[b-greek] Re: Discreet vs prototypical categorization

From: clayton stirling bartholomew (c.s.bartholomew@worldnet.att.net)
Date: Thu Oct 19 2000 - 14:52:22 EDT

Whew! How am I going to answer all this mail? Well first of all thanks to
Randall Buth for his insightful analysis (off list), and also to Mark Beatty
and Kimmo Huovila who clearly have a grasp of what is at stake in this

Now for a few clarifications. Note the qualifier (quote below) in: "the
WHOLESALE abandonment of discrete functions." I am not denying the existence
of scalar properties in natural language or that scalars are a useful method
for language analysis. What I am questioning is the use of scalars as a
quick fix for all the problems we have with discrete categories. When we run
into difficulties with fuzzy boundaries on our functional categories we
should not just automatically accept a scalar solution just because it
appears to solve our problem. In the long run the scalar quick fix may not
be a solution at all but simply a covert means of postponing our problem, a
means of avoiding the more difficult problem of reworking our taxonomy of

CSB said:

>> My thinking on this is that the wholesale abandonment of discrete functions
>> is the equivalent of intellectual suicide. It is jumping ship in the midst
>> of a gale. The weather was getting rough with discreet functions so lets
>> jump overboard and drown in scalars.

on 10/19/00 7:21 AM, Kimmo Huovila wrote:

> . . . Languages tend to describe
> prototypical categories in discrete terms (ie, a construction is or is
> not used to describe the category, which itself is not discrete), with
> the result that most linguistic categories are somewhat fuzzy.

Agreed. They are fuzzy, some of them are so fuzzy they are useless.

>Not to
> recognize this is to force the language to suit the model built by the
> analyst, but one that does not really help discover the original meaning
> (either of the text nor of the grammatical categories).

Total agreement. When I hear a second year seminarian all in a quandary over
which little pigeon hole to stuff this constituent into I want to say "STOP"
this is a futile activity. The pigeon hole is an artificial construct and
needs to be understood as such.

>This is a
> problem I feel many theologians often fall victim to (just my opinion).

Why pick on theologians? I think that NT scholars are probably just as
guilty of this as anyone.

> If you proceed from the assumption that linguistic categories describe
> discreet categories (as opposed to linguistic categorization by
> prototype), and you try to define your passage in terms of that
> categorization, you may very well end up making questions that were not
> relevant at all to the original speaker or the audience


>, especially if
> you tend to multiply categories ad infinitum. It is more useful to see
> the range of typical and possible meanings and relate this to the
> context.

Yes, but isn't this either/or a bit hasty? Should we not challenge this
either/or? This is the problem I am addressing. The syllogism:
If your categories or multiplying out of control you must be dealing with a
scalar language phenomena. I don't think this is beyond dispute.

Returning to Greek exegetical problems:

How do you make exegetical decisions based on hypothetical position on a
cline or a scalar range? We are not just talking about one range. We are
talking about a single constituent in context, say a genitive substantive,
having multiple semantic/pragmatic properties all of which fall on a
different scalar range.

So we end up with having to think about our genitive substantive in context
as a point of intersection in a space with N dimensions where all of these
scalar ranges meet. How on earth are you going to do exegesis with a model
like that?

Fuzzy categories, with all their limitations are useful for the purpose of
discussion. I cannot see how we can even discuss a genitive substantive in
context using a scalar model. This is a pragmatic objection.

We also need to ask the question if genitive substantive in context really
has all these scalar properties. We could pose this as a question: Is the
Greek genitive case a scalar or non-scalar phenomenon? The obvious answer to
this is that Greek cases really don't fall into categories such as
scalar/non-scalar but are located somewhere along a cline between these
points. In other words scalar/non-scalar is not a discrete category.

I hope this suggestion is sufficiently horrifying to help some people get
the point I am making here.

>This does not deny the possibility to use discrete categories
> to describe them, but one must be careful to realize that the discreet
> metalanguage does not make the phenomenon itself discrete. I assume most
> linguists working with actual corpora are familiar with the difficulty
> of trying to define where to draw the line where a phenomenon passes
> from one category to another.

Agreed, this difficulty is the reason for this discussion. If there was no
difficulty there would be nothing to discuss.

> I am not trying to get rid of discrete functions altogether (and I do
> not think that the trend has been that, either),

True, even T. Givon who is deep into the scalar approach recognizes the need
to sort out which language phenomena are scalar and which are not.

>but I cannot see why it
> would be an intellectual suicide to realize the scalar (or prototypical)
> nature of language, and come fully to grips with it

Again, the word "WHOLESALE" is being over looked. I am objecting to the
uncritical acceptance of the scalar solution as a cure all for our problems
with discrete categories. When we find a category has fuzzy boundaries we
should not just leap immediately to the conclusion that we have encountered
a scalar language phenomenon. I am not accusing you of doing this, but it
seems that "scalar" has become an axiom and I don't think it deserves the
status of a axiom. Every time someone suggests that we see some language
phenomenon as scalar I think we should demand a defense of that statement,
not just accept it.

 Kimmo, Randall, Mark Beatty

Thanks for the discussion. It has helped clarify things.


Clayton Stirling Bartholomew
Three Tree Point
P.O. Box 255 Seahurst WA 98062

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 - 15:36:39 EDT