What linguists call "Escher sentences" at first seem sensible even though, on inspection, their apparent soundness dissolves. A common example is:
More people have been to Berlin than I have.
Maybe the way to approach the weirdness of this sentence is to note that there is an implied ellipses at the end, and that the common way to clarify the intended meaning would be to understand:
More people have been to Berlin than I have [been to Berlin].
But this isn't helpful—in fact, it's easier to see that there's something wrong with the "clarification." Two things are being compared but what are they? If more people have done X than Y, then you should indicate what X and Y are. In this case, though, there's only one thing mentioned—"been to Berlin." If "been to Berlin" is not what belongs in the placeholder at the end, then anything goes:
More people have been to Berlin than I have [knocked up].
If it can mean anything, then it means nothing, and no one will understand.
Yet no one on the receiving end of
More people have been to Berlin than I have
is apt to object or even experience confusion. Ordinary language rules may decree that the sentence has no meaning, but everyone thinks they understand, and they probably do—which is also pretty weird. The leading candidate is
Some people have been to Berlin more often than I [have been to Berlin].
There are other candidates—for example:
I'm not the only person who's been to Berlin.
The first interpretation makes the assumption that the comparison at issue is not quantity of people but rather quantity of trips. The second interpretation makes the assumption that the issue at hand does not pertain to any kind of quantitative comparison at all. The way you settle on one option or the other, or some option I haven't thought of, probably depends mostly upon context, which is the main reason linguists are interested in Escher sentences in the first place. It seems there’s a lot more to language comprehension than grasping language rules. Consider these two sentences:
(A) No head injury is too minor to be ignored.
(B) No head injury is too minor to be taken seriously.
These sentences must have opposite meanings, right? Which one means that a minor head injury should be taken seriously, not ignored? By ordinary language rules, there's a case to be made for (B), the import of which might be clarified and amplified:
Is the injury too minor to be taken seriously? Then it's not a head injury.
Since the sentences have opposite meanings, and (B) means that even minor head injuries should be taken seriously, it follows that (A) means that minor head injuries can be ignored. But doesn't everyone agree that (A) says the opposite of that? It seems that we might be able to agree that (A) and (B) mean the same thing, despite the fact that "ignored" is the opposite of "take seriously." I'd tell you what's going on here if I understood it myself. Do ordinary language rules require (A) to mean the opposite of what everyone "feels" it means? I don't think that's clearly the case, but I will note that if you try to clarify the meaning of (A) in the same way I clarified the meaning of (B)—
Is the injury too minor to be ignored? . . .
—you introduce a problem that didn't arise before, since the meaning of something "too minor to be ignored" is known only to Yogi Berra and a few other philosophers.