It doesn’t, because we don’t have a baseline level of effectiveness. And we don’t have curves indicating how long it takes to get a response, etc.
But it remains true that in societies where we have decided that executing people is wrong, the efficacy of executions (either in terms of recidivism or deterrence) is not debated. We don’t execute people because doing so is wrong, not because it is ineffective.
Again, I don’t think this study actually shows that. I mean, per this study, showing people evidence is actually counterproductive, and that to get the quickest results you should just have a chit-chat and not show them what you know and can already prove. Oh, but that’s because maybe you only show recalcitrant prisoners the evidence, and evidence-showing maybe isn’t actually an independent variable… which one could just as plausibly say of torture itself.
And that’s without actually going into the study and seeing what kind of disclosures were made early, what the definition of “early” is, or anything like that. A p[roper experiment on this would be extremely unethical, but would probably be necessary in order to properly understand what is happening.
I’m not going to “urgently” read a 56 page law review article. You could try summarizing things if you think it’s essential reading.
But here’s what I got from a quick look at the conclusions:
Permitting torture in the hypothetical case, in other words, does not answer what one should counsel in the present debate over coercive interrogation techniques in the “war on terrorism.” The hypothetical only highlights the consequentialist balancing of lives in a context cleansed of all other crucial factors. It tells us nothing more than what we already knew – that a tiny risk of catastrophe can swamp all other concerns in a moral judgment. Sensible moral judgment requires a better sociological analysis.
[…]
I submit that one can be a consequentialist and say that torture should be permissible in the hypothetical case, while still holding firm to the view that,nonetheless, torture should be absolutely prohibited, even in the present situation of terrorist threat.
So, your urgently-recommended article goes further than I do, and says that torture should be permissible in a ticking-time bomb situation (I don’t think ti should be, for the moral reasons the author alludes to), but that in practice it should be prohibited for both moral and the real-world practical inability to accurately measure the costs and benefits.
Presumably, so long as we could tighten up on the decision-making process and reduce it to a quantifiable process, objections to torture would disappear, especially since the author concedes that there are some situations where torture does indeed work.