Re: Trigger language question concerning the use of "to be"

Chris Bates wrote at 2005-05-09 23:56:41 (+0100)

 > Yep, this is the way I took the language when I learned some
 > (although when to use ligatures gave me problems). It's supported
 > by the fact that: a) obliques are marked in the same way as
 > possessives b) "verbs" when used as arguments are treated exactly
 > like nouns in the markers (ng or ang mainly) that they take, and
 > finally c) "adjectives" in fact seem to be treated like nouns in
 > most respect, with order of noun and adjective (if there even is
 > truly such a distinct class) in the NP and their general behavoir
 > seeming extremely similar if not identical. It'd be funny if this
 > analysis is correct, since creating all noun languages is a regular
 > "wacky no natlang does it" conlang idea. The one problem I can see
 > is that, while you can argue all occuring verb forms are
 > nominalizations, there are certainly verb roots which are not in
 > the same class as noun roots, since I don't believe nouns can
 > freely take the same morphology as verbs.

Not so.  Roots differ in terms of which voice/focus/nominalization
affixes they can take, but they don't split cleanly into verbs and
nouns.  Nearly all roots can take at least some, and very few can take
all.

e.g.
_bahay_ "house" _magbahay_ "to build one's house"
_anak_  "child" _maganak_ "to breed" _ianak_ "to give birth to someone"
_walis_ "broom" _magwalis_ "to sweep"

(_mag-_ "actor voice", _i-_ "conveyance voice")

The paper to read on this is Nikolaus Himmelmann's "Lexical categories
and voice in Tagalog".

http://www.linguistics.ruhr-uni-bochum.de/~himmelmann/LEXCAT.pdf

He doesn't actually call Tagalog words "nouns", on the grounds that if
you only have one category it doesn't make sense to identify it as
nominal or verbal.  He does note that you might choose to call
voice-marked words "verbs" on the understanding that it's a purely
morpho-lexical and not a syntactic category.

Section 3.1 is particularly relevant to the matter under discussion.

Some of his points:

* Almost all Tagalog roots are glossed as English nouns, adjectives or
  participles rather than finite verbs.

* There are four basic morphosyntactic functions in Tagalog clauses:
  predicate, subject, non-subject argument or adjunct, and modifier.
  Tagalog content words are not subcategorised with regard to terminal
  syntactic categories.  All phrasal categories with the exception of
  the predicate (when clause-initial) are composed of a function word
  which indicates the category and a content word.  Content words are
  not subcategorised with regard to the content words with which they
  may appear, and all content words may serve as predicate.  Fully
  inflected ACTION-words (that is, words denoting actions,
  prototypically verbal concepts) can serve in nominal constructions
  (subject, adjunct, modifier, with quantifiers) without any special
  treatment.

* Tagalog roots definitely do, nonetheless, fall into lexical
  categories with regard to the affix-sets with which they may occur,
  and the meaning of the resulting affixed forms.

* Practically all Tagalog roots (and many derived stems) can occur
  with at least one "voice" affix (and concomitant aspect/mood
  inflection) without any further derivation.  Conversely, roots are
  not bound - almost all roots may occur in unaffixed form, though
  some do so more naturally than others.

* Voice marking is essentially derivational in character.

Reply