In chapter 2 of The Lie Behind the Lie Detector
(https://antipolygraph.org/pubs.shtml), Gino Scalabrini and I
wrote:
In 1995, a "walk-in" approached the Central
Intelligence Agency outside of the PRC and provided
an official PRC document classified "Secret" that
contained design information on the W-88 Trident
D-5 warhead, the most modern in the U.S. arsenal,
as well as technical information concerning other
thermonuclear warheads.
Thus began an ongoing investigation of suspected Chinese
espionage within the Department of Energy, according to
chapter 2 of the report of the House Select Committee on U.S.
National Security and Military/Commercial Concerns with the
People's Republic of China, more commonly known as the "Cox
Report." But in the very next paragraph, the Cox Report
notes:
The CIA later determined that the "walk-in" was
directed by the PRC intelligence services.
Nonetheless, the CIA and other Intelligence
Community analysts that reviewed the document
concluded that it contained U.S. thermonuclear
warhead design information.
The Cox Report does not disclose how the CIA determined that
the "walk-in" was "directed by the PRC intelligence
services." Nor does the Cox Report offer any insight into why
the PRC intelligence services would provide the CIA with
documents that could reasonably be expected to compromise
their own sources and methods.
Could it be that the CIA determined that the "walk-in" was
directed by the PRC intelligence services because a CIA
polygrapher found portents of prevarication when he gazed
into the polygraph charts? As previously noted (p. 16),
hundreds of CIA employees were unable to pass their polygraph
screening exams in the wake of Aldrich Ames' arrest in 1994,
and the 1995 "walk-in" incident occurred squarely in that
wake. If the CIA did terminate its relationship with the
"walk-in" based on the voodoo science of polygraphy, then it
committed a blunder of monumental proportions.
The Cox Report is available on-line at:
http://www.fas.org/spp/starwars/congress/1999_r/cox/
Support for our speculation on the role of polygraphy in CIA's
determination that the "walk-in" was a double agent appears in
today's (Thursday, 19 Oct. 2000) Washington Post in an article by
Walter Pincus and Vernon Loeb entitled, "Spy Probe Shifts to
Missiles":
http://washingtonpost.com/ac2/wp-dyn/A34437-2000Oct18?language=printer
Pincus and Loeb write:
Because of the CIA's belief that the walk-in was a double
agent, a full translation of the documents seemed less
pressing. "He failed an agency polygraph," one intelligence
official explained. The CIA's suspicions about the informant
also slowed the FBI's already limited investigation at Los
Alamos of Wen Ho Lee.
The only "evidence" thus far publicly provided for CIA's
conclusion seems to be that he "failed an agency polygraph."
However, because polygraph "tests" are easy to beat (see chapter 4
of The Lie Behind the Lie Detector), and foreign intelligence
services are likely to know this, and because these "tests" also
have a built-in bias against truthful persons (chapter 3), the
fact that the walk-in "failed" seems to be pretty good evidence
that he was not, after all, a Chinese double agent.
The FBI does not believe that the defector was a double agent.
Pincus and Loeb report:
The CIA concluded several years ago that the defector who
supplied the documents was a Chinese double agent, casting
doubt on the information he delivered and delaying its
translation from Mandarin to English. But the FBI, which has
interviewed the defector in the United States, believes that
he is legitimate. The CIA now says the evidence about the
defector is "inconclusive," but agrees that the information
he handed over has proven accurate, a senior government
official said this week.
This seems to be yet another case where the U.S. Government's
reliance on unreliable polygraph "testing" has caused serious
damage to America's national security. It's high time that those
responsible for our reliance on the voodoo science of polygraphy
be held accountable.
George Maschke
ps: I'll be without Internet access for the next 3 weeks,
but replies are welcome. I'll respond when I return.