Re: a case-free language?

Thomas R. Wier wrote:
>
> As I was trying to show in my discussion about Georgian, something
> *can* indeed phonologically merge with a word, but it must not
> *morphologically* merge with it.
Interesting.  That might be what happened with the case endings
in my conlang Sohlob.  They modify NPs rather than words, and go
back to old postpositions, but at the same time they trigger
umlaut in the words they are attached to, and undergo vowel height
harmony themselves, as well as losing their own vowel.

(All the Kejeb -- the protolang -- postpositions are
*(S)C(R)V [1]-- i.e. a legal syllable onset plus a vowel,
but the original V gets lost.  OTOH the final V of the
word which the postposition is attached to resurfaces
(properly umlauted and harmonized) and is Sohlob-
synchronically reinterpreted as the initial vowel of
the "suffix".

[1] In Sohlob a suffix may surface with an extra vowel
since a final Cr or Cw gets a svarabhakti vowel.

/BP 8^)
--
Benct Philip Jonsson -- melroch at melroch dot se

         Solitudinem faciunt pacem appellant!
                                             (Tacitus)