Re: CHAT: query: where to start?

dirk elzinga wrote:

> On Wed, 9 Aug 2000, J Matthew Pearson wrote:
>
> > Speaking of non-concatenative morphology, I've recently begun toying with a new
> > scheme for marking certain cases in Tokana.  Currently, the dative case is marked
> > by adding "-i" (/j/) to words ending in a vowel and "-e" to words ending in a
> > consonant:
> >
> >   Absolutive   Dative
> >
> >   tomla        tomlai [tomlaj]    "mountain"
> >   uosu         uosoi [wosoj]      "pebble"
> >   esian        esiane [ESjanE]    "name"
> >   totsat       totsate [totsatE]  "table"
> >
> > Now I'm thinking of changing things so that the dative is always marked by an
> > "-i" glide, which gets added to the end of a vowel-final stem, and before the
> > final consonant of a consonant-final stem (infixation):
> >
> >   Absolutive   Dative
> >
> >   tomla        tomlai [tomlaj]    "mountain"
> >   uosu         uosoi [wosoj]      "pebble"
> >   esian        esiain [ESjajn]    "name"
> >   totsat       totsait [totsajt]  "table"
> >
> > Not sure if I'll go with that or not, though...
>
> Go for it! I'm still smarting from your abandonment of reduplication
> in Tokana; it'd be nice to see some noncat morphology again ... :-)
I think my sentiments are leaning towards those of Doug Koller:  Keep the original
system for the most part, but introduce some random 'exceptions' which feature the
infixation version.  In fact, I've already begun to do this.  A very small number of
CVC nouns form what I call "locative adverbs" by adding the oblique prefix "i-" and the
infix "-i-".  These locative adverbs tend to have an idiosyncratic meaning, as opposed
to the regular datives, which have a compositional meaning:

  lan     "path"
  ilain   "en route, on the way (somewhere)"
  lane    "to/on the path"

  mok     "hearth"
  imoik   "home, to/at home" (as in "I went/stayed home")
  moke    "to/at the hearth"

Perhaps I'll just add some more of these to the list and be content with that.

My major reason for resisting the 'infixing dative' idea is not that I dislike
non-concatenative morphology, but that it would spoil the symmetry in the current case
system.  Right now the five oblique cases are marked by a set of suffixes which
generally have two allomorphs--the first used with stems ending in a vowel, and the
second used with stems ending in a consonant or glide:

                 After V               After C/G
  Dative         uoso-i    [wosoj]     totsat-e   [totsatE]
  Allative       uoso-in   [wosojn]    totsat-ne  [totsatnE]
  Instrumental   uosou-a   [wosowa]    totsat-a   [totsata]
  Ablative       uoso-u    [wosow]     totsat-u   [totsatu]
  Comitative     uoso-um   [wosowm]    totsat-mo  [totsatmo]

Suppose I were to replace dative "-e" with an infixed "-i-" glide and ablative "-u"
with an infixed "-u-" glide (giving "totsait" [totsajt] and "totsaut" [totsawt] in
place of "totsate" and "totsatu").  What would I do about the allative and comitative
cases, which seem to be morphologically related to the dative and ablative under the
current system?  I think I'll just leave things the way they are, and perhaps search
for some other way to introduce non-concatenative morphology into Tokana.

Matt.