Posting again because I realised Shodan is using another horrific and flawed assumption: That one reported murder equals one convicted, incarcerated and released felon.
In fact the USDOJ website shows that over the period 1981-1996, the USA managed to convict roughly 400 people for every 1,000 murders committed, reducing Shodan’s still-fallacious recidivism effect by more than half. Unless, that is, he wants to blame unsolved murders on a failure to apply the DP, which would seem rather capricious.
So that’s 187,000 multiplied by 0.4, giving 75,000. We’ve then got to subtract from that all the murderers who were executed (1,000), all of those who died in prison (?), and then we’ll have a figure that still is too high because it still ignores the alternative of LWOP and so forth. Are you sure you’re still willing to hang your hat on this argument, Shodan?
Not his death, his murder. It is a case where putting someone in prison (Christopher Scarver) did not prevent him from committing another murder. So, by extension, putting someone in prison for life without parole would not remove the opportunity of further murders, even of people who were serving time for other, non-capital crimes.
Sure they’re relevant. One of the advantages (as well as the disadvantage) of the DP is that it is irreversible. Once you have executed someone, you cannot get their life back. You also cannot change your mind and release him, as was done with Ed Wein and Willie Horton.
OK, how about Gino Colon? He is allegedly responsible for many of the 37 murders committed in prison in Illinois during the last ten years, including the murder of a prison guard. Or Terry Lee Farmer if you want a cite for prisoners who present a danger to the public in other ways. See also Clarence Ray Allen.
It seems you need to propose a system of mandatory life without parole, remove the option of commuting a sentence from governors, and solitary confinement with no contact with guards, for all murders. Is that your alternative?
Why exactly are you subtracting deaths in prison from the total? The DOJ statistic of 1.2% is of convicted murderers who are re-arrested for murder within three years of their release. Obviously those folks didn’t die in prison, and they were convicted of murder, so that statistic wouldn’t be affected by unsolved murders, or by those who died in prison.
And your 400-out-of-a-thousand figure is not quite accurate. As your cite mentions,
So your figures are off by an undetermined amount.
It is unclear whether your statistic means “1.2% of all murderers are re-arrested within 3 years of release if released”, or “1.2% of all murderers that are released are re-arrested within 3 years”. I assumed the latter, since to me “within three years of their release” implies that it considers only those convicts who are in fact released. Therefore, the need to subtract from the convicts figure those convicts who are never released, either because they are executed or because they are simply never let out. If you can provide the cite for the 1.2% figure then we can hopefully disambiguate matters. It is, however, a triviality, since as pointed out this figure is still irrelevant as regards determining the use of the DP as currently practiced in the USA, and there exists no usable DP error rate data for assessing the system you propose.
Yes, it seems I misread “convictions per 1,000 murderers” as “per 1,000 murders”. However, this indicates that the true figure you should be using is still lower, as this number then merely represents the conviction rate of those cases that make it to trial, and does not include all those murders which go unsolved. Regarding the effect of multiple offenders, just as one murder can be committed by more than one person, so can one person commit more than one crime. Unless it is your contention that every murder is committed, on average, by 2.5 unique felons, then you cannot contend that my figures are so wrong as to make your original claim correct.
My figures are indeed out by some margin, but there are two counteracting effects at work, and in the absence of data, it seems unreasonable to assume that one would be so vastly greater than the other. My data still represents a better attempt at representing the truth than yours, and demands consideration. Frankly I’m a bit surprised at the trouble I’m having simply finding the conviction numbers themselves, but I would hope at this point you would agree that your bald assumption that 187,000 murders means 187,000 convicted and released killers is hopelessly mistaken.
And of course, this argument is still academic, since the total number of convicted and released murderers over this time is not the number we should be using, as both LHoD and I have pointed out.
I believe it is 1.2% of those who are released. If I understand your distinction, which I think I don’t.
Obviously those who are released didn’t die in prison. But see my cites for some examples of murders committed by those who were not released.
I would think that those that died in prison tended to be those with longer sentences, which correlates roughly with egregiousness of crime, which correlates roughly with likelihood to re-murder. Thus I bet that those who are released are expected to be better bets not to re-offend, and those that are not released would skew the re-offense statistic upwards. And keep in mind that the 1.2% does not include those who committed murder in prison and were therefore not released.
I don’t have any data on the average number of people involved per murder. But, as you mention, more than one murder can be committed by one murderer, and therefore the two counteracting effects are also at work there. But if one person commits more than one murder, that would also tend to skew the 1.2% figure upwards as well. That is to say, we may be preventing more than 1.2% innocent lives by executing murderers. The 1.2% is of murderers who re-murder, with no reference to whether they killed one person or twenty. Ed Kemp, for instance, killed two people (his grandparents) and, after being released, killed nine more people. Jack Abbott killed a man, went to prison, was released, killed another, went to prison, and then was released again and killed again. So he would be counted as part of the 1.2%, even though he killed multiple times.
And again, see my cites for people who were never released and still went on to kill again.
I don’t necessarily agree for the reasons listed above, but if you like, let’s say that it is really your figure of 74,000 and include my cites of people who killed while in prison or furloughed. In order to invalidate my argument, you need to come up with 888 wrongful executions before you can claim that, overall, we would not save innocent lives by using the DP. So far, we have - maybe - one, and that one is debatable.
The best statistic I can find regarding prosecutions per murder case is an incidental one contained in the abstract for this report:
So being generous and assuming that both figures represent exactly half their respective national ones (and making the more dubious assumption that 1988 is representative of the whole 30-year period), that would indicate a prosecution to offence ratio of about 1.25 in murder cases. This corroborates BJS statistics indicating that multiple offender cases are more common than multiple victim cases (although this presumably ignores the instances of the same person being involved in multiple, separate homicides). All of this means that we’d have to multiply our ~0.4 conviction rate by the prosecution ratio of 1.25, giving an eventual conviction to (prosecuted) offence ratio of 0.5. So your figure was “only” out by a factor of two, notwithstanding the fact that it ignores cases that don’t come to court at all, and notwithstanding the fact that it’s still irrelevant to the DP as practised anyway.
Well, it is that you seem to be using the (well, a) number for total convicts, then taking a percentage of that number, when the percentage is in fact a percentage of a smaller number (those convicts that are released at some stage). For an exaggerated example, say we have 100 convicts, and 50 are released. If we have a recidivism rate of 10% for parolees, then that quite clearly means that 5 of those parolees will commit offences. You are doing the equivalent of applying the 10% to the whole number, including those who don’t get released, boosting your number of recidivist offences in contrary to the statistic. Murders while in prison are another matter, and would seem to be something that can be addressed in other ways. Additionally, given that even those sentenced to death tend to spend a considerable number of years in jail, it’s not like they don’t get the chance to commit crimes while inside either.
No, and I do not need to, for the reasons below (and here again I must deal with your equivocation of DP as practiced, and your proposal).
DP as practiced:
I had hoped I wouldn’t regret debating relatively minor points of accuracy, ignoring the elephant in the room, but I now do. Here, even the 74,000 figure is massively overstated, as it takes into account thousands upon thousands of convicts who were convicted and released in a DP system. It is therefore vastly unfair to blame their release on the lack of a DP, since they were released in spite of it. Therefore, we must consider only those parolees who would have been eligible for the DP as currently used, but were convicted in a State which did not have it. Their recidivism rate must be compared with the projected error rate were the DP uniformly applied. This is a far, far cry from the calculation you have performed, and even then still (and you have persisted in failing to address this point) ignores the alternative of LWOP. Consequently, the true number of errors necessary to balance the equation is still vastly fewer than the 887 you are presently insisting upon.
Regarding your confidence in the error rate of the DP-as-practiced, I again point out that there is no legal recourse for the dead, and thus the number of errors is going to remain forever legally zero. The fact that the prosecutor him self has said that the DP should never have been sought in Cantu’s case, and that the sole witness has recanted should be enough to consider Cantu’s execution an error in itself, and your willingness to cavil on this point merely serves to illustrate just how difficult it is to assess a genuine error rate. The system has no means for even identifying mistakes, let alone rectifying them; your confidence that the error rate is low rests largely on the fact that the system does not look for them.
The DP as proposed by you:
Here it is the DP’s own error rate which is again at issue, even more so than in the present climate. You are resting entirely on the unwarranted assumption that you could increase the number of executions by two orders of magnitude without an increase in error rate (or for that matter, without the judicial system buckling under the strain). This is an entirely unjustified claim, and one which runs entirely contrary to common sense. This does not mean it’s untrue, but it does mean that it is rather incumbent upon you to demonstrate that your system could cope without increased error. Given the difficulty in assessing even the error rate of the system we can observe, it seems rather unlikely that you’re going to achieve this to any degree of statistical satisfaction. So far, you haven’t even tried.
If you’d like to debate the DP as practiced, then you simply must address the fact that most of the 74,000 were either ineligible for the DP while in a DP state, or were eligible and it was not used, and therefore cannot be considered a failure to have DP on the statutes. On the other hand, if you just want to debate your own proposal of mandatory DP for murder, then you have to address your baseless assumption that a near-zero error rate can be maintained while killing 6,000 inmates a year.
That sounds good to me. Add in the ability for a retrial should significant flaws emergence in the original evidence, or significant new evidence/new evidence collecting techniques be found, and i’m happy.
This also excludes the other less populous counties, regardless of their crime rates, multiple people being tried for one murder at one trial, and does not include (as mentioned) those that murder in prison. And therefore is a tenuous assumption at best, and not particularly probative of much.
But let’s move the goal posts as far as we reasonably can. Assume that the 1.25 figure is right. So the 187,000 murders represent 233,750 murderers. On average, 1.2% of the rest will re-murder in three years or less.
But, instead, we kill them all. What percentage are you willing to assume will be reached of innocent executions if we increase the number of executions that much? Twice as many? Five times as many? Let’s be generous - ten times as many mistakes will be made. So 2,340 of the executed were innocent.
So that’s 2,793 innocent deaths being prevented at a cost of 2,340 innocents executed, for a net savings of 633 lives.
So, why is the more moral system the one where 633 people die needlessly?
Does Shodan ever have contact with actual human beings or is this all academic for him?
Would you be willing to kill a basically harmless person begging for their life because you felt they needed a good killing for something they had done in the past, even if they regretted it?
Well, you see, killing people actually, you know, involves killing people. Did you forget about that?
Is it really your position that only way to better prevent murders is to kill people? We could prevent the sport of baseball from occuring the same way… but there are less costly options we could also look into, maybe?
I mean a percentage where you, personally, draw the line. Not some weighted system of net cost or benefit to society based on hypothetical post-release murder rates, but exactly how many innocent people you are willing to have put to death in order to maintain capital punishment.
As in, “I, Shodan, am comfortable with the notion that ______ out of every 100 executed murderers are actually innocent, but no more.”
For me it’s any percentage greater than zero. What about you?
Because it’s the system in which we’re not making a deliberate calculation to kill people, that’s why. It’s the same reason why we don’t kill one healthy person in order to provide organs to five sick people: a person has an inherent dignity of life, and we may not kill them in order to promote someone else’s welfare.
I agree with her statement, am not in favor of executing murderers, and have no idea how you could unintentionally misunderstand her statement as advocating execution of murderers.
The multiple defendants issue is irrelevant, since the link stated the number of defendants, not trials. The fact that other counties were excluded is irrelevant, since the proportion of the total number is indicated, and that’s all we’re interested in. Once you get some data (as opposed to individual cases) for murder rates in prison, we can include it and examine how the DP measures up, when solitary is an option and DP convicts live in prison for many years anyway.
No! The 187,000 murders represent 233,750 accusations, representing (here comes the 0.4 figure) 93,500 convictions, resulting in some lesser number of parolees (since again you ignore LWOP), representing an absolute maximum (if every single murderer is released, which of course they are not) of 1122 parolees being re-arrested for murder (not even convicted). And since we now know that multiple-offender murders are the most common, that number of arrests most likely represents a still-lower number of victims.
Do you see now how all of these figures are hopelessly malleable? I think your estimation of the increase in error rates under your regime is probably high, but then I think that our present error rate could easily be ten or more times what we perceive it to be, because there is no systemic error checking once the sentence is carried out. For you to set this up as some accurate equation is simply not supportable by the statistics, as at every stage we see vast uncertainties, to the point where nearly nothing is deducable.
Having shown your figures to be repeatedly dubious, and your assumptions about the alternatives to DP to be insistently misrepresentative, it’s frustrating that you’re still clinging to this level of accuracy. It simply is not justified. I think your figure for the victims of recidivism is out by more than a factor of two as demonstrated above, and is then still subject to an uncertainty of easily a factor of two each way. I believe our estimation of the error rates has an uncertainty of an order of magnitude, and as such do not merely believe that these calculations favour abolishing the DP; I believe that they are so uncertain as to be useless as a basis for either position. I am not trying to demonstrate how more innocents are executed than are murdered by recidivists - I am trying to show that you are pushing the figures beyond all statistical reason.
I will happily admit that even were the figures to show a statistically valid net benefit for the DP (which I do not accept), I would still oppose the DP. I do not believe it is the state’s position to decide that it will take it upon itself to execute a certain number of innocents for the greater good; I think it is the state’s duty to serve justice not according to some pragmatic calculation of deaths caused, but by giving each individual case the scrutiny it deserves, and making sure that restitution can be made in the event of a mistake. You assume that our lives are the state’s domain, to be juggled according to some calculation of collective utility bestowed from above. I will enter into a debate on your terms for the purpose of showing that your figures are fallacious, but I nonetheless think that the very effort is anathema to what I would like our society to be. By your reasoning we could open up our very standard of proof to an optimisation equation; we might find that fewer people overall die if we enact the DP if we’re only “pretty sure” that the accused is guilty. Should we? Again, I deny that the state ought to be engaging in this sort of sickening bartering of lives.
Your 2,000 (1,000, who knows?) recidivism victims are undoubtedly tragedies; their deaths should be averted if possible, murdered by people who should be kept off the streets (and indeed can be without resort to the DP). The 2,000 (1,000, who knows?) innocent victims of the DP are strapped down and murdered in cold blood by an uncaring paternalist state for some arbitrary definition of the greater good. Both are evils, but I believe the latter to be the greater evil, for we commit it in full foreknowledge of the results of our actions. We can fight the former while still accepting that we should not commit the latter.