Fool proof 100% accurate lie detector

[bengangmo]

So a totally accurate, fool proof lie detector is invented. It's non ivasive, and doesn't need any sort of trained monitoring at all - giving a simple yes he's lying type response.
Its cheap enough and portable enough to give one to every Police Officer.

1) How and where should it be used in law enforcement?
2) Would you need any changes to "probable cause" laws to use it.
3) What changes to current laws would you like to make to take best advantage of such a device.

Imagine a few scenarios.

a) Traffic stops - do you have any drugs in the car? Could this question be asked even absent probably cause, with teh results of the lie detector then used to justify a search?
b) How about totally unmitigated fishing epeditions. You know that somebody in an office of 300 people have been stealing stuff. Can you question all 300 with the device?

What other sorts of changes would it bring about?

[Khampelf]

Such a device is so prone to abuse of civil liberties that the prototype should be destroyed, and the inventor incarcerated indefinitely, however viciously ironic that would be.

How do you objectively determine "truth", anyway? I know it's a given of the hypothetical, but still. How does it work on sociopaths who have no concept of truth, or Alzheimer victims who flat don't remember?

[Muskrat Love]

Such a device is so prone to abuse of civil liberties that the prototype should be destroyed, and the inventor incarcerated indefinitely, however viciously ironic that would be.

How do you objectively determine "truth", anyway? I know it's a given of the hypothetical, but still. How does it work on sociopaths who have no concept of truth, or Alzheimer victims who flat don't remember?

Sociopaths know what the truth is, we just don't see anything wrong in lying.

[Moon Dog]

It may be able to tell you are lying. But can it tell you what they are lying about or why. Sometimes the lie isn't just a direct opposite response to a question but a diversion away from some minor point or subject you don't wish revealed.

[Uthrecht]

But can it tell you what they are lying about or why.

Sure, but that's what the waterboarding is for.

[Moon Dog]

Sure, but that's what the waterboarding is for.

D'oh :smack:

I mean you can do it over 100 times in a month to the same person and it's still not torture is it ?*

*Blatant political comment sidetracking from main question which should be ignored.

[Uthrecht]

Anyways, to answer the OP. I'm torn on it. I mean, the police are given training on recognizing an evasive or lying suspect. Is there anything blatantly *wrong* with a tool that doesn't come into contact with the subject and has a highly accurate success rate? Perhaps it could be likened to a breathalyzer. I believe that most states allow an officer to ask/require a test if they have suspicion that the person is intoxicated. It seems that if the officer suspects the person (and with both the breathalyzer and the lie detector, that could be part of the rule set), they could have it in use during an interrogation. Rather than be allowed to just walk around with it on and ask questions.

[Khampelf]

Sociopaths know what the truth is, we just don't see anything wrong in lying.

I bow to your experience.

[Moon Dog]

Would you need any changes to "probable cause" laws to use it.
?

Of course. How else are you going to drag them away for thought crime citizen ?

[shaggy]

In a world full of gray lines, it would be excruciatingly difficult to establish boundaries. My shoot-from-the-hip initial reaction is to keep it restricted to personal harm cases (murder, rape, etc), politicians, and let parents have a version set only to detect their own kids. If I had to pick just one of those three... I would want one for my kids:D

[Muskrat Love]

Logically, I think they'd probably be legally restricted. I don't think we'd see police being allowed to use them at traffic stops or in interrogations without the person's lawyer being present. We might see the information being inadmissable in court - useful to confirm you have the right person, but you'd still have to build a case. In that case, I could see some people not wanting to use it for legal reasons. When I was a P.I., we had access to criminal records going back decades, but most of our clients in California did not want us to report anything to them that was more than 7 years old, because their state laws prohibited them from using that information in hiring decisions. In one case a client was very angry because we accidentally reported a rape conviction from 10 years ago, they still had to hire the person because they had no legal reason to deny them employment. If there are restrictions on the admissability of lie detector evidence, some people won't want to use it for these reasons. Prosecutors might decide that it hurts their case if they knew someone did something, because it would affect and color their evidence gathering.

[bengangmo]

I would think it would be invaluable for establishing facts:
1) Where were you on Saturday night?
2) Were you speeding?
3) Did she say no?

The correct protocols would of course need to be in place to avoid "double questions" (or two issues being present in one question)

[Algorithm]

So are we assuming the 5th amendment (and likely 6th) is repealed?

[Uthrecht]

Why wouldn't an individual still be allowed to not answer questions under this scenario? The upside is that if an officer asks someone a question right now, they have to rely on their own interpretation of the the person's response to decide what's what - this would tell them if the suspect is lying. The downside is that it wouldn't be able to tell them why they're lying, or potentially exactly what part of the question they're lying about. So the officer would still have to rely on judgement. The further downside is, having a piece of equipment that is accurate in some limited respects could lead to misinterpretations, which is why its use would have to be regulated, and I would think recorded.

[bengangmo]

So are we assuming the 5th amendment (and likely 6th) is repealed?

5th is self-incrimination right? (not so sure about 6th). But yeah, that is what I am trying to ask. If we could know for certainty the veracity of any statement, why shouldn't somebody that has "done wrong" be called to account for it?

As I understand it, many of the rules are in place as a protection against some form of abuse.

If we always know the incontrovertible proof, then isn't the requirenment of protection from abuse no longer relevant?

[Algorithm]

5th is self-incrimination right? (not so sure about 6th). But yeah, that is what I am trying to ask. If we could know for certainty the veracity of any statement, why shouldn't somebody that has "done wrong" be called to account for it?

As I understand it, many of the rules are in place as a protection against some form of abuse.

If we always know the incontrovertible proof, then isn't the requirenment of protection from abuse no longer relevant?

Right to counsel. If police officers are allowed access to such a tool, can compel any civilian to answer a question to their satisfaction with no probable cause nor reasonable suspicion a crime has occurred, and the evidence gathered necessarily holds up in court, right to counsel is rendered irrelevant.

Generally, I see the potential for abuse to be high. A series of questions, answered truthfully, can still make an innocent person seem guilty, or vice versa. Even if the entire population had faith in the device's effectiveness, you have to assume a non-exploitable system could be set up to control access and use of the device in court and society at large for it to be fair and useful, and that's far too broad an assumption for me to make. It's hard to have a meaningful discussion when the proposition relies on an assumption so far outside the realm of possibility.

[hajario]

This is a very interesting thread and I think it will do even better in PPR.

[Moon Dog]

Possibly a hijack from the initial intent but how do your prove your lie detector is 100% accurate ? Is it even possible to prove 100% accuracy ?

[Muskrat Love]

This could possibly hurt the police's ability to gather information from suspects. Criminals often talk to the police because they think they can deceive them or at least avoid some of the blame of what they did. If it was common knowledge that any lies told to the police would be detected, nobody would ever talk to them.

[Acid Lamp]

Possibly a hijack from the initial intent but how do your prove your lie detector is 100% accurate ? Is it even possible to prove 100% accuracy ?

Probably irrelevant. We could assume the same level of accuracy as DNA evidence, IE it might be wrong, but it's millions to one odds it is.

The device would be heavily restricted in its uses. Most likely only in cases with a murder, rape, national security, or public safety emergency pending. I also see this tool as being quite useful during the first critical hours of an investigation into a missing person. The usual suspects could be eliminated quickly, thus allocating valuable resources to where they ought to be.

[Iris]

I see it being even more restricted, at least as it's first introduced. As in: only allowed when the subject has agreed, even if there's probable cause.

On the other hand, it would clear out a whole bunch of court cases quickly when they did agree.

"Did you do it?'
"No."
"The court finds the defendant not guilty. Next!"

[bengangmo]

I would wonder how far down the "criminal" scale it should go...

It would be great to see such a thing used to establish the facts in the case of a murder, assault, theft etc. But what about speeding, or littering?

Would you be ok with such a thing if it was used like:
1) Did you ever litter?
2) Have you ever been speeding?
3) Jaywalking?

It also has really good potential for traffic accidents -
"what were you doing immediately prior to the accident?"

I don't see the idea of people having to testify against themselves as being too much of an issue - if you have committed a crime* then I don't see that neccessarily the onus is on the prosecution to prove it if there are other ways we can incontrovertibly establish the facts with input from the defendant. (note: I understand why the law is constructed that way, and think it is for the "best" given current situations)

[Sleel]

No fishing expeditions, same as the rules we've got now for testimony and evidence, no change there. So bengangmo's questions wouldn't be allowed.

Under questioning, I'd expect the questions allowed to be very closely circumscribed to a greater extent than current polygraph tests. Only questions relevant to the exact crime being investigated would be allowed.

Not used unless you have probable cause to suspect them of a crime. There's no reason to believe people would completely give up their rights to privacy just because of this invention, and our justice system has a long history of preserving these kinds of limitations on questioning and searches.

I would expect it to become a positive right, that you can ask to take the test to clear you of a crime, but not be forced to take it to implicate yourself. Unless you can push through an amendment to repeal or abrogate the 5th, you're not going to be able to (legally) force someone to submit to the Truth Detector unless they've waived the right by current or previous agreement.

Most innocent people would readily elect to take the fast way out, as long as the accuracy rate is actually 100% or close enough that a person feels that it can only help and not hurt him.

Police testimony should be subject to challenge under the Truth Detector. That would be one way to prevent abusive use of the machine to force testimony; make the street go both ways.

I can barely imagine the kind of security and confirmation I'd want in making sure no one tampers with one of the damn things. For starters, I'd have detectors selected at random, with neither the questioner or interviewee aware of exactly which machine was used until after the session was logged. The selection algorithm would have to be logged, audited, and analyzed for true random selection. Dual-machine use in every trial, or repeated trials with different machines would be one way to minimize the chance of one screwing up and giving a false positive.

Everyone in the chain of custody for every single machine would have to be questioned every single time a trial was performed, and those people's testimonies would have to be logged and subjected to the same kind of double-checking. Let's face it, if there's something that is actually that close to infallible, someone will try to substitute for it or suborn it. The only way to make sure that the machine is working as advertised would be to ensure the integrity of the machine.

[BJMoose]

Count me with those who see a huge self-incrimination problem with such a device. Of course, if such a device were practical and in common use, I believe most folks quickly would learn to keep their mouths shut!


Forgive me a digression, but this thread happens to come up at an interesting time for me. For the past year or two I've been slogging my way through the volumes of hearings and depositions published by the Warren Commission, and as it happens, right now I'm reading through the stuff on the polygraph examination that Jack Ruby insisted on being given. Interesting stuff. And lemme tell you, polygraph exams are a good deal move involved (and tedious) than the novels and TV shows ever let on. Considerable time is spent going over the questions with the subject so that it is certain that the person being examined thoroughly understands the questions so as to avoid ambiguity when the actual exam is conducted. (During the exam one is supposed to limit his answers to Yes or No - the questions are carefully crafted to make this possible. I'm making just an educated guess here, but I suspect the problem with verbosity is that it tends to mask the physiological tics that the machine is measuring and that the examiner considers when he makes his judgment as to the veracity of the subject's responses.) There was one rather funny moment when Ruby himself, of all people, spotted a huge ambiguity in the wording of one proposed question. While the "expected" answer was No, the question was worded in such a way that the only "honest" response was Yes. The question was promptly reworded.

So I suspect this would be a problem with a "perfect" lie detector. You must occasionally get false positives simply because the suspect and the machine are "not on the same page."