- Welcome to AntiPolygraph.org Message Board.
A forum for free and open discussion of polygraph-related topics
Posted by George W. Maschke
- Mar 24, 2002, 09:45 AMRock,
To be sure, my congratulations to the winners of the Bureau's polygraph lottery were intended in the way of black humor, not schadenfreude, and I in no way see their plight as a source of merriment.
There's something of the Orwellian in the reasoning behind the following line from the Associated Press article:
In the minds of FBI counterintelligence officials, those who "fail" a polygraph seance seem to be presumed guilty until proven innocent -- and the possibility that the polygraph was simply wrong seems to be excluded.
The FBI officials who told the Associated Press that those who fail "may eventually be cleared because things like medical conditions can cause anomalies on the tests" are deluded. As Dr. Alan P. Zelicoff, M.D. of the Center for National Security and Arms Control observed in his 14 June 2001 letter to DOE counterintelligence chief Michael H. Waguespack:
As I noted in a list of questions presented to panel members of the National Academy of Sciences Study to Review the Scientific Evidence on the Polygraph:
The FBI (and other agencies) seem to be unable to grasp the concept that a truthful person can indeed "fail" in the absence of some special circumstance that might explain the "failure."
It's high time that this superstitious "polygraph lottery" were ended.
To be sure, my congratulations to the winners of the Bureau's polygraph lottery were intended in the way of black humor, not schadenfreude, and I in no way see their plight as a source of merriment.
There's something of the Orwellian in the reasoning behind the following line from the Associated Press article:
QuoteOfficials said that some workers whose polygraphs raise initial concerns about deception may eventually be cleared because things like medical conditions can cause anomalies on the tests.
In the minds of FBI counterintelligence officials, those who "fail" a polygraph seance seem to be presumed guilty until proven innocent -- and the possibility that the polygraph was simply wrong seems to be excluded.
The FBI officials who told the Associated Press that those who fail "may eventually be cleared because things like medical conditions can cause anomalies on the tests" are deluded. As Dr. Alan P. Zelicoff, M.D. of the Center for National Security and Arms Control observed in his 14 June 2001 letter to DOE counterintelligence chief Michael H. Waguespack:
QuoteAs the technical staff at the Labs pointed out repeatedly during the October 1999 DOE Polygraph Hearings and subsequently to Dr. Andrew Ryan [chief of research at DoDPI] and Mr. David Renzelman [DOE polygraph program chief], there are no scientific studies that identify medical conditions of relevance to the polygraph.... Also, there are no guidelines, textbook chapters, or review articles in any medical, psychology or psychiatry journal that describe medical contra-indications to the polygraph.
As I noted in a list of questions presented to panel members of the National Academy of Sciences Study to Review the Scientific Evidence on the Polygraph:
Quote7) At January's meeting, Mr. Renzelman also said that the polygraph is not a lie detector. He explained that the polygraph is used to determine if a question "bothers" the examinee, noting, "Polygraph is only a means of...of...of looking at emotion that is taking place at the time a person listens to, thinks about, answers a question that the examiner and the person taking the test has agreed upon originally. And if the answer to that question bothers the person taking the test, then it tends to bother us. And then it's our job to find out, 'Why did that bother you?'"
Does a reaction to a relevant question necessarily indicate that it is the answer to the question that "bothers" a person (and not, for example, the mere fact of being asked the question in the first place)?
What scientific methodology do DOE (and presumably, other) polygraphers use to find out why a question "bothered" a person? What scientific research supports the validity of that procedure?
The FBI (and other agencies) seem to be unable to grasp the concept that a truthful person can indeed "fail" in the absence of some special circumstance that might explain the "failure."
It's high time that this superstitious "polygraph lottery" were ended.
Posted by therock
- Mar 23, 2002, 08:59 PMROFLMAO!!!!!
Posted by George W. Maschke
- Mar 23, 2002, 08:31 PMNewspapers across America, including The Washington Post, are carrying the Associated Press' latest article on FBI countespionage measures under the title, "Possible Problem FBI Agents ID'd."
To any of those "possible probem" FBI agents who may be reading this, congratulations on winning the FBI's version of The Lottery.
To any of those "possible probem" FBI agents who may be reading this, congratulations on winning the FBI's version of The Lottery.
Posted by George W. Maschke
- Mar 23, 2002, 06:42 AMFBI polygraphers have apparently succeeded in bringing the FBI counterintelligence-scope polygraph failure-to-pass rate in line with the percentage of FBI applicants who fail to pass the counterintelligence portion of their pre-employment polygraph interrogations. According to an Associated Press report dated 22 March 2002, the current failure rate among FBI employees is just over 1%:
http://www.worldnews.com/?action=display&article=12631598&template=worldnews/search.txt&index=recent
As noted in the message above, the FBI failure-to-pass rate had previously been on the order of 5%. (The article by Lenny Savino linked above is no longer available on the Knight-Ridder website, but an excerpt is available here
.)
That any of the 700 FBI employees polygraphed since Hanssen's arrest actually are spies is an unlikely proposition to begin with. But bearing in mind that polygraph screening has absolutely no validity whatsoever, and the base rate of guilt is quite low when it comes to espionage, what is the likelihood that any spies among the 700 employees are among the roughly one-in-a-hundred who "failed" the polygraph? Is it not more likely that any genuine spies will have learned how to pass the polygraph, like Aldrich Ames, Larry Wu-tai Chin, and (apparently) Ana Belen Montes did?
As Drew Richardson noted in his 1997 Senate testimony, to the extent that the FBI relies on the polygraph to direct its counterespionage efforts, it is undermining, not enhancing, national security.
http://www.worldnews.com/?action=display&article=12631598&template=worldnews/search.txt&index=recent
As noted in the message above, the FBI failure-to-pass rate had previously been on the order of 5%. (The article by Lenny Savino linked above is no longer available on the Knight-Ridder website, but an excerpt is available here
.)
That any of the 700 FBI employees polygraphed since Hanssen's arrest actually are spies is an unlikely proposition to begin with. But bearing in mind that polygraph screening has absolutely no validity whatsoever, and the base rate of guilt is quite low when it comes to espionage, what is the likelihood that any spies among the 700 employees are among the roughly one-in-a-hundred who "failed" the polygraph? Is it not more likely that any genuine spies will have learned how to pass the polygraph, like Aldrich Ames, Larry Wu-tai Chin, and (apparently) Ana Belen Montes did?
As Drew Richardson noted in his 1997 Senate testimony, to the extent that the FBI relies on the polygraph to direct its counterespionage efforts, it is undermining, not enhancing, national security.
Posted by George W. Maschke
- Jul 30, 2001, 02:29 PMIn a 28 July 2001 article titled "FBI describes polygraph failure rate as 'surprisingly low,'" Lenny Savino of the Knight Ridder Washington bureau reports that fewer than 25 of the more than 500 FBI personnel recently polygraphed had failed to pass. This article may be read on-line at:
http://www.krwashington.com/content/krwashington/2001/07/28/washington/BC_FBI_EXCLUSIVE_WA_national.htm?template=aprint.htm
Savino reports, "There's no way to tell what a normal failure rate would be for the FBI, but a senior official characterized the preliminary failure rate as 'surprisingly low.'"
Creating the public impression that the FBI polygraph failure rate is "surprisingly low" just before confirmation hearings (which may address polygraph screening) begin for FBI Director designate Robert S. Mueller III seems to be the very purpose for which Savino's anonymous FBI sources leaked this information.
But does less than 25 "deception indicated" or "inconclusive" outcomes (both constitute failure to pass) out of approximately 500 counterintelligence-scope polygraph interrogations constitute a "low" failure rate? It could only be considered "low" in absolute terms if one expected to find more than 25 spies among 500 FBI agents!
In relative terms, let's compare FBI's failure-to-pass rate of roughly 5% with the failure rate in FBI pre-employment polygraph "testing." According to C.S. "Steve" Rogers, a retired FBI polygrapher who is now Counterintelligence Officer at the Office of Internal Security, Los Alamos National Laboratory (LANL), less than 1% of FBI applicants polygraphed failed the counterintelligence portion of their pre-employment polygraph examinations. (Note that for applicants, an outcome of either "deception indicated" or "inconclusive" constitutes failure to pass and results in rejection of one's application.) Rogers is reported to have made this point at an 18 January 2001 meeting of the LANL Employee Advisory Committee, the minutes of which are available on-line at:
http://www.lanl.gov/committees/eac/minutes/0001Minutes/18Jan01.html
If high-level FBI employees are failing to pass their polygraph interrogations on counterintelligence issues at five times the rate of applicants, can such a rate truly be considered "low?" Perhaps only by a self-interested FBI official hiding behind the cloak of anonymity as he/she tries to manage public opinion.
According to the minutes of the LANL Employee Advisory Committee, Steve Rogers also told the Committee that the Department of Energy's (DOE's) Albuquerque test center "hasn't had a false positive result in the over 1800 tests they have performed." The FBI's roughly 25/500 failure rate seems rather high by comparison.
If the FBI were to require periodic polygraph screening of all employees, then a 5% failure rate applied to a work force of roughly 28,000 would mean some 1,400 failures to pass. And this situation would be repeated every five years with successive rounds of polygraph interrogations.
If FBI management treats those who fail to pass their pseudoscientific truth test the same way they treated former FBI Special Agent Mark Mallah, they are going to have a serious morale, retention, and recruitment problem. For more on the case of Mark Mallah, see his recent letter to the Senate Committee on the Judiciary at:
http://antipolygraph.org/articles/article-016.shtml
and his earlier public statement at:
http://antipolygraph.org/statements/statement-002.shtml
http://www.krwashington.com/content/krwashington/2001/07/28/washington/BC_FBI_EXCLUSIVE_WA_national.htm?template=aprint.htm
Savino reports, "There's no way to tell what a normal failure rate would be for the FBI, but a senior official characterized the preliminary failure rate as 'surprisingly low.'"
Creating the public impression that the FBI polygraph failure rate is "surprisingly low" just before confirmation hearings (which may address polygraph screening) begin for FBI Director designate Robert S. Mueller III seems to be the very purpose for which Savino's anonymous FBI sources leaked this information.
But does less than 25 "deception indicated" or "inconclusive" outcomes (both constitute failure to pass) out of approximately 500 counterintelligence-scope polygraph interrogations constitute a "low" failure rate? It could only be considered "low" in absolute terms if one expected to find more than 25 spies among 500 FBI agents!
In relative terms, let's compare FBI's failure-to-pass rate of roughly 5% with the failure rate in FBI pre-employment polygraph "testing." According to C.S. "Steve" Rogers, a retired FBI polygrapher who is now Counterintelligence Officer at the Office of Internal Security, Los Alamos National Laboratory (LANL), less than 1% of FBI applicants polygraphed failed the counterintelligence portion of their pre-employment polygraph examinations. (Note that for applicants, an outcome of either "deception indicated" or "inconclusive" constitutes failure to pass and results in rejection of one's application.) Rogers is reported to have made this point at an 18 January 2001 meeting of the LANL Employee Advisory Committee, the minutes of which are available on-line at:
http://www.lanl.gov/committees/eac/minutes/0001Minutes/18Jan01.html
If high-level FBI employees are failing to pass their polygraph interrogations on counterintelligence issues at five times the rate of applicants, can such a rate truly be considered "low?" Perhaps only by a self-interested FBI official hiding behind the cloak of anonymity as he/she tries to manage public opinion.
According to the minutes of the LANL Employee Advisory Committee, Steve Rogers also told the Committee that the Department of Energy's (DOE's) Albuquerque test center "hasn't had a false positive result in the over 1800 tests they have performed." The FBI's roughly 25/500 failure rate seems rather high by comparison.
If the FBI were to require periodic polygraph screening of all employees, then a 5% failure rate applied to a work force of roughly 28,000 would mean some 1,400 failures to pass. And this situation would be repeated every five years with successive rounds of polygraph interrogations.
If FBI management treats those who fail to pass their pseudoscientific truth test the same way they treated former FBI Special Agent Mark Mallah, they are going to have a serious morale, retention, and recruitment problem. For more on the case of Mark Mallah, see his recent letter to the Senate Committee on the Judiciary at:
http://antipolygraph.org/articles/article-016.shtml
and his earlier public statement at:
http://antipolygraph.org/statements/statement-002.shtml