The basic contradiction at the heart of the "it's not torture" argument

There are two lines that torture apologists take: either they say that torture is justified in certain circumstances ("it works"), or they maintain that what the US did was not torture ("the United States doesn't torture").

Addressing the latter argument, I think a fundamental problem with it is this: if the methods under consideration--waterboarding, sleep deprivation, exploiting insect phobias, "walling", etc.--are not "harsh" enough to rise to the level of torture, then why do they yield results? If forced standing is really no different than Rumsfeld "being on his feet" for most of the workday--in other words, if forced standing is no big deal--then why would we expect it to make a hardened terrorist spill the beans? So the contradiction is this: if the methods are lax enough to not be torture, then it doesn't make sense that they would succeed in wresting information from an incooperative detainee, and so you would be against using these methods on efficacy grounds. But if they are effective in coercing information from detainees, then there must have been a point at which the detainee decided that he would rather stop the abuse than keep mum about what he knows. But pushing a person beyond this "breaking point" pretty much just is torture.