On Thu, 17 Aug 2017 10:12:39 +0100, Raymond Brown <[log in to unmask]> wrote:

>The thing with these languages and IIRC the Gaelic languages is the
>habit of using the subject pronouns immediately after the verb virtually
>as affixes; this sometime affected the endings.
>The singular ended in -m in early Brittonic which softened (lenited) to
>-ṽ (still written -m in early Welsh.  Welsh and Cornish simply
>denasalized [ṽ] to to [v], spelled _f_ in Welsh, whereas in Breton the
>nasal prevailed and [v] was lost hence [jaṽ] -> [jan].

I knew _f_ was a spelling of /v/ which could arise from lenition of /m/.  The step of this process that seems noteworthy to me is not any of that but [m] > [n].  How did that happen?  Not a sound change, if the Breton first plural still ends in /-m/.  Not interaction with a pronoun, unless there was a first singular pronoun starting in a dental...

On Thu, 17 Aug 2017 14:46:44 +0100, Raymond Brown <[log in to unmask]> wrote:

>>> I'm put in mind of the Tocharian person endings, which are also
>>> puzzling; maybe Brittonic could illuminate the sorts of processes
>>> to look at.
>I meant to add "I doubt it" before the snip.  :)

The sort of thing I meant is: the short answer for why Brittonic did this is interaction with following subject pronouns.  So did Tocharian of some phase have VS word order with pronominal subjects?  (No time to attempt to look that up now.)  If so we've got a candidate explanation which to me seems better than the current one.