>To my view, with respect to this rule, Mr. Jens Elemegaard Rasmusen
>gives very clear and argumented defination that Albanian possesive
>pronoun <(i,e) jonë> `our' has no other possibility to be derived
>otherwise than from compound of pronominal stems *saH2 + *nas > jonë
>(*s > j and aH2 > a: > o). First part is feminine form of pronominal
>stem *so- `he', until *nas is pronominal stem of pronoun in plurals
>*nes. Our opinion that possesive pronoun <jonë> `our' is really a
>compoun, but where second part should be probably in zero-grade form
>*n.s to be explained the lack of rhotasism in Tosk dialect, was not
>disputed. I think also that as a compound could be treated other
>Albanian possesive pronouns <tonë/tanë> `ours', derived from
>*taH2/to + *n.s.
You are right related to the fact that we have compound forms here.
But the derivation is wrong.
The problem with your derivation is that jonë has the masc. form ynë
(and in addition we have the gen-dat. form sonë too)
so is more logic to consider:
1. jonë (fem.) = (a)jo + në < PAlb *a-ja: + *na
(pseudo) Eng. 'she-one that is ours'
in Rom. 'aia a noastrã'
See Alb. ajo 'she / that one'
2. ynë (masc.) = (a)y + në < PAlb *a-j(u) + *na
(pseudo) Eng. 'he-one that is ours'
in Rom. 'ãla al nostru'
See Alb. aj,ay 'he / that one'
Ex:
Alb. 'djali ynë' Rom. 'bãiatul ãla al nostru'
Eng. 'the boy, that-one that is ours'
3. sonë (dat.gen. form) = so + në < PAlb *tsja: + *na
(pseudo) Eng. 'that-one (fem.) that is ours'
in Rom. 'cea-ia a noastrã'
Note:
The lack of rhotacism in Tosk dialect is explain by the fact that
these coumpounds are 'recent' formations (as a single word, I mean)
so: 'jonë' was '*(a)jo në' when the Tosk rothacism happened.
Best Regards,
Marius Alexandru
P.S. : Also your possible *saH2 + *nas would give *gjonë and not
*jonë (because *saH2 is accented and sa:- is not lost in this
position in Albanian)