Closed solmsdorf closed 7 years ago
Regarding copular uses of 'dug with adjectival predication, we can analyze them as we analyze copular uses of yin. So yi ge is not arg1
, at least not of the VERB
. Rather, it is an argument of the ADJ
. 'dug pa therefore depends on rnams khra lam me via the cop
relation.
I realize that the example is on Mila 55b, but if you can annotate that example anyway in BRAT, then I can copy it into the documentation.
Ok, copula use of 'dug should be implemented. The above-mentioned example cannot be annotated that way, it's not possible to draw relation-arcs from an A
.
I had configured the cop
relation to only target words with the AUX
POS tag, like yin. I have now broadened the relation to target any VERB
. Most verbs can't be copulas, but 'dug can.
You should now be able to annotate the example on Mila 55b. You won't be able to make the NOUN
depend on the ADJ
, but that's okay, because for the time being we are only annotating verbs.
Can I close this issue? Can you do what you need to do?
Can 'dug function as a light verb, similar to yod (cf. https://github.com/tibetan-nlp/lim-annodoc/issues/3)?
E. g. from Mila 11b: gzhan (
arg1
) rnams kyang shugs ring (arg2
) nar nar 'dug (V
) "the others, too, sighed deeply".=> 'dug may at times also be interpreted as a
cop
(= testimonial use of 'dug, cf. Hill 2013)! Which may also be the case here.E. g. from Mila 55b: yi ge (
N
=arg1
) rnams khra lam me (A
) 'dug pa (cop
) "the letters were clear".