Troy Davis

So, the main thing to say is that it's horrible, a tragedy, an atrocity, one of the worst miscarriages of justice I've ever heard of, and kind of disgusts me with the entire American criminal justice system. I'm also proud to live in two states that don't have the death penalty, and even prouder to have contributed to making that so in New Jersey. Last night was one of my least favorite experiences ever.*

(I wrote this part yesterday, Thursday, but then stalled because of lots of work to do and I'm only getting around to finishing this tonight. I'm not changing the time frame, though.)

But I want to do a little more than just say that. I think there is one very specific lesson to draw from the whole Troy Davis thing, and I think it would be nice if we were to draw it as a nation.


The argument, morally if not legally, against the killing of Troy Davis was very simple: he was innocent. At least, we have every reason to think he was innocent, and at most we have no reason to think he's guilty. And in general the idea of the government killing a random innocent black man is considered, well, not cool. That's quite a simple train of thought, I'd say, with rather obvious intuitive moral appeal. But it's not true that there is no argument on the other side. There is, in fact, an argument on the other side, and it revolves around the notion of wanting to have finality in the system.

Because, these people say, at some point you just have to trust the process. We design ourselves a nice criminal justice system, with ample (some say too ample) protections for and asymmetries in favor of the defendant, we let those convicted of a crime appeal their sentence, if the government conducts the trial in a way that violates the rights of the accused then we overturn the verdict, etc. That system will never be perfect: it doesn't even try to make the number of guilty people who aren't convicted get close to zero, and though it does try to avoid even one false conviction it does not, cannot, and will never succeed at meeting that threshold. That's just a fact of life, and one we more or less have to accept. If it were the case that if, any time after a defendant was convicted, they could show that now there was reasonable doubt as to their guilt, we released them, then the trial would have very little meaning at all. Maybe a key witness died. Maybe there was a flood in the evidence building, and the forensics were destroyed. Evidence decays, as all things do: if the government needed first to prove guilt beyond a reasonable doubt and then to continue to be able to prove guilt beyond a reasonable doubt in perpetuity, we'd get chaos. One particular brand of chaos we would get is that every convict would make infinite appeals on the grounds of innocence post-conviction, clogging up the system horribly.

I think that's a compelling argument for finality in the system. Requiring that the prosecution stay above the reasonable doubt threshold forever is clearly unworkable. There has to be some sort of burden on the convicted criminal to demonstrate that the conviction was wrong. After all, the reason we have the standard of reasonable doubt, or in the old days "moral certainty," is that prior to being convicted a person is presumed innocence. But, uh, after you've been convicted you're not presumed innocent, you're presumed guilty, and rightly so! What a verdict of guilty is is a legal finding of fact that you are guilty; if the law does not then treat you as guilty, then what's the point? Now, some people may want to say that you just are guilty, and get over it; Antonin Scalia, in particular, seems to think that innocence per se can never be the grounds for setting aside a conviction, if there wasn't some constitutional error in the process. Maybe we don't want to go as far as Scalia (in fact, we don't!), but I think it's reasonable to think that there is some burden on the de jure guilty man to prove his innocence. Maybe that's not "beyond a reasonable doubt" or "to a moral certainty." Maybe it's "preponderance of the evidence." Maybe it's more, maybe it's less. I don't know exactly, but I think it's reasonable that there be such a standard. Perhaps if they can show that there's strong evidence that one very specific other person did the deed instead, such that a prosecutor would be likely to get a conviction against that person, that should carry extra weight.

But here's my point: all of the logic of the above paragraph, which I think is completely valid, is utterly meaningless in capital cases. As Nicholas Kristof said on Twitter Wednesday night, "When smart people debate whether or not a man should be executed, that's a good reason not to execute him." In a capital case, I argue, you can't be any less sensitive to newly-unearthed exculpatory evidence than you would have been at the original trial. With capital cases, the entire dynamic is different because the sentence does not cover a period of time, it occurs at one very specific point in time, and can never be revoked or amended in any way beyond that point. How can we carry out an execution if we are not certain we ought do so, to the same standard we would use at the trial for determining if we should carry out an execution? An innocent person in prison who uncovers a modest piece of evidence showing their innocence can investigate further, and hope to uncover more evidence to eventually meet the burden of proof to demonstrate their innocence (side note: I think there might want to be some sort of right to counsel even post-conviction post-appeal). One who is executed when reasonable doubt has been introduced post-conviction is just gone, dead, forever. If we aren't convinced of guilt beyond a moral certainty, how can we justify killing someone?*

So here's what I think: it's all very well and good to want finality in the system, but if you want that then you have to accept not having a death penalty. Or, at the very least, you need not to insist upon that same finality in capital cases. When we're talking about killing, we need to be damn sure, both when we convict and when we execute. We weren't damn sure that Troy Davis was guilty on Wednesday, but we killed him anyway. That's just not an acceptable system. Or, if you find that system acceptable, then you appear not to have much problem with killing someone for what you yourself admit might just be no reason whatsoever (because, let's remember, if Troy Davis was innocent then we killed him for literally no reason at all). And if that's your moral position on the issue, then I honestly just disagree with you and find that opinion beyond the pale.



*I don't think we can justify executing anyone. I'm just saying that if you want capital punishment you need to abandon finality in the system, at least for capital cases.