and weapons components at the Department of Energy’s Pantex Plant outside Amarillo, Texas, led to 1,414 weapons being disassembled and ultimately disposed of while another 2,449 were disassembled and reassembled using original or replacement parts. The weapons in this latter category were then redeployed.

The real issue is not whether the 12,000 nuclear weapons now in the U.S. stockpile will explode but whether they will explode at their designated yield. Significantly, only seven U.S. nuclear tests have “fizzled,” either failing to detonate or exploding at a yield much lower than predicted (and nearly all of these tests–the last of which was in 1978–were of experimental devices, not actual production models). With the end of the Cold War, the military requirement to guarantee the destruction of superhardened targets such as missile silos, and thus for precise yields, is decidedly less critical. Relatively minor potential degradations in the explosive power of some of our nuclear weapons are unlikely to weaken deterrence (or, as Mr. Schlesinger suggests, spur proliferation), not least because our adversaries and allies will have no idea which weapons, if any, are less “reliable” and, if so, to what degree.

Get daily updates from Brookings