ustoo Dead

Sites to see:

Essential
Almost Essential
From the Left
From the Right
Magazines and Journals
Various Weblogs and Pundits
Think Tanks
Data
Vast Right Wing Conspiracy
Shrill Leftist Garbage

 

Saturday, March 05, 2005

Why I think juries should be larger, randomly chosen, and shouldn't require unanimous verdicts (all in one blog post). 

A few minutes ago Brandon and I were discussing Wittgenstein over backgammon. The conversation proceeded to capital defense and I had the--what I'm sure isn't novel at all--thought that juries ought to be larger.

The jury can make one of two mistakes: it can convict an innocent man (in statistical parlance, this would be a type I error) or it can acquit a guilty man (a type II error). The size of the jury ought to be decided (in part, at least) with the goal of minimizing the chances of these two errors (note that ceteris paribus decreasing the chanes of one type of error, usually results in increasing the chances of the other).

It seems like in some cases there are people who would convict the defendant regardless of the evidence. For argument's sake, you can throw out numbers above the probability that these people would get on a jury (and I do below), but it's clear that the smaller the jury is, the more likely you are to end up in the anomalous situation where the jury is dominated by people who are itching to convict the defendant.

Similarly, there may be cases where a stubborn juror will acquit the accused no matter what. A jury system that uses larger juries and requires more than eleven-twelfths of the jurors to vote for conviction in order to return a conviction will experience fewer type II errors caused by the stubborn juror than our current system does. (This is true under the modest assumption that stubborn jurors who will only acquit are all independently and equally likely to occur among the jurors.)

The numbers I pasted in below refer to the problem of the juror who wants to convict no matters what (I truncated the tables so as not to include a bunch of outcomes with probabilites that are essentially zero). Assuming this man pops up 20% of the time, the tables show the number of "always convicts" jurors on the left with the associated probability of that number of jurors coming up on the right. Simple stuff I know. The assumptions are so simplistic that this isn't exactly applicable to the real world, but it's still a fair approximation of what would really happen and the results are striking. Under the 20% assumption, a 12 man juror has nearly a 2% chance of being at least half-filled with people who will always convict. This isn't a big number, but given how many jury trials there are, it's significant. With 24 and 36 person juries, the same number drops to .1% and .08% respectively. You double the size of the jury and suddenly your're nearly 20 times less likely to experience the bad scenario mentioned above.

I imagine, that given the thousands of people who are paid to sit around and think of the law, I'm not the first person to think about this. One of Ziggy's philosophy professor's did similar calculations, but if I remember right, they assumed that all jurors were equally like to vote for a conviction. Some of the contributors to this blog are interested in capital punishment (Alex wants to set them free and Ziggy wants to kill them all). Have you gusy read anything about optimal jury sizes? Given the benefits of 24 person juries over the 12 man juries, it seems like 12 is a very bad number indeed. Did we just inherit it from those bints and blokes in England?

I think juries ought to be bigger in capital cases. As Brandon pointed out to me, the remedy would have to be legislative. I don't think too many lawmkers want to be seen as sympathetic to someone even accused of a capital crime, so jury resizing will probably never come up for debate, but it would be great if it did. With some data, lawmakers could thing about the costs of larger juries and type I and type II errors and then move on to the question of the size of the jury and how many members should have to vote to convict.

12 Jurors
0 6.872%
1 20.616%
2 28.347%
3 23.622%
4 13.288%
5 5.315%
6 1.550%
7 0.332%
8 0.052%
9 0.006%

24 Jurors
0 0.472%
1 2.833%
2 8.146%
3 14.934%
4 19.602%
5 19.602%
6 15.518%
7 9.976%
8 5.300%
9 2.355%
10 0.883%
11 0.281%
12 0.076%
13 0.018%
14 0.003%
15 0.001%

36 Jurors
0 0.032%
1 0.292%
2 1.278%
3 3.620%
4 7.467%
5 11.947%
6 15.432%
7 16.534%
8 14.984%
9 11.654%
10 7.867%
11 4.649%
12 2.421%
13 1.117%
14 0.459%
15 0.168%
16 0.055%
17 0.016%
18 0.004%
19 0.001%


Oh yeah, I don't think lawyers should be allowed to dismiss jurors. It seems like the small size of the jury makes jury selection necessary in order to weed out peole the defense and prosecution think aren't fair-minded, but it causes other problems. Larges juries would work to decrease the influence of peopel who already have their minds made up about the case.

-Daddy Brooklyn  15:07 EST | |

About us:

This weblog is an ongoing, if periodic, effort by several friends to stay in touch, in reading material, and in ideas.

Lucky Luciano is a former Italian Stallion real estate hustler and Benedict Arnold CEO turned shady lawyer-to-be. He lives in Denver.

Ben is a Paramedic and would-be philantropist who lives in Denver. He knows everything about nothing.

Fuzzy Dunlop lives in Manhattan. He is more than capable of standing up to the stresses of a high crime urban environment.

Jess is a teacher. But have YOU given her an apple? No, you haven't. You should be ashamed of yourself. This crazy feminist currently rests her copy of Awakening in Jersey City.

Matt is a pariah, iconoclast, and professor of gambling living in Oakland.

Miguel Sanchez is not Lionel Hutz.

Daddy Brooklyn lives in Brooklyn. He hates Republicans, though he wouldn't mind being ensconced in the landed elite of New York City.

Paul just smoked my eyelids and punched my cigarette.

Ziggy Stardust has no past.

Archives
Powered by Blogger eXTReMe Tracker Weblog Commenting by HaloScan.com