Lie Detectors - The Industry, the Technology and the Victims.
The first lie detector, employed centuries ago, was a handful of rice dropped into the mouth of a suspect. If the rice stayed dry while he answered questions, he clearly was a liar — exposed under the questionable theory that a liar's salivary glands would dry up when gripped by fear. The lie detector used most commonly today is far more sophisticated. Developed by the psychologist and criminologist Leonard Keeler almost forty years ago, it comprises a pneumatic tube that fits across a subject's chest to measure breathing, an inflatable rubber cuff that wraps around the arm to measure blood pressure and a pair of electrodes that touch the fingers and, by the flow of current, measure the dampness of the palm. These instruments activate pens that draw wiggles and waves on a rolling sheet of paper — a process that gives the lie detector its modern name, polygraph, Greek for "many writings." In theory, an examiner can look at the chart, note any unusual wiggles and waves, and nab his man. This polygraph, obviously more complicated than a few grains of rice, is also touted as more accurate. In truth, it is not.
In America, the polygraph has become big business and a fixed part of the national consciousness. Local police officers, tired of more disagreeable and fatiguing methods of questioning, use lie detectors on the husband who glibly announces that he was upstairs snoozing when someone downstairs whacked his wife's head with a bat. Politicians deny they are scalawags and scoundrels and then offer to take a lie-detector test to prove it. Big stores use lie detectors to discover whether prospective employees are, ever have been or ever will be pilferers, perverts or pinkos.
The federal government is devoted to polygraphs. A recent survey by the House Subcommittee on Government Operations, headed by Rep. John E. Moss (D., Calif.) revealed that nineteen federal agencies use them. Two of these — the Central Intelligence Agency and the National Security Agency — characteristically classify their use of polygraphs. The other seventeen, according to the report, gave 19,122 lie-detector tests in fiscal 1963. The federal government in that same year owned 512 lie detectors worth $428,066, and had 639 employees, paid $4.3 million a year, who were trained, in some fashion, to operate them. In addition, federal agencies employed private investigators to conduct 322 tests during the year. The government uses the polygraph for many reasons: to investigate leaks of information and other security cases; to question Vietcong guerrillas captured in Vietnam; to investigate crimes at military bases, post offices and other federal jurisdictions; and to screen applicants for federal jobs. A year ago, the Federal Aviation Agency hoped to adopt some simple device to detect bomb-carrying passengers. Under the plan, a passenger would check his baggage at the airline counter, put his hand on a lie detector, and swear he had no bomb. The FAA, however, dropped the plan when no manufacturer could meet the specifications. There have been more extreme proposals for government use of the polygraph. A Rand Corporation study several years ago suggested that they might be used to police an arms-control agreement with the Russians. The U.S. would periodically examine high Russian officials with lie detectors to make certain that they were living up to the agreement.
No thorough survey has been made on the prevalence of lie detectors in state and local government and in private business, but it must greatly exceed the use within the federal government. There are now 500 lie-detector firms, employing 1,000 to 1,500 polygraph examiners, who question subjects at a cost of about $35 a test. Business Week says that 80 per cent of their income comes from business firms that want to know if their employees or prospective employees are honest. The income of one of the major security firms, John E. Reid & Associates, increased 26 per cent in 1960. Other figures give some hint of the money in lie detection: the two chief manufacturers, the Stoelting Company and Associated Research, both of Chicago, charge from $900 to $1,800 for a polygraph. At least three private schools train operators. A typical one, Cleve Backster's School of Lie Detection in New York, charges $500 for its six-week course.
The polygraph industry has developed a scientific veneer and a jargon. The operators belong to the American Academy of Polygraph Examiners, the Academy for Scientific Interrogation and the National Board of Polygraph Examiners; in a handful of states they are licensed. They talk about "peak of tension" tests and "control type" questions and "preventive security," and they boast, as George Lindberg of John E. Reid & Associates did in testimony before the Moss subcommittee, "we can make decisions in better than 90 per cent of the cases tested . . . [with] an accuracy capability of less than 1 per cent error. . . ."
The obsession of lie detectives with scientific language emerges in an exchange during the Moss subcommittee hearings — hearings, incidentally, that revealed more about the practice of lie detection in America than had ever been put on the public record:
Moss: Those questions are developed by the operator?
Cleve Backster: By the examiner, sir.
Moss: The examiner. Is he different than the operator?
Backster : He certainly is.
Backster: Yes, in our language he is.
Moss: ... In your own personal experience in the type of schooling that you provide and that you recommended here today, is there a difference?
Backster: Yes, sir; because when a person says operator it costs him 10 cents each time at our school.
Moss : Are we in a matter of personal pride or are we in a matter of semantics?
Backster: We are in a matter of standard phraseology used in a highly refined field.
The lie-detector experts have developed a complex system of questioning designed to exclude the possibility that the wiggles and waves on a chart could come from honest jitters instead of guilt. The examiner includes a control question designed to elicit some guilt feelings from a person who is telling the truth on the really important question. If a person's emotions produce physiological excitement on the control question but not on the hot question, he is normal and innocent. If, on the other hand, his emotions spark on the hot question but not on the control question, he must be lying. This theory was discussed at length with Backster at the subcommittee hearings:
Backster: That [the control question] would be such as "during the first 18 years of your life, do you remember ever doing something to hurt somebody who trusted you?"
Rep. Henry S. Reuss (D.,Wis.): . . . Then you get a little upward squiggle of the needle at that point, and you are thus inferring that the fellow must be leveling with you because he gets a little perturbation because, as you know, he must have done something to hurt somebody who loved him?
Backster : Sir, we expected him to respond to that particular question if he were not wrong on the relevant question. His psychological perception set if he were wrong on the target issue [i.e., a liar] . . . would be aimed toward that which could hurt him most and he would go right by the [control question] . . . showing no significant response. This is a verified principle and validated.
Reuss: Where did you get this great truth from?
Backster: Mr. John Beid [the Chicago polygraph expert] is the one who got the initial great truth regarding this, and I am very happy for it because it was the first breakthrough in lie detection for years.
Fred E. Inbau of the Northwestern University School of Law, co-author with Reid of the standard text in the polygraph field, defended this "great truth" and others like it by analogy. He reminded the subcommittee : "At one time it was said that the bumblebee according to the laws of aerodynamics should not he able to fly, but the bumblebee hadn't heard about this and went ahead flying anyway."
But logic like this has not impressed Congress. "I have yet to see any convincing evidence," said Representative Reuss, "that the so-called lie detector is of any legitimate use at all other than in its psychological effects on the subject, and this could be done with a box and some whirring noises and lights quite as well. . . ." Reuss examined the Inbau and Reid statistics closely and discovered that, although the two polygraph experts claimed an error of only 1 per cent, they could verify only 18.9 per cent of their cases.
The fact is that no experimental research has been done to prove that lie detectors work or that they don't. Dr. John I. Lacey, a psycho-physiologist at Antioch College, told the subcommittee that "the field of lie detection has . . . developed outside of the confines of any of the recognized scientific disciplines." After an exhaustive study for the Department of Defense two years ago, Dr. Jesse Orlansky of the Institute for Defense Analyses wrote: "Although the polygraph enjoys wide usage, we are not able to estimate its value. It is possible that the regard in which it is held is due largely to the ability of the examiners to conduct effective interviews and only slightly to the polygraph instrument itself; or the reverse. We do not know." Ironically, the Department of Defense, one of the federal government's most prolific users of lie detectors, received the report, classified it and then ignored it. Moss subcommittee investigators later uncovered it.
Some doubt has also fallen upon the instrument itself. It may not be as technically sophisticated as some polygraph experts pretend. Dr. Lacey, for example, described it as "a fairly crude piece of instrumentation." He said the manufacturers have made "no attempt to take advantage of the many, many things we have learned in the past two decades about the recording of physiological responses . . . the nature of the examiner-examinee interaction . . . new methods of display, new methods of computation." And a recent article in the Harvard Business Review concluded that the polygraph can be unreliable unless the operator takes into account room temperature, humidity, time of day, air content and the recent activities of the subject.
While the polygraph industry's claims to high science have been shattered by the available psychiatric and physiological evidence, the instrument does seem to have a limited value. The scientists who testified before the subcommittee agreed that polygraphs work better than chance — perhaps with 70 percent accuracy — and, in the hands of trained and competent operators, could turn up useful clues for further investigation.
But very few polygraph operators have the competence and training. "There are a number of individuals," Dr. Joseph F. Kubis, a Fordham University psychologist, told the subcommittee, "who buy these instruments, do not take any training, set up, as it were, a shingle, and operate with this instrument. . . . There are more of them than we can see and identify." Even Professor Inbau admitted that 80 per cent of the operators are insufficiently trained.
In a field so unsupervised, the Moss subcommittee had difficulty finding out just how much training a polygraph examiner required. "The more training the better, obviously." said Dr. Martin L. Orne, Harvard psychiatrist. "But there would be no point in saying that you should have only Ph.D. psychologists. They are not available." But in general the scientists who testified tended to see a need for far more training than did the representatives of the polygraph industry. Dr. Kubis thought that operators should have a minimum of six months' training in use of the machine and then a year of further experience interpreting the results. Dr. Lacey said he could discuss standards only by describing the background of those who read the physiological machines in his laboratories. Nurses, X-ray technicians and college graduates, he testified, were allowed to operate the machines in his lab after at least six months' training. But he went on, "I am not comfortable leaving them alone . . . for the complete interpretation of a set of data ... in three years."
Within the industry, Inbau, Reid and Reid's associate, George Lindberg, prepared a paper for the Moss subcommittee setting their proposed minimum requirements as a college degree, six months' training in use of the polygraph, and then six months' personal supervision of the student as he conducts examinations. These standards, however, did not please Backster, who runs a six-week school. "I don't understand," he said, "by virtue of these standards that are so stipulated how with less than perhaps fifty people who have acquired their training by this method, the other 2,000-some examiners have managed to get along."
At the moment, Backster's standards prevail, particularly in the federal government. A former employee of the CIA, Backster set up its lie-detector program and, as he testified, "on one occasion or another I have been a consultant to most every government agency that is using the polygraph today." Almost half the federal agencies require no further background of a trainee than a high school diploma, and the extent of federal training ranges from the maximum of a seven-week course in the Army to the five- to eight-day course given by the Navy. This sorry record prompted Moss to conclude, "I have a real concern, in fact, a firm conviction, of the complete inadequacy of the training."
An imperfect instrument in the hands of an ill-trained man who believes that both he and the instrument are infallible can lead in only one direction — to the entrapment of the innocent. While polygraph experts like John Reid maintain that they err only in occasionally freeing a guilty man, never in ensnaring an innocent one, the evidence proves otherwise. Even Professor Inbau acknowledged it. "If you get an examiner," he testified, "who will browbeat everybody who takes the test hoping he will get a confession from the guilty ones and be able to report these people are lying, these things can happen."
A year ago, in an article written for the American Journal of Psychiatry, Dr. H. B. Dearman, a psychiatrist, and Dr. B. M. Smith, a psychologist, reported a case in which the polygraph had hurt an innocent subject. A young bank vice president, submitting to his firm's semi-annual routine lie-detector examination, failed the question, "Have you ever stolen any money from the bank or its customers?" He was given a test four times, and he failed the question each time. Convinced of the machine's infallibility, thoroughly confused, knowing that he had padded some expense accounts a bit, the young man finally broke down and confessed to the crime. He had stolen $1,000 — a figure arrived at through some other questions on the polygraph tests. The bank books were audited. No money was missing. The bank president sent the young man to Dr. Dearman. The psychiatrist discovered that he was disturbed about his feelings toward his wife and mother, both of whom were customers of the bank, and that he felt guilty about some financial transactions with them amounting to approximately $1,000. Dr. Dearman decided that the young man's physiological changes, as recorded by the polygraph, reflected these emotional problems at home.
It is in personnel screening, however, not crime detection, that the polygraph can do the most damage. There is absolutely no recourse for a man who has failed a lie-detector test while applying for a job. He has no way of proving that the lie detector — which allegedly measured his future truthfulness and honesty — is wrong. He could do that only by staying truthful and honest on the job. But he has no job. Moreover, his failure on the lie-detector test may be noted in his personnel file and in government dossiers and scar him forever. Representatives of the polygraph industry estimate that lie detectors reject from 25 to 30 per cent of the job applicants who take the examinations. These figures grow starker when the polygraph experts boast that these rejectees probably would have been hired if management had been dumb enough to screen them by the usual technique of interviews and reference checks.
"I would judge or estimate," Cleve Backster testified, "that about one out of every four that would undergo and have gone on the job by virtue of the conventional forms of screening would have disqualified themselves by their own assertions during the polygraph examination."
"Does this strike you as a high figure?" asked Rep. Ogden Reid (R., N.Y.).
"I would say initially it struck me as a high figure; yes, sir," Backster replied. "I am not shocked by it any more."
Since personnel screening is not designed to uncover a specific act, but only a tendency to act, polygraph operators can accomplish less and boast more in this field than in any other. The work can be sloppier than usual, the machine can take down less information — no one need know. A trade magazine, Bus Transportation, published an article in 1956 entitled, "Electronic Marvel Weeds Out Dishonest and Unfit Applicants," which described a new invention of Cleve Backster's. "With a special machine known as the Backster electronic evaluator it's now possible for a personnel executive to find out, in two hours if necessary, just about everything of importance in the background of a prospective employee," the article said. In its description of the "electronic marvel," the article said it comprised no more than the electrodes that measure the electric currents across the palm of the hand. Even Backster, in his testimony before the Moss subcommittee, acknowledged that this kind of measurement by a galvanometer alone would tell an examiner next to nothing about a subject. In fact, he told the subcommittee, an examiner would have to be at least "partially without sound reasoning" to depend on a galvanometer. Confronted with the Bus Transportation article, which he had reprinted and distributed for promotion, Backster said the writer had omitted the fact that his "marvel" measured breathing as well as palm sweating.
"You get these articles," Moss said at the hearings, "that say the scientific wonder, and with it all you have to do is set an employee down and ask him a series of questions and when you are finished you know if he has problems, you know if he is truthful, you know if he is honest, all of these things, you know. The fact is you know nothing. You may have many suspicions, but no more knowledge than when you started. ... I suspect . . . that this little device has robbed many a person of an opportunity for employment or promotion."
Even harsher criticism of the use of the polygraph in personnel work came from Dr. Kubis, the Fordham psychologist who has spent much of his career experimenting with lie detectors. "The growing use of lie detectors in many vital aspects of our life is unwarranted, dangerous and degrading," Dr. Kubis testified. He condemned the lie detector, when used by inexperienced and unscrupulous men to check employees and prospective employees, as "a blindly probing instrument that can severely damage the inner life and reputation of the person tested."
Certain practices which have nothing to do with the machine itself, but which nevertheless have grown into the polygraph tradition, are distasteful to anyone concerned with the rights of American citizens. Most examination rooms, for example, have two-way mirrors behind which other investigators can watch the proceedings. Many rooms have hidden microphones.
Until very recently, the armed services never told a subject about the mirror and the microphone. One subcommittee witness, Maj. Gen. Ralph J. Roberts, the Army Provost Marshal General, didn't think it was necessary to tell a soldier anything. "It is my experience in thirty-six years in the Army that all of the soldiers that go into CID [Criminal Investigations Detachment] know that they are monitored and Big Brother is watching," he said. Similar testimony came from J. M. Barron of the Office of Naval Intelligence, who told the subcommittee that the Navy doesn't volunteer information about the rights of a citizen to a polygraph subject because "he is an American citizen. He should know." This brought a bristling reply from Congressman Reuss: "I would comment that I think the day will never come when the American citizens, as an incident of their citizenship, have to be aware of the fact that the room is bugged and the mirror is a device to enable Big Brother to see them. . . ." The anger from the Moss subcommittee persuaded Deputy Secretary of Defense Cyrus Vance to issue a new directive that ordered Pentagon polygraph examiners to tell their subjects of any mirrors and microphones. However, mirrors and mikes would stay.
But even without mirrors and microphones, there is serious question about the constitutionality of any polygraph test. The polygraph industry defends the use of the machine by pointing out that polygraph examinations are not admissible in most courts as evidence and that no one has to take the test if he does not want to. In this way, the industry maintains, the Fifth Amendment is not violated. But Rep. Cornelius E. Gallagher (D., N.J.) questions whether the examination violates the Fourth Amendment protection against unreasonable search and seizure: "Is not the Fourth Amendment violated if a condition precedent to employment is the subjection of a polygraph test, a test that seeks in some instances ... to determine not only the veracity but to probe the very thought process of an individual?"
Dr. Dearman, the psychiatrist who handled the case of the harassed bank vice president, agrees with Representative Gallagher that the Fourth Amendment is violated, but adds that the Fifth Amendment is violated as well. "I am aware," he said in a letter to Congressman Moss, "that proponents of the use of the polygraph fall back on the statement that no one is forced to take a polygraph test. . . . However, this is of little or no avail because the examinee does not realize that not only will his conscious thoughts and his autonomic responses to them be recorded but his unconscious thoughts will also be delved into and consequently he will give autonomic responses to . . . thoughts of which he is totally unaware. . . . It is my opinion that the polygraph is used mainly as a mental blackjack to obtain a confession."
Whether unconstitutional or not, the polygraph is certainly unjust, imperfect and dangerous. It may have some value in the hands of a highly trained and competent criminal investigator who is seeking leads, but that is the limit of its usefulness. Even there, its scientific validity is so far from perfect and its potential for harm so great that any citizen — no matter how secure in his innocence — is well advised to refuse the test. As a personnel screening device, it can produce nothing but harm (and profits for polygraphists) and Congress should demand the end of its use in federal government personnel offices immediately and should exert all its pressure to persuade private firms to discontinue it. If necessary, it should pass legislation to that end.
The polygraph is a pernicious instrument that has been seized upon by a society obsessed with gadgetry. It should be relegated to a Smithsonian Institution exhibit case as a monument to an American craze.
[Stanley Meisler is a Washington newsman.]