Yes, you're right that a mechanism is required to get beyond mere correlation. There is one, I just didn't go into it. It turns out that the foreskin is an entirely different type of tissue; it's very porous and supposedly provides a great breeding ground for the HIV virus. And I think the same circumcised/noncircumcised split in AIDS prevalence has been observed in India. I don't mean that I necessarily subscribe to the HIV theory per se; I'm not sure all the alternatives have been explored. About 10 years ago Walter Ascher (I think his name is) at Berkeley developed a theory that HIV produces something that mimics T cells. The body then thinks it has lots of T cells and shuts down its own production. That squares with Duesberg's contention that the HIV virus is puny, and can't kill anything. But I've never seen any follow-up of Ascher's work.