From: Piotr Gasiorowski
Message: 29183
Date: 2004-01-06
> Even in the earliest records, patronymics in <-o/ev-ic^I> (from (j)o-This can be nicely illustrated with reference to Vladimir Ilyich
> declension stems), <-in-ic^I> (a- and i-declension) and <-ic^I> (all
> declensions) already form a rather independent grammatical category,
> while patronymics in <-ovU>, <-inU>, *-jI originally were still
> possessive adjectives, requiring determinable words like <synU>,
> <dUc^i>, <vUnukU> -- probably this is somehow related to the facts
> that this latter group has been transformed into last names, while
> the first has retained its purely patronymic function.