The topic of CNN's news program The Point with Greta Van Susteren on Monday, 9 July 2001 was "Lie Detectors: Separating Fact From Fiction." Unfortunately, Van Susteren did a poor job of separating fact from fiction, allowing untrue statements about polygraphy to go unchallenged, chief among them being guest Ralph Nieves' claim of 97% accuracy, which is wholly unsupported by peer-reviewed scientific research. Nieves also made a claim to the effect that although the examiner can be beaten, the test itself can't. (Those who have read
The Lie Behind the Lie Detector will know better.)
Van Susteren neglected to mention that her guest, Ralph G. Nieves, Sr., is himself a long-time polygrapher whose livelihood depends on the public's belief in polygraphy. You'll find a biographical sketch of Mr. Nieves on the Integrated Security Services website at:
http://www.intesecurity.com/Pages/b_ralp.html You can help Greta Van Susteren separate polygraph fact from fiction by sending her an e-mail at askgreta@cnn.com. The following is an excerpt from the program transcript, which is available on-line at: http://www.cnn.com/TRANSCRIPTS/0107/09/tpt.00.html VAN SUSTEREN: ... Lie detectors work great in TV movies and detective novels, but do they really prove anything? Joining me from New York to put polygraphs on the hot seat, former New York Police detective Ralph Nieves.
Ralph, thanks for joining me this evening.
RALPH NIEVES, FORMER NEW YORK CITY POLICE DETECTIVE: Good evening.
VAN SUSTEREN: OK, Ralph, if you were working with the Levy family, what questions would you wants to pose to the congressman during a lie detector session?
NIEVES: Well, I would certainly look to put the issue of any physical violence to the victim or to the alleged, the missing person to make sure that he was at least not participating in any physical violence toward her. Certainly, questions as to his responsibility for her missing would be too vague. I mean, you'd have to be very narrow in scope and specific to the issues. At least that one issue can be put to rest.
In addition...
VAN SUSTEREN: OK. So give me an idea. I understand that the vagueness wouldn't be good, but I mean like specifically, what questions...
NIEVES: Did you -- did you participate in any physical violence toward Ms. Levy? Did you or any of your agents participate in any physical violence toward Ms. Levy? Did you participate in her abduction? Questions of that nature. Obviously, the question formulization would have to be done by the examiner at the time of the test.
In addition to that, you have some excellent examiners in Washington. I mean, the government -- the United States government is the largest user of lie detectors. They do close to a million examinations a year. And certainly if the congressman wished (UNINTELLIGIBLE) he could do that. But obviously, attorneys will advise him maybe contrary to that, because we all know, as you know, attorneys and clients, sometimes clients are not forthright in all their answers, and they have to protect his interests also.
VAN SUSTEREN: Ralph -- Ralph, what's the margin of error in a polygraph test?
NIEVES: Well, with the algorithms that have been designed and the computer systems -- the photograph you have is an analog instrument. The algorithm, we have about a 97 percent accuracy factor with that.
It's a great tool to exclude a person during an investigation. It's a wonderful investigator's tool.
VAN SUSTEREN: But it's not perfect, Ralph. I mean, where are its -- where are its problems?
NIEVES: Well, nothing -- nothing is perfect. I mean, obviously, it, once again, it's a tool. When it works, it works very effectively, but it just doesn't work all the time.
VAN SUSTEREN: But -- but then that raises a good question: When does it work and when doesn't it work?
NIEVES: Well...
VAN SUSTEREN: I mean, I understand if the questions are vague it doesn't work. When else doesn't it work?
NIEVES: Well, it doesn't work depending upon the examiner that's administering the test. I mean, people always state that they -- they can best the test. You don't beat the test: You beat the examiner. And once again, the FBI examiners are probably the finest examiners in the world. And I mean, that -- those are areas that once a decision is made to conduct the examination, as far as him beating the test, if he's going to practice deception, he's a liar. If he's forthright -- and once again, if you had nothing to do with this issue, I certainly would do anything to possible to put that one issue to rest. And if a lie detector would answer that question, that's fine. There's no exposure.
VAN SUSTEREN: When, Ralph -- when, Ralph -- I mean, I have seen some lie detector tests where the results are called "inconclusive." What does that mean?
NIEVES: Inconclusive means that we can't determine whether the person is being truthful or deceptive. That happens on occasion. Like anything else, there's nothing that's a perfect answer.
It's just a wonderful tool. In situations where the investigation calls for it, it works very effectively.
And once again, the new systems that they have now and algorithms and the computerized systems that the federal government has developed themselves actually, because that's the ones we use now in the private sector, are extremely reliable. And it puts issues to rest. And certainly the issue of any physical violence toward Ms. Levy is of primary importance to the family. And that's an issue that can very easily be at least indicating that the congressman has no involvement in that area, and the efforts could be going into other areas and look at the investigation -- look at other parts of the investigation to see where they can put a solution to this matter.
VAN SUSTEREN: Ralph, someone who has obviously been somewhat deceptive is certainly the congressman in this instance -- at least apparently wanted to conceal his personal relationship with the intern. After all, he is married. Does that level sort of a guilty conscience, having tried to hide that type of matter, would that in any way inflict some sort of inability of an examiner to get truthful answers on the question you posed, violence or no violence?
NIEVES: Not at all, because any examiner who conducts this test will spend anywhere from three to four hours with the congressman, and all those issues would be resolved then, and the focus would be simply, and my suggestion would be (UNINTELLIGIBLE) did he participate in any physical violence or the abduction of Ms. Levy. That's the issue, and that's solely the issue.
So his affair -- his love affair with her, that should not impact upon the examination, because the question isn't about that. It's about whether he hurt that lady or had any -- any of his agents to anything to that young lady, which I think is of primary importance.
VAN SUSTEREN: OK. Assume that -- assume -- well, let's do a hypothetical. Someone agrees to be -- to submit to a polygraph examination. You walk into the room -- take me step-by-step through the procedure.
NIEVES: Well, there's a whole pretest interview. The fact pattern of the case has to be thoroughly reviewed by the examiners. They walk in, there's a whole...
VAN SUSTEREN: How do you hook them up?
NIEVES: Well, when you attach an individual -- the actual running of the test takes maybe 20 minutes. It's all the preliminary work that goes into -- that's entailed in the investigation: reviewing the facts of the case, what do the investigators have so far, is there any inconsistency in the statement, and just focusing on one particular issue. And that's the simple issue, did he hit, rob, shoot, stab Ms. Levy, or did he know anyone that participated in that? That's the issue, and that's the steps that you take.
To get into the polygraph test, it's too extensive for me at this time to explain to you every step of the process. But it will take a considerable amount of time.
VAN SUSTEREN: Well, let me ask you, interpreting -- is there any level of subjectivity in interpreting the results?
NIEVES: Well, the subjectivity that used to exist with the analog instrumentation has been minimized because of the computer algorithms that we use now. So a lot of that guesswork or the evaluation has been minimized.
You never give up your responsibility of reviewing the charts physically and scoring them, but the algorithms, which are the computer systems that were designed initially through NASA, the Space Administration -- technology from them -- eliminates a lot of that guesswork.
VAN SUSTEREN: In light of the fact they say eliminates all the guesswork, why do you think, Ralph, that they still are not uniformly accepted in court as evidence?
NIEVES: Well, that's not true. That's not true at all.
VAN SUSTEREN: Uniformly. They're not accepted in all courts.
NIEVES: Well, you get stipulations all the time for the testing process. The United States government uses it extensively. When we come into court on that level as a defendant and we use...
VAN SUSTEREN: But you say stipulation: That's when both sides agree.
NIEVES: Correct.
VAN SUSTEREN: Uniformly, you can't impose it upon -- in every court and have it admissible.
NIEVES: Well, we're imposing it upon, but we're using it as a tool where it's selectively decided to stipulate to, and the defense and the prosecutor and the judge agree. Look, we have a test, they both agree, the results are accepted.
On a federal level, obviously they're locked in. I mean, they do close to a million examinations a year. The research and development that went into these systems was extensive. The taxpayer dollars was phenomenal. If it works for the government, it should work for the defense.
VAN SUSTEREN: Except for I've got -- I've got to tell you one thing, Ralph: When you say it works for the government, that's oftentimes in terms of employment decisions. On something so important as whether or not we take away someone's liberty, we sometimes subject it to higher standards, and I think that probably is the reason why they've experienced...
NIEVES: Well, you....
VAN SUSTEREN: ... that's why they're not uniformly accepted.
NIEVES: Well, there are intelligence agencies that are put under that -- under that microscope every day. So that's...
VAN SUSTEREN: All right, Ralph, I'm going -- you and I could battle about this for a long time.
NIEVES: Oh, you've got that right.
VAN SUSTEREN: I'm going to give you the last -- I'm going to give you the last word on that, though.
NIEVES: Sure.
VAN SUSTEREN: Thanks to Ralph Nieves for joining us this evening.
NIEVES: You're very welcome.
