There is a little known device used in interrogations in police departments throughout the country. The examiner attaches to the subject a microphone, which is connected to a laptop computer with special software. During the interrogation, the examiner charts on the computer the voice pattern of each answer. By the analyzing the patterns, the examiner can learn on which answers the subject was likely deceptive.
Seeking greater versatility in the field than the polygraph, retired Army officers developed voice stress analysis technology in the 1960s and 1970s. Like the polygraph, the newer Computer Voice Stress Analyzer (CVSA) is based on the theory that a subject will exhibit stress when he’s deceptive. But proponents of the CVSA claim that it’s a more reliable truth verification system. With the cost of purchase and training at thousands of dollars less than the polygraph, it’s used by more than 1,000 domestic law enforcement agencies.
With the onset of the War on Terror and the high demand for quality intelligence and frequent interrogations, some military units desire access to this tool that their civilian counterparts so widely tout. The manufacturer of the CVSA, the National Institute for Truth Verification, is only too happy to oblige as it aggressively seeks new federal and military markets. A demand and a supply — simple enough, right?
Not in the bureaucratic world of military procurement. The United States Special Operations Command has pursued voice stress analysis technology over the last two years. Finding promise in the CVSA, SOCOM initiated research and development into a device that would miniaturize it into a handheld Field Interrogation Support Tool. SOCOM budgeted for the program and was ready to go, until the Pentagon halted it.
While it’s tarred as scientifically unproven and opposed by the Pentagon bureaucracy, two things are clear: cops stateside love it, and troops want it. From the Polygraph Institute to the undersecretary of defense for intelligence, Stephen A. Cambone, the Department of Defense is denying our very best the equipment they want to pursue the War on Terror. Is a Pentagon turf war preventing troops from being better prepared?
THE EQUIPMENT’S UTILITY is self-evident, and its potential implications substantial: troops require quality intelligence in times of war. Defeating terrorists in Iraq and preventing future domestic attacks will require information from detainees at American facilities worldwide and at Guantanamo Bay.
A July 2003 Washington Times story detailed the perils of inadequate interrogation equipment, reporting that Saddam Hussein loyalists were thwarting polygraphs. In one reported case, the polygraph did not catch a detainee blatantly lying to interrogators about his involvement in weapons programs because he had been trained in countermeasures by the old Special Security Organization.
While the Pentagon has officially declared the CVSA off-limits and thus discouraged most units from using it, those who have tried it like what they see. After-action reviews and interviews with interrogation personnel show that special operations forces and counterintelligence units are eager to use it.
Three unclassified after-action reviews from the War on Terror, provided to TAS by the National Institute for Truth Verification (NITV), strongly recommend using the CVSA in detainee interrogations. The assessment by a Qatar counterintelligence staff sergeant found that the CVSA “can be a virtually irreplaceable tool at the field operator level….This is a system that can be integrated into operations at all echelons and will prove to be an invaluable addition to the Army inventory.” Similar test periods at Guantanamo Bay in 2003 and at Bagram Collection Point, Afghanistan, in June 2004 were also successful, deeming the CVSA an “invaluable tool” with “outstanding results.”
A military supervisor of interrogations at Guantanamo, who spoke with TAS on the condition of anonymity, called the CVSA “a useful tool in the interrogation of detainees.” Because it is more flexible, more precise, and less invasive, he said, “It was more useful than the polygraph.” As a result of using it, the supervisor said that interrogators obtained actionable intelligence from detainees, and now believes strongly that military personnel should have access to the CVSA. “We’ve got a global war on terrorism going on here, and the guys need all the tools they can get. There aren’t enough polygraphers to go around, and the system’s cumbersome,” he said. A contract CVSA examiner told TAS that the Southern Command has halted use of the devise at Guantanamo, but such reports could not be verified.
Bill Endler, a civilian CVSA examiner who was on contract with the military in Iraq through NITV, ran about 50 examinations from October 2003 to January 2004 at Camp Slayer, Abu Ghraib prison, and the Green Zone. When he first arrived, interrogators were reluctant to use Endler, “but once I showed them how much I could help,” they came around. Detainees came to call Endler “the truth man,” he said.
Among his interrogations, Endler noted two interviews with high value detainees. Taha Yasin Ramadan, Iraqi vice president and the “Ten of Diamonds” in the coalition’s deck of most wanted officials, broke down after Endler began the interview. He admitted plotting attacks on coalition forces. The CVSA succeeded, Endler said, “once I proved to him that it worked.” Endler said the only unsuccessful CVSA interrogation was of Tariq Aziz, who proved uncooperative. Endler has not been called back to Iraq. Since his August interview with TAS, Endler has become NITV’s director.
While military experience with the CVSA is limited, those who use it say it works. NITV exhibited dozens of letters of appreciation for the equipment from hundreds of law enforcement agencies. Agencies contacted by TAS shared this enthusiasm, touting the CVSA as a valuable tool. Detective Peter Rago, of the Schaumburg Police Department in Illinois, said the CVSA has been “extremely valuable.” Rago pointed to difficulties with the polygraph, including its high rate of inconclusive results and a lack of certified examiners in Illinois law enforcement. Detective Corporal Jeanne Landis of the Flathead County (Montana) Sheriff’s Office and Detective Bill Case, of the Reading (Pennsylvania) Police Department, echoed Rago’s success. Case mentioned its value in eliciting a recent confession from a man who raped his own stepdaughter. “I haven’t had anything where a person showed deceptive and then was exonerated or was found to be innocent,” Case said.
DESPITE THESE SUCCESSES domestically and abroad, there is skepticism in the scientific community concerning the Computer Voice Stress Analyzer. Critics, including those within the Department of Defense, dismiss the CVSA as scientifically unproven. Undersecretary of Defense for Intelligence Stephen A. Cambone justified his decision to bar DoD use of the equipment in a letter last year to Sen. George Allen (R-Va.), citing studies that “…concluded there was little or no evidence, scientific or otherwise, for the applications of voice stress analysis….”
However, scientific studies of the CVSA are inconclusive. Various studies have confirmed the existence of voice stress. Whether the CVSA can accurately measure that stress and discern deception is another question. The National Institute for Truth Verification argues that no study has been performed using manufacturers’ protocols: the CVSA can only be accurately tested using real world scenarios and “known conclusion” cases. Without the stress of actual deception, study participants don’t fear the consequences of lying.
Dr. John Morgan, assistant director for science and technology at the National Institute of Justice, agrees that the CVSA has not been subjected to a thorough scientific evaluation. “There is no objective data to provide the basis for voice stress or polygraph as a reliable method for detecting deception or stress,” Morgan told TAS.
The most rigorous studies on the CVSA, such as Victor L. Cestaro’s for the Department of Defense Polygraph Institute and the Air Force Research Laboratory’s for the National Institute of Justice, have not thoroughly tested the device. The Cestaro study relied on a mock crime scenario “in the absence of jeopardy.” The National Institute of Justice study had one real world component which found that the CVSA accurately detected stress in 45 out of 45 cases. Both studies found that while it detects stress, they could not confirm the detection of deception.
To the strict empiricists, the lack of scientific verification of the CVSA is a sufficient case against its use. John Palmatier, a Fort Lauderdale, Florida polygraph examiner and critic of the new technology, argues that without scientific backing, the CVSA could be a gimmick. While it may well elicit confessions, that “does not mean that the device itself discriminates between what is true and what is deceptive,” Palmatier wrote in an article for an American Bar Association journal.
Still, Palmatier admits in the article “there is as yet no definitive answer” whether the CVSA is “a modern technological innovation or ‘the Emperor’s New Clothes.'” Palmatier’s primary concern is that the CVSA could be a mere prop used to browbeat confessions from interviewees. He wrote, “Almost anything can be portrayed as a magic device that will enable its user to see what is true.” While Palmatier depends on the polygraph for his livelihood, he told TAS that “if the CVSA worked I’d be using it.” Could the CVSA ever be a proper tool? “Given the proper research and the proper development, maybe someday, but not right now,” he said.
Even if the CVSA is a scam, its devoted users find it worth spending thousands on training and multiple machines. “If it’s a prop, it’s an expensive prop,” Humble said. The aforementioned Guantanamo supervisor pointed to its utility, “If it’s a prop, and it works, oh well.” Schaumburg Detective Rago said of the prop accusation, “For me, it works, so I might as well use it.” Though critics are dismissive of the CVSA, the conflict over the device boils down to very different perspectives: utility versus scientific reliability.
WHILE CRITICS DISMISS voice stress as scientifically unreliable, the polygraph — the Pentagon’s equipment of choice — offers little or no advantages. The polygraph is known for its unreliability, from the countermeasures which can render exams inconclusive to its failures to catch notorious interviewees, such as Aldrich Ames.
In the case proscribing polygraph exams as admissible evidence, Justice Clarence Thomas wrote, “To this day, the scientific community remains extremely polarized about the reliability of polygraph techniques….[T]here is simply no way to know in a particular case whether a polygraph examiner’s conclusion is accurate, because certain doubts and uncertainties plague even the best polygraph exams.”
Compounding widespread scientific doubt of the polygraph is the National Academy of Sciences findings on the polygraph’s use for security screening. Commissioned by the Department of Energy, the 2003 study found evidence for the polygraph’s validity “scanty and scientifically weak.”
The Academy found that most government polygraph research is conducted by those invested in the polygraph’s success. Pointing specifically to the Department of Defense Polygraph Institute (DoDPI), the study reported that
U.S. agencies charged with initiating and sponsoring polygraph research…are also charged with the mission of training polygraph examiners and developing new polygraph applications. The dual mission of acting as a sponsor for polygraph research and as a sponsor for polygraph practice creates an obvious conflict of interest. Any reasonable investigator would anticipate that certain research questions…or certain patterns of results…will be less welcome by such research sponsors than empirical demonstrations that the polygraph “works.”
These pointed questions about the polygraph’s scientific validity are nearly identical to the Pentagon’s arguments against the CVSA.
DoD personnel may help explain this inconsistency in DoD policy. DoDPI’s chief of special studies, Frank Horvath, is the former president of the American Polygraph Association, the professional organization of polygraph examiners. The office upon which the DoD relies most for its scientific studies on the polygraph and the CVSA is a steadfast defender of the polygraph. And other federal agencies cite DoDPI studies in rejecting the CVSA.
Humble and others interviewed by TAS believe most polygraph examiners are biased in favor of the polygraph even after their careers wind down, as most law enforcement and military polygraphers open private practices after retirement. If CVSA gains in use and credibility, many examiners like Palmatier could be out of business. “You’re talking about guys who are attached to this as a retirement bonus program,” said the Guantanamo supervisor, “You’re immediately in a rice bowl issue and fur starts to fly. DoDPI’s going to protect itself there and they have a monopoly.”
While the CVSA has its skeptics, the former executive director of the National Institute for Truth Verification is still a believer. David Hughes, a former police captain who is no longer with NITV as when TAS began reporting on the story, continues his support for CVSA, despite being engaged in a legal dispute with Humble.
THE PENTAGON HAS THWARTED military attempts to expand CVSA use. The United States Special Operations Command (SOCOM) showed a particular interest in developing the CVSA for the field, receiving an NITV briefing August 24, 2004. SOCOM requested and received funding to miniaturize the equipment into a handheld Field Interrogation Support Tool. This program is listed in the fiscal year 2005 budget for the Defense Acquisition Challenge Program (DACP), and was reported by Special Operations Technology magazine and as recently as October 15 in the St. Petersburg Times.
Sensing DoD opposition in the early months of 2005, SOCOM commissioned a survey of domestic law enforcement agencies utilizing the CVSA. In March 2005, independent researchers contacted agencies across eight states. The researchers reported to SOCOM that domestic agencies warmly recommend the CVSA as an investigative tool. A researcher familiar with the survey said that, based on this research, “Without this tool, it slows down criminal investigations, and somebody’s kid is not going to come home, that’s domestically and internationally.” On average, those surveyed perceived the CVSA as more than 91% reliable.
Despite SOCOM’s apparent interest in the CVSA, the Pentagon cancelled the project. After researchers briefed SOCOM on April 4, SOCOM officers attempted to brief Carol Haave, deputy undersecretary of defense for counterintelligence and security. She delayed two meetings and cancelled a third in May and June. Humble says he learned in June that Haave terminated the project. Lt. Commander Steven Mavica, a SOCOM media relations officer, confirmed for TAS that at Haave’s request SOCOM suspended the project in the third quarter of fiscal year 2005.
Undersecretary Cambone is telling Congress a different story. He wrote in an August 26, 2005 letter to Sen. Rick Santorum that “USSOCOM reviewed the proposal and ultimately declined to fund the initiative.” Pentagon officials did not respond to questions about this inconsistency or its decision to terminate the project.
The Pentagon is also actively halting use of the CVSA in the field. When interrogators attended a training class for the CVSA this fall at Qatar’s Camp As-Sayliyah, they were required to sign an August 28, 2005, memorandum acknowledging DoD’s policy barring use of the CVSA.
IN THE FUTURE, the Pentagon may authorize use of the CVSA. But some Senate military aides believe the DoD is dragging its feet. Through Humble’s contacts with Senate Republicans as a member of the Senatorial Trust (the membership price tag is $15,000), he has rallied political backing. A handful of Republican Senators have repeatedly written DoD and other federal departments in support of the CVSA receiving a fair examination. Secretary of Defense Donald Rumsfeld is aware of their inquiries, but has referred the issue to Undersecretary Cambone.
Behind the scenes, their military aides have actively examined DoD policy, especially after Cambone’s June 8, 2004 memorandum barring the use of any “credibility assessment instrument” other than the polygraph. He also promised that his staff will research technologies beyond the polygraph but, until accuracy can be independently determined, “the polygraph will remain the sole instrument.” Upset with the memo, aides held a July 13 2004, meeting with DoD counterintelligence and polygraph officials. A Senate aide in the meeting says they were told by Toby Sullivan, DoD director of counterintelligence, to expect a memo clarifying that other interrogation devices may be used, but that only the polygraph may be used for truth verification. Congressional aides haven’t seen such a memo.
While aides emphasize their desire to ensure all available technology receives a fair review, some are growing impatient with the DoD. They were promised at the July 13 meeting that the DoD would commission an independent study of the CVSA.
The DoD study is being performed by Dr. Harry Hollein at the University of Florida. Hollein declined to discuss the study with TAS until it is complete. Pentagon officials confirmed that Dr. John Capps, brother of former DoDPI director Michael Capps, is overseeing the Florida study.
One Senate military liaison is unimpressed with the CVSA and his colleagues’ efforts on its behalf. “I personally doubt the War on Terror is going to be won or lost based on this product. If you’re going to pick multiple dark arts, why spend thousands of dollars on a machine that doesn’t work as it’s supposed to work?”
Another Senate military aide said he has confirmed the CVSA’s success with interrogators and officers in the field. “If it actually proves to be a truth telling device, we’ll have missed an opportunity,” one Senate military aide said. “If it’s useful, we’ll have denied these people a useful tool. Commanders in the field ought to make these decisions….[DoD officials] don’t trust their commanders in the field.”