Alex Fink, On 06/05/2012 03:57:
> On Sun, 6 May 2012 01:18:39 +0100, And Rosta<[log in to unmask]>  wrote:
>> Unfortunately I don't have sound intuitions about languages other
>> than English. But I've spent well over 20 years thinking about
>> "They [=bathroom scales] are" and "It [=balance] is", "They [=oats]
>> are" and "It [=wheat] is".
> But then in some Englishes we can get things like "My bank are".
> Another incursion of semantic agreement where formal morphosyntactic
> features would expect to hold sway, isn't it?

Yes, "My bank have written to me" and "Five minutes is enough". Semantically-driven 'agreement' (i.e. it's not agreement but semantically-based number-marking on the verb) is straightforward, as is a mixed system where there can also be agreement with the morphosyntactic number of the subject, and morphosyntactic number can be lexically-specified.

The thing that has long perplexed me, rather, is what determines the number of the pronoun:
"They [=bathroom scales]"
"It [=balance]"
"They [=oats]"
"It [=wheat]"
I see essentially two possibilities. Firstly, there could be a *semantic* (encyclopedic) category 'pluraloid', whose members prototypically have plurality, but can also be simply stipulated to be members of the category. The linguistic evidence of "The bathroom scales are" and "The oats are" (-- "one oat" is a kind of oat, not a grain of oat) would then tell us that they are semantically pluraloid. So oats are referred to by _they_ because oats are semantically pluraloid, and _they_ means only something pluraloid. The other possibility is that third-person  pronouns consist syntactically of a phrase containing an ellipted common noun, whose lexicaly-specified number can then inflectionally determine the morphological form of the phrase when the common noun is ellipted.

>> As for me (& English), I don't think null Ds are warranted, and
>> indeed I don't believe in the existence of a category Determiner or
>> of a category Pronoun as anything more inclusive than Personal
>> Prounoun. Sometimes I think personal pronouns are a subtype of
>> common noun, and sometimes I think they are a fusion of _the_ and a
>> common noun. The 'left-periphery' of the English NP is an area of
>> many mysteries.
> In what sense a fusion

Two syntactic sentence-words corresponding to one phonological sentence-word.

>(or, dare I ask, with what degree of  reality)?

I'd say the discipline of linguistics is not even sure yet what constitutes linguistic reality, let alone able to assess degrees of reality. That said, I imagine that every linguist intends their proposals to be a description of reality.