> Utter nonsense. The regular development of aerá:m- (with
unstressed /ae/
> and long stressed /a:/!) would have been into *erám-, which could
> undergo assimilation to /arám-/ at _any_ time in its early history.
The
> fact that we seem to be dealing with sporadic rather than regular
> assimilation does not indicate _anything_ as regards its age.
>
> Piotr
>
Not true, Piotr.
Please read Rosetti Piotr, (I will quote it in the evening)
the "assimilation in Rom. aramã is ae<->a => aa->a and took place on
Latin ae Not on Latin e"
This because there is no assimilation e<->a => a<->a ...
see: Rom. iarbã 'grass' => Latin herba => shows an e in Romanian
e >je>jea>ja
So the timeframe of this assimilation is clear before ae>e in
Latin and took place on ae => any Latinist (Miguel?) will say you this
Best Regards,
marius
P.S. Also an assimilation is not 'sporadic' Piotr it happens in a
right and specific context.