The question of the polygraph's reliability as a lie detector runs throughout history; a history dominated by a single case that took place, oddly enough, before the instrument was invented. This is the Frye Case.1
On November 25, 1920, almost a year after John Larson had joined the Berkeley Police Department and a few months before he had read William Marston's article on lie detection, a young black man named James Frye shot and killed a wealthy physician, also black, in Washington, D.C. The victim, Dr. Robert W. Brown, was murdered in his office at about 8:45 in the evening. Another physician was in the office and witnessed the shooting. Frye ran out of the office with the eyewitness running after him. The chase ended abruptly however when Frye took a couple of shots at his pursuer. Since the witness didn't know Frye, the police had no idea who had committed the crime.
Seven months later, Frye committed an armed robbery which led to his arrest on August 21, 1921. The police questioned him about the robbery and he confessed to it and to the murder of Doctor Brown.
From this point on, the facts of the Frye Case, as evolved over the years, have become more myth than reality. In 1981, a law professor at George Washington University named James E. Starrs set the record straight when he presented a paper about the Frye Case at the annual meeting of the American Academy of Forensic Sciences.2 According to the Frye myth, and the popularly accepted version of the case, Frye confessed on the advice of a friend who told him he would get part of the reward put up by the victim's family if he took the blame for the killing. Frye later repudiated his confession, as the story goes, when he learned that he had been duped. It was at this point his attorney brought William Marston into the case to test his honesty.
Continuing with the popular version of the case, the results of Marston's lie detection test indicated Frye had told the truth when he denied killing the doctor. But the trial court refused to admit Marston's lie detection evidence, so Frye was convicted and sentenced to life in prison. According to the myth, the friend who had talked Frye into confessing later admitted killing the physician. As a result, Frye was freed after living three years behind bars.
The above account not only makes a good story, it sheds favorable light on scientific lie detection. If the trial judge had been more open minded, an innocent man would not have been convicted.
According to Professor Starrs, however, the above version of the Frye Case is grossly inaccurate. What really happened was that Frye took back his confession when his court appointed attorney, Richard V. Mattingly, told him to. By the time the case went to trial, Frye had come up with an alibi. He said he'd been visiting a woman named Essie Watson.
In his book, The Lie Detector Test, Marston states that he was called into the case by Mattingly a few weeks before the trial because the attorney couldn't find any witnesses to support Frye's alibi. Marston went to the jail on June 10, 1922 to give Frye his systolic blood pressure test, a primitive method that involved nothing more than a standard medical blood pressure cuff and a physician's stethoscope. After each question put to Frye, Marston simply took his blood pressure. Compared to Larson's polygraph that at the time was a year old, Marston's test was crude and unreliable.
After Marston had administered his test, he was convinced of Frye's innocence. He wrote: “No one could have been more surprised than myself to find that Frye's final story of innocence was entirely truthful! His confession to the Brown Murder was a lie from start to finish.”3
The murder trial began on July 17, 1922 in a Washington, D.C. court before Judge William McCoy. The defense was based upon Frye's alibi witnesses and Marston's lie detection test. When the defense attorney tried to put Marston on the stand so he could explain his lie detection test results, Judge McCoy objected. The judge also denied the lawyer's request to allow Marston to give an open court demonstration.
The trial lasted four days, and without the lie detection evidence, Frye didn't have much of a defense. According to Marston, someone had frightened off his alibi witnesses. Frye therefore had no choice but to take the stand and testify on his own behalf.
Despite the fact Judge McCoy had made a decision that would become an important and historical legal precedent, his ruling did not affect Frye as one would expect. Since the argument over the admissibility of Marston's test was conducted in front of the jury, the jurors knew that Marston's procedure had corroborated Frye's claim of innocence.
The jury deliberated three hours then returned with a verdict of guilty. But instead of finding Frye guilty of first degree murder, a crime that called for the death penalty, they found him guilty of murder in the second degree, an offense that brought a life sentence. As Marston put it: “As far as James Frye was concerned, the test undoubtedly saved his life. No jury could help being influenced by the knowledge that Frye’s story had been proved truthful by the lie detector.”4
The irony of the Frye Case can be summed up as follows: A guilty man's life was saved by the erroneous results of a lie detector test that had been ruled inadmissible because of its scientific unreliability.
The influence the lie detector had on the Frye jury was exactly the type of thing Judge McCoy was trying to avoid. If lie detection results were offered by the prosecution as evidence of an innocent man's guilt, the results could be disastrous.
Richard Mattingly appealed Fry's conviction on the grounds Judge McCoy had erred in excluding Marston's test. Marston, himself a lawyer, followed the case closely. In a letter to John Larson he mentioned that he had read in the newspaper that Larson had gotten the polygraph evidence into court. “If this is so,” he wrote, “will you not give me details and citation, if possible, to help us in the Frye Case? We are after a Supreme Court precedent in that case, and are carrying it up, after conviction of second degree murder, squarely on the deception test exception, as you may have noted.”5
Larson replied a few weeks later: “I wish to say that the papers were mistaken, for no attempt has yet been made to introduce our deception records into court. I might add that the chief reason has been that we have not been invited to do so.” Later in his letter, Larson made a statement that must have unsettled Marston a bit: “Whenever I can find time to write up the cases, I will send copies of the records of all types of cases and of individuals who have lied without any appreciable change of blood pressure, as well as those who have shown extreme changes.”6
In 1923, the Circuit Court of Appeals in the District of Columbia upheld Judge McCoy's exclusion of Marston's lie detection results. Judge Van Orsdel wrote the appellate court's opinion which established a test for the admission of expert testimony based upon new scientific principles, a standard that is used by the courts today. The portion of the Frye opinion most courts cite when denying the introduction of evidence they don't like, including polygraph evidence, is:
Just when a scientific principle or discovery crosses the line between the experimental and demonstrative stages is difficult to define. Somewhere in this twilight zone the evidential force of the principle must be reorganized, and while courts will go a long way in admitting expert testimony deduced from a well recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have general acceptance in the particular field in which it belongs.
We think the systolic blood pressure deception test has not gained such standing and scientific recognition among physiological and psychological authorities as would justify the courts in admitting expert testimony deduced from the discovery, development, and experiments thus far made.7
Although Judge Van Orsdel set a general standard for the admission of new scientific evidence, his opinion doesn't indicate exactly what he objected to in Marston's lie detection procedure. It is not clear whether the judge questioned the principle that lying causing changes in a person's blood pressure or if he objected to Marston's systolic blood pressure test as a method of gathering and recording data for interpretation. He may have rejected both the scientific principle behind Marston’s test and the technique itself.
If the Frye court's rejection primarily involved the lie detection technique rather than the scientific principle behind it, then it was Marston's systolic blood pressure evidence, not Larson's polygraph that was being ruled inadmissible in the Frye Case. If so, it could be argued that the Frye decision has been inappropriately cited all of these years as precedent for the exclusion of polygraph evidence.
It's interesting to speculate on what would have happened if Judge McCoy had been confronted with the polygraph instead of Marston's procedure. In all probability the result would have been the same. In Larson's case, however, the defense may not have pushed to get the results into court since Larson's results probably would have shown that Frye was lying.
At any rate, it is clear that Marston, in attempting to get his evidence into court, had jumped the gun, and in so doing, had forced the court to reject scientific lie detection. As a result, a legal precedent was established that has haunted the polygraph field for over sixty years.
As for James Frye, he was paroled from the District of Columbia Prison at Lorton, Virginia on June 17, 1939. He had served eighteen years in prison. Frye died in 1953 at the age of fifty-eight. If it hadn't been for William Marston, he may have died a lot sooner.
1. Frye V. United States, 54 App. D.C. 46, 293 F. 1013, 34 A.L.R. 145 (1923)
2. Professor Starr's version of the Frye Case can be found in his article, “’A Still-Life Watercolor’: Frye V. Unites States,” Polygraph (June, 1983).
3. Marston, William, The Lie Detection Test, (NY: Richard R. Smith, 1938), p. 71.
4. Marston, p. 73.
5. William Marston to John Larson, October 25, 1922. Vollmer papers, Bancroft Library, University of California at Berkeley.
6. John Larson to William Marston, November 2, 1922. Bancroft Library.
7. 293 F. at 1014.