Let’s talk about Amanda Knox instead.

Amanda Knox is an American woman who was jointly convicted, with her boyfriend at the time Raffaele Sollecito, of the sexual assault and murder of Meredith Kercher in Perugia, Italy, on 1 November 2007.

Meredith Kercher, a 21 year old British university exchange student from Coulsdon, South London, was found dead on the floor of her bedroom with stab wounds to the throat. Some of her belongings were missing, including cash, two credit cards, two mobile phones, and her house keys.

Rudy Guede, an Ivory Coast native raised in Perugia, was convicted in October 2008 of having sexually assaulted and murdered Kercher, and was sentenced to 30 years, reduced on appeal to 16 years in December 2009.

Also tried were Knox, an American exchange student and flatmate of Kercher, and Knox’s then-boyfriend, Sollecito, an Italian student. Knox and Sollecito were convicted on charges of sexual assault and murder in December 2009, and sentenced to 26 and 25 years respectively.

Their convictions were overturned on appeal on 3 October 2011 by a panel of six jurors and two judges. In an official statement of their grounds for overturning the convictions the judges wrote there was a “material non-existence” of evidence to support the guilty verdicts at the trial. The appeal judges further stated that the prosecution’s theory of an association between Sollecito, Knox and Guede was “not corroborated by any evidence” and “far from probable”.

I first heard of the case—and it stuck in my mind ever since—when I read this

Two intelligent young people with previously bright futures, named Amanda and Raffaele, are now seven days into spending the next quarter-century of their lives behind bars for a crime they almost certainly did not commit.

on LessWrong.com, a community blog devoted to refining the art of human rationality.
The Amanda Knox Test: How an Hour on the Internet Beats a Year in the Courtroom.

The author wielded something called “the Sword of Bayes” and, to the following propositions

1. Amanda Knox is guilty (of killing Meredith Kercher)
2. Raffaele Sollecito is guilty (of killing Meredith Kercher)
3. Rudy Guédé is guilty (of killing Meredith Kercher)

assigned the following probabilities.

1. Small. Something on the order of 0.01 or 0.1 at most.
2. Ditto.
3. About as high as the other two numbers are low. 0.99 as a (probably weak) lower bound.

The author continues

Needless to say, this differs markedly from the consensus of the jury in Perugia, Italy.

How could this be?

Am I really suggesting that the estimates of eight jurors — among whom two professional judges — who heard the case for a year, along with something like 60% of the Italian public and probably half the Internet (and a significantly larger fraction of the non-American Internet), could be off by such a large amount?

Of course, the author really was suggesting exactly that. (It’s tempting to say—but, for obvious reasons, I won’t—that the LessWrong.com author was vindicated by the verdict of the appeal court in October 2011, that overturned Knox’s and Sollecito’s convictions.)

I won’t go into the nitty gritty details of the case. If you’re interested in further reading, Injustice in Perugia is a website set up by a Knox and Sollecito supporter, documenting the case.

This post’s take-home messages are two: read LessWrong.com, and learn the Bayesian Way.