Are there any unanswerable questions or just incomprehensible answers?
Logic and computational theory tell us that there are three general classes of problem.
Those that are computable by some shortcut algorithm;
Those that are only computable by iterating every step in the process (like the Mandelbrot set, or the digits of Pi); and
Those that have no computable solution (the algorithm continues forever without terminating).
The latter of these three classes give rise to what is known in computational circles as the “halting problem” – how do you decide whether to keep on computing in the hope that the problem resolves, or to stop computing that problem and move on to something else.
Comprehension is a different notion completely.
For some people comprehension relies on being able to tie together groups of ideas with a linear sequence of steps.
For others of us, comprehension relies on being able to use our intuition reliably to get useful outcomes.
For all normal people, comprehension is impeded if essential concepts for relating items in the sequence are missing.
So any of us will find any concept incomprehensible if we are missing any of the concepts in the chain of concepts leading from our agreed assumptions to the question at issue.
A further source of incomprehension is simply error. Someone may have an explanation that seems to make perfect sense to them, but contains some false steps in the logic.
Some people are simply unskilled in the disciplines of logic, and make pseudo-scientific claims.
Some people are deliberately deceptive, and use pseudo-scientific mumbo-jumbo to make themselves sound credible.
So the answer is very clear and simple in a sense, there are both unanswerable questions and incomprehensible answers (for a whole host of different reasons).
You raise an interesting aspect.
As humans we tend to answer questions quickly. Often we are less concerned with the accuracy of the answer than we are with the social utility of being seen to be attentive to the questioner.
That ordinary mode of being human contrasts sharply with the modes of being human involved in activities where accuracy of information is more important that social status aspects of responses. When one is dealing with computers or technology (either alone or in groups), social aspects of communication are of little or no value, and only accuracy and timeliness come into consideration.
When I am writing computer code, timeliness is not an issue, only accuracy, and the computer is maddeningly insistent on absolute accuracy (it simply will not work otherwise). After spending a few hours in that mode, it can take me some 10s of minutes to switch back to normal human modes of communication (which my wife finds extremely frustrating, as in the transition I take everything she says absolutely literally). I strongly suspect the same effect sometimes colours my written communications in this forum also.
If someone asks a question we don’t wish to address we can respond saying we are not going to address that question, which is, in a strict sense, a comprehensible non-answer. It is a response, yet it quite explicitly does not directly answer the question. (Politicians tend to specialise in such answers.)
So yes – it is a highly dimensional issue.