Feds Target Instructors of Polygraph-Beating Methods

schwit1 writes "Federal agents have launched a criminal investigation of instructors who claim they can teach job applicants how to pass lie detector tests as part of the Obama administration's unprecedented crackdown on security violators and leakers. The criminal inquiry, which hasn't been acknowledged publicly, is aimed at discouraging criminals and spies from infiltrating the U.S. government by using the polygraph-beating techniques, which are said to include controlled breathing, muscle tensing, tongue biting and mental arithmetic. So far, authorities have targeted at least two instructors, one of whom has pleaded guilty to federal charges, several people familiar with the investigation told McClatchy. Investigators confiscated business records from the two men, which included the names of as many as 5,000 people who'd sought polygraph-beating advice. U.S. agencies have determined that at least 20 of them applied for government and federal contracting jobs, and at least half of that group was hired, including by the National Security Agency. By attempting to prosecute the instructors, federal officials are adopting a controversial legal stance that sharing such information should be treated as a crime and isn't protected under the First Amendment in some circumstances."

Comments

Only if they have a phrenology test

[Anonymous Coward]

I mean if we are going to go with the crackpot solutions we wouldnt want phrenology to feel left out, i believe it has some valuable insight and wait till i tell you about alchemy and auras.

Re:Only if they have a phrenology test

[Opportunist]

What, no jobs for ESPers? What kinda prejudiced quackery is that?

Re:Only if they have a phrenology test

[camperdave]

Well, the horoscope said it was a bad time to hire psychics.

Re:Only if they have a phrenology test

[Anonymous Coward]

Phrenology is a bad example for crackpot science: in a time when all psychology was still stated in religious terms such as "soul" it was one of the first attempts to come up with something rational & measurable.

Phrenology turned out to be wrong, because it was falsifiable. Mainstream psychology at that time wasn't even wrong.

Re:Only if they have a phrenology test

[squiggleslash]

The only science that's been found to be behind it is that people are slightly less likely to lie if they think that a lie detector will call them out on it.

Monitoring breathing, et al, doesn't mean it is capable of detecting lies. Me saying "molecules", "atoms", and "memory" doesn't make homeopathy have a science between it either...

Re:Only if they have a phrenology test

[TarPitt]

There were plenty of DoD specifically for ESPers at one point:

http://en.wikipedia.org/wiki/Stargate_Project

As long as you didn't mind staring at goats

Re:Only if they have a phrenology test

[VortexCortex]

I still don't understand why people assume monitoring breathing, heart rate, and skin conductivity is a 'crackpot' solution.

OT3 here. Allow me to clean this misconception up. FYI: e-meters work the same way.

Re:Only if they have a phrenology test

[spire3661]

Its NOT science, in that you can never come up with a true reading. All you can say is a human responded in certain ways. You have NO WAY of knowing how accurate the machine is. Polygraphs, as used in law enforcement might as well be a crystal ball for all its legal usefulness.

Re:Only if they have a phrenology test

[Anonymous Coward]

It is a near-cracpot technique, because it measure nervousness, not lying.

Of course some liars are nervous about being caught - especially when the interviewers are using this fancy scientific equipment on them. So it might help, except that the polygraph is "so old hat" and it is known that you beat it easily enough by "relaxing and pretending your lie is true".

On the other hand, some people are nervous for reasons orthogonal to the interview. They may worry about being illegally parked, or worry that an unexpected personal question might mark them as homosexual or some such. Or simply nervous about screwing up their interview because they are not good talkers.

Not all polygraph owners/users know what the machine really do. If they think it really is a "lie detector" you have crackpottery indeed. A question might remind them about something important they forgot - the polygraph notices a reaction, mis-interpretation could easily happen.

And there is the issue of scale. Someone being polygraphed in a murder investigation might simply worry that his dope or thieving might get discovered. Bad, but nowhere near a reason to jail hin for murder though.

Re:Only if they have a phrenology test

[MightyMartian]

It's not that polygraph machines are sometimes unreliable, it's that they are utterly unreliable. For all the reliance law enforcement has had on them over the years, they are essentially voodoo. The simply fact is that they don't work, and what little they do do is placebic, in that the ignorant may fear being hooked up to a machine about as useful as L Ron Hubbard's e-meter and thus be spooked into confessing.

Re:Only if they have a phrenology test

[Seumas]

"Polygraph machines" are nothing but a tool that facilitates the accusation and judgement of the person operating it. If the person wants to find you suspicious, they will conclude that the results from the machine indicate suspicion. This has been demonstrated repeatedly and there's a strong reason why polygraphs are not admissible in court (except the few states where they are -- but only if all parties involved agree to it).

Re:Only if they have a phrenology test

[Zemran]

There is a scientific basis behind homoeopathy but it is still sugar pills with no meaningful medicinal purpose except as placebo. The main difference between alternative medicine and medicine is that when an alternative or herbal remedy is proven to work it becomes medicine.

Re:Only if they have a phrenology test

[Jane Q. Public]

"I still don't understand why people assume monitoring breathing, heart rate, and skin conductivity is a 'crackpot' solution. There is a scientific basis behind it, unlike most actual 'crackpot' areas. It doesn't ALWAYS work, and it's (clearly) beatable, but it's still a science."

No, it isn't. This is why:

The THEORY behind polygraphs is rational. But theories (by definition) are testable!

The REALITY behind polygraphs is that they have been tested. And tested. And tested again. And again. And again. And not one responsibly performed "blind" study has EVER shown polygraphs to work worth a damn.

So: the idea behind polygraphs IS "science" in a sense. But only in the sense that the actual science shows they don't work.

Okay? Understand the difference?

Re:Only if they have a phrenology test

[nbauman]

I still don't understand why people assume monitoring breathing, heart rate, and skin conductivity is a 'crackpot' solution. There is a scientific basis behind it, unlike most actual 'crackpot' areas. It doesn't ALWAYS work, and it's (clearly) beatable, but it's still a science.

That's a fair question. Suppose you have a technique that was developed by scientific exploration, it's tested, and it turns out not to work. Is it science (but discredited science), or is it just not science at all?

It's not like they're praying to the aliens in orbit to read the person's mind and tell them if they're lying or not.

I'd like to see a controlled trial in which one team reads peoples' minds by praying to aliens, and the other team uses a lie detector. Which team will be better at detecting lies?

Re:Only if they have a phrenology test

[nbauman]

As AntiPolygraph.org has pointed out, one use of the polygraphs is simply to conduct interviews with subjects without a lawyer present, so that the examiners can use interviewing techniques with unrestricted questions that an informed, rational person would never submit to.

If interviewers started asking questions about your sex life, a lot of applicants would walk out there, and it would usually be illegal. But in a polygraph exam, they might permit it.

Re:Only if they have a phrenology test

[nbauman]

Homeopathy is a practice with a plausible scientific mechanism behind it. That's the first requirement of medical science.

The second requirement of medical science is that somebody tests it in a randomized, controlled trial to see if it works. At that point, homeopathy failed.

(Or I think it did. I can't cite off the top of my head a randomized trial in which homeopathy was ineffective. I assume there is one.)

Re:Only if they have a phrenology test

[Shadow of Eternity]

Homeopathy's mechanism is that you cure something by putting a drop of what *causes* the symptoms you want to cure into water, shake it in a bunch of directions while "intending" to make a cure, and then repeat the dilution and shaking process about 30 more times. That's neither plausible nor scientific. You've got a better chance of finding a single grain of rice in a ball of water the size of the moon than you do an atom of anything but water in a homeopathic remedy.

Re:Only if they have a phrenology test

[sjames]

Because controlled studies have shown that none of that allows a polygraph operator to determine if the subject is telling the truth. It turns out that it works a lot like tarot cards. The operator forms an opinion and uses the readout as a way to justify it.

there's a lot of stuff out there that uses the trappings of science to explain it's amazing capabilities that turn out to be bunk. Phrenology for example was based on the 'obvious' fact that more developed areas of the brain would result in a palpably raised area of the skull and that you wanted to hire people with the more virtuous areas developed and avoid those who had well developed areas for crime.

Re:Only if they have a phrenology test

[whoever57]

The only science that's been found to be behind it is that people are slightly less likely to lie if they think that a lie detector will call them out on it.

Penn and Teller did an program about lie detectors. Firstly, why are they not being prosecuted, since they explained how to beat them?

They explained that what happens is that the lie detector is a BS machine, but after the test is "over", the interviewer tells them that the lie detector showed that they were lying and that they should come clean. Many people then tell the interviewer the truth.

According to Penn and Teller, fooling the lie detector is simple: spoof the results by contracting and releasing a large muscle (they suggested the sphincter muscles) during the interview. This will destroy the value of any baseline measures.

Re:Only if they have a phrenology test

[dcollins117]

As AntiPolygraph.org has pointed out, one use of the polygraphs is simply to conduct interviews with subjects without a lawyer present, so that the examiners can use interviewing techniques with unrestricted questions that an informed, rational person would never submit to.

If you submit to a polygraph test you've just established exactly who is dominant and who is submissive. You've already lost before the first question is asked.

Re:Only if they have a phrenology test

[rtb61]

Not to forget, lie detector test have zero impact on psychopaths, so basically the worst of the worst will pass, kind of making the lie detector scam pointless, or more accurately the question reaction flim flam show pointless. This being the reason they are banned in most countries. This really stinks of the FBI intending to use fake like detector tests to incriminate any one they want too and these plans are threatened by the exposure.

Re:Only if they have a phrenology test

[Velex]

Sadly, psychology still isn't even wrong in our time.

Re:Only if they have a phrenology test

[nbauman]

"Plausible mechanism" is a pretty low barrier.

Suppose instead of water, they were using beef broth, and had infected the beef broth with just a few bacteria. They could dilute and shake 30 times and still get infected beef broth at the end.

They could hypothesis that there is some similar mechanism in water which will still be found after 30 dilutions.

Suppose the ancient Greeks were considering homeopathy, before they had confirmed the existence of molecules and so forth. Homeopathy could have been plausible to them.

Rather than argue about how plausible that is, it's easier to try to confirm or disconfirm it using the experimental method.

It's good to examine mistaken ideas because they force you to define the scientific method very precisely.

Re:Only if they have a phrenology test

[nbauman]

That's another way to put it. Cops like to dominate people.

Take a look at that memo from the New York City stop-and-frisk lawsuit that we discussed on Slashdot yesterday.

Re:Only if they have a phrenology test

[jamstar7]

Its NOT science, in that you can never come up with a true reading. All you can say is a human responded in certain ways. You have NO WAY of knowing how accurate the machine is. Polygraphs, as used in law enforcement might as well be a crystal ball for all its legal usefulness.

Which is why they're inadmissible as evidence in a trial. The damned things don't work. Period. They can be beat easily. Hell, just smoke a joint first, you'll beat it. You'll either be so mellow that nothing bothers you, showing no changes, or so paranoid that your readings will be all over the scale and thus useless.

Why they're cracking down on somebody who shows how to beat a bullshit test using readily available information, just collated for an easy instructional session, is beyond me. They have something against private enterprise or something??

Re:Only if they have a phrenology test

[Shadow of Eternity]

Except you're forgetting the whole part about using the same thing which causes the symptoms you want to cure. If someone can't sleep you use caffiene but you "intend" to make a cure, basically believe really hard. If your idea of "plausible mechanism" is to go back to magic and superstition then you've essentially moved the goalposts to such a degree that there's not one thing which can be considered implausible.

Re:Only if they have a phrenology test

[nbauman]

A lot of the plausible mechanisms that science tested in the 17th century did come from magic and superstition.

Earth, air, fire and water. Sounds plausible. You couldn't know unless you tested it.

Actually they got the concept right. It just turned out that there were 92 elements instead of 4. Why is 92 elements more plausible than 4?

Re:Only if they have a phrenology test

[ultranova]

Suppose the ancient Greeks were considering homeopathy, before they had confirmed the existence of molecules and so forth. Homeopathy could have been plausible to them.

Not really. Ancient Greeks had wine, and were familiar with the practice of diluting it to control (lessen) the effects. It seems unlikely they wouldn't had made the obvious connection.

Re:Only if they have a phrenology test

[Sabriel]

Not to forget, lie detector test have zero impact on psychopaths, so basically the worst of the worst will pass, kind of making the lie detector scam pointless,

Actually, that would be worse than pointless: such a false negative bias (all psychopaths pass, some non-psychopaths do not pass) would increase the ratio of psychopathy in those organizations. Then look at the organisations that use the test: defence, security and intelligence agencies. Aesop version: We chickens, afraid of being attacked by foxes, have been hiring snakes.

Re:Only if they have a phrenology test

[The Wild Norseman]

Not to forget, lie detector test have zero impact on psychopaths, so basically the worst of the worst will pass, kind of making the lie detector scam pointless, or more accurately the question reaction flim flam show pointless. This being the reason they are banned in most countries. This really stinks of the FBI intending to use fake like detector tests to incriminate any one they want too and these plans are threatened by the exposure.

And also remember that the most infamous and successful spies in the United States passed their polygraphs handily. Polygraphs are a dangerous distraction when relied upon as they have been in the past and apparently continuing into the future.

Re:Only if they have a phrenology test

[niftymitch]

As AntiPolygraph.org has pointed out, one use of the polygraphs is simply to conduct interviews with subjects without a lawyer present, so that the examiners can use interviewing techniques with unrestricted questions that an informed, rational person would never submit to.

If interviewers started asking questions about your sex life, a lot of applicants would walk out there, and it would usually be illegal. But in a polygraph exam, they might permit it.

Bingo.... and yes the issue of power applies.

I suspect that no legal counsel is sufficiently schooled in polygraphs to be of any value.
So few are fully competent of the law in a dynamic situation as to be of any help.
This is in contrast to court where rehearsal and research accompanied by a team is possible.

I might consider subtle changes in the body language of the agent conducting the polygraph as bias.
Flash a tattoo on an arm unbutton a jacket, lean forward, lean back, smile, no smile, square smile....
If the operator of the polygraph has knowledge of the suspected crime I am convinced he is in
such a position of absolute power that I cannot convince myself that the polygraph has any value
beyond hearding sheep.

Re:Only if they have a phrenology test

[niftymitch]

"Plausible mechanism" is a pretty low barrier.

Suppose instead of water, they were ......

At the one molecule per gallon dilution levels it is hokum.

Today it is common to get "shots" to desensitize individuals from allergy triggering substances.
The one molecule per gallon dilution (or more) places homeopathy outside of the reach of the FDA.

At the concentrations found in the bubble patch tests used for TB and grid on the back testing it
is conceivable that the triggered response that a doctor reads be the upper bounds for homeopathy.

Sadly triggered responses by some are quite serious and permitting material to be sold over the counter
is contrary to the medical care that one in many might need. We see this now in response to the astounding
jump in the number of immunizations children receive.

N.B. Immunization and homeopathy are birds of the same feather. One is at a level requiring regulation. The other not.

Re:Only if they have a phrenology test

[Raenex]

"Penn and Teller did an program about lie detectors. Firstly, why are they not being prosecuted, since they explained how to beat them?"

From the article:

In his speech in June, Customs official Schwartz acknowledged that teaching the techniques _ known in polygraph circles as "countermeasures" _ isn't always illegal and might be protected under the First Amendment in some situations.

"I'm teaching about countermeasures right now. The polygraph schools are supposed to be teaching about countermeasures," he said. "So teaching about countermeasures in and of itself certainly is not only not illegal, it's protected. You have a right to free speech in this country."

But instructors may be prosecuted if they know that the people they're teaching plan to lie about crimes during federal polygraphs, he said.

In that scenario, prosecutors may pursue charges of false statements, wire fraud, obstructing an agency proceeding and "misprision of felony," which is defined as having knowledge of serious criminal conduct and attempting to conceal it.

Oh Americans

[Anonymous Coward]

Oh Americans and their liberty - hail our new Polymorphic overloards.
Discussing how to avoid them is a crime.
They should come for martial arts instructors too, can be used to escape the police [state] :)

It would be fun to have instruction for such avoidance discussed on Darknets like Freenet or i2p forum.

Re:Oh Americans

[Joce640k]

I don't think the "polymorphic overlords" are here yet... unless Obama really is a giant weasel and not just pretending to be one.

First the came for the poly dudes

[smittyoneeach]

But I was not a poly dude, so I was all: 'Meh'.
Then they came for the yoga instructors, since relaxation is where it's at, and I was kinda: 'Urf?'
Then they came for my surf board.

Oh man, oh man

[Anonymous Coward]

The Obama admin is sure pulling out all the stops on the full retard organ, this time.

Re:Oh man, oh man

[__aaltlg1547]

This comment is offensive to retarded people. They shouldn't be tainted with the same brush as jackbooted thugs.

Re:Oh man, oh man

[nitehawk214]

This comment is offensive to retarded people. They shouldn't be tainted with the same brush as jackbooted thugs.

Also an insult to lump them in with people that believe in polygraph, dowsing and homeopathy.

Re:Oh man, oh man

[Zemran]

This comment is obviously offensive to jackbooted thugs as it clearly implies that they are lower in the social order than retarded people. This is clearly not the case as it takes considerable intellect to work out how to put on a pair of jackboots or even know what they are. You cannot go into your local mall and ask for a pair of jackboots, they never have them in stock. So simply being able to purchase a pair of jack boots shows considerable intellect.

Be interesting if the course were a book

[transporter_ii]

Be interesting if the course were a book and they sold it on Amazon instead of teaching a class. Make the 1st Amendment kick in a little harder.

Re:Be interesting if the course were a book

[transporter_ii]

A quick search on Amazon turned up:

How to Beat a Lie Detector Test (Secrets Series) by Steve Gillman (Jul 20, 2010)
Beat the box: The insider's guide to outwitting the lie detector by Vlad Kalashnikov (1983)
Deception Detection: Winning The Polygraph Game by Charles Clifton (May 1991)

Re:Be interesting if the course were a book

[TarPitt]

"Applied Cryptography" used exactly this method when crypto algorithms were subject to export controls.

You couldn't export say the source code for DES, but you could include the source code in a book on crypto, as first amendment protections applied.

The first amendment even protected use of an OCR friendly font for the source code.

Re:Be interesting if the course were a book

[jeti]

And you can probably forget about your security clearance if you order any of these books.

Re:Be interesting if the course were a book

[VortexCortex]

What if you read them at the library? Oh! Or what if you borrow them from a friend who checked them out?

Quick! Someone calculate the 6 degrees of separation between anyone who may have come in contact with this information! We might be able to solve this NSA thing after all!

Re:Be interesting if the course were a book

[mrbester]

Most police procedurals have had plentiful episodes where polygraph tests are manipulated detailing these same methods so you don't need any book. Perhaps the feds should go after the writers from the last 20 years for their "dissemination of classified information" as well. Dicks.

Re:Be interesting if the course were a book

[Teun]

Or you could look up some official papers like court documents and the studies they're based on from countries that outlawed polygraphy for legal evidence.

Re:Be interesting if the course were a book

[nbauman]

The way they got the instructors was by having undercover agents tell them that they wanted to beat the polygraph to facilitate a crime.

It's a crime to help somebody else commit a crime. http://lawcomic.net/guide/?p=446

If the agents had simply asked the instructors to teach them how to beat a polygraph test, it would have been hard to get a conviction. Instead, they told the instructors that they wanted to beat a polygraph test for the purpose of committing a crime.

Several people familiar with the investigation said Dixon and Williams had agreed to meet with undercover agents and teach them how to pass polygraph tests for a fee. The agents then posed as people connected to a drug trafficker and as a correctional officer who’d smuggled drugs into a jail and had received a sexual favor from an underage girl.

As soon as the "customers" started to talk about illegal activities, the instructors should have refused to deal with them and given them their money back.

Chad Dixon, who wasn't too sophisticated about these things, might have fallen for it. But I don't know how Doug Williams, who used to work for the cops, could fall for it.

I don't know if there's any legal duty to report your suspicion of the commission of a crime. I don't think so, but they included child sex into it for a reason. It might be good for them to say, "You know what? I think you're an undercover agent." Then they could (correctly) argue later that they didn't think the claim of committing a crime was credible.

Re:Be interesting if the course were a book

[nbauman]

RTFA. It suggests that prosecution is possible only if the instructor has specific knowledge that one of his students intends to use the abilities taught to hide evidence of a criminal offence, so clearly authors of a book would not be prosecutable in this fashion.

That's an important point.

Several people familiar with the investigation said Dixon and Williams had agreed to meet with undercover agents and teach them how to pass polygraph tests for a fee. The agents then posed as people connected to a drug trafficker and as a correctional officer who’d smuggled drugs into a jail and had received a sexual favor from an underage girl.

Re:Be interesting if the course were a book

[nbauman]

What if you read them at the library? Oh! Or what if you borrow them from a friend who checked them out?

It wouldn't work. They would ask you, "Have you ever read a book on how to beat the polygraph?"

Re:Be interesting if the course were a book

[nbauman]

If an undercover agent went into a bookstore and said, "Do you have any books on cryptography? I have to send some nuclear bomb secrets to America's enemies," and they sold him a book, they could probably prosecute that.

Re:Be interesting if the course were a book

[icebike]

What if you read them at the library? Oh! Or what if you borrow them from a friend who checked them out?

It wouldn't work. They would ask you, "Have you ever read a book on how to beat the polygraph?"

But you'd know how to beat that question, so.....

Re:Be interesting if the course were a book

[tragedy]

Yes, they still allow libraries, but if they use a networked electronic checkout system, every book you check out is reported to the police.

Re:Be interesting if the course were a book

[gregor-e]

Prosecuting for knowledge of criminal intent is what allowed us to imprison someone for ten years for selling lights.

Re:Be interesting if the course were a book

[niftymitch]

So is the NSA downloading Amazon's sales records to look for terrorists? I used to think the first amendment would say "no", but now I'm not so sure they'd interpret it the same way I and the courts do.

I think the tense is wrong.

They already have the records... further the records are cross referenced to Visa/ Mastercard etc.

And they have all the meta data to build a connection graph from you to anyone else on the planet.
With modern connectivity the reach illuminated by the Kevin Bacon exercise has collapsed. Today most of us are connected to FB friends some like K. Kardashian have millions of connections (+13 million). Mailing lists like kernel.org and OLPC are global and large. This alone places the vast majority of us within the three connections the NSA has acknowledged they search.

The point is that "Six degrees of separation" is no longer true (IMO) in the hands of massive data swept piles where connectivity rules are morphable/ fragile and can be selected by an agency to show anything.

One pool of common connections is the set of humans and dogs with data records in the union of FBI, CIA, NSA, Google, Facebook, Comcast, .... . Thus I assert that connectivity of three sounds safe but can still be abused to connect anyone to anyone or anything..

BTW: Boo the dog has half the connections on FB that Kim has. +7million is a lot and there are a lot of "famous" folks

Re:Be interesting if the course were a book

[Hognoxious]

It's like asking me to teach you how to put the evil eye on people versus asking me to tell you how to put the evil eye on people so you can kill your boss.

There's zero difference because there's no such thing as the evil egdéj()* ;
no carrier

How about

[Anonymous Coward]

Just admitting that Polygraphs are not reliable indicators of truthfulness?

Re:How about

[__aaltlg1547]

According to an NAS study, they're something like 85% reliable. The problem with an 85% reliable test is that it will produce a lot of false positives and false negatives. People you should have hired will be screened out and people you shouldn't have hired will be accepted. Older-fashioned methods work better. Interview the person, the family members, long time acquaintences and co-workers. Ask open-ended questions about the person's relationships, how they work with others, how they view authority, what they do in the community, etc. You'll discover anything that's relevant before long.

Subjecting people to lie detectors is all about threats and intimidation. They probably deter more bad people from even applying than they screen out in the test, but they also deter good people who have no confidence in polygraphs. But those people are also detered by the prospect of somebody prying into their life like they do in DOD type security screenings.

Re: How about

[nospam007]

"Just admitting that Polygraphs are not reliable indicators of truthfulness?"

If they do that, they have to stop using the non-functioning bomb- detectors as well, we can't have that!

Re:How about

[camperdave]

Lie detectors are 100% reliable. If I see one at a job interview, it is a sure sign that I don't want to work there.

Re:How about

[jonbryce]

And lets look at this 85% reliability more carefully.

Supposing you have 1,000,000 non-terrorists and 100 terrorists. You ask them if the are a terrorist, and use the lie detector to determine whether or not they are telling the truth. Everyone says they are not a terrorist. The lie detector will identify 150,085 people as terrorists, of which only 85 are actually terrorists. In otherwords, if the lie detector says you are a terrorist, there is a 0.057% probability that you are actually a terrorist.

How do these figures work?

Of the 1,000,000 non-terrorists, it will correctly identify 850,000 of them as being non-terrorists, and incorrectly identify 150,000 as being terrorists. Of the 100 terrorists, it will correctly identify 85 of them as being terrorists, and incorrectly identify 15 of them as not being terrorists. A total of 150,085 people identified as terrorists, only 85 actually are.

Re:How about

[Anonymous Coward]

According to an NAS study, they're something like 85% reliable.

According to the American Psychological Association, "There is no evidence that any pattern of physiological reactions is unique to deception. [...] polygraph research has not separated placebo-like effects (the subject's belief in the efficacy of the procedure) from the actual relationship between deception and their physiological responses [...] virtually no research assesses the type of test and procedure used to screen individuals for jobs and security clearances [...] Most psychologists and other scientists agree that there is little basis for the validity of polygraph tests."

The 85% figure you quote refers to what is called "specific incident testing", i.e. investigating whether the subject performed some particular action about which details are known. This is not the kind of testing used in security clearance tests of the kind discussed by TFA; those tests are a whole lot less reliable.

Re:How about

[Anonymous Coward]

The parent article needs to be modded down for in-correct application of math/stats (slashdot math/stats types will be easily able to confirm this). While false positives and false negatives are very important to discuss with testing (of basically any sort), the author does not understand that false positive probability does not (generally) equal false negative probability, or the application of 'reliability' to this problem. There is a trade off between the two errors. If you are changing thresholds on a test (very precise), then if you want fewer false positives (false accusations) then you will have more false positives*.

Not enough information is given on the 85% "reliability", but generally one might assume this applies to the false negative rate, i.e. the test will detect 85% of terrorists, and 15% of terrorists will not be detected. It does not however say anything about mistaking non-terrorists for terrorists. (We could switch this assumption around). A low 'reliability' in this assumption, generally means few false positives - the test is sacrificing the ability to detect terrorists in order to not falsely accuse non-terrorists (in layman's terms this is 'giving the benefit of the doubt').

http://en.wikipedia.org/wiki/False_positives#Type_I_error

* A test can still be improved for both error types, but that would come about from improving the test vice changing thresholds of 'guilt'.

Re:How about

[Znork]

Well, unless you actually act on the tool and treat those 150085 people as terrorists, in which case 10% of them will actually become terrorists and each will drag at least one other person into it through loyalty and/or family ties. In which case you now have more than 30k terrorists of which only half is believed to possibly be a terrorist.

Not a solution, no, but part of the problem.

Re:How about

[mpeskett]

Was it this one? http://yudkowsky.net/rational/bayes

Cancer testing is a fairly easy example of something with a fairly low incidence in the general population, and where we would really like to avoid false results (so we don't miss a tumour or put people through unnecessary chemo).

Always important to remember that if your error rate is higher than the background rate of how many people have cancer, then you're going to end up flagging more people without cancer than with. Takes a very precise test to raise your confidence level much away from the highly-probable assumption of "not cancer" (or "not terrorist").

Re:How about

[517714]

I worked for a retail company that applied lie detector tests to their employees. They had a written test for aptitude and general truthfulness followed by a polygraph test. The manager said I was their only employee who was not required to take the polygraph based on the written test results. I wonder how I would do today on that same test thirty years later when the world isn't so black and white in my mind.

Re:How about

[nbauman]

Some lawyers take the position that it is unethical for them to allow a client to commit perjury on the stand. There are 2 cases:

(1) Lawyers tell their clients that they don't want to hear any admissions of guilt. Sometimes this results in a lawyer thinking that his client is guilty, when the client is really innocent, as Jesse Friedman claimed. http://en.wikipedia.org/wiki/Capturing_the_Friedmans

(2) Lawyers occasionally have reported their clients' admissions to the court. There was one case in which a client had stabbed somebody, and he told his lawyer that he was going to testify that he saw something silvery and thought it was a gun, so he acted in self-defense.

Re:How about

[jonbryce]

There were 100 terrorists, and it caught 85 of them.

Re:How about

[jsepeta]

still hoping to swing that three-way? good luck.

Re:How about

[quarterbuck]

That is not too bad if you combine with "camperdave"s comment that interviews work better.
Suppose interviews with family etc. work at 86% accuracy and the errors in that are independent of errors in polygraph. In that case interviews + polygraph give you a combined 2% error rate (for false negatives, which are the dangerous ones in case of terrorists). False positives do go up a lot to about quarter of the initial population, but if you have a large enough population to work with, you do not suffer in terms of quality when it comes to the final selected population of spooks.
People who fail polygraph tests, who are too short or look ugly are not a protected class - so you can legally discriminate against them. Is it justified? I am not too sure.

Re:How about

[__aaltlg1547]

I think NSA are looking for spies more than they're looking for terrorists.

Re:How about

[niftymitch]

According to an NAS study, they're something like 85% reliable. The problem with an 85% reliable test is

Gack....

The issue is that 85% reliable implies a control and standard method. i.e. a sufficiently large known sample base.

My problem with this is that I cannot imagine a control that I would trust.

Further the number of bad guys needs to be largish. This might be possible in retail but not in national security. In national security the population distribution would contain a lot of pathology. To be good one must have a degree of paranoia and an adventurous mind looking for handy for criminal activities.

Protecting a lie

[Anonymous Coward]

IALA

The real crime here is that law enforcement agencies are using such a notoriously unreliabletechnology for investigatory and evidentiary purposes. Polygraphs have absolutely no place in the modern justice system.

Re:Protecting a lie

[wiredlogic]

There are no (as in NONE) states that require you take one.

The issue is when a polygraph is a defacto requirement for getting certain jobs even if you have a right to refuse.

Re:Protecting a lie

[interval1066]

The real crime here is that law enforcement agencies are...

THE REAL crime here is that there is NO WAY this Fed action passes the first amendment smell test. ANYONE has an ABSOLUTE CONSTITUTIONAL RIGHT to say whatever they want about lie detectors, yet no one seems to give a wiff.

Re:Protecting a lie

[tqk]

There are no (as in NONE) states that require you take one.

The issue is when a polygraph is a defacto requirement for getting certain jobs even if you have a right to refuse.

That's not the only issue. Plenty of LEOs assume failing the polygraph indicates guilt so they'll double down on proving guilt on the part of their suspect instead of widening their search to find the real perpetrator.

Re:Protecting a lie

[nbauman]

The issue isn't any of the First Amendment rights. The issue is that the undercover agents tricked the instructors into believing that they were helping people commit a crime.

Several people familiar with the investigation said Dixon and Williams had agreed to meet with undercover agents and teach them how to pass polygraph tests for a fee. The agents then posed as people connected to a drug trafficker and as a correctional officer who’d smuggled drugs into a jail and had received a sexual favor from an underage girl.

I think it's entrapment, but the Supreme Court doesn't agree.

Re:Protecting a lie

[interval1066]

Several people familiar with the investigation said Dixon and Williams had agreed to meet with undercover agents and teach them how to pass polygraph tests for a fee.

It is, in fact, a first amendment right to discuss ways of beating a polygraph, just becuase the media hasn't figured that out (they are stupid anyway) deosn't mean it isn't so. Any lawyer working this case should be able to glom right on to that fact. Also; I defy you to find any statute anywhere that makes doing this a crime. Why in the world anyone would need to keep such instruction on the low down is beyond me.

Re:Protecting a lie

[nbauman]

We don't know exactly what the conversation was between the undercover agents and the instructors, but judging by past cases, they set these things up very carefully to get the suspect to cross the line into illegality.

They must have said something like, "I'm committing a crime and I want to know how to beat the lie detector in case the cops question me about it," that would cross the line. The instructor would believe that he was helping the a criminal to break the law.

There have been cases where a drug dealer's girlfriend picked up the phone, wrote down a message, and was convicted of drug dealing.

Re:Protecting a lie

[interval1066]

There have been cases where a drug dealer's girlfriend picked up the phone, wrote down a message, and was convicted of drug dealing.

I would hope not, such cases would more likely be a situation where the girlfriend knew what was going on, had past dealings with the law regarding, etc. Any public defender worth making his salary should be able to show a message being taken by an innocent bystander, I've seen such scenarios play out in courts. Inncent bystanders usually don't get rail roaded, at least as a matter of prosecutorial policy. Even the real crooked prosecutors get warned by judges to lay off innocent people from time to time, if its plain to see they are in fact innocent.

Re:Protecting a lie

[nbauman]

Any public defender worth making his salary should be able to show a message being taken by an innocent bystander, I've seen such scenarios play out in courts. Inncent bystanders usually don't get rail roaded, at least as a matter of prosecutorial policy.

Not usually, but it does happen. Some prosecutors think that their job is to prosecute as many people as they can and get the longest sentences they can. When defense lawyers start collecting cases of unjust sentences, there are a lot of cases of drug dealers' girlfriends who were peripherally involved but got longer sentences than the actual dealers, because the girlfriends had nothing to offer in a deal.

The quality (and salary) of public defenders varies greatly.

Even the real crooked prosecutors get warned by judges to lay off innocent people from time to time, if its plain to see they are in fact innocent.

Maybe it's just the outrageous cases that wind up in the newspapers, but I read about a lot of them. America seems to have turned into the world's biggest police state. College kids get 10, 20 years for dealing drugs? It's sin and punishment run amuck.

So it's come to this

[Opportunist]

Finally, we have a case for information being outlawed.

Re:So it's come to this

[transporter_ii]

I've had a website devoted to alternative cancer treatments almost since the start of the Internet. I wonder if they will knock, or just kick the door in?

Re:So it's come to this

[BrokenHalo]

Well, kicking the door in is just an advanced form of knocking.

Re:So it's come to this

[camperdave]

Well, kicking the door in is just an advanced form of knocking.

Jusk ask Chuck Norris.

Re:So it's come to this

[cellocgw]

I've had a website devoted to alternative cancer treatments almost since the start of the Internet. I wonder if they will knock, or just kick the door in?

You have it exactly backwards. Polygraphs *and* "alternative [anything]" are the fakes. It's more like, if I published a book on why so-called alternative treatments were complete bunkum (which they are) and the Feds wanted to shut me down.
As Iain whatsisname said, "If it worked, we'd call it a treatment. It's called 'alternative treatment' because it DOESN'T."

Re:So it's come to this

[mspohr]

They will most likely send a SWAT team (without a search warrant):
Texas SWAT raid destroys organic farm:
http://www.huffingtonpost.com/2013/08/15/texas-swat-team-conducts-_n_3764951.html

Re:So it's come to this

[Opportunist]

Still, it should not be illegal to publish the information just because the information is bull.

If it is, I demand that the next thing outlawed are Bibles and other "holy books".

Obstructing an agency proceeding and wire fraud?

[lamer01]

Really? Talk about overreaching....

QL'EB?

[Hognoxious]

It's like attacking tarot readers for claiming they can work out when palmists are making shit up.

Polygraphs

[onyxruby]

Why the hell are polygraphs still being used in the 21st century? They aren't admissible in a court of law for a damned good reason. They are junk science and no better than a voodoo board. The only thing they do is tell whether or not your nervous. They are a perfect example of something that provides a false sense of security as Ames and your other famous spies all /passed/ their lie detector tests. These things need placed in the museum of junk science post haste.

Re:Polygraphs

[Impy the Impiuos Imp]

They're more useful as a scare technique since the common man thinks they work well.

They're also useful, like the "anonymous" tip, to generate further trumped-up reason to investigate someone.

Re:Polygraphs

[onyxruby]

There's a reason people sell alternative medicine as herbal supplements. This way you sidestep the FDA and get to avoid any type of medical scrutiny. It's a legal way of selling snake oil. They need to change the law so that herbal supplements, homeopathic and the like all require FDA approval before making claims.

The funny thing is a lot of our medicine did trace back to these old practices. However with medical science they were able to find the parts that actually worked, study them and those get submitted as medicines. The rest are modern snake oils that take consumers in for billions of dollars every year.

Re:Polygraphs

[LetterRip]

Why the hell are polygraphs still being used in the 21st century? They aren't admissible in a court of law for a damned good reason. They are junk science and no better than a voodoo board.

Voodoo is a rather apt analogy. The reason they are used is that they help amplify the belief of the individual that they will get caught in a lie. Thus the reason the FBI are angry is that this teaching will negate the belief that you'll get caught and defeats the psychological manipulation.

Ie if the vooodoo man casts a hex on you, and you believe in voodoo - then you might engage in behavior that makes the hex self fulfilling; but if another voodoo man sells you a talisman to ward off the hex - your belief in the second voodoo man cancels the belief in the first voodoo man.

Re:Polygraphs

[Nidi62]

The only thing they do is tell whether or not your nervous.

Not even nervous. I took a polygraph (well, voice stress analysis) as part of the hiring process for a fairly large metropolitan police force (with a Masters degree I would have started out at roughly $45k per year base, as opposed to the roughly 25k I am making at my current job. Yay shitty economy). One question was so absurd (have I ever hired a prostitute) that I laughed as I replied in the negative. Of course the readout then showed "stress" in my voice. However the baseline tests (which were the exact same questions)showed I was being truthful. In the end, after going through the whole hiring process, passing the physical test and everything, they decided not to hire me. In the end I think it was a good thing though, because this particular police department is not the most reputable in my city, and now I can see why.

Re:Polygraphs

[jonbryce]

That is the case in Europe. They still sell all the stuff, but there is nothing at the point of sale to say what it is supposed to do.

Re:Polygraphs

[jopsen]

Why is that question absurd? Many people have hired prostitutes?

And in many places it's perfectly legal to do so... Furthermore, if you asked me that question at an interview, I'm sure my union would smell blood a sue the hell out of you...
Though to be fair, my union would probably sue you over the mere suggestion that I submit to a polygraph test :)

In many countries it is very limited what details about a candidates personal life you are allowed to inquire about, as any such inquiry opens you to law suits for discrimination, should you choose not the hire the candidate for any reason what so ever.

Re:Polygraphs

[Rockoon]

That is the case in Europe. They still sell all the stuff, but there is nothing at the point of sale to say what it is supposed to do.

The thing is that at one time our FDA operated in the same fashion. Their only goal was to ensure the safety of the product, not its efficacy.

But then the large pharmaceutical companies lobbied to corrupt the system, pushing for the requirement that efficacy also be proven. The reason for this is as both an expensive barrier to entry as well as a delay tactic.

Re:Polygraphs

[Zero__Kelvin]

There is plenty of alternative medicine that works just fine; in some cases it works better. I used to take 150mg of Zantac twice a day due to GERD. I needed it every day for years. At some point someone turned me on to Cayenne pepper tablets. I took them before meals for a short period of time and guess what ... I didn't need Zantac anymore. This is but one of hundreds of examples. Evidently you weren't aware that a great deal of medicine is created by taking a chemical that occurs in nature, tweaking the molecule so they can patent it, and charging your insurance company an arm and a leg for it (excuse the pun.) Off the top of my head, one medicine that is a direct derivative of naturally occurring chemicals is Aspirin.

Re:Polygraphs

[winwar]

You know what alternative medicine that works is called?

Medicine.

If it is alternative, then it isn't medicine.

Re:Polygraphs

[Zero__Kelvin]

So you don't have any idea what it means, then. I can accept that. For future reference. Some work, and some don't. Acupuncture absolutely works. In other words: "There are more things in heaven and earth, Horatio, Than are dreamt of in your philosophy. "

Re:Polygraphs

[Cederic]

How did you assure that she wasn't duped into entering the country to work in the sex trade?
How did you assure that she's disease free?
How did you assure that she's not being forced into prostitution?
How did you assure that she's over 18?
How did you assure that she's a she?

I have no issue with people choosing to work as prostitutes, but I have contempt for those that hire them.

Re:Polygraphs

[intermodal]

When I worked for a municipal fire department, I had to take a polygraph during the hiring process. The biggest thing I learned in my reading beforehand was that they're basically quackery, and that the only way they work even a little is if you believe they do.

One of his first questions during the test was if I had done any reading on how polygraphs work before coming there. I responded of course. He asked disapprovingly of why, and I told him that only a fool would fail to research every part of a hiring process.

And then I proceeded to not give a crap throughout the entire "test." I even blatantly lied to him about my name during the baseline, and he got mad at me. I laughed and told him I thought he needed a lie to work from. Both he and I knew that he was being screwed with.

Yes, I passed.

Re:Polygraphs

[Gibgezr]

From following the top line (Wikipedia) and then clicking on the reference it offers (http://www.nsf.gov/statistics/seind02/c7/c7s5.htm):
"Belief in Alternative Medicine

Alternative medicine is another concern. As used here, alternative medicine refers to all treatments that have not been proven effective using scientific methods. A scientist's view of the situation appeared in a recent book (Park 2000b):

Between homeopathy and herbal therapy lies a bewildering array of untested and unregulated treatments, all labeled alternative by their proponents. Alternative seems to define a culture rather than a field of medicine—a culture that is not scientifically demanding. It is a culture in which ancient traditions are given more weight than biological science, and anecdotes are preferred over clinical trials. Alternative therapies steadfastly resist change, often for centuries or even millennia, unaffected by scientific advances in the understanding of physiology or disease. Incredible explanations invoking modern physics are sometimes offered for how alternative therapies might work, but there seems to be little interest in testing these speculations scientifically.[59]
In response to the 2001 NSF survey, an overwhelming majority (88 percent) agreed that "there are some good ways of treating sickness that medical science does not recognize." (See appendix table 7-58.) The American Medical Association defines alternative medicine as any diagnostic method, treatment, or therapy that is "neither taught widely in U.S. medical schools nor generally available in U.S. hospitals." However, at least 60 percent of U.S. medical schools devote classroom time to the teaching of alternative therapies, generating controversy within the scientific community. Critics have also been quick to note that one of these therapies, "therapeutic touch," was taught at more than 100 colleges and universities in 75 countries before the practice was debunked by a nine-year-old child for a school science project (Rosa 1998)."

Now, as to your claim that acupuncture absolutely works:
Acupuncture works...as a placebo. That's it.

The National Council Against Health Fraud in the US on acupuncture (http://www.ncahf.org/pp/acu.html):

"NCAHF believes:

Acupuncture is an unproven modality of treatment;
Its theory and practice are based on primitive and fanciful concepts of health and disease that bear no relationship to present scientific knowledge;
Research during the past twenty years has failed to demonstrate that acupuncture is effective against any disease;
Perceived effects of acupuncture are probably due to a combination of expectation, suggestion, counter- irritation, operant conditioning, and other psychological mechanisms;
The use of acupuncture should be restricted to appropriate research settings;
Insurance companies should not be required by law to cover acupuncture treatment; and
Licensure of lay acupuncturists should be phased out."

Re:Polygraphs

[Zero__Kelvin]

"Now, as to your claim that acupuncture absolutely works: Acupuncture works...as a placebo. That's it."

That is straight bullshit. I know many, many people who tried it in a situation where they did not believe it would work, but then later had to admit that it did in fact do so. I, myself, tried it with absolutely no expectation of it working. Furthermore, working is working. Saying that it works by the placebo effect is in and of itself an admission that it works. From there, it is academic as to how, and science has always had a problem with things they couldn't yet explain. A scientist from 200 years ago would likely try to hang me as a witch if I showed up with a working radio, for example.

You also evidently don't realize that science is only now beginning to realize that the neural network as mapped out by ancient Yogis thousands of years ago turn out to be extremely accurate.

" As used here, alternative medicine refers to all treatments that have not been proven effective using scientific methods."

You seem to be mistakenly of the belief that this sentence says that science proves that these methods don't work.

"Perceived effects of acupuncture are probably due to a combination of expectation, suggestion, counter- irritation, operant conditioning, and other psychological mechanisms;"

Nothing is quite as scientific as a paper that uses the word probably, and phrases like "NCAHF believes" :-) ROTLMAO.

Our President

[Anonymous Coward]

Is now more immoral and corrupt than his predecessor. That is quite a feat for anyone.

Re:Our President

[fustakrakich]

That will never be proven. And the degree of corruption is hardly important.

Re:Our President

[jsepeta]

Mitt has too much money to be proven evil. But Obama is sure as hell a disappointment.

Re:Our President

[fustakrakich]

But Obama is sure as hell a disappointment.

Only to the naive. He performed exactly as expected by the rest of us.

Bad summary is bad.

[Anonymous Coward]

They aren't arresting people for just teaching the methods. The instructor they arrested had trained two undercover agents posing as criminals that wanted to lie on the exam. One was a drug trafficker and the other a correctional officer that smuggled drugs into prison and received sexual favors from an underage girl. The instructor taught them how to cover up those crimes. Seems pretty simple to me. If you say you want to rob a bank, and I give you a gun to do it I'm criminally liable for it. Why isn't fraud the same? It would be one thing if the instructor didn't know they were criminals, but he did. The summary makes it sound as if they're wantonly arresting people.

Re:Bad summary is bad.

[sqlrob]

Except they don't use polygraphs in criminal investigations, so what's the problem?

Re:Bad summary is bad.

[Anonymous Coward]

It doesn't matter. The law applies in cases where you know they are criminals or believe them to be criminals. What you're saying is that undercover investigations don't work ever. Police put undercovers out on the street for prostitution stings all the time. You're still guilty even though the person was not a prostitute. You thought they were and if they had been you would have done it.

Re:Bad summary is bad.

[ebno-10db]

Yes it sounds like they're going after them for conspiracy rather than simply teaching these techniques, which is the sort of legal technicality beloved of prosecutors, but you're missing the bigger point. This is not analogous to someone selling a gun to a person who says they want to rob a bank; it's analogous to letting someone take your chemistry class even though they say they want to make a bomb to blow open a bank safe. This is stopping the dissemination of information because it could be used for nefarious purposes.

Additionally, the undercover agents said that they already did commit these crimes, not that they were planning on using these techniques to commit crimes in the future. If potentially helping somebody to beat the charges is a crime, then why are defense attorneys legal?

Re:Bad summary is bad.

[Anonymous Coward]

They weren't taking the class to beat the crime. They were taking it to get a job in federal service. That means the crime of fraud is about to be committed. Defense attorneys defend past crimes not future ones. If these agents went to an attorney and said, "I committed these crimes, but I want a job in federal service how do I beat the lie detector by hiding these?" and he helped them he would be guilty of fraud or conspiracy or whatever your jurisdiction says.

People seem to be confusing what's going on here. The agents posed as criminals that wanted to hide their past to receive a benefit. It's fraud, and the instructor taught them for the purpose of committing fraud. This isn't new territory. If I claimed to have a purple heart, I can. I can go on TV and claim it. I can go to veteran functions and claim it. But once I attempt to receive a benefit based upon that fraudulent claim I am guilty of fraud. The agents were trying to lie (hiding criminal infractions that prohibit them from public service) to receive a benefit (employment). It's fraud, and anyone knowingly assisting them to achieve such criminal goals is guilty. If the instructor didn't know about the past then he's clear. But he did.

Re:Bad summary is bad.

[BrokenHalo]

They aren't arresting people for just teaching the methods. The instructor they arrested had trained two undercover agents posing as criminals that wanted to lie on the exam. One was a drug trafficker and the other a correctional officer that smuggled drugs into prison... ...The summary makes it sound as if they're wantonly arresting people.

Those cases are what they're citing as justification (for what they're worth, which isn't much), but you're missing the main point, which was made earlier in the article that you appear to have actually read(!)

I'll remind you:

By attempting to prosecute the instructors, federal officials are adopting a controversial legal stance that sharing such information should be treated as a crime and isn’t protected under the First Amendment in some circumstances.

Re:Bad summary is bad.

[camperdave]

They aren't arresting people for just teaching the methods. The instructor they arrested had trained two undercover agents posing as criminals that wanted to lie on the exam. One was a drug trafficker and the other a correctional officer that smuggled drugs into prison and received sexual favors from an underage girl. The instructor taught them how to cover up those crimes. Seems pretty simple to me. If you say you want to rob a bank, and I give you a gun to do it I'm criminally liable for it. Why isn't fraud the same? It would be one thing if the instructor didn't know they were criminals, but he did. The summary makes it sound as if they're wantonly arresting people.

Thing is... robbing a bank is a crime. Lying on a job interview isn't.

Re:Bad summary is bad.

[Zero__Kelvin]

The problem is that you are confusing what happens in a criminal investigation with what is admissible in a court of law. Plenty of people have, no doubt, been convicted of crimes they didn't commit because they took a polygraph and threw a false positive, at which point the "investigators" stopped looking for the real criminal and started fabricating ... er ... gathering evidence that tends to convict the innocent party and ignoring evidence that might point toward exoneration.

Re:Bad summary is bad.

[Zero__Kelvin]

"; it's analogous to letting someone take your chemistry class even though they say they want to make a bomb to blow open a bank safe."

In the US, this is known as a consiracy, so yes, you are correct. If a person makes clear that they are seeking information from you that will help them commit a crime, and you give them that information, then you are indeed complicit in that crime under US law. For example, if someone asks you where a person will be at a certain time and makes it clear that they want to know so they can kill them, and you tell them, you are as guilty as they are of murder.

Re:Bad summary is bad.

[ISoldat53]

Too bad he didn't use a lie detector on them.

Re:Bad summary is bad.

[StikyPad]

Really? Because the right to counsel is specified in the 6th Amendment, and even that protection only applied in federal courts until Gideon v. Wainwright in the 60s. In other words, because people fought long and hard for the right, much like many of the other rights that we, quite literally, take for granted these days. That's why.

Re:Bad summary is bad.

[tomhath]

The linked article is obviously very biased, as is the slashdot headline. Why this made the front page is a mystery. One quote form the article stands out:

Dixon, 34, also declined to provide specifics on his guilty plea but he said he’d become an instructor because he couldn’t find work as an electrical contractor. During the investigation, his house went into foreclosure. “My wife and I are terrified,” he said. “I stumbled into this. I’m a Little League coach in Indiana. I don’t have any law enforcement background.”

In other words, the guy was committing fraud by charging for this "instruction". He was convicted of fraud. The Big Brother angle is all hype and speculation.

Re:Bad summary is bad.

[cheekyjohnson]

Police put undercovers out on the street for prostitution stings all the time.

The government is corrupt. What else is new?

You thought they were and if they had been you would have done it.

That seems impossible to know.

Re:Bad summary is bad.

[The_Revelation]

If the polygraph were in any way effective, it would have outed the two agents to the instructor the first time they were hooked up to the machine. Seems like a bit of a double-edged case they have walked into. Either (A) The agents were outed early by actual science behind the 'lie detector', thus the Instructors weren't doing anything wrong or (B) No crime was committed because a company was being paid to train someone on the use of a product that doesn't actually do anything. They may as well have been training them on using Astrology to beat the police, or Psychic redirection.

Re:Bad summary is bad.

[infinitelink]

What I find funny is...fundamentally speaking, given that lie detectors are themselves deceptions and frauds, teaching someone how to "beat" the detection/fraud is itself being fake and, therefore, even if they had [pretended--not even knowing they are pretending] to teach someone how to beat these fake tests thinking that those persons would use those techniques to "beat" tests/evaluations, there would be no crime: just faking with no actual harms resulting. So what you've written...bullshit. (No offense.)

But isn't the real point....

[3seas]

...Polygraphs can be beaten and as such are not reliable!
Deniability is man most powerful tool. So really its all about abstraction. What definition do you apply to the questions or do you simply deny the questioner over your own internal thoughts?

The ability of beat a polygraph might actually be a quality the government is looking for....... considering all the lies they have told and certainly spying would find the ability to beat a polygraph an asset.

So you see, its really all null and void this polygraph issue.

Now what more does anyone need to consider in their mental state to beat a polygraph?

Re:But isn't the real point....

[dbIII]

considering all the lies they have told and certainly spying would find the ability to beat a polygraph an asset.

Only if you are spying on a place that puts value in the polygraph and the US government is just about alone in the world there.

20th Century Witchcraft

[rwyoder]

Over the years I've seen 3 investigative reports on TV, and read many articles on the topic. It all comes down to the same thing: The polygraph is just a stage prop in an interrogation, for the purpose of scaring the ignorant into confessing. Here is Penn & Tellers report:

http://www.youtube.com/watch?v=8NLf7XwLpyQ

Re:20th Century Witchcraft

[Zimluura]

and BS was strictly an entertainment show, without any shred of information applicable to reality...that way they can't be sued for libel at least. i guess since they're entertainers that is really all they're ever allowed to do, entertain always, never inform.

Would have had to charge many of my co-workers

[StatureOfLiberty]

I worked in electronics sales in the early 80s. In San Antonio, TX at the time you had to take a polygraph to work almost anywhere (for example, Radio Shack was one). As soon as I was hired in most places, my new co-workers started telling me how to beat the polygraph. (I had no reason to worry, but they told me anyway). In the end I found out that many of these folks were robbing the employer blind. And all had passed a polygraph.

Of course, your ability to beat the polygraph probably has a lot to do with who was administering the test. Since so manyl employers back then required polygraphs, you ended up with a bunch of 'Polygraph Marts' who had people administering the tests who really weren't qualified to do so.

Streisand Effect, anyone?

[CaptainOfSpray]

Let's spread the news of how to beat polygraphs as widely as possible. Now we have the government banning it, that makes it desirable knowledge, OK?

From TFA: "Charles Honts, a psychology professor at Boise State University, said laboratory studies he’d conducted showed that countermeasures could be taught in one-on-one sessions to about 25 percent of the people who were tested. Polygraphers have no reliable way to detect someone who’s using the techniques, he said. In fact, he concluded that a significant number of people are wrongfully accused."
Mirror these sites and anything else you feel relevant
http://www.wikihow.com/Cheat-a-Polygraph-Test-(Lie-Detector)
https://antipolygraph.org/articles/article-034.shtml

Re:Streisand Effect, anyone?

[jsepeta]

I'm just waiting for Obama to ban books and for Mitch McConnell to rally up the Tea Party to create a group of Firemen.
http://www.sparknotes.com/lit/451/summary.html

Re:Obstructing an agency proceeding and wire fraud

[serviscope_minor]

Wire fraud is a great crime.

Basically it captures any thing in the internet which even might involve money at some point, e.g. fines for copyright infringement, payment for services. And has huge maximum terms. So since almost everything involves money at some point and many things happen over the internet it allows them to add almost arbitrarily long sentances to something that would otherwise get almost nothing.

Basically the perfect legislation as far as they are concerned.

Polygraphs don't work. They should use e-meters.

[Anonymous Coward]

Free gullibility test

So

[Impy the Impiuos Imp]

> Investigators confiscated business records from the two men, which included the names of as many as
> 5,000 people who'd sought polygraph-beating advice.

Which was, of course, the real goal. Much like seizing the records of companies that sell hydroponics equipment.

So what has this incident taught these instructors, whether they be good or evil?

1. Cash-only and don't use records.
2. If someone says they want to do evil, give them their money back and kick them from the class. Otherwise, don't ask, don't tell.

Feds Instructors of Polygraph-Beating Methods

[drinkypoo]

Is what I read. As in, teaching how to get the desired polygraph results from a suspect through beating

Re:Seems to me...

[ebno-10db]

By your reasoning defense attorneys are also guilty of "knowingly [teaching] them to hide that information".

mike judge

[Joe_Dragon]

http://vimeo.com/50020343

I hope they succeed

[houghi]

We can not have foreigners spying on our people. These jobs should not go to foreigners. They belong to Americans. They should spy on our people.

(Uh wait!)

USA! USA! USA!

(Phew, that was close.)

If there's evidence of efficency

[goldcd]

Then it wouldn't be alternate, would it?

Re:If there's evidence of efficency

[Zero__Kelvin]

Yes. It would. You are leaving out the profit part of the equation. For example Licorice Root works as a decent antibiotic. Guess how much money Big Pharm makes if you use Licorice Root, as I have successfully done on multiple occasions.

Re:Seems to me...

[Anonymous Coward]

No. First, defendants normally don't take the stand. Second, if they do, the attorney cannot tell them to lie (like this instructor did). If a defense attorney tells them to lie he's guilty of suborning perjury. They can spin facts and frame things but they cannot tell their client to lie. If you were at the location in question an attorney cannot tell his client to say you weren't there. The instructors in this case said "lie." Defense attorneys don't tell their client to lie, what they normally do is say, "tell me what happened." Which just means "lie to me then I'll try and make it work even though it won't because what you're about to say won't make any sense and by the way you're not testifying." But again, even though they might think their client is lying they don't know for sure. If the client said no I really wasn't at home I was there the attorney can't say we'll go with you not being there. They can't. It doesn't mean they don't, but the law says they can't. But just because defense attorneys get away with it doesn't mean everyone should.

Meta-Crime?

[Pope Raymond Lama]

The TFS gives away the "criminal" practices - "polygraph-beating techniques, which are said to include controlled breathing, muscle tensing, tongue biting and mental arithmetic. " - so now they will come after /. as well... :-)

And maybe, commenters who quote TFS...

Fortunately when the sit me down for interrogation, now I know all that is needed is byte the tongue for not giving away the ID numbers of my fellow /.ers; So, don't worry!

Hitchhiker's Guide to Becoming a Police State

[FuzzNugget]

Step 1: claim to champion freedom of speech, but oppress it when is inconvenient for the establishment.

Are they going to go after that episode of P&T's Bullshit where they say you can beat the box by clenching your ass?

Free Speech

[nurb432]

It's quite clear it no longer applies here. Unless your speech is 'state approved', better watch your back.

What crime?

[MobyDisk]

Teaching someone how to beat a polygraph is not a crime.

Liars to fedgov ARE criminal

[Etherwalk]

But they were not criminals, the lied to the instructor, so the instuctor was training liars not criminals.

Lies to fedgov are not protected by the first amendment, and fedgov makes job applicants waive their rights anyway. It is a crime to lie on a security clearance application, and a crime to lie to a federal agent. Helping someone lie to a federal agent is therefore also a crime.

Re:Liars to fedgov ARE criminal

[mbone]

Note, however, that is it not a crime for a federal agent to lie to you. Symmetry does not apply.

Job interview

[Etherwalk]

It depends on the job interview. If the person interviewing you is a federal agent, it's a crime.

Re:Job interview

[dbIII]

How about scamming the government by selling a service that pretends to detect lies but doesn't? Shouldn't that be a crime?
Obviously it isn't but these scammers are consuming a lot of taxpayers money and providing nothing other than a dangerous false sense of security.

Next they'll issue a warrant for

[mark_reh]

George Costanza: "It isn't a lie if you believe it"

Political trials

[mbone]

I don't know what else you can call this. Note that (according to McClatchy) they are not charging that instructing people how to beat a polygraph is a crime (as far as I know it isn't), they are targeting people who instruct this with whatever random crime they can come up with, and probably using entrapment to do it :

In the last year, authorities have launched stings targeting Doug Williams, a former Oklahoma City police polygrapher, and Chad Dixon, an Indiana man who’s said to have been inspired by Williams’ book on the techniques, people who are familiar with the investigation told McClatchy. Dixon has pleaded guilty to federal charges of obstructing an agency proceeding and wire fraud.

That this sort of gross misuse of the prosecutorial power is a danger to freedom hardly needs to be said.

Re:Teaching someone to beat pseudoscience?

[Nadaka]

Yes, that is the point. Do anything that sheds light on the incompetence, illegality or malevolence of the us intelligence industry makes you an enemy of the state.

Re:So when are they going to target...

[Zero__Kelvin]

In Soviet USA, Church of Scientology targets them!

Re:Seems to me...

[mbone]

The difference is that the US Government does not generally sent undercover agents to pretend to be defendants and entrap defense attorneys. Although, given what they are doing here, I would expect to see that soon enough.

Re:Seems to me...

[amorsen]

That makes no sense. Defendants cannot commit perjury, they have the right to lie. Therefore a defence attorney cannot be guilty of subornation of perjury by telling the defendant to lie.

really? REALLY?

[slashmydots]

Cheating at a pretend test? Now THAT is low. Seriously, the only thing protecting us from terrorists and sociopaths in the FBI or whatever is a polygraph? We're fucked.

I will just leave this here to anger my fellow /.

[wbr1]

www.antipolygraph.org

Re:Every psychopath will pass the polygraph

[PPH]

This explains the NSA.

quick lesson

[stenvar]

A president who stands for the rule of law and liberties: "Polygraph tests are unreliable and have little scientific data to back them up; I am immediately ending their use by government by executive order and working towards making them illegal as part of job applications."

A totalitarian-leaning president with a disregard for the rule of law and the Constitution: "Let's prosecute people who teach others to get around our unreliable and unproven interrogation tactics."

It's clear what kind of president we have. Guys, don't elect such a loser and liar again. At least by 2012 it should have been clear to everybody what kind of president he was.

Clench your anus

[Karmashock]

That's how you beat the polygraph. Thank you and good night

Polygraphs are voodoo.

[jcr]

There is no such thing as a "lie detector".

-jcr

Re:Specificity and Sensitivity

[Immerman]

Indeed. It a shame more practicing doctors don't understand the math better, or explain . Perhaps it would improve things if all lab results came with a probability breakdown instead of just "positive/negative"

Instead of "You tested positive for slashdotitis" it would read "Test positive: independent probability of slashdotitis is 57.3%, follow-up test recommended", or "Test negative: independent probability of slashdotitis 1.4%". Of course then you'd also have to teach doctors how to combine independent probabilities for an accurate diagnosis and good luck with that. I suppose it would be easy enough to write an app for it though - you don't need to know how to do the math, just that it's too complicated for you to do in your head.

They should prosecute for fraud

[davidwr]

* Prosecute those who teach you to "beat lie detectors," giving a pass only to those who have a 100% success rate or who advertise a success rate lower than their actual success rate.

* Prosecute those who sell or market polygraphs as having a success rate higher than they actually do, those who materially misrepresent the tool's reliability in a given situation, or those who, by omission, imply it has a given reliability in situations where its reliability is lower.

Oh, and the article is not about arresting those who, in general, teach how to beat the system but about arresting those who knowingly teach people who have said "I need to lie in this specific situation and get away with it," or something close to that. It's the moral equivalent of prosecuting a pharmacist on conspiracy charges for selling a box of behind-the-counter cold medicine to someone who walks in and says "I need some pseudo ephedrine so I can make some meth."

If it can't be used as evidence in a courtroom

[Anonymous Coward]

Then why on Earth do we even bother using lie detectors at all ? The bigger question being how can teaching someone to obscure their answers even be a crime at all if the GD courts don't consider it reliable enough to use at all ?

Re:If it can't be used as evidence in a courtroom

[tragedy]

Then why on Earth do we even bother using lie detectors at all?

As others have pointed it, they can be used as a scam to interrogate people without their lawyer present. They can also be used for other types of scams. Like police dogs which will signal the presence of drugs at their handlers command rather than when they actually scent drugs (or perhaps it's more that they avoid signalling that they scent drugs until they get their handlers commands because their noses are so sensitive they smell drugs on just about everyone from the money in their pockets and just from walking around on the streets) in order to manufacture probable cause. Also as a scam to force a confession like the traditional, "your partner ratted you out, sign this confession and you'll walk in a couple of years, otherwise we'll see to it that you get the chair!". There may be other reasons as well, but one of the most important things to remember is that fake "bomb detectors" which were actually re-branded "golfball detectors" (which didn't work to find golfballs either) were bought in large quantities by law enforcement agencies and, even after the scam was revealed, some of those agencies stubbornly continued to use them. Draw from that what you will.

Re:Seems to me...

[Mr. Slippery]

Defendants cannot commit perjury, they have the right to lie.

I'm afraid you are confused. While rarely charged, defendants in criminal cases swear or affirm to speak the truth when testifying and are subject to charges of perjury.

Wasn't that Bernstein, not Schneier?

[Beryllium Sphere(tm)]

At least djb is the one who had a lawsuit about it.

Re:And?

[EmagGeek]

They are also used to screen applicants for a security clearance. Yep, you heard that right. Our government uses debunked, junk science to determine who to trust with our most vital state secrets.

Re:Seems to me...

[tqk]

Defendants cannot commit perjury, they have the right to lie.

I believe you're mistaken. There's plenty of things like interfering with an investigation or the course of justice that they can charge you with for lying. You (generally) have the right to remain silent (in the USA). Use it.

Re:Seems to me...

[KGIII]

If your lawyer asks you if you committed the crime then fire them. Go on, ask me how I know this... (No, don't really ask. It's a good story but too long to type and you wouldn't actually read it anyhow.)

Re:Seems to me...

[KGIII]

You might want to review this one with a lawyer. Really... Your lawyer can't tell you to lie. Your lawyer can't even KNOW you're going to lie or that you are lying. Again, if your lawyer asks if you're guilty of the crime (they won't) fire them.

Re:Seems to me...

[amorsen]

Indeed, it appears I am mistaken. The US "justice" system is even crazier than I thought.

Re:Seems to me...

[amorsen]

Yes I'm wrong. I believed that such a basic right would be recognized in a democracy, but it is not.

What's alternative about that?

[goldcd]

http://en.wikipedia.org/wiki/Liquorice#Medicine

That's just medicine.

I could still always go and chew the bark off a willow tree when my head hurts, but I prefer to give big pharma their dollar to get my hands on those more convenient aspirin tablets.

Re:What's alternative about that?

[Zero__Kelvin]

A little slow today are we? From your linked article: The compound glycyrrhizic acid, found in liquorice, is now routinely used throughout Japan. Again, as with aspirin, they extract a compound and package it for sale. They aren't handing out Licorice Root and billing an insurance company for it.

Polygraphs

[jcochran]

I had a job some time back and was required to take a polygraph exam. During the exam, I got a rather strong feeling that a 'game' was being played and that no one informed me as to the rules. The results were inconclusive and I was scheduled for a re-examination. Given my feelings on the subject, I decided to 'learn the rules' and research polygraphy before my reexamination. Learned quite a bit on the subject, one element of which was that there was a classified government study on the effectiveness of polygraphy. I never saw the contents of that study, but if I were a classification authority and if the study reflected what's available in the public literature, then I too would classify the study. The reason is simple since the public literature on polygraphy summed up as follows.

As a means of determining lies from the truth, polygraphy is totally useless. However, as a means of eliciting voluntary confessions from naive subjects, it is extremely effective.

Let's just say on the follow up exam, I enjoyed myself far more than the examiner.

Lie detectors don't work ..

[dgharmon]

"Federal agents have launched a criminal investigation of instructors who claim they can teach job applicants how to pass lie detector tests"

Lie detectors don't work, all it does is give a pretext for the testor to claim you lied. If you believe that the machine and tester can detect lies then you are more likly to tell the truth or cop to lying. Lie detector machines are pseudoscience at its worst.
-------

The Ontario Skeptic, Volume 16, Number 3 (Fall 2003) pp.1, 6.

'Prof. Furedy disputes the value of this procedure, known as the Control Question Test (CQT).

"It is not a test at all in the sense that, say, an IQ test is a test," he says.

The validity of IQ tests in determining intelligence may be controversial, but at least they are scientifically based and use standardized procedures, so the results found by one competent operator will be the same as those found by any other, says Furedy.

However, the so-called control questions of the CQT are designed by the individual examiner, based on discussions with the subject, and the entire examination can vary greatly in length and subject matter. Much of the procedure is often spent not trying to determine whether the subject is telling the truth but trying to elicit a confession of guilt. As a result it cannot be called a scientific or standardized test.

Even when administered by an "expert", the polygraph fails to distinguish between an anxious-but innocent person and an anxious-but guilty person, says Furedy'.

Can we believe the prosecutors?

[dgharmon]

> The instructor they arrested had trained two undercover agents posing as criminals that wanted to lie on the exam ..

"Documents in Dixon's case are filed under seal in federal court, and prosecutors didn't return calls seeking comment".

Lying or Deception?

[almechist]

OK, I get that they aren’t claiming the teaching of the technique itself is illegal (not yet, anyway, it sounds like they’d like it to be, so that may be next on their agenda), what they’re actually claiming is that it’s criminal fraud to knowingly help someone lie to the government. I will take their word that lying to the feds on an application or during a job interview is indeed a criminal offense, but even granting that morally dubious proposition, is beating a notoriously ineffective and scientifically unsound technology really the same thing as telling an outright lie? It seems to me that equating the two would be a pretty tough sell in a courtroom where the technology itself is banned because it’s so unreliable. Wouldn’t what was proposed by the undercover operatives in this case more accurately be called deception rather than lying? There is a subtle difference between deception and telling an outright lie. And if we’re really going to make deception a crime, where does it end? What about prevarication? Misdirection? Simple undecided-ness? This is a dangerous slope we are treading with this type of prosecution. Hopefully it will eventually be declared an unconstitutional overreach by prosecutors.

Re:Seems to me...

[zippthorne]

They have a right to say nothing. They can even state the truth in a self-serving way. But if they lie, they are committing perjury.

Junk Science

[Anonymous Coward]

If the polygraph really worked, then you couldn't "beat it". But it doesn't work. It is junk science.

The polygraph measures proxies for anxiety. They can tell that you are nervous or anxious. But they can't tell why. They can't tell whether it is because your career and future livelihood depend upon what the examiner thinks of you, or because you are a real spy.

When this is all over, the current US government is going to be just as much of a laughing-stock as the Nazis were, and as the East German Stazi were. All use of polygraphs is a waste of taxpayer money -- especially since we now know that the NSA has far better records on all of us, having violated the 4th Amendment with willful malice.

Why not just use all those records that the NSA has already collected, and do away with this polygraph nonsense?

Better yet, lets prosecute the NSA and its employees for violating the highest law of the land.

But I want spies in my government

[Anonymous Coward]

is aimed at discouraging criminals and spies from infiltrating the U.S. government

I want spies to infiltrate the U.S. government. That's the only way we can learn just what the fuck is going on.

e-meter /= polygraph

[globaljustin]

e-meters work the same way.

right...I thought e-meters only measured skin conductivity

either way, I don't understand why this fact isn't reported in the discussion more...

the existence of the e-meter proves that polygraphs dont work and people can train themselves to control the phsysiological variables

you'd think critics of the polygraph would bring it up more often

Just abandon the silly toy

[dbIII]

Face it, J. Edgar Hoover was known for taking kickbacks and the toy doesn't work. It's been years and it still doesn't work. Give up on it.

This is highly cynical

[WOOFYGOOFY]

1) because these same people know that there are technologies on the very very near horizon which are much better than current lie detectors which is little more than a GSR reader.

2) because of 1), it's merely a stalking horse for general the desire to enshrine into law the sequestration of general knowledge or facts about the general world they wish people didn't know.

3) establishing 2) above would lead to contempt and ,disrespect for the government, the widespread perception of illegitimacy of the government by the governed. This is THE ONLY way terrorists can actually destroy this nation. Even a biological attack isn't going to make the nation actually end. Dissolution and disunion will be self-inflicted.

We didn't evolve to accept intellectual feudalism or some kind of knowledge Forest Law in any form.

Re:Seems to me...

[Cederic]

erm. Why?

The justice system seeks to establish the truthful facts. Allowing someone to lie jeopardises that goal.

In the UK people have gone to prison for perjury due to their statements in their defence in trials for other things. This is good.

Pucker factor

[CHIT2ME]

I received training in beating polygraphs in the military. I would not use it now because I don't have anything to hide. But, the truth is they are easy to beat, as long as you don't "cave" when the they tell you afterwards that you failed. Polygraphs should be seen for what they are....."pseudoscience"!!! They don't work, are easy to beat, and with so many false positives they can destroy the lives of innocent people. I'm all for a ban on polygraphs!