Polygraph testing. DOE says fewer workers will face the machine.

Abstract

POLYGRAPH TESTING Bowing to mounting criticism, the Department of Energy (DOE) will curtail its use of polygraph tests among nuclear weapons scientists and security officials. Last week DOE Undersecretary Kyle McSlarrow told a Senate panel that the agency will shrink the potential pool of employees required to take lie detector tests from about 20,000 to 4600. But critics say the government is still putting too much faith in a device that has little scientific credibility. DOE “cannot continue to hinge the careers of scientists on voodoo science, no matter how few or great the number,” says Representative Ellen Tauscher (R-CA), whose district includes Lawrence Livermore National Laboratory. Polygraph machines monitor subjects' blood pressure, pulse, respiration, and skin conductivity as they answer questions. Advocates say the machine can tell if someone is lying. The U.S. government has long required polygraph tests for scientists, soldiers, and spies who handle sensitive information, even though most courts bar the results from trials because of doubts about reliability. Spy scandals at Los Alamos National Laboratory in New Mexico in the late 1990s led DOE to expand its polygraph program, despite widespread opposition from researchers. DOE is now having second thoughts about policies it adopted earlier this year, McSlarrow told the Senate Committee on Energy and Natural Resources on 4 September. One factor is a recent report from the U.S. National Academy of Sciences (NAS) that found polygraphs were bound to produce faulty results and could give counterintelligence officials a false sense of security ( Science , 25 April, p. 577). “I found many of the NAS's concerns about the ‘validity’ of polygraph testing to be well taken,” McSlarrow said in outlining the revised plan, which will cover an estimated 4600 employees with access to the most sensitive weapons and intelligence information. “This category will not include everyone with a ‘Q’ or a Top Secret clearance, nor will it include all weapons scientists,” he noted. In addition, the agency will conduct random tests among about 6000 other employees. McSlarrow also emphasized that failing a single test wouldn't be grounds for dismissal, although it would trigger a broader investigation. “This is a smart decision by DOE,” says Senator Pete Domenici (R-NM), a polygraph critic. McSlarrow expects the latest plan to be available for public comment by the end of the year. But the shift may not actually cut DOE's current use of polygraphs, warns Jeffrey Colvin, a physicist at Livermore and a leader of the Society of Professional Scientists and Engineers, a union opposed to testing. The department has never tested all potential candidates, he says, so the switch could simply codify the current program. Steven Aftergood of the Federation of American Scientists in Washington, D.C., questions the usefulness of the new random testing program. “It is sort of a homeopathic approach to security policy,” he says, “in which the mere specter of a polygraph test … is believed to have a deterrent effect.”