Nobelist Kary Mullis once asked for a reference paper with the simple statement "HIV causes AIDS." This article reviews the modern argument for the HIV/AIDS hypothesis, covering main lines of evidence from human epidemiology and experimental animal virus research. Special attention is paid to the issue of how AIDS may be defined so that the possibility of AIDS without HIV may still be theoretically discussed. Major emphasis throughout this article is placed on the arguments of modern HIV/AIDS skeptics, Peter Duesberg and Robert Root-Bernstein, who do not believe that HIV has a central role in AIDS. It is concluded that HIV/AIDS skeptics have chosen overly broad definitions of AIDS which are not clinically useful, and which would, if employed, result in many confusing diagnoses of "AIDS" and "HIV-free AIDS" in people with good prognoses. HIV is one of a closely-related family of viruses which causes AIDS-like immunodeficiency diseases in a number of animals species, and HIV/AIDS skeptics have ignored or minimized this research in order to construct needlessly complicated alternative hypotheses for the cause of AIDS. These alternative views are based on correlations between AIDS and toxin exposure shown by epidemiologists to be artificial a decade ago, but which skeptics still refuse to abandon. Examination of the HIV/AIDS controversy thus allows us to draw some general lessons about how skepticism in science works, and the ways in which it can go pathologically awry.
