The path to nuclear test ban treaties was fraught with complex verification challenges, particularly concerning the detection of underground nuclear explosions. The ability to reliably monitor compliance was a central concern for all parties, leading to extensive technical discussions and political impasses. These challenges ultimately influenced the scope and limitations of early agreements, including the decision to pursue a partial rather than
a comprehensive test ban.
The Geneva System and its Limitations
In 1958, the Conference of Experts in Geneva convened to study methods for detecting nuclear tests. This international gathering of scientists devised an extensive control program known as the "Geneva System." This system proposed 160–170 land-based monitoring posts, 10 sea-based monitors, and occasional flights over land following suspicious events. The experts concluded that this scheme could detect 90% of underground detonations down to 5 kilotons and atmospheric tests with a minimum yield of 1 kiloton. This technical finding was initially well-received and endorsed by the U.S. and UK as a basis for test-ban negotiations.
However, the Geneva System soon faced scrutiny. The experts' report did not address crucial political questions, such as who would conduct the monitoring or when on-site inspections, a key U.S. demand, would be permitted. Furthermore, detecting outer-space tests (above 50 kilometers) was deemed impractical. The sheer size and potential cost of the Geneva System also raised concerns about its feasibility. More significantly, data from the 1958 Hardtack operations, particularly an underground test called Rainier, led U.S. scientists, including Hans Bethe, to believe that the Geneva findings were too optimistic regarding underground test detection. They revised their estimate, suggesting the system could only detect tests down to 20 kilotons, significantly higher than the initial 5-kiloton estimate. This implied a need for a much larger and more expensive detection regime, including new monitoring posts within the Soviet Union, which the Soviets dismissed as a pretext.
The Problem of Decoupling and On-Site Inspections
Further complicating verification efforts was the concept of "decoupling." In June 1959, analysis conducted by the Livermore National Laboratory and RAND Corporation, at the instruction of Edward Teller, revealed that the seismic effect of an underground test could be artificially dampened. This meant a 300-kiloton detonation could be made to appear seismically as a one-kiloton detonation, making detection extremely difficult. These findings were largely affirmed even by pro-ban scientists like Bethe. A panel chaired by Robert Bacher delivered another blow, concluding that even on-site inspections would face significant difficulties in definitively determining whether an underground test had occurred.
These technical uncertainties fueled political disagreements over inspection quotas and the authority of an international supervisory body. The Soviet Union's proposal for verification was seen by Western nations as too reliant on self-inspection, with control posts primarily staffed by citizens of the host country and minimal international oversight. The West insisted on a more robust system, with half of a control post's staff drawn from another nuclear state and half from neutral parties. The Soviet Union also demanded that the Control Commission, the international supervisory body, require unanimity for its actions, effectively giving Moscow a veto. The West rejected this, advocating for a system that would not be easily paralyzed by a single nation. These persistent disputes over the technical and political aspects of verification ultimately contributed to the abandonment of a comprehensive test ban in favor of a partial one, as the complexities of reliably detecting all forms of nuclear tests proved too great to overcome in the early Cold War environment.
