from the FAS Project on Government Secrecy
December 6, 2000


Now that Congress has drastically expanded the number of national security personnel who are subject to polygraph testing, the government is belatedly initiating new studies to test the validity of the polygraph.

The Department of Energy will release $860,000 to the National Academy of Sciences for an 18 month study, the Albuquerque Journal reported December 5:

Senator Jeff Bingaman, who had requested the study over a year ago, said in a press release: "The distinguished scientists and engineers who work at Sandia and Los Alamos deserve to know whether polygraphs produce valid results and this study will help make that determination." See:

Meanwhile, the Department of Defense announced in Commerce Business Daily today that it is seeking a contractor to perform a new study on polygraph validity.

"The goal of this project is to manipulate volunteers into telling specific lies during polygraph examination to test the accuracy of the polygraph examination procedure," the DoD announcement said. See:

But there is a methodological difficulty in such tests that tends to render them useless: There is no way to ensure that a volunteer who is instructed to lie in an artificial setting will present the same physiological signs as an actual liar in a setting where the stakes are genuine.

The irony of these new polygraph studies is that they follow, rather than precede, the expansion of polygraph testing. This reflects the fact that the congressional radicals who are driving security policy are largely indifferent to questions of scientific validity.

Worse, the new National Academy study might make it more difficult to repeal the latest increase in polygraph testing in the short term. This is because proponents will be able to argue that no action should be taken during the eighteen months that the matter is under review by the Academy.


The hundreds of secret compartmented programs at the Pentagon are each assigned a two word nickname and a classified codeword. The names are selected and combined, more or less arbitrarily, from lists of pre-approved words, giving rise to an oddly suggestive form of Pentagon poetry.

William M. Arkin, the independent analyst and author, picked out the names of 26 particularly resonant Pentagon programs -- DIAGONAL GLANCE, CHALK POINSETTIA, CREDIBLE WOLF -- for a feature he wrote in last Sunday’s New York Times Magazine.

Several of these terms have not appeared previously in the public domain. Furthermore, in several cases Arkin has gone where no uncleared person has gone before and has provided a thumbnail description of the contents of the secret programs. Someone, somewhere in the Pentagon, will be grinding his teeth at the impudence. See:

An additional layer of obscurity is attached to these names because of the fact that over time they may be reused for other secret programs.

For example, Arkin identifies CLOUD GAP as a "classified Air Force program." In the 1960s, however, CLOUD GAP was a joint program of the Department of Defense and the Arms Control and Disarmament Agency "to test the feasibility of theoretical arms control and disarmament measures." The earlier edition of CLOUD GAP, now practically forgotten, culminated in a test to demonstrate the verifiable destruction of nuclear warheads.


To subscribe to Secrecy News, send email to with this command in the body of the message:

To unsubscribe, send email to with this command in the body of the message: Secrecy News is archived at: