To Catch a Lie

A concept which goes hand-in-hand with the nature of satisfactory evidence is the credibility of witnesses. How do you know when someone's perceptions are skewed? How do you know when they are biased? How do you know when they are lying? Its this last question I want to address here.

It is the rare case where someone, sometime is not lying. Everyone lies from time to time. For instance, I'm a inchoate bike mechanic. A few weeks back, I asked the tech at my local bike shop about work stands. He pointed them out, and I promptly searched for them on line, where I found one much cheaper. When I went in to buy other parts, he asked me about the work stand. I felt bad for going elsewhere, so told him, "oh, I just picked up a used one." Lie.

Had he inquired further, I would have had to make a choice: come up with a backstory about my lie ("Oh, I just searched Craig's List...") or tell the truth ("I couldn't help myself, Amazon's prices are 20% less than yours..."). And therein lies one secret to separate lies from truth. This is a secret I use — as does your average pulp-fiction detective — all the time in my practice: both with reticent clients and with adverse witnesses: every statement has a backstory.

As described in 1903 by Francis L. Wellman in The Art of Cross Examination:

No one can frequent our courts of justice for any length of time without finding himself aghast at the daily spectacle presented by seemingly honest and intelligent men and women who array themselves upon opposite sides of a case and testify under oath to what appear to be absolutely contradictory statements of fact. ... The inquiry is most germane to what has preceded, for unless the advocate comprehends something of the sources of the fallacies of testimony, it surely would become a hopeless task for him to try to illuminate them by his cross-examinations.

Cross-examination is the process by which an attorney questions the witness about his story — hopefully catching the witness in a lie (or possibly reinforcing his story with additional detail). There are many famous cross examinations. Here is one by our president, Abraham Lincoln defending an alleged murderer (follow link for additional detail):

By Mr. Lincoln:  Did you actually see the fight?
By Mr. Allen:  Yes.

Q:  And you stood very near to them?
A:  No, it was one-hundred fifty feet or more.

Q:  In the open field?
A:  No, in the timber.

Q:  What kind of timber?
A:  Beech timber.

Q:  Leaves on it are rather thick in August?
A:  It looks like it.

Q:  What time did all this take place?
A:  Eleven o'clock at night.

Q:  Did you have a candle there?
A:  No, what would I want a candle for?

Q:  How could you see from a distance of one-hundred fifty feet or more, without a candle, at eleven o'clock at night?
A:  The moon was shining real bright.

Q:  Full moon?
A:  Yes, a full moon.

Q:  [Consulting an almanac.] Does not the almanac say that on August 29th the moon was barely past the first quarter instead of being full?
A:  [No answer.]

Q:  Does not the almanac also say that the moon had disappeared by eleven o'clock?
A:  [No answer.]

Q:  Is it not a fact that it was too dark to see anything from so far away, let alone one-hundred fifty feet?
A:  [No answer.]

Verdict: not guilty. Mr. Allen may well have believed he witnessed the fight — but the facts tell otherwise.

Radiolab recently recorded a show on the very topic of lying: the Fact of the Matter. As always, this Radiolab episode is well written and well orchestrated. My main problem with Radiolab is that they never seem to ask the next question I would have asked and seem content to end the inquiry without the firm answer which is alluringly close — I suspect they do that on purpose (hey, I'm just asking? or maybe they want you to come to your own conclusions on less-than-complete information).

Anyway, in that episode, they interview a Hmong survivor about the genocide which the Laos government committed against the Hmong post-Vietnam. He claimed he saw the aftermath of a chemical attack (an alleged mycotoxin) and the destruction it brought. He also admitted that regular weapons were used to kill his people.

Radiolab is really only after the answer to one question: were the Hmong attacked with chemical weapons? The mycotoxin came in the form of a yellow liquid or powder, and caused the US government to accuse the Soviet government of chemical warfare (causing the US to stockpile chemical weapons). The powder turned out to be uncontaminated bee feces and it was revealed that the Laos government had no capacity to produce or spread such weapons.

During the interview, the witness became agitated when confronted with these facts. He stopped the interview and said, essentially: you're playing a semantic game, the important point is that the Hmong were massacred, what difference does it make that it was chemicals or bullets? (Radiolab got grief from listeners about being too harsh, and provided this thoughtful defense.)

As seekers of the truth, it matters a great deal. However, the interviewers, instead, considered whether there were multiple "truths" at issue (the lack of presence of mycotoxin and the emotional "truth" that the mechanism of death is unimportant) and left it up to the listener to decide how to resolve the conflict.

I think that was a copout by Radiolab. No, they did not need to force the witness into a confession that he was wrong about the chemical weapons. But, they did need to draw the reasonable inference without softening the blow. The reasonable inference under the facts is that the reprehensible acts of the Laos government could not have included the chemical weapons reported.

Sometimes catching a lie is uncomfortable. Sometimes the witness blames the interviewer. It is a true pain of this profession. Especially for someone like me who wishes to please everyone.

Your, Bear