The Judicial Process -Creating the output model

So in order to construct our model we need a pain scale and we need to be able to place a victim's suffering on this pain scale. If this was being done to determine public policy, by a public commission, all sorts experts would be used to do the necessary evaluations, but since the purpose of this model is clarify my understanding of the problem the pain scale will just reflect my judgments and information I can easily gather. I will use a pain scale from zero to twenty. I will anchor the scale in my mind by assigning a value of twenty to the worst pain I can think of, which is losing a child. With this in mind I personally would value the pain of a woman raped at about seventeen. I'm a male so all you women out there feel free to criticize this. Now comes an issue of how complex do we want to make the model. We could consider things like age of the victim and marital status and come up with different pain levels, but I'm going to regard that as beyond the scope of this exercise and stick with the single value of seventeen. I will also arbitrarily use the single pain value of two for the victims of a burglary although various factors could affect the pain level of a burglary. The pain value of an innocent man sent to jail has to be more complex because it will vary with time spent in prison and we have to take this into account in order to examine the length of sentence questions. So I make the following pain assignments with respect to pain for time served. I assign a pain of seventeen to serving twenty years unjustly, twelve to serving ten years unjustly, eight to serving five years unjustly, and six to serving one year unjustly. Any amount of time between two of these anchor points will have pain proportional to the total time between the two points. Sorry for the math. Now we have my scale. Feel free to criticize it and give your own thoughts. In my next post we will begin to discuss how we will use this scale and these judgments to clarify the answers to our questions

The Judicial Process Model-The output

With the Judicial Process model we strive to answer three questions:

1) How many guilty people are we willing to let go free to prevent one innocent person from being convicted?

2) How long a sentence is appropriate if convicted?

3) If there are mitigating circumstances, how much if any should the sentence reduction be?

It seems obvious, to me at least, that the answer is different depending on the crime. So I propose to look at two very different crimes. The crime of simple unarmed burglary and the crime of rape.

Since Risk Analysis is a numeric technique it requires a numeric output. So the question is what shall be our output. The original quote from William Gladstone talks of the suffering of the convicted innocent person. This suggests to me that a measure of "degree of suffering" should be the output of the model. If I were developing the model I would therefore create a numeric "Pain Index" and the output of the model would be in terms of pain as defined by this index. We would then design our judicial system to create the least pain. However, In doing this we will consider not just the pain of the innocent convicted person, we will also consider the pain of victims and their loved ones. Included in the group of victims is not just the victim of the case being tried, but possible future victim of a guilty person.

You may be of the opinion that we can't quantify pain. But all this means is that it is difficult and personal. When you have to assign a number to the amount of pain caused to an innocent man by spending a year in jail and then must assign a number to the pain of a woman raped it causes you to decide, are these really equal, or is one worse than the other. If so how much worse. Ten percent, a hundred percent and etc.

I will go into more details on how we might develop the output and what issues developing the output will force us to think about in the next post.

Politics and Decision Making

Are you a Republican, a Democrat , or an Independent. Are you liberal or conservative. The answer to this question depends on the decisions you would make on various issues. Pro choice or pro life. For gay marriage or against gay marriage. For immediate withdrawal from Iraq against a too quick withdrawal. For keeping the Bush tax cuts against keeping the Bush tax cuts. Now my position is there are very few, if any, decisions which are clearly right or wrong. I have this position for two reasons.

1) Since the world is random and chaotic and the outcome of almost all decisions depend on uncertain factors even the best decision can work out badly.

2) Since different decision makers have different values with respect to the possible outcomes of a decision, the same decision with the same result can turn out to be wrong for one decision maker and right for a different decision maker.

My purpose on this blog is help people understand better both their own decisions and the decisions other people would make that might differ from their own decisions. My tool for this will a probabilistic modeling technique called Risk Analysis. At this point, I can hear the door slamming as everybody leaves the blog, but I promise you there will be very little mathematics on the blog, because the value of the technique is in how it makes you think about the factors that go into the decision more than the numeric results that come out of the model.

“Better that ten guilty persons escape, than that one innocent suffer,” said
English jurist William Blackstone.

Other people have suggested other ratios. Also related to this issue is the question of appropriate sentences, and reasons for reducing the sentence after conviction. This is the first decision, or set of decisions, I am going to look at through th eyes of the Risk Analysis technique.

I hope you will come back and at least give it chance. I won't say it will make you change your mind on any decision, but it should at least make you think about your position and see that perfectly reasonable and good people might have a different position than you.