Plague and polygraph

Following the Crooked Timber seminar on The Republican War on Science I heard from John Mangels, science writer for the Cleveland Plain Dealer, who pointed me to this series of reports (free registration required) on Dr Thomas Butler, an infectious disease researcher who (apparently mistakenly) reported missing 30 vials of plague bacteria, and ended up being railroaded into prison by an FBI determined to get a conviction even after it became apparent that the events they were supposedly investigating had never occurred.

It’s an amazing story, which as Mangels says is a metaphor for the clash between science and the Bush administration, and between fear and reason in the post-9/11 world. Much of is the kind of thing that can happen anywhere once the wheels of criminal investigation are set turning.

I was struck, though, by one particularly American feature of the story – the crucial role of the polygraph or “lie detector”. This method is (literally) a piece of witchdoctor magic, tricked out with enough electronic gadgetry to impress the class of believers in technology, as opposed to science, we discussed in the seminar. This group plays a much bigger role in the US than elsewhere, which may be why the polygraph is taken seriously only in the US.

16 thoughts on “Plague and polygraph

  1. I’ve just been reading Titan by Stephen Baxter (1997). Its set ten years later. Its the most prescient SF I’ve read about our current decade.

    One of the baddies is a shutddown science and tech fundamentalist president and another is a USAF high ranking China hating dude who quotes Trotsky.

    Given recent events in Iraq this linkage between violence-loving (world revolution) Trots who ‘convert’ to violence-loving(if-it-moves-invade!) is way ahead of its time. Baxter describes accurately the intoxication of power and fear in US politics. (How can they be so dumb and so bright at the same time. Maybe Burke was right about mob-rule democracy.)

    If it wasn’t for 9/11 when Bush had to talk like the Fonze when he was wrong, “Chinuh—is— uh—guhd —fffffriend.” then the book would be quite accurate in flavour.

  2. Despite your scepticism, polygraphs (and there are other versions now) are not witchdoctor magic. They might not be reliable, but various reviews (just look through pubmed) suggest that there is some correlation between responses and people’s interpretations of whether they are not lying or not.

    This isn’t surprising, because the fact that something different happens in some people’s brains when they lie or tell the truth seems unquestionable (otherwise they wouldn’t know they are lying) , and for some groups of people, there are going to be large differences when lying (and some no measurable differences with current technology), just as have been found for many other things that evoke potentially emotive responses, like neutral vs. emotional words, different facial expressions etc. .

  3. Witchdoctor magic normally has a positive correlation with the desired outcome. After all, witchdoctors weren’t/aren’t fools and they are more likely to stick with methods that (appear to) produce the desired outcome.

    In the case of lie detection, witchdoctors, traditional and modern, are well aware of the fact that people behave differently when they are lying, for example by sweating, and have learned this response can be enhanced by the use of a supposedly infallible detection technique.

    But a positive correlation is not enough to make a tool useful in criminal investigation. If the investigator overestimates the reliability, as US police forces clearly do, the tool does more harm than good.

  4. All a polygraph does is measure whether a person is nervous and fearful or not. Now they could be nervous and fearful because they’re telling a lie, or it could be because they’re strapped up to a machine receiving hostile interrogation. And if they’re non-responsive, it could be because they’re telling the truth, or it could be that they know the machine is fallible and they can beat it, or they’re habitual liars.

    It puts a veneer of objective truth finding over a vast human psychological conundrum. I’d much rather have experienced effective interviewers probing, without relying on a flawed technology.

  5. Disturbing. How vindictive of a society have we become- so eager to sell our neighbors down the river. I was especially shocked at the behavior of the University where he worked. Then you have the polygraph- itself an awe-inspiring supernova of naivete, (to say nothing about the ninny who gave the exam)- and the end result is trashing the life and career of one of your best and brightest. The cast of villains on this one would put Nixon’s plumbers to shame: the ‘tough on crime’ judge, the ‘homeland defense association’ feds, the casual jury, the collaborating University. At least the Romans and the Brits fought their demise. Would appear we’ve embraced ours.

  6. The scientologists use a version of the polygraph in order to detect the presence of alien thought control. L Ron apparently vouches for its efficacy, so this is clearly a technological marvel and only the hide-bound and self-interested nature of the legal profession prevents its use to replace the discredited jury system. (Warning: the above comment may contain traces of irony)

  7. I don’t know how accurate lie detectors are but I can’t accept that most interrogators are really that much better at interpreting responses that are outside their expected “norms” -not a good basis for conclusions of guilt or innocence, specially if, for whatever reason they have already formed opinions of guilt or innocence. Try being a black person in australia suspected of theft, one who has retained some of their traditonal manners and modes of respect such as politely refraining from looking people in the eye “ah… he won’t even look at me – must be guilty”. I imagine it’s the people who are outside the norm that are most vulnerable to bing misread – by both machine and interrogator – especially those with undiagnosed mental illnesses.
    A truly accurate lie detector could be quite a boon but I think I’d start with our elected leaders “seen any cables or emails about bribes lately?” and our judges and our police – “taken any bribes lately?” – since they are the people we most depend on en-masse to be doing the right thing whilst simultaneously being ideally placed to avoid detection and prosecution if they aren’t.

  8. I think you’ll find that recent fMRI studies looking at lie detection in normals are now claiming around 90% accuracy in lie detection, which I presume compares favourably with witchdoctors and many other types of evidence. In addition, since its possible to run replications of the same study in different labs, unlike many other types of evidence (like, say, eyewitness testimony, which we know is inaccurate and susceptible to bias), it means that people know exactly how such evidence should be weighted.

  9. A disease researcher who is merely imprisoned ought to consider himself lucky:

    Yet another possible sphere of subterranean military/intel/corporate skulduggery. You wouldn’t read about it, no fear, not for a decade or two anyway. I’m sure there’s a convincing explanation.

    Being a whistleblower here would be bad enough, but imagine the balls you’d need for it there.

  10. “This group plays a much bigger role in the US than elsewhere, which may be why the polygraph is taken seriously only in the US. ”

    This statement is, at best, debateable.

    “In the United States of America, (where polygraph testing is a growth industry) the admissibility of lie detector test results is determined by courts and legislators on a State by State basis.

    In the Federal legal system, test results are inadmissible as substantive evidence.[6] Whilst some States have allowed test results in criminal trials, States such as such as California have prohibited the admission of such evidence unless all parties consent to its admission.[7] Other States such as Illinios completely bar the use of such testing in criminal trials. This prohibition extends to requesting, requiring or suggesting that a defendant submit to such a test.[8]

    The preponderance of authority in the United States is against the admission of polygraph evidence with a variety of grounds having been asserted for refusing its admissibility including: –

    1. It intrudes on the ultimate issue which the Court must determine.
    2. It does not fulfil the criteria of the Supreme Court of the United States test in Frye Case[9] with regard to admitting scientific evidence.
    3. It is hearsay evidence.
    4. It relates to the credit of witnesses not suffering psychiatric illnesses and is therefore not a proper matter for expert evidence.
    5. The elicitation of the responses is unfair because of the trickery and deceit necessary to obtain responses.
    6. The testimony is self-serving for the Defendant.[10]

    In Fryes Case[11] the court held that evidence obtainable from the use of scientific instruments or techniques is admissible if the instrument or technique has a reasonable measure of precision and is accepted in its scientific field or profession.

    More recently, the approach of US courts has been to admit evidence where there is recognition by specialists within a profession or field of science, even though the wider professional or scientific group may be unfamiliar with the technique.[12]

  11. ‘“This group plays a much bigger role in the US than elsewhere, which may be why the polygraph is taken seriously only in the US. â€?

    ‘This statement is, at best, debateable.’

    Avaroo doesn’t appear to appreciate the difference between a comparative statement and an absolute statement.

    (Hands up if this surprises you … I didn’t think it would.)

    As the ever-handy Wikipedia indicates, the US is the only country in which polygraphs are used extensively:

    QED, polygraphs are used more extensively and intensively in the US than elsewhere.

  12. For purposes of blogging, wikipedia suffers no defects that are not inherently present in a blog anyway. That is, its unreliability (not that bad, according to some studies) only comes from being an internet resource, and that happens with everything you see on a blog. For greater blog reliability you have to check it independently yourself, so it’s no worse whether your search starts with the internet or with what someone just decides to post.

  13. Anyone can post a definition on wikipedia. It isn’t a definitive source for information on any subject. And it certainly cannot be used to prove anything.

  14. The point is, no blog discussion can prove anything by itself anyway – unless you are willing to double check it elsewhere. Citing wikipedia is just shorthand for spelling out everything in the reference as though it originated with you – but no better or worse for all that.

  15. No, the point is that the statement in question is debateable. Citing wikipedia is meaningless and adds not one thing to the debate.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s