On the "standard conception" of scientific theories