From: Iver Larsen (firstname.lastname@example.org)
Date: Mon Mar 26 2001 - 04:29:20 EST
> > > Luke 22:45 hEUREN KOIMWMENOUS AUTOUS 'he found them SLEEPING!'
> > > [Moon]
> > >
> > > I would like to ask why you consider KOIMWMENOUS AUTOUS as an NP.
> > Mainly because it agrees with AUTOUS in number, gender and case. It
> cannot refer
> > to the implied subject (Jesus) and it is not subordinated to the
> finite verb.
> > >
> But the agreement with AUTOUS in number, gender and case is not only
> the property of modifiers but also that of predicates.
It seems that we have different definitions of what a predicate is. A predicate
to me is the key or nuclear slot of a clause, which is normally filled by a
verb. A predicate is combined with one or more arguments that function as
subject, object or indirect object. I think what you are referring to is what
may be called "predicate attribute". The German grammar I have access to makes a
distinction between Prädikat and Prädikativum (Prädikatsattribut).
> > > It is based on the basic principle:
> > > (1) The modifiers of the head noun contributes
> > > to determining the referent of the NP.
> > > (2) Predicates explain the referent of the NP, which is
> > > already determined.
> > I am not familiar with such a principle, either from general linguistics or
> > Greek grammar. Can you explain the reason and rationale for it? It
> seems to me
> > that the principle uses a pragmatic criteria to determine a syntactical
> > question, and that, to me, is questionable.
> 1) The principle is that a constituent is a modifier of a head noun if it
> contributes to the determination of the referent of the head noun. It is
> simply the definition of a modifier.
> I did not learn it from any book or teacher. But I always thought it was
> an obvious principle.
I am afraid I disagree that this is simply a definition of a modifier. I am
reluctant to accept your principle because it is far from being obvious to me. A
modifier modifies the head noun. It will give additional information about the
head noun, but whether this information helps to determine the referent or adds
further descriptive information should not IMO determine a syntactical analysis.
In many cases it would difficult to draw the line between these two.
> But first of all, I do not consider KOIMWMENOUS AUTOUS as an NP
> because the following analysis comes to my mind.
> This analysis has been long taught in English school grammar.
> What is the problem with this analysis?
> hEUREN KOIMWMENOUS AUTOUS
> V NP(O)-complement NP(O)
> 'he found them SLEEPING
> NP(S) V NP(O) NP(O)-Complement
> In this analysis, NP(O)-Comp is a predicate that
> explains NP(O), rather than a modifier of NP(O).
I would not consider an NP(O)-complement a predicate, but I am willing to accept
it as a predicate attribute. I am not sure I would call "sleeping" an NP at all.
It is normally called a "participial adjective". If you go to the underlying
structure, you will find a predicate in the dependent clause, as in: "he found
them as they were sleeping."
I suppose this is the background for calling it a predicate attribute. The
important point is that it relates to the object and gives further information
about the object rather than the subject and the verb/predicate.
I agree that English grammar requires predicate attributes as in "He found Peter
awake". Some English grammars call "awake" a "predicate adjective", others just
So, I think you are right. It is probably better to call it a complement to the
NP(O) rather than a constituent within the NP. A complement to an NP does not
need to be an NP itself. What was my main interest, though, was that normally
such complements follow the NP they relate to, just as adjectives and other
modifiers tend to do, but in this particular case, it precedes, and this implies
that the complement has greater relative focus than the pronoun it complements.
B-Greek home page: http://metalab.unc.edu/bgreek
You are currently subscribed to b-greek as: [email@example.com]
To unsubscribe, forward this message to leave-b-greek-327Q@franklin.oit.unc.edu
To subscribe, send a message to firstname.lastname@example.org
This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:36:54 EDT