Closed mgeorgescu closed 10 years ago
We had some discussion on this before, deciding that commanders as a military rank should not be decomposed. But based on occurrences of commander in our corpus, we have a range of usages:
(1) specific military rank, e.g. in U.S. Navy (corresponding to Lieutenant Colonel in Army) (2) somewhat generic term for high-ranking officers, e.g. Medvedev asked military commanders to come up with a plan. (3) with :ARG1, e.g. company commander, Ivanov, commander of Russian Military Space Force
I would recommend to
(2) somewhat generic term for high-ranking officers, e.g. Medvedev asked military commanders to come up with a plan. (3) with :ARG1, e.g. company commander, Ivanov, commander of Russian Military Space Force
It seems to me that commanders in military commanders and company commanders has the same meaning, but relates differently to the modifier. So I think I'd prefer decomposing both to person :ARG0-of command-02
:
military commanders
(p / person
:ARG0-of (c / command-02)
:ARG0-of (h / have-org-role-91 :ARG1 (m / military)))
company commanders
(p / person
:ARG0-of (c / command-02
:ARG1 (c2 / company)))
Should we decompose commander to "person :arg0-of command" or should we leave it non decomposed?
Most contexts where it appears are military contexts.
We would be in favor not to decompose it but there there are some chases where you have the "thing :arg1-of command" and it's a pity if we don't decompose it, we cannot use the arg structure. Any feedback is welcome. Thank you