Killing people is the best solution.

So, to take this to its logical conclusion, is there then no evidence that punishment is an effective deterrent to law-breaking, Drewdad? Or is it just this form of punishment that is not a deterrent at all?

Good, lord. Are you really that idiotically pedantic?

OK, then.

"The preponderance of evidence indicates that there is no scientific support for the idea that the death penalty is a more effective deterrent than other commonly applied punishments for murder."

So if there is no scientific support for the idea that the death penalty is a more effective deterrent than other punishments, then I take it that severity of punishment has no effect upon lawbreaking, is that right? I am just trying to understand your method of logical reasoning here.

I do think there are other factors involved that perhaps have not been properly addressed here besides severity of punishment, including the swiftness and sureness of any form of punishment.

Maybe I did make assumptions that are false. That doesn't mean that information wasn't gained. If this were real, as soon as an anomaly that broke my assumption was found, I would re-think things. This is how science works.

How about this: I'll make a new problem: "You have a stream of random, real world input, that is converted through a consistent process into one of the upper case letters from the English alphabet. It has been proven that a correlation exists in the real world, such that after the conversion process the letter A is always followed by the letter B (they are 100% correlated). This is a one way correlation."

When you look at a single B, you cannot prove what it follows. However, the correlation gives you information that it is more likely an A then any other character. In fact, this correlation lets you know it will be an A roughly 50% of the time. If this correlation didn't exist, it would follow A roughly 3.8% of the time. Clearly I have gained knowledge. If not from the fact that they are correlated, where does this extra information come from?

I am looking forward to your explanation.

In statistics, it's called a "ceiling effect." Once you reach a certain level, the outcome remains the same no matter how much more severe you make the penalty, or how much more attractive you make the reward.

Edit: I have heard it referred to as a ceiling effect, and I've heard it in regard to statistics. I suppose "ceiling effect" could be used in other contexts.

But Nick, I'm not discussing whether correlations can be informative. Of course correlations can give you information. It's just that when a correlation is found, information about causation is not information that is gained.

Damn you Robert, getting me involved with talk of programming and such! I'm going to try to get out of this thread. I'm not debating anything at this point, just making a statement:

Mathematically speaking, A cannot cause B unless A is 100% correlated to B. Given that, I will say that finding anything less then 100% correlation tells you that no causation exists. If 100% correlation is found, then it is a candidate to test for causation.

Keep in mind, 100% correlation doesn't mean there have to be an equal number of A and Bs. It just means that a consistent percentage increase in A yields a consistent percentage increase in B. If you triple A, then B has to triple, even if the ratio is 35:1 or something.

Alright, hopefully I will stay out of this now, until the next time someone talk about programming. (is there a correlation between discussion involving programming, and my participation on a2k?)

Ah....

I'm not sure that's the case.

Some people get lung cancer, even though they don't smoke. Not everyone who smokes gets lung cancer....

I said "mathematically speaking" because in the real world I think its likely people say "A causes B" when that isn't necessarily the case. I'm not convinced that smoking causes cancer, I think smoking causes something to happen in the human body which makes it very likely for cancer to develop over time. However I don't know nearly enough about cancer, smoking, or the human body to be able to argue that point.

If it is true that smoking causes lung cancer, then people who smoke and don't get lung cancer must die from some other cause, before the lung cancer has a chance to develop.

People who don't smoke and get lung cancer must get it from another cause (just because A causes B, doesn't mean that C can't cause B as well).

(technically this is a slightly different topic, so its ok that I replied )

But that will still mess up your 100% correlation theory.

Causality requires a correlation; it does not require a 100% correlation.

So the way I read that, smoking doesn't cause cancer. Benzo[a]pyrene causes damage to the P53 gene. There are still a number of variables here:

1. The amount of Benzo[a]pyrene that enters the system with each cigarette smoked.
2. The amount of damage caused to the P53 gene (possibly variable by person)
3. The speed it takes benzo[a]pyrene to damage the P53 gene (possibly variable by person)
4. The amount of damage a P53 gene can have before cancer develops (possibly variable by person)

Again, I am admittedly speaking beyond my level of knowledge here, and just making guesses. I think they say "Smoking causes cancer" because it is a strong enough correlation. It is perfectly possible that none of that is variable, and that if a single particle of benzo[a]pyrene enters the body, it will undoubtedly cause cancer if given enough time.

If that is the case, then they are 100% correlated, and it doesn't matter that some people die before it has a chance to happen. What matters is that B will happen as a direct result of A, if given enough time. Not that it actually does happen.

Perhaps this is my opinion of how things should be defined, though, and not the way things actually are defined.

<sob>

(hysterical sympathetic lol)

I have not read all of the posts here, in regard to statistics, but one old saying I will offer in that regard, "figures don't lie, but liars will figure."

Now with that said, I can believe a ceiling effect, that sounds quite reasonable, however I would doubt that there is an absolute ceiling effect in every case between life imprisonment and the death penalty for example.

Just using common sense here in regard to a possibly larger point in this discussion, I suspect that the swift and sure factor in regard to the punishment may overshadow the severity of the punishment to a considerable extent. For example, if a criminal knows he has lots of "ifs" in order to receive the punishment, first being if he is even caught in the first place, then haggling in court in regard to plea bargains and appeals even if he is convicted, the criminal knows he probably has years before any potential of being executed, I think all of the statistics are therefore somewhat worthless in terms of evaluating the death penalty as a deterrent. We therefore cannot use statistical data very well in terms of capital punishment being a deterrent because the punishment is not swift and sure, far from it. There are other factors that have to be considered as part of the mix of factors, but that doesn't mean you have to throw out common sense as one way to look at this issue. Even with all of the above, if you wipe out all fear of the death penalty, I fail to see how any violent criminal should have any fear at all of any punishment.

It is difficult to compare countries in terms of murder rates and punishment because cultures differ so widely, however take a look at Saudi Arabia if you doubt the death penalty would have no effect. What do you think would happen there in regard to crime rates if capital punishment was eliminated?

That should read as follows:
.....however take a look at Saudi Arabia if you think the death penalty would have no effect. .....

attention: when someone is desperate and commits a crime leading to a possible sentence of death, guess what, they get more violent and desperate.

the problem with society and crime is that, society forces crime to exist.period.

get real, get the big picture.

Okie, do you have a point? Your argument here does not seem to support the death penalty. If "swift and sure" is your goal, then you should want to avoid the death penalty, and all of the automatic appeals that go with it.

You still seem to be under the misapprehension that the choice is between the death penalty and no punishment at all. Whether this is a strawman argument, or the result of very sloppy thinking, it is not persuasive.

That's where it started (in so far as the possible deterrent factor), and all any of us can do is weigh the preponderance of evidence, and use our best judgment to form an opinion. But that is hardly the only implied benefit anyway. Dozens of people have been killed by murderers who’d previously been convicted of murder. Surely you’ll concede capital punishment makes recidivism pretty unlikely.

Even so: Pretending that murderers don’t consider the Death Penalty worse than Life Sentences is an exercise in denial anyway. It is a rare perpetrator indeed that chooses the Death Penalty over Life. Suicide is as easy as it is unusual among convicted murderers. The very threat of a Death Sentence provides incentive for many murderers to forfeit their right to be presumed innocent until proven guilty, by way of a guilty plea in exchange for a life sentence. This is also evidenced by the preponderance of people on Death Row who will exhaust every possible appeal before succumbing to their fate. Clearly, they fear Death worse than Life Imprisonment.

Next consider the fact that many murderers are not beginner criminals. More than half of all people arrested in any given year are repeat offenders and some 20% of murderers are already on probation, parole, or out on bail at the time of their offense. Out of 1,408,337 violent crimes in this country last year, 16,929 were murders. But not one person was deterred? Are we really expected to believe that with the tens of millions violent crimes that have been committed since the Death Penalty’s reinstatement, that the odds against a would-be murderer being deterred by the Death Penalty is greater than 500,000 to 1 among would be murderers?