November 22, 2005

Kill Them All

Hatched by Dafydd

Patterico is always fascinating to read... even when he's dead wrong, as he is today.

Under the title An Innocent Person Executed? [Post 1], Pat writes:

I can’t definitively say that this is the first example of a clearly innocent person having been executed in the United States — but it sure seems that way. At the very least, he shouldn’t have been convicted. This is why I have argued [Post 2 - DaH] that no death sentence should be imposed unless the defendant’s guilt is proved beyond all possible doubt. As more cases like this crop up, more people will agree with me.

I could not possibly disagree more. The problem, of course, is that the standard Patterico suggests -- that we should impose the death penalty only when the proof of guilt is "beyond all possible doubt" -- is probably impossible to achieve in this world. (Patterico disagrees; I will deal with this below.)

Under this standard, you could not even execute Osama bin Laden: after all, he wasn't even present at the scene of the crime... how can you say that beyond all possible doubt he ordered the attack?

The sad but inescapable fact is that every possible system for determining guilt or innocence will fail... and will fail in both directions. That means that some innocent people will be convicted, and also some guilty people will be acquitted.

Not "may" fail; the odds are 100% that any system will fail. This is because every system, no matter how well thought out, depends ultimately upon fallible and at times corrupt human beings. Thus, Patterico's first suggestion is clearly wrong: we have definitely executed completely innocent people before, and we shall do so again -- unless we ban capital punishment entirely.

And that is the only way to adhere to the Patterico Principle that no innocent person should ever be executed. The Patterico Principle is functionally equivalent to banning capital punishment altogether. Any judicial system that would not be able to execute bin Laden for the 9/11 murders of nearly 3,000 people could not execute any but a tiny handful of people... and even then only when the jury violates its oath and actually sentences on the basis of the reasonable-doubt standard, not the possible-doubt standard. Below a certain level of death sentences, we cannot fairly be said to have capital punishment at all.

Here is the problem, however. The damage caused when the system fails in one direction, when an innocent person is executed, is easy to see. But when it fails in the opposite direction, it's much harder to tell whether damage has been caused; we can never really know that a murderer set free did not kill again. We only know that he has not been arrested for killing again. And considering how many murders go unsolved each year, that's cold comfort indeed.

Besides, we know of many cases where a murderer loosed by a disfunctional CJS did kill again; that scenario is far more common than even the most exaggerated claims that an "innocent man was executed." In fact, I suspect that we actually know of more people killed every year by murderers who somehow get out of prison than the total number of people executed in all fifty states in that same year, including all the guilty ones. So it's very likely that more innocent life is taken because our system is too lax on murderers than is taken because it's too harsh.

This is the side of the equation that folks who call for liberalization of the CJS rarely consider with the seriousness it deserves. Pat is no exception; he discusses the possibility on another post [Post 3], but the discussion is facile and explicitly avoids quantification:

Is the death penalty a deterrent? I think it is. However, I am not convinced by the study cited by Xrlq. I have not reviewed the study, but I am highly, highly skeptical of any “study” that purports to quantify the number of lives saved by each execution — just as I am very skeptical of “studies” that purport to show that the death penalty has no deterrent effect. I simply don’t believe that any study can quantify such intangibles with anything approaching scientific precision. I am confident that any reasonably competent expert could probably take apart the study cited by Xrlq and demonstrate it to be junk science.

Patterico, who is not a statistician or even a mathematician, and who was not even aware of this study before Xrlq drew it to his attention, is nevertheless "very skeptical" of it or of any other study that "purport[s] to quantify the number of lives saved by each execution."

Why is he skeptical? Certainly not on the basis of having studied them with sufficient expertise to pass judgment on the methodology or analysis. It's hard to escape the conclusion that Pat is "skeptical" because the study clearly conflicts with the decision Pat had already arrived at before he saw the study. (I'm sure that Pat is being completely honest in saying he is skeptical; I'm questioning his accuracy, not his veracity!)

Here is the core of Pat's argument why he thinks we need this likely impossible standard of "beyond all possible doubt," or (as he phrases it later), "proof to an absolute certainty" -- which is, of course, just as bad. Early in [2], he argues:

[I]f it is ever shown that we have executed an innocent person, that could be the beginning of the end of the death penalty in this country. [Emphasis added, here and elsewhere except as noted]

But later, he "repeats" (those are air-quotes) the argument thus:

[I]f it is ever proven, with rock-solid evidence, that an innocent person has been put to death, that will be the beginning of the end of the death penalty in this country. Poll numbers already suggest that the public has concerns about innocents being wrongfully convicted. Common sense says that, if a concrete example of an executed innocent came to light and were widely publicized, the polls would swing wildly against the death penalty. It would take time, but such an example would (in my opinion) mark the beginning of the end of executions in this country.

This is a famous fallacy in rhetoric; I call it the Accretion of Certainty Phenomenon: one begins by saying some event might happen; the next time it rolls around (in the same piece), the event now will probably happen; and at last, in the closing argument, we discover that the event will happen.

No evidence has been adduced to explain this magical transsubstantiation from the wafer of possibility to the body of certainty. It's just the "snark" principle, from the Lewis Carroll mock-epic the Hunting of the Snark: "what I tell you three times is true." Simply repeating something makes it sound more definite, concrete, emphatic.

But in fact, Patterico gives us no reason to believe that he is correct. If you polled 1500 Americans and asked them if they believe that any state, ever in the history of the United States, has executed an innocent man, I suspect the number who would say "yes" would approach 100%. I haven't done this study, but you can try just asking around among your friends. Ask them, "do you believe that sometime since 1789, when the Constitution was finally ratified, that some state has wrongfully executed an innocent man?" If you can find even one person who says "no, no state has ever in our history executed anyone who wasn't guilty," then preserve him in formaldehyde, because you have a rare specimen of credulity indeed!

And yet, the American public still strongly supports capital punishment -- probably less than in the 1800s, but certainly more than in the 1960s. Many folks will even point to cases that they believe were wrongful executions: blacks hanged in the Jim-Crow South after a kangaroo-court conviction; Sacco and Vanzetti; Caryl Chessman, supposedly the "Red-Light Bandit;" Ethyl Rosenberg (some would even argue Julius; I personally believe both were guilty and deserved to die). Folks by and large know that dreadful mistakes happen and that innocent people have almost certainly been executed, but all but a handful of states support capital punishment anyway.

So why would "common sense" dictate that one more such case "would" result in "the beginning of the end of the death penalty in this country?" (Especially if the "innocent" person wrongfully executed was no choirboy, given how many people believe thugs who steal their cars should be executed.) This strikes me as the same sort of "common sense" that tells us that the world is flat.

But let's get back to the forbidden subject: those who die because a murderer was not executed. In this imperfect world, there are no solutions; there are only trade-offs. Every policy decision is a trade-off: you reduce aspect X only by accepting a higher incidence of aspect Y. In this case, Patterico wants to reduce the number of death sentences; but he can only do so, whether he admits it or not, by accepting an increase in the number of innocents killed by convicted murderers who somehow get loose, either into the general prison population or else back into society itself.

Like it or not, even a sentence of life without possibility of parole (LWOP) does not mean the person will actually serve life. There are a number of situations that can allow that person back into society at some point in the future:

  • A future court might rule that LWOP constitutes "cruel and unusual punishment" or that it was unequally applied.
  • A future liberal legislature might remove that punishment from the code.
  • The prisoner might be released by accident, or he might escape.
  • A future governor might pardon him or commute the sentence, even without bothering to review the individual cases (as Gov. George Ryan did in Illinois in 2003, and Gov. Winthrope Rockefeller did in Arkansas in 1971, in both cases commuting death sentences to ordinary life sentences -- not LWOP).

Each of these has happened a number of times in the past. Every few years, Charles Manson comes up before the parole board. Why? He was originally sentenced to death; but the California Supreme Court, under Chief Justice Rose Bird, nullified the death penalty and changed all the death sentences to ordinary life (not LWOP, which did not exist as a distinct sentence at that point). Manson is still in prison; he's so high-profile, he'll likely never get out. But how many death-row inmates were released on parole following the Bird decision? And how many of them went on to kill again?

Unless Patterico has definite evidence that this number is significantly lower than the number of innocent people executed in California during that same period, he is wrong to dismiss the argument merely on his gut-feeling of skepticism.

So let us not make the same mistake that Patterico made; we'll leave the statement one of probability: it is likely that significantly more innocent people have been killed by murderers who eventually got out of prison than by the wrongful execution of innocent people. If Pat wants to dispute this specific point, perhaps he can start by estimating how many innocent people are typically executed per year.

Now to the point I said I would discuss below; this is "below" enough, so let's get into it! I said that the standard Patterico suggested be required for the death penalty, "proof beyond all possible doubt" (a.k.a. "proof to an absolute certainty"), was in fact functionally impossible to achieve (by which I mean you might get lucky occasionally, but not as a rule). Not surprisingly, Xrlq made the same point. Here is Patterico's rebuttal [3]:

I think Roberta, a commenter to my original post, understands what I mean. In a comment to that post, Roberta says she prosecuted a death penalty case while personally applying the “beyond all possible doubt” standard — and she obtained a death verdict. All the while, she presented the case to the jurors just as if she had been required to prove the defendant guilty beyond all possible doubt. Even if the jury had been instructed according to my proposed standard, I’d bet that she would have obtained the same result.

And the conclusion of that paragraph?

I think Roberta’s experience indicates that my more stringent standard could work in the real world.

But the point is that Roberta did not test the Patterico Principle in "the real world," because that is not the standard the jury was instructed to apply. She may have thought in her own mind that's what she was proving; but how can we possibly know that's what the jury thought? Do we know that they didn't have doubts, discuss them in the jury room, and conclude that their doubts were not reasonable? How would Roberta know that... was she present while the jury deliberated?

Even Patterico was only willing -- at first -- to say he would "bet" that she would still have gotten her death sentence. But after the Accretion of Certainty, he concludes that her experience "indicates" that his standard would work. "Indicates" is too strong a word for this vague chain of assumptions, however. This example is meaningless as a test, because the whole point is whether a jury would ever impose the death penalty if they had to decide guilt to the preposterous standard he proposes, not what the DDA muses about while she's presenting evidence.

So in fact, Patterico has no evidence at all, not a shred, that any jury would ever find a defendant so guilty that the death penalty would apply. And certainly, even Patterico would readily admit they would do so far less frequently than they do under the reasonable-doubt standard... which means fewer death-penalty convictions, hence more murderers with life or LWOP -- hence likely more murderers let loose eventually. And some portion of those would kill the innocent again. The only question is how many, and how many wrongful executions of the innocent would be prevented by the Patterico Principle.

Given that I think it highly probable that virtually no jury anywhere could find a defendant "guilty beyond all possible doubt," which means a de-facto nullification of the death penalty, I find Patterico's alarmist conclusion amusing (in a head-shaking sort of way):

[I]f it is ever proven, with rock-solid evidence, that an innocent person has been put to death, that will be the beginning of the end of the death penalty in this country.

In fact, the Patterico Principle itself will be "the beginning of the end of the death penalty," if it's ever adopted. Since I know Pat supports the death penalty, I suggest he simply has not correctly thought through the implications of his own proposal.

By contrast, I want to see executions of those on death row easier and quicker than they are now. I believe that tremendously more innocent people are killed by our catch-and-release policy for murderers than are wrongfully executed by out of control cops, prosecutors, judges, and juries. I believe deterrence is only effective insofar as people see the punishment actually being meted out. (I also want to see caning, à la Singapore, for those habituals who see prison as free R&R with three hots and a cot thrown in.)

I want to see inmates of death row being executed at a faster pace than new ones are sent up, so we whittle down the huge surplus of dead men walking.

Will this mean some innocents are executed? Oh, without a doubt. But I believe the trade-off will be fewer innocents killed overall. And unlike the Patterico Principle, that is a trade-off that actually benefits both society and also the individuals within it.

Note: This is my conservative suggestion. If anyone cares to see my libertarian suggestion, be forewarned: it's even more gruesome and troubling to folks like Patterico than this'n!

Hatched by Dafydd on this day, November 22, 2005, at the time of 6:28 PM

Trackback Pings

TrackBack URL for this hissing: http://biglizards.net/mt3.36/earendiltrack.cgi/252

Listed below are links to weblogs that reference Kill Them All:

» So what's the libertarian solution? from TriggerFinger
Dafydd, the Big Lizard, has what he calls a conservative solution for the death penalty (in response to an article from Patterico).... [Read More]

Tracked on November 26, 2005 5:47 PM

» Is Fear of Executing the Innocent Driving Down Death Penalty Support? from Big Lizards
This appears to be the general worry underlying Patterico's proposal, over on his blog almost two years ago, that we only allow executions when defendants are found guilty "beyond all possible doubt," rather than merely "beyond reasonable doubt." He an... [Read More]

Tracked on November 28, 2005 8:19 PM

Comments

The following hissed in response by: cdquarles

Excellent post Dafydd.

I believe that you understand Type I errors and Type II errors in statistics (or medicine). To me, the criminal justice system must operate using procedures that minimize both kinds. Type I errors in this instance are executing people not guilty of a capital crime (false positives). Type II errors are not executing people who are guilty of capital crimes (false negatives).

My experiences with lawyers and the justice system show that they don't comprehend the differences and the consequences for justice of said lack of comprehension (positive predictive value). Highly sensitive tests often yield lots of false positives. Highly specific tests often yield lots of false negatives. Tests typically fall into either camp. It is rare for a test to be both (and when it is, this is something highly situational...that's life in the statistical universe when drawing inferences or making conclusions).

Our system is already biased toward Type II errors. Pat's system simply isn't workable in the setting of capital punishment. Capital punishment is justified. A murderer should be executed, period. A spy or a traitor in wartime should be executed, period (after extracting intel).

I would be interested in your libertarian suggestion, Dafydd (just to see if it comports with mine).

The above hissed in response by: cdquarles [TypeKey Profile Page] at November 22, 2005 10:11 PM

The following hissed in response by: Dafydd ab Hugh

Cdquarles:

Oh, my basic libertarian response to this is to make it an affirmative defense against a charge of murder or homicide to argue that the victim "needed killin'."

It would be an affirmative defense; the burden of proof would be on the defendant to prove this was true. And of course, there would have to be some sort of review to ensure that you didn't have racist lynchings where the racist defendant assures the Klansmen jury that the uppity black needed killin' cuz'n he had a white girlfriend.

But with a few safeguards, I do believe there are a lot of folks who just plain need to be shot -- and I'd rather they were shot by the people who know them and have personal knowledge of their crimes and perfidies than executed by a faceless bureaucracy that can be led around by the nose.

I have the feeling that if Patterico reads this far (into the comments, I mean), his shriek -- like that from deep 'neath the crypt of St. Giles -- will resound for miles. (Though for a different reason, one hopes.)

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 22, 2005 11:05 PM

The following hissed in response by: Patterico

I'll respond more fully when I have time. For now, I will simply encourage anyone who finds my argument wanting, based on Dafydd's description of my posts, to be sure to read my posts in their entirety before coming to a firm conclusion.

In a nutshell, I think Dafydd gives too much weight to utilitarian arguments and too little to the context of the fact that the government is putting people to death. We could avoid many more murders still by executing criminals who have not murdered; the issue is whether society would find it just. I also find his criticism of my various tenses (could vs. will, etc.) to be excessively nitpicky -- and would rather see more factual refutation and less guessing about my motives in making a particular argument or another. Finally, I think the concern is about putting innocent people to death in the modern era. Of course everyone believes injustices happened in the era of lynchings.

The above hissed in response by: Patterico [TypeKey Profile Page] at November 22, 2005 11:18 PM

The following hissed in response by: Patterico

For example, I wonder if people reading Dafydd's post would assume that I am a skeptic on the very idea of the death penalty as a deterrent. If I had not written my posts, and were familiar with them only through Dafydd's description and selective quotes, I think I might well conclude that Patterico rejects the concept of executions as a deterrent. I wonder if your readers would be very surprised to learn that, after I rejected Xrlq's study as likely placing too definitive a quantity on something unquantifiable, I went on to say in the very same post:

That said, I agree with the assertion that the death penalty likely deters murders. I find this intuitively obvious. For one thing, attaching a high price to premeditated murder shows that we take the crime seriously. Inevitably, this will have an immeasurable but real effect in shaping society’s attitudes towards that crime. Also, there is no question that the death penalty specifically deters murder in some cases. Murderers sometimes murder others in prison, or escape and murder people outside of prison. If they’re dead, they can’t do that.

Viewed from a strictly rational viewpoint, if we believe capital punishment deters, perhaps we should be willing to live with some risk of mistake.

But politics is not always rational, and we have to be realistic. If evidence of one or more wrongful executions comes to light, we may well lose the death penalty, as I argued in my post. I just don’t think the public would stand for many such examples.

I stand by that statement, though I should note that I am referring to the modern era. Raise your hand if you would be disturbed to learn that we had executed a completely innocent person in the modern era. I intend to ask just that question on my own blog.

The above hissed in response by: Patterico [TypeKey Profile Page] at November 22, 2005 11:27 PM

The following hissed in response by: Patterico

But thanks for saying I am fascinating to read!

The above hissed in response by: Patterico [TypeKey Profile Page] at November 22, 2005 11:42 PM

The following hissed in response by: Jay Tea

Dafydd, excellent post. But you overlooked one aspect of the LWOP sentence: the inmate is literally free to do whatever he wishes, without fear of further sanction. Such a scenario is being played out in Massachusetts right now, where LWOP murderer Joseph Druce is facing trial for murdering a convicted pedophile priest in the ex-priest's cell, after stalking him, following him into his cell, jamming the door shut, and brutally beating him to death in full sight of guards stymied by the jammed door. (Massachusetts has no death penalty.)

In a perverse sense, Druce is being rewarded for this crime. Instead of day after day of the same monotonous routine, he now looks forward to being taken out of prison and taken to court. He gets to change his clothes, get some fresh air, see new faces, all the things that take away some of the sheer hell of oblivion he was suffering before.

And what is the end gain to society for the expense of the trial and giving him his "day in court?" At worst, he'll be given a second LWOP sentence -- a difference without distinction. What matters it how many LWOP's one is currently serving? One is not locked up until death, then left in the cell for another lifespan -- and even if one was, would it really matter to them?

Since executing Druce wasn't an option, I thought they simply should indict him, then simply hold off prosecuting as long as possible. That way, in the incredibly off chance his original conviction was challenged, they could use this murder to keep him locked up. Otherwise, use their limited resources on cases where they can actually make a difference.

And toss Druce down a deep, dark hole for the rest of his days.

J.

The above hissed in response by: Jay Tea [TypeKey Profile Page] at November 22, 2005 11:58 PM

The following hissed in response by: cdquarles

Dafydd,

Your libertarian response does match mine.

Patterico,

I read your post, and I agree with Dafydd. The justice system needs to maintain due process with Type I and Type II errors in mind. The system currently is biased to make Type II errors. If, after full due process someone who is truly innocent has been executed, that's a tragedy. Personally, I doubt that this has happened in the modern era (given the many years it takes to execute someone these days); and it wouldn't change my mind about the death penalty a whit if it has.

The above hissed in response by: cdquarles [TypeKey Profile Page] at November 23, 2005 12:41 AM

The following hissed in response by: Dafydd ab Hugh

Patterico:

Pat, you completely missed the point of the section of my post you grouse about.

It's irrelevant whether you believe capital punishment deters; I know you support it, and I said so right here. But you refuse to quantify the extent to which it deters... indeed, you go so far as to express skepticism (based upon nothing but your gut feeling) that it's even possible to quantify the deterrence factor.

But if you will not quantify how many innocent lives are saved by the death penalty (let's call this D), then you cannot compare that figure to the number of innocent people wrongly executed (a number which you also refrain from hazarding a guess at, but let's dub it E).

I didn't read Xrlq's argument, but I would imagine it was the same as mine: that the figure D is considerably greater than the other figure, E. That is, that with all its flaws, I believe the death penalty saves far more innocent lives than it takes.

You seem greatly exercised by the fact that figure E represents lives taken by the government, while ~D (those killed because we don't execute murderers) represents lives merely taken by liberated criminals. I admit, you have me there; since I fail to see the distinction -- except the deaths in E are much kinder than the deaths in ~D -- I can't for the life of me get your point.

I know there is a legal aphorism that it's better that a hundred guilty men go free than a single innocent man be wrongly convicted. But would you push that so far as to say it's better that a hundred innocent people be murdered by a killer set free -- than that a single innocent person be wrongly executed?

This seems a perverse accounting of the value of innocent life indeed, Pat: my life has a different value depending who exactly is killing me. In fact, this smacks of special pleading: "we the people" need not feel any responsibility for those murdered by criminals, even if it was our actions that let those criminals loose in the first place; but we must feel completely responsible for wrongful executions, even though they were done purely by accident by people who believed they were executing the actual evil-doer.

This double-standard for the killing of innocents renders your argument incoherent, as there is no way to respond to the suggestion that life is valued, not by how it is lived, but by who terminates it.

I'll stick with my "utilitarian" argument: more innocent people killed = bad; fewer innocent people killed = good.

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 23, 2005 5:41 AM

The following hissed in response by: Patterico

A utilitarian perspective is a perfectly fine basis for a philosophical argument, but it is also completely at odds with the philosophical basis underlying our criminal justice system. Our system is based on the idea that guilt must be proved beyond a reasonable doubt.

If the defendant sitting before you *probably* committed one or more heinous murders, and will *probably* kill again if released -- but his guilt cannot be proved beyond a reasonable doubt -- then he must be acquitted under our system. Looking at matters from a purely utilitarian perspective, he must be convicted and executed.

So your problem is not merely with my argument, but with the entire philosophical construct of our criminal justice system.

Now, you may cheerfully admit that -- and if you do, then you have a sound and consistent theoretical argument. But you need to understand the full implications of your fully utilitarian view -- a view which, by the way, counsels for a *lower* burden of proof in murder cases than in normal cases.

Back to reality: most citizens are largely sympathetic with the principle of reasonable doubt, because Americans tend to be distrustful of government. You could achieve fewer murders in L.A. by passing a law authorizing execution of each defendant charged with any gang-related crime, whether it's selling drugs, tagging up a wall, or murder. Since roughly half of all homicides in L.A. are gang-related, mass executions of this sort would certainly have the effect of lowering the murder rate. (You'd find that you ended up executing many of your future victims as well as your future murderers too, but that's another story.) I would certainly oppose any such regime, despite the utilitarian benefits, and so would most citizens. I'm not sure where a pure utilitarian like you would come down. At what point do issues of excessive punishment for the crime enter your analysis, if at all?

The above hissed in response by: Patterico [TypeKey Profile Page] at November 23, 2005 6:59 AM

The following hissed in response by: Dafydd ab Hugh

Patterico:

I didn't call my argument utilitarian; I called it "utilitarian," as I was simply quoting you!

This is another fallacy: the straw-man argument. You know very well that my argument is precisely for maintaining the reasonable-doubt standard. You are the one who dubbed it utilitarian, then used your personal definition of utilitarian to make the silly claim that since you called my argument utilitarian, and since you believe a real utilitarian would use a lower standard than reasonable-doubt, that therefore I must be arguing for that lower standard.

Instead of dancing around my actual argument, why don't you confront it head-on?

1) How many innocents does capital punishment save each year?

2) How many innocents are executed each year?

3) Is (1) - (2) a positive or a negative number?

4) How many actual murderers would get a sentence other than death if juries were forced to decide on the basis of "beyond all possible doubt" instead of the reasonable-doubt standard?

5) Realistically speaking, how many of these killers would eventually end up back on the streets after they didn't get the death sentence?

These are the questions you need to confront, not the silly semantic game of trying to make it appear as though I'm arguing for a standard other than reasonable-doubt.

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 23, 2005 1:24 PM

The following hissed in response by: Patterico

I am not dancing around a thing, my friend, but rather trying to get *you* to confront issues about your own argument that you have thus far failed to confront. (I won't say you are "dancing around them" as such talk is counterproductive.)

Your own definition is "more innocent people killed = bad; fewer innocent people killed = good." Whether you dub this utilitarianism (with or without quotes) or something else entirely is irrelevant to me. (It seems to me to be squarely utilitarian, and I'm not sure why you seem to shy about acknowledging that, but never mind that for now.) We'll go with your definition.

What you should acknowledge is that the reasonable doubt standard, which you say you support, most definitely results in more innocent people killed than would a lower standard of proof. It also means more more innocent people burglarized, more innocent people robbed, more innocent people raped, and so on and so forth.

However do you justify retention of the reasonable doubt standard if these are your only guiding principles?? If these are your principles, you should explain why they don't lead you to argue for a weakening of the standard.

*My* articulated principle is attempting to avoid execution of the innocent while trying to retain the death penalty. My standard -- which I believe to be workable -- is the best way we can attempt to achieve this idea in the real world, in my opinion.

My concern is twofold: 1) a desire to avoid the moral horror of the state deliberately putting someone to death who is factually innocent of a murder; and 2) a concern that the horror many in our society would likely feel upon encountering proof of #1 could have the effect of losing us the death penalty. I do not see either of my concerns addressed in your analysis, except in the portion of your post which mocks my word usage of could vs. would and so forth.

I recognize that *my* standards may result in more innocent people dying, though I can't say how many. It's all a trade-off.

Then again, my standards might save lives. If we lose the death penalty entirely, because society junks it due to a concern for innocents being executed, then more innocent victims will die as a result, I think we agree. That should factor into the analysis.

You mock my skepticism about studies that purport to quantify the number of innocent lives saved due to the death penalty, by noting that I am neither a mathematician nor a statistician. This appears to be a suggestion that bloggers' opinions on matters outside their professional ken should be discounted. I reject that argument, but if we are going to apply it, then I will note that Roberta and I have more experience with the criminal justice system than you. We believe that my standard is workable and you do not. But you have tried zero murder cases.

Enough for now. I will be curious to see whether you confront my actual points head-on, as you accuse me of failing to do with you.

The above hissed in response by: Patterico [TypeKey Profile Page] at November 23, 2005 4:17 PM

The following hissed in response by: Dafydd ab Hugh

Patterico:

However do you justify retention of the reasonable doubt standard if these are your only guiding principles?? If these are your principles, you should explain why they don't lead you to argue for a weakening of the standard.

This is (again) the entire point of me denying that I am a Utilitarian: I do have principles beyond the Jeremy Benthemism... hence the scare-quotes when I used the word that you (not I) had picked out for me.

I believe in abstract justice, decency, courage, and so forth, even when they conflict with happiness. But since I'm not in the habit of writing 3000-word essays as comments to my own blog, I didn't think it necessary to spell out everything in detail. But since you insist, it is precisely those other principles that prevent me from lowering the standard from reasonable-doubt to likely-guilty or possibly-guilty or not-proven-innocent.

Oh, and I likewise dispute this claim of yours:

What you should acknowledge is that the reasonable doubt standard, which you say you support, most definitely results in more innocent people killed than would a lower standard of proof. It also means more more innocent people burglarized, more innocent people robbed, more innocent people raped, and so on and so forth.

No, I don't believe it does. A pluralistic society such as ours that jails people left and right, in a way that causes the populace to lose confidence in the basic equity of the system, will lead to far more crime than a system where people do have such confidence.

But by the same token, the system must not only give folks confidence that the innocent won't be wrongly punished -- it must also give citizens confidence that the guilty will not be wrongly set free or given too light a punishment for their crimes. And this, at the moment, is unquestionably what agitates the citizenry the most: the idea (real or imagined) that criminals are quite literally getting away with murder -- as well as getting away with robbery, rape, assault, theft, fraud, income-tax evasion, and so forth.

*My* articulated principle is attempting to avoid execution of the innocent while trying to retain the death penalty. My standard -- which I believe to be workable -- is the best way we can attempt to achieve this idea in the real world, in my opinion.

But I think you're wrong on both counts, and that was the point of my original post.

My concern is twofold: 1) a desire to avoid the moral horror of the state deliberately putting someone to death who is factually innocent of a murder; and 2) a concern that the horror many in our society would likely feel upon encountering proof of #1 could have the effect of losing us the death penalty. I do not see either of my concerns addressed in your analysis, except in the portion of your post which mocks my word usage of could vs. would and so forth.

Sure you do; you just don't recognize it when you see it.

I noted that we certainly can quantize the number of people saved by the death penalty (because we have). The quantization may be flawed, but that's why you use multiple methods and something like regression analysis to get a reasonable range.

We then estimate how many are wrongly executed each year and compare the numbers. If they're close, you might have an argument. But if the first is thousands of time more than the second, I don't think you have a leg to stand on.

Here's more addressing: I'm all in favor of using DNA evidence whenever it's available, and I certainly wouldn't want a system that didn't allow for appeals. But on the other side of the coin, sentences should mean something, they should be harsh and unpleasant, they should not be indefinitely delayed by two dozen appeals, they should be public (so folks can see that justice is being done), and they should of course be consistent among all perpetrators of the same or similar crimes... Joe shouldn't get a lighter sentence for tax evasion than Pete merely because Joe is in all the right clubs, while Pete is an annoying science-fiction fan.

The standard of guilt should be neither too low nor too high. For centuries, we have been satisfied with the "beyond a reasonable doubt" standard; and barring some extraordinarily persuasive argument that we should change it, I prefer we keep it.

Then again, my standards might save lives. If we lose the death penalty entirely, because society junks it due to a concern for innocents being executed, then more innocent victims will die as a result, I think we agree. That should factor into the analysis.

Sure thing. Do you see this happening right now? I don't see any evidence of it.

You mock my skepticism about studies that purport to quantify the number of innocent lives saved due to the death penalty, by noting that I am neither a mathematician nor a statistician.

Pat, you tend to see any counterargument as "mocking" you. I stated then and I state now that you have no technical expertise to analyze these studies. Even I barely do, because my specialty wasn't sadistics.

If you had cited other criminologists who rejected the studies, that would be rather better; but you didn't.

This appears to be a suggestion that bloggers' opinions on matters outside their professional ken should be discounted. I reject that argument, but if we are going to apply it, then I will note that Roberta and I have more experience with the criminal justice system than you. We believe that my standard is workable and you do not. But you have tried zero murder cases.

But murder cases are not tried before a panel of lawyers... they're tried before ordinary citizens -- people like me -- and I know for a fact that if I were required to find someone guilty "beyond all possible doubt," I could never do it; I would vote "not guilty" every time.

Suppose we were discussing instead the plausibility of a particular motion to be granted by a judge. In that case, your reductio above would in fact be correct: as a non-lawyer, it would be absurd for me to tell you whether a motion to suppress blood evidence found outside the residence of the defendant was going to be granted or denied.

But lawyers don't own the judicial system; it's for everyone, which is why we have juries. And I have as much authority to say what I think is moral as a lawyer, a doctor, or a Marine-Corps colonel does.

This is in stark contrast to a purely technical skill like statistical analysis... that is something you must be trained to do. To suggest equivalence is as wrong as saying that since I wouldn't tell a neurosurgeon how to remove a brain tumor, I have no right to tell a priest he shouldn't steal from the poorbox.

To make it absolutely plain:

1) The reasonable-doubt standard is traditional, accepted, and we have much experience with it... not just the lawyers trying cases but the juries hearing them.

2) We all know that murderers often kill again, if given the chance.

3) We all know that people sent to prison often get out again, even if they were supposedly sent up for "life."

4) We all know that some number of factually innocent people will be executed every decade (I say decade because it may well be less than one per year).

5) Many people, including myself, believe that changing to a super-hard standard of guilt will result in many fewer death sentences.

6) Many people, including myself, believe that fewer executions will lead to more murders.

7) Therefore, many people, including myself, oppose the idea of setting the super-high standard you propose.

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 23, 2005 9:57 PM

The following hissed in response by: Patterico

Dafydd,

You say:

I believe in abstract justice, decency, courage, and so forth, even when they conflict with happiness. But since I'm not in the habit of writing 3000-word essays as comments to my own blog, I didn't think it necessary to spell out everything in detail. But since you insist, it is precisely those other principles that prevent me from lowering the standard from reasonable-doubt to likely-guilty or possibly-guilty or not-proven-innocent.

I believe in the same things, and they lead me to propose the standard I have proposed. We obviously disagree about where to draw the line, and that's fine. But it is not some devastating argument-ending point to note that my standard would cost more lives than your alternative. Your standard would cost more lives than a "clear and convincing" alternative standard of proof. These are all trade-offs.

I have an argument that my standard could also save lives. You demand evidence. I can't prove it definitively, but I note that in comments on my blog, many people seem to argee with me that execution of a clearly innocent person could undermine public support for the death penalty.

You have an idea that lowering the standard of proof below reasonable doubt would be tantamount to jailing people willy-nilly, leading to an undermining of confidence in the system, and to more crime. This is every bit as speculative as my argument, if not more so. Moreover, a "clear and convincing evidence" standard would hardly lead to a system that "jails people left and right, in a way that causes the populace to lose confidence in the basic equity of the system." It would simply lead to more people being jailed.

I have far less confidence in the ability to determine with precision the number of lives saved by the death penalty. You imply in my post that because I had not reviewed a particular study cited by Xrlq, that I am unfamiliar with attempts to do so. I have read several books on the death penalty, including discussions of studies on deterrence. I have never seen any discussion that convinces me that it is possible to quantize such issues; hence my skepticism regarding Xrlq's study.

I fully agree that "lawyers don't own the judicial system; it's for everyone, which is why we have juries." Don't you remember? I said I *rejected* the argument from authority you seemed to make regarding my not being a mathematician or statistician. I said only that if we are going to accept that argument -- which, remember, I explicitly *reject* -- then it applies to discount your observation regarding how a juror would apply my proposed standard.

Your argument does not persuade me otherwise. My expertise is not limited to points of law. I assure you that, whether you realize it or not, I have far more experience in assessing how a jury is going to react to a particular fact pattern, given applicable law. I have tried over 40 jury trials; you have tried none. The art of being a trial lawyer is far more than knowing the law that the judge will use in a motion; it is also making judgments, based on experience, as to how 12 ordinary citizens will vote based on a certain fact pattern and a certain set of instructions. I am hardly infallible on that point, but it is a skill set that I have that you don't.

Lack of a skill set does not make an opinion invalid. But a person who has a skill set who gives a firm opinion on an area within their expertise should not be lightly dismissed, if that person has built up credibility. So, for example, if you simply dismiss my opinion on a death penalty study because I am not a statistician, I won't pay much attention. But, since you have credibility with me, if you use your mathematical background to assess one of these studies and you claim it has merit (something you haven't done), I will listen and not reject your opinions out of hand.

As for how citizens would actually apply my standard, I can't say for sure. I invite you to opine on it, because I reject arguments from authority, but I also ask you not to discount my opinions and those of Roberta simply because you would always vote a certain way.

Your assertion that you would not vote a certain way given my proposed instruction does not answer the question of how 12 ordinary citizens would vote. I mean no offense by this, but you wouldn't likely end up as a juror on a murder case anyway. You are very smart, very conservative, very talkative and outspoken, and don't have an entirely conventional appearance. It's almost certain that you would be stricken by one side or another.

In any event, I have a post that says how someone who actually has been on a jury that assessed the death penalty would apply my standard. It turns out this juror would have convicted using the "beyond all possible doubt" standard. I ask you also not to easily discount this citizen's comments. He hasn't built up credibility by repeated commentary, but he is the only person I have seen weigh in on the subject so far who has actually been there.

I apologize for the disjointed and unorganized comments. I'm on vacation and don't have sufficient time to organize my thoughts better.

The above hissed in response by: Patterico [TypeKey Profile Page] at November 24, 2005 3:10 PM

The following hissed in response by: Dafydd ab Hugh

Patterico:

I have far more experience in assessing how a jury is going to react to a particular fact pattern, given applicable law. I have tried over 40 jury trials; you have tried none.

How many have you tried under the standard you propose? I would guess the answer is "none."

Unless you have actually tried cases where the judge used the standard "guilty beyond all possible doubt," you really have no idea how each member of a jury would take that standard.

Besides, if I take what you said above as my guide, then what would you say if I produced a prosecutor who had far more experience than you, but who flatly disagreed with you. Would you withdraw this argument?

If not, then you're not actually claiming expertise in this area gives you special insight.

I really believe this all boils down to your personal moral code: you are uncomfortable asking for the death penalty unless you are convinced "beyond all possible doubt," however you define that. That is a perfectly legitimate argument, but it's entirely subjective: my moral code tells me that BARD is sufficient.

But, since you have credibility with me, if you use your mathematical background to assess one of these studies and you claim it has merit (something you haven't done), I will listen and not reject your opinions out of hand.

Post a link, and I'll take a look.

I have a post that says how someone who actually has been on a jury that assessed the death penalty would apply my standard. It turns out this juror would have convicted using the "beyond all possible doubt" standard. I ask you also not to easily discount this citizen's comments.

I don't discount it at all: I'm certain there are some jurors who would be convinced to vote guilty under that standard. The problem is that you need all twelve to be so utterly convinced.

My friend Lee has about the same opinion as I regarding most legal stuff; if anything, he is more conservative. And he looks pretty geeky. Yet he has been on four juries, three of them criminal.

He's on holiday now, but when he returns, I'll ask him about your proposed standard and see what he says -- I'll ask what sort of evidence would he require to convict for murder if the standard he had to apply was "beyond all possible doubt," rather than "beyond a reasonable doubt."

(If you want to reword my question, I'll ask your version instead.)

None of this matters, really, as I'm sure you agree that significantly fewer defendants will be convicted under the stricter standard (if you didn't, why would you bother pushing for it anyway?) That means fewer death sentences. You do agree, right?

The deterrant value of a sentence like the death penalty is lessened, not increased, by reducing its likelihood. Your only real argument is your idea that any day now, an utterly clear case that an innocent man was put to death will so enrage the populace that death penalties will be overturned across the country.

Thus, you appear to argue, it's better to keep a tiny number of death sentences for cases, not necessarily of the most egregious or lurid murders, but rather of the murders that happen to occur in front of witnesses or security cameras. But to do this, you sacrifice what will probably turn out to be the lion's share, or at least a very significant chunk of murder cases that today would be considered capital murder: all those in which at least one juror was only convinced beyond a reasonable doubt, not beyond unreasonable doubts.

This is an awfully long chain of contingency, when there has been no sign that (for example) the American people's support for a war is in any way correlated to proven cases where we killed the wrong person... but instead to the loss of American life and the perception of whether we're winning or losing.

My experience tells me that, in fact, the huge majority of people in this country are actually impatient with all the rights we afford to accused criminals -- not fearful that we may accidentally convict an innocent man. And they're certainly not demanding more and more procedural safeguards against such a thing. Can you point to any evidence of such a shift in the public mood about crime, deterrance, or capital punishment?

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 26, 2005 11:22 PM

The following hissed in response by: Patterico

Would you withdraw this argument?

Which one? The argument from authority that I reject anyway?

Can you point to any evidence of such a shift in the public mood about crime, deterrance, or capital punishment?

Not really. My posts on the issue have links to a poll showing a softening of support based on recent evidence of innocents released from Death Row. It still shows strong support for the death penalty, but shows much less support if substantial numbers of innocents were assumed to be wrongfully convicted (it also says those results are unreliable). I simply suspect that the fickle American public might not have the stomach to continue to support the death penalty forever if evidence of a truly innocent person executed emerged.

You're right that my moral code plays a part in this. I do encourage you to read books about wrongfully convicted people. You could start by renting the movie The Thin Blue Line. Then read Adams v. Texas, on which it's based. I have other suggestions for books to read in my posts on the issue.

The above hissed in response by: Patterico [TypeKey Profile Page] at November 27, 2005 3:40 AM

The following hissed in response by: Dafydd ab Hugh

Patterico:

You're right that my moral code plays a part in this. I do encourage you to read books about wrongfully convicted people. You could start by renting the movie The Thin Blue Line. Then read Adams v. Texas, on which it's based. I have other suggestions for books to read in my posts on the issue.

It's irrelevant. I have always assumed that some innocents will wrongly be executed, no matter how careful we are with the death penalty; I've always believed that Caryl Chessman was innocent of that kidnapping/rape, and my father first told me about that case back in the 1960s (my father was an attorney, and he observed several of Chessman's appeals).

That appalling case resulted in the repeal of the "Little Lindbergh" Law, but not the repeal of the death penalty itself for, say, murder. If that tragic circus act could not shake Californians' support of the death penalty, I do not believe a case like the one you discussed would even come close.

Reading about heart-rending cases can only serve to inflame my sentiments and passions -- and I choose not to base my morality on passion or sentiment.

We have struck a dynamically stable moral position with the "reasonable doubt" standard plus updating our views of old cases whenever a fundamentally new forensic technology becomes available. And for deterrence, if anything, we err too much on the side of caution.

I'll wait to embrace a change away from that standard when I start seeing significant shifting of the public's attitude towards the death penalty... not in a poll but in voting patterns. Not before then.

I think it would actually be fairly easy to cut through an initiative or legislative effort to eliminate the death penalty: a television commercial would show a few of the most appalling, revolting mass and serial murderers we've had in the past twenty years, from Richard Ramirez, John Wayne Gacy, Ted Bundy, and Jeffrey Dahmer to Timothy McVeigh, Musab Zarqawi, and Osama bin Laden, each with the number of innocent people they butchered superimposed over their faces. The VO says "which of these killers would you select to live?"

The initiative would go down in flames; and if the lege enacted a ban, that would go down as soon as an initiative could be sent round and put on the ballot.

As I said, if this is primarily a moral position of yours, there is of course no argument possible: your position is prima facie moral and rational, even if I don't exactly share it (so's mine, even if you don't embrace it).

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 27, 2005 6:23 AM

The following hissed in response by: Patterico

Reading about heart-rending cases can only serve to inflame my sentiments and passions -- and I choose not to base my morality on passion or sentiment.

It could also serve to inform you as to how and why wrongful convictions happen, and to whom. Isn't more information always helpful? Are you so doubtful of your own commitment to rational thought that you are unwilling to further inform yourself because of the danger that doing so might inflame your passions?

The above hissed in response by: Patterico [TypeKey Profile Page] at November 27, 2005 11:47 AM

The following hissed in response by: Dafydd ab Hugh

Patterico:

It could also serve to inform you as to how and why wrongful convictions happen, and to whom. Isn't more information always helpful?

No. Why should it be? More food, more water, and more air are not always helpful.

In this case, what difference does it make? I'm not responsible for fixing the problems of the CJS, so it's irrelevant to me exactly why some bloke was wrongly convicted in the 1970s (because the cops fabricated evidence). I'm perfectly willing to believe that happens more than even you think it does.

But that doesn't alter the fact that we need a death penalty -- which of course you too support -- and that the standard I support is "beyond a reasonable doubt." How can it possibly aid either justice or society to refuse to execute people based upon unreasonable doubts?

(By the way... how would your BAPD standard prevent the cops from literally fabricating evidence? If the standard is attainable at all, then it's just as susceptible to fraudulent testimony as the BARD standard.)

Are you so doubtful of your own commitment to rational thought that you are unwilling to further inform yourself because of the danger that doing so might inflame your passions?

Not fear; I simply use the right tool for the right job. I know that passion and rationality are inimicable: the more the former, the less the latter. I don't avoid either sentiment or passion... what I said was I refuse to allow either to decide my moral code.

"Affirmative action" comes from allowing passion for racial equality to override one's sense of justice. Welfarism derives from the sentiment that all those people just need a "helping hand." In both cases, logic indicates they don't work -- but when the heart just bleeds for people, logic is out the window, and soon you have all the things we both hate. Passion creates the overwhelming sense that we have to do something about some state of affairs -- even when "doing something" will only make things worse.

I save my passion and sentiment for daily life, for Sachi and for movies, literature, art, music, and food, where it belongs -- not for deciding moral and ethical issues or scientific questions.

(By the way, I have actually seen the Thin Blue Line and also read a book about the Chessman case. And the Serpico case. And I followed with great assiduity the wave of fraudulent child-molestation cases similar to the McMartin Day Care cases. I'm hardly unfamiliar with the ways in which the legal system can go wrong. But that, too, is irrelevant to this point; all human systems -- including the one you propose -- will "gang aft agley" in some calculable number of cases.)

Dafydd

The above hissed in response by: Dafydd ab Hugh [TypeKey Profile Page] at November 27, 2005 1:12 PM

Post a comment

Thanks for hissing in, . Now you can slither in with a comment, o wise. (sign out)

(If you haven't hissed a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Hang loose; don't shed your skin!)


Remember me unto the end of days?


© 2005-2009 by Dafydd ab Hugh - All Rights Reserved