[jbovlaste] Re: grammatical constituents

[Pierre Abbat <phma@phma.optus.nu>]

On Thursday 03 September 2009 16:32:48 MorphemeAddict@wmconnect.com wrote:
> I don't think they should be numbered at all.  And subject and object and
> indirect object are parts of sentences, so something with component "jufra"
> 'sentence' (juf ju'a. x1 (text) is a sentence/statement about x2
> [topic/subject/predicate/focus] in language x3        1h 111    [phrase (=
> jufpau, suzrelvla, suzvla, gensle)])

Subjects and objects are parts of clauses, not sentences. A clause is a bridi; 
a jufra can contain several of them, or none at all.

As to numbering them, I'm taking the viewpoint of someone who knows Lojban 
grammar and tries to describe a natlang in similar terms. Of course, such a 
grammarian will run into parts of speech, such as adverbs and adjectives, 
that don't exist in Lojban, but verb arguments are familiar, and he will 
number them as he is used to.
I (1) run.
I (1) hit the ball (2).
I (1) give Tom (3) the ball (2).
I (1) bet you (3) ten quatloos (2) (that you (1) can't find a tritransitive 
verb (2)) (4).
The last sentence has two subjects and two direct objects, because it has two 
clauses.

The terms "subject" and "object" are artifacts of the Greek and Latin 
grammarians whose foundation we build on. Had English grammar been based on 
Sanskrit grammar, we would probably have numbered places. Sanskrit has eight 
cases, which are numbered 1-7 in Sanskrit grammar (the vocative considered 
something else and not assigned a number).

Pierre