[b-greek] Re: Predicative/attributive - position and function

From: Iver Larsen (alice-iver_larsen@wycliffe.org)
Date: Tue Jan 23 2001 - 13:35:09 EST


> [Moon]
>
> Let me ask two questions first.
> 1)
> Should TO KALON receive more emphasis than TO DENDRON
> when we consider TO KALTON an apposition to TO DENDRON?
> If so, it is contrary to the implication
> of TO DENDRON TO KALON which does not put emphasis on KALON, and
> my analysis is wrong. But if we think that the apposition between
> the two NP is very weak, my analysis makes sense.
>
> 2)
> DENDRON TO KALON is also possible, and seems to be less specific than
> TO DENDRON TO KALON. If we think one as " a tree, i.e. the good one" and

> the other as "the tree, i.e. the good one", the degree of specificity can
> be explained. If we consider TO KALON to directly modify DENDRON, TO in
> front
> of DENDRON seems redundant.

[IVER]
I have enjoyed our dialogue, but I need to draw it to a close soon, because I am
spending too much time on this list.

I am not presenting the result of some finished research but thinking through
the issues as we go along. Through the organisation that I work with we are
trained to analyse and describe the structure and grammar of
never-before-studied languages. We do this, of course, without reference to
written grammars or dictionaries, because they don't exist until we produce
them. So, I tend to come to Greek in the same manner, using some of the same
analytical tools and methods, without ignoring all the earlier research others
have done.

To study the structure of an NP and the function of its constituents we would
collect/elicit as many NPs as possible with as many modifiers as possible. We
would then sort them in like groups and check which modifiers can co-occur and
which modifiers can never occur together within the same NP. We would also study
the relative order of the modifiers. I have not done that for Greek, and I do
not know if some of my colleagues have. If I get the time I might do it. The old
Greek grammars could not benefit from the advances in descriptive linguistics
that have taken place the last 50-60 years.

A certain description of, say, the Greek NPs is not so much false or true as it
may be more or less adequate. The simpler it is and the more it can explain the
more adequate it is.

You have noted that NPs like DENDRON TO KALON are also possible. I had not
thought of that, but you are right. They are very rare, but they are possible.
They are fairly common with a proper noun, since a proper noun often does not
have the article, e.g. Acts 5:37 IOUDAS hO GALILAIOS. I was only able to locate
two instances in the NT with a common noun, but there may be more:
John 14:27 EIRHNHN AFIHMI hUMIN, EIRHNHN THN EMHN DIDWMI hUMIN
(Peace I leave you, my peace I give to you)
Acts 2:20 PRIN ELQEIN hHMERAN KURIOU THN MEGALHN KAI EPIFANH
(Before the great and glorious day of the Lord comes)
In the first instance the word "peace" is a word that normally occurs without an
article. In the second instance the term "day of the Lord" is such a well known
and specific concept that it hardly needs a determiner, almost like a proper
name.

It seems to me that we have three options for the analysis of these NP
constructions.
(1) Posit two or more NPs in apposition, a weak one if you prefer, but
nevertheless several NPs in succession.
(2) Say that the article in TO DENDRON TO KALON is just repeated with no other
purpose than to show that KALON is a constituent of the NP rather than being
outside the NP and function as a predicate.
(3) Say that the article functions to rankshift the following clause in order to
make it part of the NP. This rankshifting is in most languages including English
done by a relative pronoun, but in Greek it seems to be done by the article with
a function like a relative pronoun.

The first of these options I don't like, because it seems too complicated and it
would result in an inordinate amount of appositional NPs. Such appositions in
any language tend to be rare and special cases. It also splits up the head noun
from its various modifiers/attributes.
The second option I am now abandoning, partly because of your second question.
This leaves me with the third option. So let me defend that one.

In most if not all languages it is possible to have a whole clause function as
part of a NP. I mentioned above "never-before-studied languages". The hyphens
are used to make a whole clause an attribute of the head noun, but the more
common (grammatically correct?) form would be "languages which/that have never
before been studied". This whole thing is one NP. The head noun is modified by a
whole clause, and the relative pronoun "which/that" functions to indicate that
the clause is part of and modifies the head noun.

Let me quote from Gal 1:11, which I used in a lecture today:
TO EUAGGELION TO EUAGGELISQEN hUP' EMOU OUK ESTIN KATA ANQRWPWN
The gospel (which/that) is proclaimed by me is not according to people

In English, if I say "I preach the gospel" it is a full clause NP(subject) VP NP
(object).
If I say "the gospel I preach" it is an NP. The same words, but the
different order indicates that the first is a clause, the second is an NP. The
rankshifting marker "which/that" is not even needed in English, because the word
order is fixed, so the order alone gives enough grammatical clues that this is
an NP with a rankshifted clause in it.

My suggestion now is that the article can function as a rankshifter in Greek.
This description can take care of a whole variety of rankshifted clauses,
including a very short clause consisting of only an adjective in the predicate
position, e.g.
TO DENDRON TO KALON would then be analysed as "the tree which (is) good"
DENDRON TO KALON would then be "a tree which (is) good."

This makes sense to me, because it fits with the relative function of the
article in Greek. I would be interested in whether you think it will work. I may
have overlooked something, and as I say I am thinking aloud here, not presenting
the results of finished research.

This does not mean that such constructions have to be translated with a relative
clause in English. If the rankshifted clause only consists of an adjective it is
simpler to move it into the adjective position before the noun in English and
say "the good tree" or "a good tree". Consider also more complex constructions
like:

Rev 1:5 hO MARTUS Ho PISTOS Ho PRWTOTOKOS TWN NEKRWN
          The faithful witness, the first-one (raised) of the dead

This could be taken as three NPs, the last two in apposition to the first as
suggested by the UBS and Nestle-Aland Greek text. Or it could be taken as one NP
(the faithful witness) followed by another NP in apposition. English would
prefer that as witnessed by most translations. Or it could be taken as one
complex NP with two rankshifted clauses modifying the head noun. I doubt it
makes much difference in Greek. In the Sabaot language I mentioned earlier as a
never-studied-before-we-did language, there is no attribute position of
adjectives whatsoever, so this would become "the witness who is faithful who is
the first...". We can have a whole string of such relative constructions in
Sabaot.

Here ends the story, as they say in Sabaot. I think we are closer together in
the end than when we started.

Thanks,
Iver Larsen
Kolding, Denmark
alice-iver_larsen@wycliffe.org


---
B-Greek home page: http://metalab.unc.edu/bgreek
You are currently subscribed to b-greek as: [jwrobie@mindspring.com]
To unsubscribe, forward this message to leave-b-greek-327Q@franklin.oit.unc.edu
To subscribe, send a message to subscribe-b-greek@franklin.oit.unc.edu




This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:36:48 EDT