Is there anything that forbids administering lie detector tests on criminal defendants?

Does self-incrimination kick in?

Google returned nothing.

Polygraph results are inadmissible as evidence in most US courts, so it would be a waste of effort.

Polygraph lie detectors were ruled inadmissible in court in Frye vs. United States in 1923 (if I googled correctly). Later rulings have allowed polygraph results to be admissible under certain circumstances.

Polygraph tests can still be administered by law enforcement officials, but the person taking the test has to consent to it, and even then the results can’t be used in court.

A hypothetical example is Mrs. X ends up dead. The police obviously suspect Mr. X since most of the time the husband is the killer. Mr. X. claims he is innocent. He voluntarily takes a lie detector test. If he passes it, that may be enough (along with whatever other evidence, or lack thereof, is present) to convince law enforcement that he didn’t do it, and they’ll pass him by and start focusing on finding other suspects. If he fails, though, that failure can’t be used in court against him.

IIRC, even the fact someone refused a polygraph cannot be used against them in court.

You cannot force someone to answer any question anyway, in court or in interrogation. There’s no crime in being totally silent during police interrogation, you are not required to cooperate; they must allow your lawyer to be present. Zipping it shut obviously won’t help your case, but if they’ve already hauled you in and pointed the big light at you, it’s long past whether you are helping your case. They have arrested you, they think they have enough evidence, if what you say helps you get off, good. If all it does is reinforce the ambiguity of your situation, or worse yet, gives them leads to follow up that may also hurt you - why talk?

If they can’t make you talk, and they can’t even use the results of a polygraph - the only logical use as mentioned above is when someone volunteers because they want to prove their innocence (or think they can beat it). he other use is to offer it to a suspect to see what he says.

The only other problem is that interpreting the tests is a matter of art. Like ink blots, it’s possible that interpreters see what the want to see; after all, you’re supposedly measuring stress changes on a person who is likely in the most stressful situation of his life, and has been for weeks or months, and assuming what you see is directly related to their current answers.

Not completely true. New Mexico allows polygraph tests to be admitted without the prior stipulation of the parties and without significant restriction. Most other states have banned it, as have federal an military courts.

But even in New Mexico, an accused can’t be compelled to submit to a polygraph exam.

The most recent SCOTUS case on the subject was US v Scheffer, with a defendant arguing that he should be allowed to introduce polygraph evidence as part of his own defense. The court ruled against him, and Justice Thomas wrote “A fundamental premise of our criminal trial system is that the jury is the lie detector.” His opinion acknowledged that polygraph exams are junk science and don’t have any value as evidence.

The first problem is that lie detectors can and have been beaten, so a negative doesn’t really prove anything; since they also measure stress, and people being interrogated by the police will naturally be stressed, there’s a high chance for false positives.

In addition, a normal police interrogation has a shockingly high false positive rate, too:

Basically, lie detectors just measure how badly stressed out you are at the moment. The presumption is that any sudden spike in stress levels is due to you lying, but it’s just that: a presumption. The machine does not and cannot know why you are stressed out. That part is determined by the polygraph technician, who is just looking at a series of lines on a graph and attempting to interpret a meaning. The comparison to ink blots is a valid one.

The charitable view of polygraphs is that they are not dependable. The uncharitable view is that it’s junk science.

This is the problem with the polygraph and most other ‘lie detector’ devices. They don’t have green and red lights to indicate whether a response is true or false. The results are examined, and then the examiner makes a decision that is simply a guess. Different examiners can arrive at mutually exclusive opinions based on the same results. That’s because there is no reliable, consistent link between telling a lie and a particular physical reaction.

The entire polygraph business is a scam. This paperconcentrates on the legal applicability of polygraph results to the law, but within it decribes the difficulties in even establishing a means of producing scientific double-blind studies that could settle questions of its efficacy.

Polygraphs do not measure stress. Polygraphs measure heart rate, electrical skin resistance and a few other things. This is then used in a non systematic way to indicate stress level which again is interpenetrated in a non systematic way to indicate that a person is telling the truth or not.

Also, since the measure physical reactions that normally indicate stress, people who don’t feel stress about their crimes tend to pass them.

So people who don’t think what they did was wrong – psychopaths, religious cult killers or fanatical bombers, etc. – may pass a lie detector test easily.

This might be of interest: Polygraph - Wikipedia

Another link of note: “How accurate are lie-detector tests?”, by Cecil Adams.

Here’s a hypothetical, I think I saw this once in a film: It was the prosecution which had administered a lie detector test to an arrested suspect, which was of course inadmissible at the trial.

The DA had one of the police officers on the stand and in questioning asked him if he had administered a lie detector test to the defendant, whereupon the witness of course answered in the affirmative. The next question by the DA to the police officer was, “…and after administering the lie detector test, did your opinion change as to the guilt or innocence of the defendant?”

Pandemonium breaks out…

Is this a possible scenario, in which the actual results of the test are not entered as evidence, but somehow are communicated?

Nope.

Basically, it would probably earn the attorney contempt of court charges, possible disbarrment or at least disciplinary action from the state bar. The trial would pretty much end right there with a mistrial.

If something is inadmissible in court, it does not matter how you get around to admitting it - you have broken the rules. The jury has been tainted and cannot possibly render a verdict impartially ignoring that critical detail. A lawyer who does not know that deserves to explore other avenues of work. When a lie detector test is “not admissible”, that does not mean you can’t present the little squiggly marks and the guy who “reads” those tea leaves. It also means you cannot mention or hint at any test taken, however roundabout or obscure the mention, and you cannot ask someone to comment on something that is not admissible.

Of course, it cuts both ways. The defense cannot introduce a positive test, by the DA or by their own trained seal, and say “see, my client’s innocent.”

Ah, law in Hollywood. Such a flexible thing… Where’s Matlock when you need him?

Also, the defense should have been on his feet with “Objection” the moment the wors “lie detector” came out. The judge should not have even waited for the objection.

Regardless, it’s a tainted jury and mistrial. You’re on the jury, the guy’s been charged, and now it seems from the question the DA gave the defendant a lie detector test and is asking about it and he’s still on trial. What inference do you draw from that?