But wait a second. An easy counter argument to the theory is that smoking has significantly decreased all over the Western world. So, if the number of schizophrenics hasn't decreased proportionally, then tobacco smoking can't be a major factor.
Agreed, with the caveat that mental illness diagnosis has become more common over the past century. (Though I'm not sure whether people have become more willing, in the past couple of decades, to diagnose schizophrenia specifically.)