bd-from-kg

Mike Rosoft:

I still don’t follow you. First off, the death penalty is certainly different from other penalties actually used in our legal system: it’s more severe. But how does it follow from this that it should never be imposed? Six months in jail is a more severe penalty than a $5 fine, but it doesn’t seem to me that it follows that a penalty of six months in jail ought never be imposed. Or, take any legal system and identify the most severe punishment used by it. Does it follow from the fact that it’s the most severe that it should never be imposed? This is absurd.

As for “there is no way to fairly decide between death penalty and life imprisonment”, is this statement even meaningful, and if so, what does it mean? What would constitute a “fair” way to decide what punishment to impose? Does one punishment have an intrinsic property of “fairness” or “justness” (in a given case) that other punishments lack?

The only meaningful way to interpret “fair” punishment is “the punishment that produces the best results overall” (or as Alonzo might say, which satisfies the most desires that a good person would have, with “good” in turn defined in terms of “overall desire satisfaction”). It’s often very difficult to decide which course of action “satisfies the most desires...”, but I don’t see how this problem is any different in the case of deciding what punishment to impose than it is for innumerable other moral questions.

Any punishment might be mistakenly imposed on an innocent person, and this possibility should of course figure into the calculus of whether it’s the one that produces (in a probabilistic sense) the best results. But how does the death penalty differ from other punishments in this respect?

As for the claim that it “doesn’t work”, I continue to be amazed at the number of people who say this almost casually, as if it were a settled fact. It’s not a settled fact; in fact it’s a highly implausible claim with very little evidence to support it. But I’ll be dealing with this at more length in other replies.

Yes. Most of the objections there are not to capital punishment per se but to inappropriate applications of it.

But this does suggest an argument against capital punishment that no one here has made explicitly, but which does have some force:

“It may be the case that capital punishment, properly administered, leads to desirable results overall. But given the nature of this particular society, it is certain that it will not be properly administered. If we consider how the death penalty would work in practice, even after making any feasible reforms to the system, it will produce results worse than no capital punishment at all.”

This is a pretty strong argument; in fact it's decisive when applied to the vast majority of countries. And I’m far from certain that it’s not valid for the U.S. as well. But I don't think that it is. Here the concern for individual rights can be expected to produce (and in fact is producing) gradual reform. And the fact that a system is imperfect doesn't prove that it should be abolished. The prison system itself is radically imperfect, but few people argue that it should be abolished on that account - that we would be better off just emptying the jails.

bd-from-kg

The AntiChris:

Not so. Incentives and disincentives have the power to alter behavior. To say that R is an incentive for X is to say that he desires R. And under appropriate conditions this fact will alter his behavior. Specifically, if he believes that doing Y will bring about R and none of the alternatives will, and as a result he believes that doing Y will satisfy more of his desires than any alternative, this will cause him to do Y, whereas he wouldn’t if this incentive didn’t exist.

To put it another way, all actions involve weighing the pros and cons of the alternatives. If you change the pros of one of the choices by adding an incentive, this will cause the agent to choose that option if it puts this choice “over the top” – i.e., if the pros for this choice outweigh the pros for all alternatives when the incentive is present, but not when it isn’t. (Of course this is oversimplified; the “cons” have to taken into account as well.) The same applies to disincentives in reverse.

Now for any particular individual, changing the incentive structure for choices that he would rarely make anyway may indeed have no effect on his actual actions. (For example, the death penalty would probably never cause me to refrain from killing someone, since it’s highly unlikely that I’d ever choose to kill anyone anyway.) But humans are almost infinitely variable, and the choices they face are also almost infinitely varied. If you change the incentive structure for everyone in a large society in a systematic way (by changing the rewards and punishments for certain kinds of actions) this is practically certain to affect some actions. If the change in incentives is substantial (e.g., paying $1000 for returning a lost wallet, or a life sentence for killing a man if you find him in bed with your wife instead of no punishment at all, or execution for certain kinds of murder instead of a life sentence) it will affect a significant number of choices, even though the chance that it will affect any specific choice by a specific individual may be small.

Yes. It should certainly alter the decisions of a good many potential murderers.

There’s no need to appeal to statistical studies and the like to be sure that the DP will deter some murders. The direct effect is as certain as anything can be in this life. Alonzo’s argument is not that this effect does not exist, but that the DP unfortunately also has an indirect effect (at least in the long run) of altering the incentive structure for at least some individuals in a perverse fashion, and that this indirect effect will actually cause more murders than the “direct deterrent” effect will prevent. I continue to find this highly implausible for reasons that I’ll go into further in my next reply to Alonzo.

It’s also far from clear, to put it mildly, that the “facts” do not support the claim that the DP tends to reduce murders overall, especially when it’s used properly. First, as I pointed out in my last post to Doubting Didymus, the facts can be reasonably interpreted as supporting a strong overall effect of the DP on murder rates: read the study that I provided a link to there. Second, there are strong reasons for regarding all such studies with more than a few grains of salt given the “state of the art” in sociology. Finally, as I’ll explain later, it’s highly doubtful that it would be possible to reliably infer any effect, whether positive or negative, at least in the United States, from the kind of statistics actually available, no matter how advanced or sophisticated the methodology.

bd-from-kg

Doubting Didymus:

I know where you’re coming from, and believe me, I sympathize. Let me try to clarify the situation a bit.

You say:

Here’s the problem. It’s well known that correlation does not prove causation (and in fact the correlation between two variables can be negative even though there is a significant positive causal relationship between them). Thus to determine whether there’s actually a causal link between two variables we have to look a lot deeper than just the raw correlation numbers. This leads directly to just the kind of multiple regression analysis that I referred to (the kind you say that you “care little about”). In other words, any serious attempt to use statistical evidence to determine whether there is a causal link between the death penalty and murder rates has to take the form, ultimately, of a multiple regression analysis. (The same is true of all kinds of other variables of interest in sociology. That’s why sociologists use this kind of analysis so extensively.)

Since I’ve already referred you to two such studies, there’s not much more to say along these lines. Either this kind of study is evidence, or the implications of well-accepted principles of human behavior are evidence, or there’s no evidence at all one way or the other. You can’t reject every kind of evidence that’s actually available and then fault someone for not providing evidence!

But of course this kind of evidence is orders of magnitude poorer than what’s available in real science. That’s because in real science it’s possible to do controlled, repeatable experiments where all variables but the ones you’re investigating are held constant. (This is oversimplified, but in all real sciences there’s some way to make controlled, repeatable observations and obtain new data to check against the old.) In sociology and the other so-called “soft sciences” this is impossible. Worse yet, in many cases (like studies of the effect of the death penalty) the researchers have no control over what data is collected. The data that is collected is often very different from what they really need, so they end up using X as a “proxy” for Y, with no way to test how good a proxy it is. And in any case, since you can’t do controlled experiments, the best you can do is multiple regression analysis, which isn’t remotely close to being as satisfactory or reliable.

For someone with a hard science background all this is extremely frustrating or amusing, or both. But that’s the way it is, and there’s nothing we can do about it. The only rational response is to lower the standard for what constitutes “acceptable” or even “relevant” evidence. Of course, if you’re a purist you can just throw up your hands. Except that important public policy decisions hinge on the answers. So we make our best educated guesses and don’t pretend to have anything like “proof” that we’re right.

Amen-Moses

Can you repeat these links please, the only evidence I've seen points to the fact that EVERY society that has removed the death penalty has either no change or a reduced murder rate.

Amen-Moses

The AntiChris

I wouldn't disagree with this.

The question is though, does this, in the case of the DP, translate into an overall reduction in the murder rate all things considered (eg the proportion of murders which are, genuinely, deterrable, the impact, if any, of the "Alonzo effect" etc)?

The study you linked to certainly comes to some remarkable conclusions - it makes for interesting, if unsettling reading. Unfortunately I'm not qualified to assess the validity of the "econometric model" used and, like you, remain slightly wary of any study which claims to reveal a "definitive" link between the DP and murder rates.

Chris

bd-from-kg

Amen-Moses:

Sure. See you and raise you two. There are the original two that I cited:

First, the classic paper (the first to attempt to apply relatively sophisticated statistical techniques to the question) was:

Erlich: The Deterrent Effect of Capital Punishment: A Question of Life and Death, American Economic Review, vol. 65, no. 3, June

(As I commented earlier, this paper doesn’t appear to be available free of charge on the web, although there are lots of sites that will let you download it for a modest fee.)

Second, the paper Does Capital Punishment Have a Deterrent Effect? New Evidence from Post-moratorium Panel Data by Dezhbakhsh, Rubin, and Shepherd, uses much better and more up-to-date data broken down into separate counties, and does its best to avoid the methodological problems cited by many critics in the original Erlich study. It concludes: "our results suggest that capital punishment has a strong deterrent effect. An increase in any of the probabilities -- arrest, sentencing or execution - tends to reduce the crime rate. In particular, each execution results, on average, in eighteen fewer murders - with a margin of error of plus or minus 10."

Now here are a couple of new ones:

The paper Pardons, Executions and Homicide, by Mocan & Gottings, reports on a study that found "a statistically significant relationship between executions, pardons and homicide. Specifically each additional execution reduces homicides by 5 to 6, and three additional pardons (commutations) generate 1 to 1.5 additional murders." Thus study was one of the first to use a new data set from the U.S dept. of Justice which contains detailed information on every one of the 6,143 death sentences between 1977 and 1997.

In the paper Execution and deterrence: a quasi-controlled group experiment , Cloninger and Marchesini looked at the murder rates before, during, and after a moratorium on executions in Texas and concludes:: "The execution hiatus (in 1996), therefore, appears to have spared few, if any, condemned prisoners while the citizens of Texas experienced a net 90 (to as many as 150) additional innocent lives lost to homicide. Politicians contemplating moratoriums may wish to consider the possibility that a seemingly innocuous moratorium on executions could very well come at a heavy cost."

Also, here are a couple of long essays that (among many other things) document some of the evidence that the death penalty is a deterrent:

Death Penalty And Sentencing Information in the United States by Dudley Sharp

Wesley Lowe's Pro Death Penalty Webpage

Here’s an interesting excerpt from the latter paper:

There’s lots more of this stuff out there. All you have to do is look for it. The anti-death-penalty folks (with the help of the liberal media) have been very busy spreading the lie that all the evidence shows that the death penalty is not an effective deterrent. Don’t believe them. The best that can be said (from the anti-capital-punishment point of view) is that the evidence is mixed: many studies have concluded that it is; others have concluded that they can’t find evidence that it is. None have claimed to find evidence that it’s not a deterrent, much less that the “Alonzo effect” exists.

In considering this welter of data and studies it’s important to keep in mind that even when the death penalty has been in widespread use, the number of murders that have resulted in execution (in the U.S.) has never been much more than 0.5% (one in two hundred). And unfortunately “capital” murders – those that might be eligible for the death penalty under typical statutes) have never been reported separately. So looking for statistical evidence of a deterrent effect, given the innumerable variables that affect the murder rate, is rather like trying to pick up a very weak radio station in a thunderstorm: the “signal-to-noise ratio” is terrible. Under these conditions the surprising thing would be if we could find clear-cut statistical evidence of a deterrent effect.

Finally, a good point that's been made by a number of people is that if there’s serious doubt as to whether there’s a deterrent effect, a reasonable person should favor capital punishment. The reason: If we execute brutal murderers and it turns out that there’s no deterrent effect, some brutal murderers will have died. If we don’t and it turns out that there is a deterrent effect, a number of innocent people will have died that we could have saved. To me, this is a no-brainer. The only possible “fly in the ointment” for this argument is the possibility that the “Alonzo effect” really exists and is stronger than the deterrent effect. I find this possibility too far-fetched to be taken seriously. I plan to give still more reasons for this assessment in my next reply to Alonzo (tomorrow, I hope).

Bill Snedden

Doesn't your argument assume that murderers not executed would simply be set free and thus could kill again? Or do you mean to say that the "innocent" who will have died are those who were executed in error? Do you really mean to just write them off so coldly?

The "deterrence" argument against capital punishment really isn't meant to show that there's no deterrent effect, but that it isn't realistically any better than the best alternative: life in prison without possibility of parole. Granted, the most unbiased reading of the studies must, as you have, conclude that studies exist supporting both sides, but this surely doesn't allow us to conclude, as you do, that "a reasonable person should favor capital punishment". I consider myself an eminently reasonable person and I most certainly don't favor it precisely because

a) Those conflicting studies don't provide clear direction one way or another

b) There is a very real possibility of executing the innocent.

c) The end result of both methods is the same: violent criminals are removed from society.

I would also say that it seems somewhat odd to support a system of justice that would have us kill someone for the crime of killing someone.

Reasonable people may agree to disagree, but to suggest that "reasonable people should support..." suggests that those of us who don't agree aren't reasonable. I disagree.

Regards,

Bill Snedden

Amen-Moses

So how does one explain the FACT that in very society that has abolished the death penalty murder rates have either stayed the same or gone down?

To me this seems counter-intuitive if there is indeed some real deterrant effect.

For example the quote you gave that seems to indicate that for every non-death penalty there would be 18 more murders. This would indicate that the UK only has to kill 5 people a year to totally eradicate murders yet since abolishing the death penalty the murder rate has dropped by far more than 5 people. I'm confused by this please enlighten me.

Amen-Moses

bd-from-kg

Bill Snedden:

??. The innocent people referred to are those whose murders would have been deterred by a capital punishment regime (if there’s a deterrence effect).

In this context “deterrence” refers to any additional incentive not to commit murder relative to the “best alternative”.

Reread the argument and tell me what you think is wrong with it.

But (1) the argument is directed precisely at those who think there’s some doubt as to whether the DP is a deterrent; and (2) statistical studies aren’t the only kind of evidence for a deterrent effect, any more than epidemiological studies are the only kind of evidence about the causes and origins of diseases. Any statistical effect is just the aggregate of the psychological effect on a number of people, so what we know about human psychology is also evidence on whether there’s a deterrent effect. And this evidence is clear and unambiguous, Alonzo’s speculations notwithstanding.

The possibility is real, but very remote. There is not a single known case of an innocent person being executed in the U.S. since 1900, in spite of the best efforts of the anti-DP crowd to find one. Of course, if we keep executing people we’re bound to execute an innocent person by mistake sooner or later. But we take these kinds of risks all the time. If you drive enough, it’s almost certain that sooner or later you’ll kill someone by mistake. But does the fact that you’re putting innocents at risk keep you from driving?

The issue is not the result of imposing a death sentence in a particular case, but the overall effect of a capital punishment regime. In the same way, the main point of arresting a shoplifter is not to stop that particular person from stealing for a while, but to discourage people from shoplifting. Surely this isn’t a subtlety that you never thought of before?

Is it odd to imprison someone for the crime of imprisoning someone? Is it odd to fine a merchant for cheating his customers out of their money? I don’t see anything odd in the fact that a punishment often superficially resembles the crime. The moral nature of an act is not determined by its physical character, but by the context.

As John Stuart Mill put it:

• “Does fining a criminal show want of respect for property, or imprisoning him, for personal freedom? Just as unreasonable it is to think that to take the life of a man who has taken that of another is to show want of regard for human life. We show, on the contrary...our regard for it, by the adoption of a rule that he who violates that right in another forfeits it for himself and that while no other crime that he can commit deprives him of his right to live, this shall.”

The point is that “reasonable” people will presumably feel that the death of a murderer does not have the same weight in the scales of justice as the death of an innocent. Or as Alonzo might put it, the death of a brutal murderer won’t thwart nearly as many “good” desires as the death of an innocent.

everlastingtongue

That statistic exists, mainly because very little effort is spent exonerating prisoners already executed. Are you really so sure? From where does this confidence come from? If our legal system is so fallible otherwise, how is it possible that in capital cases it is flawless? You can’t deny that incorrect convictions occur at all levels of crime. It’s just that, while the prisoner is still alive, there is a motivation for them, or those working on their behalf, to seek exoneration. There is no motivation if they are dead. This is one of the primary reasons I’m against the death penalty.

Just a couple of examples from:http://www.deathpenaltyinfo.org/arti...d=111#executed
Look, I just don’t get it. There is no proof that the death penalty is a deterrent, and many claims that discontinuation of it lowers murder rates. Also, there IS proof that it is more costly, what gives? Moral arguments aside, suppose that there were two brands of laundry soap, brand A and B for example, that did the exact same job of cleaning your clothes. Furthermore, there was evidence, perhaps even disputed, that Brand A had some other positive effects, such as making your clothes last longer. Brand B costs more than A. Who would buy brand B?