The central deception involved in the polygraph examination is the notion that it detects lies.

In his secret testimony before the Warren Commission in 1964, J. Edgar Hoover, made the point that the "polygraph, often referred to as a lie detector, is not in fact such a device." So has every serious study conducted by government agencies has confirmed Hoover's assessment. For example, the Office of Technology Assessment issued a massive report in 1983 entitled "Scientific Validity of Polygraph Testing: A Research Review and Evaluation," concluding that "the available research evidence does not establish the scientific validity of the polygraph test for personnel security screening." There are still no peer-reviewed test using randomly distributed true and false answers. As recently as 2000, the U.S. Supreme Court, in a decision upholding the absolute ban on the use of polygraph tests in court-martial proceedings said, "There is simply no consensus that polygraph evidence is reliable." In fact, there is no epistemological basis for believing that the polygraph machine is even designed to detect lies. It measure emotional responses that can proceed from many factors including fear, anxiety and nervousness. Whether or not such emotions are connected to deception cannot be determined from the data.

Not uncommonly, the emotional response may be to the question but to the tone and manner of the interrogator. Consider, for example, one of the most important polygraphs examinations ever conducted by the CIA, the interrogations of Yuri Nosenko. Nosenko had been a KGB defector who claimed to be Lee Harvey Oswald's case officer when de defected to the US. In 1966, the CIA believed he was a false defector and wanted to break him. The polygraph operator asked Nosenko a dozen key questions and assessed that he was lying in his most important answers. Then, in 1968, there was a new regime in the CIA that believed Nosenko was a genuine defector and wanted to rehabilitate him. He was given a second polygraph and asked substantially the same questions and gave the same answers. But this time, it was assessed Nosenko was truthful in all his responses. What had happened? The variations in Nosenko's emotional reaction between the 1966 and 1968 polygraph were attributed to the very different attitudes of his interrogators. The first had sounded hostile and provoked a "lie" response, the second sounded friendly and provoked a "truth" response.

The polygraph, though not a lie detector, is an inquisition tool. As such, it can elicit a confession if the subject can be deceived into believing that it is a lie detector. He may then assume that since he cannot conceal the truth, his best option is to reveal it. Polygraphs are thus designed, not just to measure responses, but to deceive subjects that they possess omniscient power to detect lies. To this end, the polygraph is embellished with multiple sensors. The CIA's polygraph machine, for example, uses rubber hoses on both the chest and abdomen to measure changes in respiration, electrodes, which are attached to fingertips, to measure changes in sweat, skin temperature and vessel constriction and a blood pressure arm cuff to measure changes in the heart rate. If any of the three sensors were actually capable of measuring deception (which they are not), the other two devices would be redundant. But three sensors— a polygraph instead of a monograph— helps the deception by dazzling subjects. If only one sensor was used— say, for, example, a blood pressure cup— subjects might focus on the issue of what blood pressure changes actually measure. Triple instruments make such a focus more difficult (which is why dazzling, or overloading the brain, is an effective technique in deception).

The polygraph examiner purports to interpret the data while asking a series of questions by allegedly comparing the reactions to each set of questions. A common technique to deceive subjects into believing the machine is omniscient is to ask a question early on to which the interrogator knows the answer and which the subject will answer deceptively. He can then pretend that the machine caught the lie. In Nosenko's initial polygraph, for example, Nosenko's case officer fed the polygraph operator a question about his rank in the KGB to which Nosenko had previously exaggerated (according to documentary evidence in the hands of the CIA) When he repeated this exaggeration on the polygraph, the operator, following a script, stopped the machine and told him he was lying.

Such deception techniques does not require a polygraph machine. Law enforcement agents have gotten similar confessions by convincing suspects that a office Xerox machine was a lie detector or, in other cases, that they had ESP. The polygraph simply combines these techniques in a package of science.

Of course, the CIA, FBIand other Intelligence agencies know that the polygraph machine is not an effective lie-detector. If it actually worked, they would be out of the espionage business. The CIA, for example, would have to assume that any mole that it recruited in the Russian intelligence services would be revealed by a polygraph test that the Russians would admininister.The Russian intelligence services would conversely assume all their spies would be caught by the CIA's regular polygraph examination. But, as Aldrich Ames, Robert Hanssen and many other KGB moles proved is that they could lie year after year without it being detected by CIA and FBI polygraphs.

So all the government agencies that regularly use this device— the FBI, CIA, Secret Service, NSA, DEA, Department of Energy (to examine thousands of scientists)-- are fully aware that it is not a lie detector. Nevertheless, the virtue of this deception is that, so long as deceives its targets, it provides an incentive to tell the truth.

Questions? Email me at
This website is still (heavily) under construction. The webmistress can be reached at