On 17 July 2006, the New Jersey Supreme Court ruled in State v. Castagna et al. that the trial court erred in not permitting the defendants, who were charged for the killing of a bar patron, to present evidence that a prosecution witness, Violet Arias, had failed a polygraph examination administered by New Jersey State Police detective Laurent Gauthier. The Court ruled in relevant part:

We are convinced that the limitation placed on defendants’ right of cross-examination did not serve the interests of fairness and reliability. To be sure, the evidence that Arias failed her polygraph examination did not establish that defendants did not commit the crimes charged. However, the polygraph evidence was important to assist the factfinder in assessing the credibility of one of the State’s key witnesses. Unlike in McDavitt, here the reliability of the polygraph test results was not important. It was Arias’s belief that the
polygraph test results revealed she had not told the truth in her second statement that was crucial. It was apparent that Arias believed she needed to change her story for the State to accept her statement and to agree to offer her a plea agreement. We hold that the trial court erred in denying defendants the right to cross-examine Arias concerning the polygraph test results, not because those results were reliable, but because the test results caused Arias to change her statement.

The New Jersey Supreme Court’s ruling in State v. Castagna et al. may be downloaded here (84kb PDF).

Associated Press writer Beth DeFalco reported on the decision in “Court ruling allows defendant’s use of polygraph results,” published by the Press of Atlantic City:

TRENTON, N.J. (AP) – A ruling by the New Jersey Supreme Court to reinstate convictions in a murder case also grants defendants new latitude to cross-examine prosecution witnesses about polygraph tests, defense attorneys said Monday.

On Monday, the high court overturned an Appellate Division decision that granted a new trial to Thomas D’Amico and a co-defendant, Josephine Castagna.

The two were among seven people charged in the beating death of Bennett Grant, 37, an Elizabeth man who got into a fight with a group of customers at Sinner’s go-go bar on Oct. 24, 1999. The father of five lapsed into a coma after he was chased, kicked and beaten by the group and died five months later.

D’Amico, an Elizabeth patrolman who was off-duty when the beating occurred, and Castagna were convicted for their role in Grant’s death, but were granted a new trial when the Appellate Division found, among other grounds, that they were denied their right to cross-examine a prosecution witness on her polygraph test.

Polygraph test results are admissible in court only when the subject of the test and prosecutors agree in writing to allow it. Defendants were not previously allowed to cross-examine prosecution witnesses about the results of their polygraph tests.

Although the high court overturned the decision to grant a new trial, it did agree that defendants should be allowed to cross-examine prosecution witnesses about lie detector tests.

In D’Amico’s and Castagna’s case, a person being investigated in the beating agreed to take three lie detector tests and agreed they could be used by prosecutors. After being told she had failed one of the tests, the woman changed her story.

As part of a plea agreement, the woman pleaded guilty to second-degree reckless manslaughter and became a key prosecution witness.

In their ruling, the Supreme Court said that the defendants should have been able to use the woman’s polygraph results and changing story to try to attack her credibility during cross-examination.

“We are convinced that the limitation placed on defendants’ right of cross-examination did not serve the interests of fairness and reliability,” Justice John Wallace Jr. wrote for the majority.

Lawrence S. Lustberg, who submitted a brief in the case on behalf the Association of Criminal Defense Lawyers of New Jersey, said that although the decision did not result in a new trial for D’Amico and Castagna, it would benefit others.

“This is an important ruling for future defendants who now know they are entitled to use polygraph results of state witnesses,” Lustberg said.

Chris Durant of the Eureka, California Times-Standard reports in “Rape sentencing awaiting polygraph results”:

EUREKA — The four men who are accused of holding a woman in their Whitethorn home in March and repeatedly raping her entered guilty pleas to some charges Monday, but their exact sentences won’t be determined until after each are polygraphed.

Nate Robin Garza, 21, Levi Cole Garza, 21, Deshawn Lee Moore, 32, and Gregory Donald Scheider, 31, were arrested within a week of the crime being reported.

[Humboldt County] Deputy District Attorney Jeff Schwartz said the concerns of the victim was a “big factor” in his decision to offer the plea agreements.

”There’s was a lot of consultation with the rape victim,” Schwartz said.

The victim testified at the preliminary hearing, but didn’t want to go through recounting the crimes in public again, Schwartz said.

Depending on the results of the yet-to-be-scheduled lie detector tests and a judge’s decision, the sentences for Nate Garza, Moore and Scheider can be up to five years in prison.

If the results of the tests are not in the three men’s favor they will be sentenced on a rape in concert charge and be sentenced to five years. They will also have a strike on their record and have to register as sex offenders wherever they live for the rest of their lives.

If the results of the tests are in their favor, the men will be sentenced on a false imprisonment charge and a judge can sentence them up to three years in prison, Schwartz said.

Levi Garza can be sentenced on kidnapping, false imprisonment and a marijuana trafficking charge if the test is favorable for him. A judge will then determine his sentence, ranging from three to eight years.

If the test is not favorable for him, he will be sentenced to five years and eight months on a rape charge, have a strike on his record and have to register as a sex offender.

The four were held to answer to various charges after a preliminary hearing in April.

Levi Garza was held to answer to a kidnapping charge, rape and false imprisonment. His brother, Nate Garza, was held to answer to a kidnapping charge, attempted oral copulation, rape and sodomy.

Moore was held to answer to rape, attempted oral copulation and sodomy.

Scheider was held to answer to rape, sodomy and possession of a controlled substance.

The polygraph is being conducted by District Attorney Investigator Jim Dawson.

”Our office is very confident in Jim Dawson to do this,” Schwartz said. “Everyone will accept his results.”

Dawson is out of town until July 24, and the next hearing is scheduled for July 26. Schwartz said he doesn’t think the results of the tests will be ready by the court date but there is a possibility.

In 1998 the United States Supreme Court ruled that the decision to allow polygraphs as evidence is up to the individual judge.

A polygraph, or lie detector, measures the body’s involuntary responses to questions. Generally, the polygraph measures reactions from the respiratory, cardiovascular and sweat gland systems.

It is a dereliction of duty to rely on polygraph results to make sentencing determinations. Polygraphy has no scientific basis, is inherently biased against the truthful, and easily manipulated through the use of simple countermeasures that polygraphers have no demonstrated ability to detect.

A committee of the Israeli parliament, or Knesset, is proceding with legislation that would ban polygraphs from the workplace. Sheera Claire Frankel reports for the Jerusalem Post:

The Knesset Labor, Health and Sports Committee moved forward on a law to ban the use of Knesset polygraph testing in the workplace on Monday. MK Zahava Gal-On (Meretz) said that polygraph constituted a basic violation of human rights.

The law would ban any workplace from using the test on potential employees during the hiring process.

And IsraelNationalNews.com reports:

(IsraelNN.com) The Knesset Labor Committee on Monday authorized moving ahead towards banning the use of a polygraph exam in the workplace, referring to potential employees who are asked to undergo the lie-detector test before being hired.

MK (Meretz-Yahad) Zahava Gal-On stated the polygraph represents a gross violation of basic rights, adding that in the coming weeks, the new regulation will be brought for ratification before the appropriate Knesset body.

Christian Toto writes for the Metro section of today’s (29 June 2006) Washington Times in “Lie catchers stand by their polygraphs.” The article is cited here in its entirety, interspersed with commentary:

    Detective Leonard Keeler developed the first lie detector — or polygraph — test in the 1930s, and the machines used today are not very dissimilar from that first model.

While in January 1931 Leonard Keeler did receive what may be the first U.S. patent (485 kb PDF) for an instrument used as a lie detector (the application for which he submitted in 1925), William Moulton Marston was using a blood pressure recording device as a lie detector as early as 1917. And John Larson was administering lie detector ”tests” in Berkeley, California in the early 1920s.

    Even the best liars can get tripped up by their own bodies as the devices detect changes in blood pressure, breathing rate and perspiration levels, which could indicate a person is being less than honest.
    Although some people can say they “beat” the polygraph on occasion, experts say it can’t be done regularly.

Polygraph operators are the wrong “experts” to ask about countermeasures, as they have the most to hide. Simple, easy-to-learn countermeasures to the polygraph are readily available. And no polygrapher has ever demonstrated any ability to detect them.

    The science may seem simple, but it’s far from foolproof. That’s why polygraph tests are not admissible in court and researchers are looking into brain-imaging technology for the next-generation lie detector.

Actually, polygraphic lie detection is not a “simple science.” It is not science at all.

    Howard Miller, head of Miller Consulting Services in Falls Church, says today’s tests are used primarily in the security industry, such as with clients operating armored-car divisions. Polygraph testing also is used in military scenarios and other intelligence areas.
    For years, Mr. Miller and his peers administered polygraph tests using analog signals, much like the medium in which most television signals once were sent. The modern polygraph, like music and television, has gone digital.
    ”The computers are very sensitive,” says Mr. Miller, who began using analog systems in the early 1980s but since has switched to digital testing. “There are things it’s reading that examiners are missing sometimes.”
    Reston resident John Sullivan, a 31-year veteran of polygraph testing for the U.S. military and author of the book “Of Spies and Lies: A CIA Lie Detector Remembers Vietnam,” says, “The basic process hasn’t changed at all.
    ”We have computerized instruments now that are helpful, but the actual process of running a test is the same,” Mr. Sullivan says.

Sullivan is correct. Polygraphy has not substantially progressed in decades.

    Typically, two tubes filled with air wrap around the person’s chest and abdomen to measure respiration levels. A standard blood pressure cuff around the arm gauges heart-rate fluctuations. Two galvanometers, or finger plates, are attached to the person’s fingers to measure the skin’s ability to conduct electricity. Hydrated, or sweaty, skin does not conduct electricity as well as dry skin.
    If the movies have taught us anything, it’s that lying during a polygraph test causes the equipment’s needle to arc wildly in response. The results are rarely that dramatic, Mr. Sullivan says, not to mention that the digital devices no longer require mechanical devices such as the quivering, ink-laden needle.
    The new equipment also enables the administrator to read the responses in order of the degree of reactivity for more simple comprehension.
    Mr. Sullivan says the next wave of lie detectors could link directly to our brains. Neurologists are working on ways to connect brain-wave activity to lying, he says.
    ”I really think that has a lot of potential, but we still have a way to go,” he says, adding that initial research shows that brain-wave changes occur when a person twists the truth.

Actually, brain-wave changes (P300) has shown little if any potential for lie detection. Rather, where it has shown promise is for the detection of concealed information.

    Former Harvard Medical School faculty member Lawrence Farwell dubs a similar process “brain fingerprinting.” This method seeks out brain-wave fluctuations once a person is given a cue related to the information about which he or she may be lying.

Dr. Farwell’s brain fingerprinting technique aims at the detection of concealed information, not lie detection.

    Polygraph tests have their skeptics, and they are legion — visit www.antipolygraph.org for a measure of the movement.

Indeed, it is not just AntiPolygraph.org that is skeptical of polygraphy. Polygraphic lie detection has been roundly rejected by the scientific community. The only remaining debate pits those who don’t yet understand that it’s junk science against the growing number of people who do.

    Lourdes Griffin, director of the Washington Hospital Center’s outpatient behavioral health service, isn’t convinced the devices can help reveal liars. The machines try to record indicators that typically reveal discomfort, but for whatever reason, “everyone responds differently,” she says.
    ”We’re so individual. Any of those stress reactions … could trigger a different stress response in someone else,” she says.
    One person might get nervous near an attractive woman, while another could exhibit stress reactions when standing in front of a large crowd, she says. A sociopath might not exhibit any outward signs of stress from lying, she adds.
    Among the biological reactions to stress is an increase in oxygen levels to the brain to let the person deal with the stressor in question.
    ”But are they linked to lying? I don’t know,” she says.
    William Chittenden, president of the Nokesville-based Virginia Polygraph Association, says a subject telling a lie might not trigger all three of the polygraph’s main measurement systems.
    ”Some examiners don’t like to deal with the breathing patterns, but it can be helpful,” Mr. Chittenden says. “I like to look at all of them. There’s information in all the channels.”

There may well be “information” on all the channels of the polygraph instrument, but none of it provides any clear indication of whether a person has spoken the truth.

    Sometimes the person being tested is medicated or took drugs thinking it could help them cheat the test.
    ”People can be on medication, and they’re still testable. Others may be really not testable if their physiology is so slowed down,” he says.
    The polygraph test isn’t the only way people are trying to separate fact from fiction.
    Clay Shields, an associate professor of computer science at Georgetown University, says a voice analyzer offers a low-tech approach to lie detection. These voice stress analysis tools, which do not need wires or any other connection to the person being interrogated, can detect levels of stress in the voice.
    ”It’s inaudible to humans,” Mr. Shields says. “Some people are marketing them as lie detectors over the telephone, and others talk of using them in airport security screenings.”
    The gadgets measure fluctuations in voice frequency levels, which some connect to dishonesty patterns.

No voice stress analyzer has ever been demonstrated to reliably detect detection. The most financially successful marketer of a voice stress analyzer, the National Institute of Truth Verification, which peddles the “Computer Voice Stress Analyzer,” has admitted in court that its device ”is not  capable of lie detection.”

    Even polygraph administrators admit some people beat the system, while others insist it can’t be done consistently.
    Mr. Miller says people can be trained to alter some biological activities such as heart rate, but even that takes several minutes to achieve, a window of time a polygraph questioning won’t permit.
    Besides, “when people are altering their biological responses, it shows in the test,” he says.

Miller is being less than honest here. Heart rate is easily increased within seconds, not minutes, using simple techniques (such as performing mental arithmetic timely with the “control” questions) that leave no tell-tale signature. For more on how anyone can fool the polygraph, see AntiPolygraph.org’s free e-book, The Lie Behind the Lie Detector (1 mb PDF).

Writer-historian Joel Shurkin comments on polygraphy in an entry on his Of Cabbages and Kings blog titled, “Liar, liar pants on fire. OK, maybe not.” Significantly, Shurkin notes that an article he once wrote for a major science magazine was suppressed because of outside pressure on the editor.

Washington Post staff writer Shankar Vedantam, who in May reported with Dan Eggen on the FBI and CIA’s questionable reliance on polygraphy, follows up with a new front page story titled, “Polygraph Readings Vary Among Agencies.” In this story, Vedantam addresses the experiences of victims of the polygraph. Excerpt:

The National Security Agency denied a top-secret clearance to David Vermette this year after two polygraph tests. But the computer programmer still has access to sensitive, classified information — from the CIA, which independently cleared him after administering its own “lie detector” test.

The FBI recently ran a background check on Wayne Johnson, which led to a five-year extension of his top-secret White House clearance. But when Johnson applied for a job at the FBI itself, the agency made him an offer — then rescinded it after a polygraph exam.

The Defense Department has long issued Tara Wilk a top-secret clearance. But when Wilk tried to get similar clearance from the NSA, she failed three tests — leaving her so frustrated she sought help from a hypnotist and a therapist.

In a region where a security clearance is a necessary ticket to countless jobs with the federal government and its thousands of contractors, it is not hard to find people caught in turf wars over clearance. Polygraph tests are often at the root of the problem.

“The CIA doesn’t respect the NSA’s polygraph and the NSA doesn’t respect the CIA’s polygraph,” said Wilk, a computer engineer from Arnold, Md. “Nobody knows who the boss is, and they all think they are the most important.”

The government recognizes the problem and plans to harmonize the process across the intelligence community, but Director of National Intelligence John D. Negroponte cannot say when that will happen, said spokesman John Callahan. “The goal is to streamline and fix things and make things better,” he said.

“The legislation which founded the DNI actually requires the DNI have as one of its goals to unify this process,” he said.

Even those who believe in the value of polygraphs acknowledge that they are far from objective. Using a polygraph device, which measures changes in heart rate and breathing as well as other cues to detect anxiety, is like searching in a dark room for an object whose shape is unknown. It is the examiner’s job not only to figure out if someone is a spy but also to search for character flaws or past actions — drug use, for instance — that might make a person unfit to handle sensitive information.

Since polygraph examiners typically do not know what to look for in a candidate, they tend to home in on anything that hints at reticence or nervousness, said John Sullivan, who spent three decades at the CIA administering the tests and still supports them. During his career, he said, he used the tests to unmask seven double agents and spotted numerous criminal and character problems.

But Sullivan said that after the agency’s polygraphers failed for years to detect the duplicity of Aldrich H. Ames, who compromised dozens of CIA operations by passing information to the Soviet Union before being sent to prison in 1994, agency examiners ratcheted up the level of intimidation during tests.

Sullivan believes polygraphers can elicit useful information without resorting to threats and harassment. But after Ames’s case, he said, CIA examiners were told that if their subjects did not complain about rough handling, the examiners probably were not doing their job correctly: “People in many cases are too aggressive . . . we were so afraid of getting beat.”

Asked why examiners disagree with one another, Sullivan said that interpreting polygraphs is more “art than science” and that examiners at different agencies range from “Rembrandts” to “finger-painters.”

“I myself and most of my colleagues caught people who passed other people’s polygraph examinations,” said Sullivan, who is retired. “I don’t want to disparage anyone else’s program, but I really feel up until Ames, [the CIA] had the best polygraph program in the government.”

While some polygraphers may be better interrogators than others, when it comes to detecting lies, they’re all fingerpainters. As Dr. Drew Richardson has aptly put it, polygraphers who administer lie tests are involved in the detection of deception in the same way that a person who jumps from a tall building is involved in flying.

As for federal agencies’ willingness to accept their own polygraph results despite conflicts with polygraph results from other agencies, the attitude seems to be, “Our juju good juju; their juju bad juju.” But polygraphy is all bad juju.

When it comes to polygraph results, the only proper way to “unify the process” across federal agencies is to heed the warning of the National Academy of Sciences, which found that “[polygraph testing's] accuracy in distinguishing actual or potential security violators from innocent test takers is insufficient to justify reliance on its use in employee security screening in federal agencies,” and end polygraph screening now.

A discussion of this article is available on the AntiPolygraph.org message board here.