Such as rampant subjectivism, or Humeian hyper-skepticism. I am
somewhat sympathetic to Hume for being rigorous, but his emphasis on
the impossibility of absolute knowledge really isn't as important as
he makes it sound. Sure, you can't be omniscient; so what? That
doesn't mean you can't be usefully knowledgeable.
There is a trend in human nature to reject extremism, and that's
good. But there is another trend that makes people think that
just because both extremes of an idea might be wrong, that that's
all there is to know; steer clear of the extremes and you'll be
"reasonable". But that's just avoiding the problem, not solving
it. The extremes are wrong because some particular point between
them is right, and simply rejecting the extremes tells you nothing
about what that point is.
To be more concrete, let's take the extremes of skepticism/credulity.
Absolutism in either is irrational, so it is correct--but not
meaningful or helpful--to say that one should be somewhere in between.
To express a meaningful and useful opinion on the matter, one must
try to identify /where/ along that spectrum you fall. I, for exmaple,
hold the opinion that 95% of humans are too credulous; that the ideal
point of view is much further toward skeptical than credulous. That
opinion has real meat; it may be wrong, but at least it commits to a
meaningful point of view.