I think about Duesberg
a lot in the context of this thread.

In fact, pretty sure he's come up before.
Actually it's potentially (?) interesting to compare Duesberg to Stan Prusiner at UCSF, who (somewhat) similarly seemed to be off the rails (my postdoc advisor, who ended up at UCSF after his time in Charm City, referred to him as
Stan the Charlatan 
), but ended up being
right -- or at least
on the right track.
I vividly remember Duesberg's logic from a long article in
Science (the front section, not the peer-reviewed stuff) back in the day. Something along the lines of
We test for the presence of HIV antibodies... if a person's made antibodies, it means that their immune system has won. (words to that effect) He thought, it seemed, that a test for antibodies for a
putative causal agent for immunodeficiency was hilariously incongruous.