Re: [tied] Nominative Loss. A strengthened theory?

--- In cybalist@yahoogroups.com, "P&G" <petegray@...> wrote:

> >we do not know which variant belonged in pausa.
>
> In Pausa is also sandhi - we cannot assume that the variant before

pausa is

> the base form. It is the underlying base form we should

reconstruct, not

> any of the phonetically conditioned variants.

What if they are all conditioned - or must appear so to us in the
present state of our knowledge? Does a good scholar then make an
arbitrary choice and report only a part of his information, or does
he leave it up to those he's talking to to take what they find
important? I did the latter.

Jens