12 ms·

This is sort of an orthogonal issue. Even without knowing Joe's gender, A would have accepted either "what did he want" or "what did she want", under the assump

This is sort of an orthogonal issue. Even without knowing Joe's gender, A would have accepted either "what did he want" or "what did she want", under the assumption that B does know Joe's gender. However, A would not have accepted "what did they want" as grammatical under any circumstances. > in a purely objective, language-design sense There is no such thing as this in linguistics. "Anatomical details" is certainly not the weirdest or most arbitrary-seeming way of assigning noun classes, across the world's languages. Compare e.g. the Ganda language. From Wikipedia: > ten classes called simply Class I to Class X and containing all sorts of arbitrary groupings but often characterised as people, long objects, animals, miscellaneous objects, large objects and liquids, small objects, languages, pejoratives, infinitives, mass nouns, plus four 'locative' classes

a1369209993 7y ago
> However, A would not have accepted "what did they want" as grammatical under any circumstances. Huh, so what do they use when B doesn't know Joe's gender? > There is no such thing as this in linguistics. Well, more pedantically: in the sense that it makes determining whether a sentence is grammically correct more dependent (or dependent, if it wasn't already) on random details of the outside world that shouldn't matter until we check for semantic correctness. > is certainly not the weirdest or most arbitrary-seeming way of assigning noun classes Of course not, natural langauges (actually by definition) are 'designed' by pseudo-evolutionary processes; if anything, it's surprising they aren't worse.