staving Dirk Elzinga:
>So "Russian, Japanese, and quite a lot of other languages" have no way to
>signal the distinction between definite and indefinite? I find that hard to
>believe. McWhorter refers simply to "fundamental features" of language; not
>fundamental morphological categories, which is how many in this thread seem
>to be reading it. A language may not have articles (the usual marker for
>definiteness) but that doesn't mean that definiteness is not marked.

Let's take an example from Latin, which is the language I know best 
out of the ones I mentioned.

puer puellam amat
puer puell-am am-at
boy.NOM.S girl-ACC.S love-3P.S.PRES.IND

{The/A} boy loves {the/a} girl.

Any of the four possible readings is valid, and when translating into 
English, you disambiguate by context, not by any morphological or 
syntactical feature of the Latin. The definite/indefinite distinction 
is not a grammatical feature of the Latin language. For McWhorter to 
be correct, there would have to be some feature the language that 
depended on whether nouns were definite or indefinite, and I know of none.