Re: Noimi Inverse Marking

> A possible exception is when both 1st and 2nd person arguments occur;
> I'm not sure whether to use
> (a) a special A2 marker for 1st person acting on 2nd person, or
In Algonquian languages, when the speech-act participants are
restricted to first and second person, special inverse forms
are used, which treat second person as higher in the hierarchy
than the first person, contrary to the generalization of
Silverstein's hierarchy:

  ke-wa:pam-i			ke-wa:pam-ene
  2-look.at-1,2.DIR		2-look.at-1,2.INV
  'You (sg) are looking at me.’	‘I am looking at you.’

  ke-wa:pam-i-pwa		ke-wa:pam-ene-pwa
  2-look.at-1,2.DIR-2Pl		2-look.at-1,2.INV-2Pl
  ‘Y’all are looking at me’	‘I am looking at y’all.’

The data is from Meskwaki, of course.  I use '1,2' to
distinguish these from the regular direct and inverse
thematic markers that are used when a third person
participant is involved. Note how every verb form has
a second person prefix, whatever the grammatical function
of that prefix.

> (b) a special A3/A1 marker for 2nd person used only with A2 = 1st person.
> I'm also not sure whether I should swap 1st and 2nd person above.
>
> At this point, I should mention inverse marking. Inverse marking is
> necessary to change which roles are mapped to which arguments. If no such
> marking appears, the word has a *direct* form and the roles R1, R2, and R3
> go with arguments A1, A2, and A3 respectively. The inverse marker will swap
> R2 with R1 if V=2 and with R3 if V=3.
>
>        Role Mapping    A1      A2      A3
>         ---------------------------------------
>        V=3  direct     R1      R2      R3
>        V=3  inverse    R1      R3      R2
>        V=2  direct     R1      R2      --
>        V=2  inverse    R2      R1      --
I had been meaning to get to this earlier, but the July 4th
celebrations interceded. :)

I should point out that the system you've created here is
AFAIK entirely unattested among inversion languages.  In
all cases that I'm familiar with (Algic, Ktunaxa, Kiowa-Tanoan,
Mapudungun, etc.), the inversion process only cares about
who's the subject and who's the object; secondary objects
don't figure into the equation.  Thus, you would only
expect an inversion between R1 and R2 (being linked to
A2 and A1 respectively), not between R2 and R3.

 =========================================================================
Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637