"Once, somebody brought it up with the colonel. 'Will [the Red Cross] ever be allowed in here?' And he said absolutely not. He had this directly from General McChrystal and the Pentagon that there's no way that the Red Cross could get in: "they won't have access and they never will. This facility was completely closed off to anybody investigating, even Army investigators." ...
Sorry, josef, didn't see this. I touched on some of the reasons why in my post above, but, coincidentally, I was just about to post this essay by Jack Goldsmith in TNR. Goldsmith was the head of the OLC, 2003-4, and has just published what looks to be a really great book (for international law geeks, at least) on his time there. Anyway, his TNR piece, "The Cheney Fallacy", is a much more comprehensive answer to your question than I could ever provide.Really? What makes you say so?
It may be true that individual torturers are sadists, or the policy-makers seek to terrorize populations of people. But torture is primarily an information-gathering tool for the US, as a matter of policy. It's an information economy. And the individual prisoners are the aggregate bits of data that comprise the information-gathering tool. The justification comes a priori: information is good; information in the name of the defense of civilization is better. But torture information is an odd entity in that much of the framework narrative of what is sought through torture comes a priori as well. Bits of data make no sense otherwise. They're meaningless bits of data. Those bits have to fit into a background framework of interpretation, beliefs, assumptions, values, goals, etc. for them to come to make sense. But the information from one torture victim is unlikely to have much useful meaning.
For torture to be effective as an information-gathering method, it has to be institutionalized and used broadly. This is what I mean by aggregation. The information comes from observation of patterns generated by aggregating what the individual victims say and less from the words of an individual torture victim. Meaningful information from torture necessarily comes from torture's institutionalization. Some information is easy enough to verify: you can indeed verify where an individual torture victim says a weapons cache is by going there and finding the weapons or not. This kind of everyday use of torture, however, hasn't been part of the American (liberal) justification of torture. It's too small-scale and doesn't rise to the level of moral urgency of the ticking time bomb scenario, for example, which is the image the institution uses to justify morally its existence. But, while you could possibly verify concrete things like a weapons cache through torture of one person, you can't verify an alleged plot and its full details without torturing broadly and collecting the bits of information - sorting out the consistent from the inconsistent - into a patterned narrative. You would probably want to do this as efficiently as possible, and this organizational and instrumental efficiency would be reflected in the physical spaces of torture.
The generic "Global War on Terror" gives at least rhetorical cover to all sorts of abuses. The torturers may not know precisely what they're torturing for in terms of information, but the War on Terror lends torture a vague sense of purpose and morally urgent necessity. You can see, then, how torture as information-gathering will always have an incentive to torture more and create an increasingly complex narrative constructed from patterns of information. Of course, the value of such a narrative is another thing altogether. The practice and the spaces may be efficient, but in the service of something profoundly confused at a moral and epistemological level. It’s really a kind of deep incompetence starting at the level of the crafting of the GWOT.
I think it's this general quest for information, however misguidedly, that serves to generate the kinds of spaces we see with US torture, rather than there being some generic torture space. If the US could somehow upload detainee's minds to a server, and develop a database out of that, that's what they'd do.
But… I have to give pause on this last point because the logic of torturing for information is ultimately incoherent. I wonder about the extent to which those who crafted the US torture regime were thinking also in terms of deterrence. If they were, then American torture was fated to be exposed to the public at some point.