number in possessive pronouns

[Nina Dobrushina or Michael Daniel]

To answer Pier Marco Bertinetto's letter:

I did not suggest that saying 'our children' (instead of 'my children')
when speaking to one's wife is obligatory was a Russian specific
feature, and suggested it was the same in at least some other languages
(although I don;t know if it is universal).

I mentioned Russian example in connection to the data like that from
Tsez because I believe there is one and the same mechanism underlying
both constraints. The facts are:

a. In Tsez, one must say 'our village' not 'my village' (when speaking
to doesn't matter who).
b. In Russian, Italian ..... one must say 'our child' not 'my child'
when speaking to a co-parent.

Now consider the following. In Russian (and probably other languages),
when a city-dweller happens to a village and talks to a man in the
street he would hardly say 'your.SG village' (it would then mean 'the
village you own'). He is much more likely to say 'your.PL village'. It
seems that the speaker is "too much inside" the village to disregard the
fact there are other inhabitants.
On the contrary, the same city-dweller talking to a person living in the
city but originating from a village would sound strange saying 'your.PL
village'; he is much more likely to say 'your.SG village'. He is "too
much distant" from it to involve other inhabitants. Of course, he can
use 'your.PL village' - but only if he, roughly speaking, keeps in mind
other individual inhabitants (meaning e.g. probably the addressee's
friends who, as he knows, are from the same village).

Consider another example. You can say: "My bus is late" in Russian. But
if you say "After school my friends and I took a bus", you can hardly go
on saying "And then my bus broke", you should say "And then our bus
broke" (at least in Russian). This is obviously because other passengers
are actualized in the context, and their relation to the bus may not be
overlooked.

So the model goes as follows - the more specific (or actualized,
intimate, in-context etc.) the other co-possessors of the item are in
the context, the more likely they will be mentioned in a possessive
construction, making the possessor NP plural. The model, to my mind,
also provides a framework to explain (a) and (b) above. Indeed:

(a) Tsez being a minority language, communication in Tsez always
naturally regards at the handful of Tsez villages from very near if not
from inside. Prototypically, it's hard to imagine a 'distant' speaker
speaking of a Tsez village in Tsez. That makes plural possession with
the word 'village' grammatically obligatory.

(b) Both the speaker and the addressee are very much in-context within
any communication. This makes any item to which both the speaker and the
addressee are related obligatorily plurally possessed (be it a child, a
bus, or money).


This is a sketch, but I think there is something to it.

Michael Daniel

PS: What we do not know is - whether there are any languages with which
even being addressee+speaker are not specific enough to require plural
possessives.