[b-greek] Word Order and Zero Anaphora

From: Wayne Leman (wleman@mcn.net)
Date: Fri Oct 06 2000 - 00:08:26 EDT

Clay responded:

> Has Givon ever been connected with SIL in any way or did you just happen
> study under him at U of Oregon?

Both, actually. I was in a regular grad program there in the Dept. of
Linguistics. Then SIl moved its N.W. school to the Univ. of Oregon campus at
the invitation of our Dept. Givon was one of the voices in the dept. who was
friendly toward that decision.

> He sounds like a bible translator, that is
> why I ask.

He is actually not, altho, because he likes SIL people (and they are quite
likeable, if I say so myself!!), he has done some field research where SIL
people work.

>There are a suspicious number of tribal languages in his examples
> and then there is Biblical Hebrew.

He simply loves to research languages. He has done much more research on
non-Indo-European languages than most linguists these days.

> > . . . I would
> > definitely not want to call NT Greek an SV (or VS) language simply
> > it has pronominal suffixes instead of prefixes.
> I totally agree. Givon is working with a model developed by Greenberg

Well, he's very aware of Greenberg's work, and would have probably have

agreed with much of Greenberg's earlier work in Africa, but I suspect they
part ways significantly on Greenberg's more recent over-arching theories
about a very small number of language families.

> Givon doesn't claim that it has anything like universal validity. He
> critques Greenberg's model on several points.

I'm not surprised. Greenberg's recent ideas are more popular in the popular
press than they are among linguists.

> >I personally don't think NT
> > Greek has any underlying *syntactic* word order . . .
> This may be the case but if NT Greek does not have some sort of "unmarked
> word order" were are you going to get your pragmatic marking for point of
> departure and so forth? You cannot have "fronted" elements unless you have
> an unmarked word order. It is like having black camels crossing a black
> desert at night. You need to have a basic pattern or you have no marking
> all.

I understand what you are saying, Clay, but I still don't think you have to
have an unmarked word order in order (!) to recognize pragmatic word
ordering. I'm not theoretically astute enuf to come up with a convincing
argument off the top of my head, but I just don't see the evidence in the
data, even tho you raise a very sound logical question about this issue. In
part I would answer that in a language where we believe there is pragmatic
constituent ordering it is quite possible that there is pragmatic "focus" of
some kind in every clause (or nearly every clause; granted, there are some
elements of syntactic constituent ordering in Greek, which is not surprising
since few languages are pure representatives of any one typological model.
Many, perhaps most, languages are mixed in terms of typology. Greek
questions words, for instance, can be viewed as being syntactically ordered
positionally, but then maybe it's still really pragmatics since questioning
of some major WH element is a form of pragmatic focus. It may sound a bit
circular but it really isn't when we get to the underlying nature of
questioning, negation, and similar functions that can be performed with

This, is, ultimately, what my wife found out for Cheyenne. None of the
standards test and procedures for determining dominant word orders worked
for Cheyenne. Finally, she was able to come up with some statistically
significant results for ordering when she used the concept of
"newsworthiness" which was promoted by linguist Marianne Mithun. It turns
out, according to these results, that each Cheyenne clause has as its
clause-initial element the constituent which is most newsworthy in that
clause. And tracking these newsworthy elements clause by clause helps one
see how the discourse is strung together pragmatically, which is something
which is critical for doing good exegesis. We really are talking about Greek
here, in case you're wondering, Carl!! IF, and that's a big if, Greek has
something similar going on, then the hypothesis would be that the
clause-"initial" element has some pragmatically higher status than other
elements in that clause. We need to chart topic and thematic continuity
within Greek episodes and pericopes, to determine what are the anaphoric
elements (old vs. new info), what is added and the special contribution of
each subsequent clause, etc. When we take a forest-discourse/pragmatic
approach to the study of Greek, we still pay attention to the trees, but we
start to see the trees are important markers within the bigger picture of
the forest. If a clausal constituent appears clause-initially in Greek,
there must be some reason for it. We don't want to be in a guessing game
about the reason since we don't want to identify functions for something and
then every time we see that element or "process" we say we have that
function, since this is the old logical fallacy of circular reasoning. But
with careful, empirical charting of Greek discourse we can begin to see
patterns which will likely hold up if we under that greatest test of all
replication. Which is where our friend, Randall Buth, must come in with his
fluent speakers so we can replicate the results!!!! :) (Seriously, I do
believe that Randall's approach can develop a level of fluency in
individuals where they, like infants do when learning a language,
subjectively internalize aspects of the grammar, and can then reproduce
proper grammar, including proper pragmatics, with newly created utterances.)

I am NOT making the simplistic suggestion that we just look at
clause-initial elements in Greek and start making linguistically significant
pronouncements about them. People have been studying Greek for too long to
be so arrogant as this. But I do think we can make some good progress at
understanding the pragmatic ordering of Greek word order, if we back up a
bit and look at rhetorical and discourse structure of logical chunks of
Greek discourse.

Somehow, in a way that I cannot yet put my finger on, I don't think we need
an underlying word order for this. For one thing, it is not necessary, of
course, to have an underlying word order to detect what is clause-initial.
That much is quite simple. There may be other issues which are not so
simple, such as what about non-clause-initial ordering. But even there, it
may be possible to observe the pragmatic rules for such ordering once we
begin to better understand the cognitive/pragmatic patterns running through
Greek discourse. We start with the assumption we do for all of language (or
at least *I* do, since I'm not a postmodernist and don't believe that
language is a theatre of the absurd with just willy-nilly utterances that
don't really mean much of anything), that the original author had some ideas
he was trying to communicate. And he followed a plan, his rhetorical plan,
for getting those ideas across, step by step, logically, or narratively
sequentially, etc.

Typically languages put either at the front or ends of sentences (or
clauses) information which is given/expected/old. We can track this kind of
"anaphoric" info in Greek pericopes. That can give us some insight into the
positioning of proper nouns which are anaphoric. Greek repeats anaphoric
proper nouns more than English does, from my observations. We can chart this
stuff and see if we can come up with statistically significant correlations
to common pragmatic functions.

I hope this makes some sense. You can probably tell that I have sat at the
foot of Givon for some time, from much of what I've said and even my slash
marks, discourse/pragmatic :)


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 - 15:36:38 EDT