Tolowa dative/benefactive

[Tom Givon]

Dear FUNK people,

I apologize to you & Vic for opening the wrong file last time. However,
now that I've opened the right one (I hope...), I still wonder whether
Tolowa abides by what Vic says about the rest of Athabaskan. The data
below is only about verbs with obligatory DAT arguments. First, for
comparison, the objects of simple transitive verbs, to establish the
morphological DIR OBJ position (data are again from a file compiled by
Dave Watters). First with zero-marked 3rd pers. object:

           'U-sh-k'Usr                 'I shave him'
           THM-1s/SUBJ-shave

           '-i~n-k'Usr                  'You shave him'
           THM-2s/SUBJ-shave

          yaa-k'Usr                     'S/he shave him' (zero subj pro)

          TR-shave

The /yU/yaa/ morpheme, currently marking transitive clauses with
third-person subject and object, still occupies the DIR OBJ pronominal
slot. So with 1st/2nd objects:

          sh-ii~-k'Usr            'You shave me'
          1s/OBJ-2s/SUBJ-shave

          nU-sh-k'Usr            'I shave you'
          2s/OBJ-1s/SUBJ-shave

And likewise in the PERFECT:

         sh-s-ii~-k'Usr              'You shaved me'
         1s/OBJ-PERF-2s/SUBJ-shave

         nn-s-ii-k'Usr               'I shaved you'
         2s/OBJ-PERF-1s/SUBJ-shave

With the bitransitive 'give', using the classifier verb (here for sg.
small object /-'a~/), it appears that Vic is right, in that the DAT
object position is distinct from (and more 'external' to) the old TR/
DIR OBJ position. Thus:

        wa-y-n-ii~-'a~                  'S/he gave it to him'
        ADV-TR-PERF-3/SUBJ-V.CL

        nn-ghaa~-n-ii-'a~              'I gave it to you'
        2s/OBJ-ADV-PERF-1s/SUBJ-V.CL

        sh-ghaa~-n-ii~-'a~              'You gave it to me'
        1s/OBJ-ADV-PERF-2s/SUBJ-V.CL

The verbs 'send' & 'bring' show a similar pattern.

However, with  a "more dative" verb such as 'tell', the DAT pronoun
clearly occupies the DIR OBJ position:

           yU-lh-nUn      'S/he told (it to) him'
           TR-L-tell/PERF

           nU-lh-nUn      'S/he told (it to) you'
           2s/OBJ-L-tell/PERF

           shU-lh-nUn     'S/he told (it to) me'
           1s/OBJ-L-tell/PERF

And as far as I can see, the same is true for 'teach' and 'show'. So at
the very least, the more prototypically-dative verbs seem to
obligatorily code the DAT argument to DO.

The data is further complicated by evidence that in some verbal
constructions, Tolowa has been pushing the DIR OBJ pronominal position
further to the left ("more external"), thus away from the old /yU/
("TR") position. So that even for 'give', the  purely
morphological criteria for what is DO are now a bit muddled. But this is
going to be too much for this discussion. Unfortunately, I cannot yet
find the file on optional BENEFACTIVE arguments. But I'll keep looking.
Anyway, thanks for y'alls patience, and Vic-- thanks for your vigilance.
Best,  TG