Sum(n Guilty Men, n=[1,10000])

September 14, 2006 3:21 PM

It's better that n guilty men go free than one innocent man go free: a historical study.
posted by apathy0o0 (28 comments total)

It's better that -1 guilty men go free than one innocent man go free.
posted by riotgrrl69 at 3:22 PM on September 14, 2006

okay...it's better that abs(n) guilty men go free than one innocent man go free.
posted by mcstayinskool at 3:28 PM on September 14, 2006

Nice title.
posted by joshuaconner at 3:28 PM on September 14, 2006

I agree - the effect of punishment, short of death, on the guilty is unquantifiable and also contentious (its purpose, its use for society), whereas the effect of punishment on the innocent is not.
posted by Flashman at 3:34 PM on September 14, 2006

At first I thought "n" to be a typo for "no" and wanted to rush in to slam this post. Then I read the article, which I found to be wonderfully enlightening. I especially liked

c squared = -1,251,677 ln n + 30,217,466

Thanks for the post and the read!
posted by F Mackenzie at 3:36 PM on September 14, 2006

Um... it's better than n guilty men go free than one innocent man go free?

Isn't that supposed to be be imprisoned?
posted by JHarris at 3:54 PM on September 14, 2006

Isn't the phrase supposed to be: "Better that n guilty men go free than one innocent man suffer." I am sorry but is there something here I am missing?
posted by MarkO at 3:54 PM on September 14, 2006

Ok, I'm confused. Don't you mean "better that n man...than one guilty man suffer (or be punished, die, whatever)"?
posted by tula at 3:55 PM on September 14, 2006

What he said.
posted by tula at 3:56 PM on September 14, 2006

Abraham's celebrated haggle from the book of Genesis, allegedly written by Moses but also attributed to God, 25 provisionally sets a value of n = (P - 10) / 10, where P is the population of Sodom.

There is just not enough out there that combines algebra and the Bible.
posted by salvia at 4:08 PM on September 14, 2006

Footnote 204

204. We have used Microsoft Excel.

And this is a paper about other people's guilt?
posted by lalochezia at 4:15 PM on September 14, 2006

...there were only four righteous people in the city, and they were all saved, though they lost their real estate.

That sentance made my day.
posted by Alex404 at 4:16 PM on September 14, 2006

I've always liked Giles Rooke's (possibly apocryphal) endorsement of n=50 in capital cases: "The humanity of the law requires, that in cases of doubt, where a man’s life is at stake, the balance should preponderate on the side of mercy. The law was made, not so much for the punishment of the guilty, as for the protection of the innocent, and it would be better that fifty men should escape, than an innocent man suffer."
posted by Phlogiston at 4:48 PM on September 14, 2006 [1 favorite]

In capital cases I think it really depends on two pieces of information: 1) what is the likelyhood of the guilty commiting another crime, and 2) how bad being the victum of a crime is relative to being jailed.

A man killed by an aquitted criminal is just as dead as a man killed by the state. I would say that in murder cases it should be 1/ (the odds of an aquitted murder striking again). I think it is maybe around 10 in cases of murder.
posted by I Foody at 5:07 PM on September 14, 2006 [1 favorite]

A man killed by an aquitted criminal is just as dead as a man killed by the state.

So is a person killed by a previously innocent man (or woman). Let's just lock everyone up!
posted by delmoi at 5:33 PM on September 14, 2006

My favorite part was the footnote on Lot:

Note, however, that while Lot lost his real estate, he did acquire condiments. "His wife looked back from behind him, and she became a pillar of salt."
posted by Amy Phillips at 6:16 PM on September 14, 2006

Delmoi, we're pretty much doing that now.... 1 in 142 US residents now in prison
posted by banished at 6:20 PM on September 14, 2006

OK, this was kind of fun, not your typical stuffy law review article. Also, it's by Eugene's brother.
posted by caddis at 6:39 PM on September 14, 2006

This just goes to prove that some people will think about anything.
posted by Pressed Rat at 7:00 PM on September 14, 2006

I don't think apathy0o0 cared enough to get his page title right.
posted by Citizen Premier at 7:25 PM on September 14, 2006

The point being that it is better that someone go free than be wrongly convicted in uncertainty, but by what margin of error? Saying that 10 guilty people should be acquitted instead of one person being wrongly convicted is a literary stab at probabillity, perhaps related to the jury size. But it is a modern question related to forensics. Do we convict someone who is accused of a crime for a proverbial 90% certainty? Or a higher degree of certainty?
posted by Brian B. at 7:59 PM on September 14, 2006

Ahhhh, damnit, I typed the wrong word in the description. That's what I get for trying to type things after pulling all-nighters.
posted by apathy0o0 at 10:06 PM on September 14, 2006

Do we convict someone who is accused of a crime for a proverbial 90% certainty? Or a higher degree of certainty?

Beyond a reasonable doubt. "In the United States, it is usually reversible error to instruct a jury that they should find guilt on a certain percentage of certainty (such as 90% certain)."
posted by kirkaracha at 5:56 AM on September 15, 2006

Doesn't 10:1 result in 91% certainty?
posted by noble_rot at 7:24 AM on September 15, 2006

The point being that there are times when the evidence is presented as reliable on a percentage basis, without assuming that the threshold is subject in the minds of jurors. Ballistics, blood tests (especially before DNA), and coincidental accidents, were all statistical margins, but used as a baseline of certainty. Finding that threshold is a statistical open question. One could argue that the jury number itself was being used as the threshold this entire time, like a vote.
posted by Brian B. at 7:57 AM on September 15, 2006

its better that stupid titles get posted than not stupid titles get non-posted.
posted by obeygiant at 9:30 AM on September 15, 2006

of course, if an innocent man is punished, dosen't it naturally follow that a gulity man has gone free as well?
posted by jonmc at 9:46 AM on September 15, 2006

Thanks jonmc - often overlooked point.
So added into the equation is always +1 - that person being whoever is actually guilty of the crime. So you not only have the injustice of making an innocent man suffer, you have the injustice of allowing a guilty man to go free. And really no matter how bad you make the innocent guy suffer, that isn’t going to deter the guilty one. Making more innocents suffer only spurs the guilty on more, ‘cause they know someone else will be taking the hit.
posted by Smedleyman at 11:08 AM on September 15, 2006