[Lingtyp] Optional determination?

[Christian Lehmann]

Am 01.09.2024 um 07:35 schrieb Martin Haspelmath via Lingtyp:
>
> It seems to me that "determiner" in Bloomfield's (1933) sense (where 
> it basically referred to articles and demonstratives) and 
> "determination" in the sense of semantics are two rather different things.
>
> Many semanticists seem to think that one needs a syntactic determiner 
> to turn a nominal expression into a referential expression, but of 
> course, many languages lack both definite and indefinite articles 
> (Grambank has 1268 languages of this type: 
> https://grambank.clld.org/combinations/GB020_GB021).
>
I fully agree. It would clarify the relevant conceptions if 
'determination' were restricted to the sense of 'providing something 
with a determiner'. The semantic operation could be called 'reference 
fixation' or something of the kind. All the while, it is clear that 
determination normally serves reference fixation.
>
> Like many other types of grammatical markers, articles are often 
> optional. So I don't really see a basis for distinguishing between 
> "maximal projection" and "non-maximal projection" in general terms. 
> (And the idea that there is a single determiner slot seems to be based 
> on English alone; even languages such as Greek and Spanish allow the 
> cooccurrence of demonstratives and articles.)
>
As far as I know, this is not so for Spanish. There is a single 
prenominal slot for determiners (comprising articles, demonstratives and 
possessive pronouns). What is possible is to occupy the slot by an 
article and to add a demonstrative or possessive pronoun in postnominal 
position.

However, it is true that there are languages which allow a sequence of 
determiners in what would otherwise be the same slot. Cabecar is one of 
them; this is why I said "most".