From: Mike Sangrey (msangrey@BlueFeltHat.org)
Date: Thu Jul 26 2001 - 11:42:26 EDT
I appreciate what Kenneth Litwak said regarding formal languages (eg
Java). I'm a Computer Engineer by trade and education and I have
liked the CLEAN distinction between syntax and semantics afforded to
computer languages. The simplicity of computer languages, relatively
speaking, is both bane and boon.
The phrase:
EN THi PROSEUCHi TOU QEOU.
is, IMO, a good test case.
PROSEUCHi is a verbal noun. BUT! that description refers to
semantics, NOT syntax. It's NOT a participle, THAT would be a verbal
noun in a syntactic sense. When we describe PROSEUCH as a verbal noun
within the semantic level, we are referring to the lexical features of
the word which require an actor to perform the action being
referenced. When we use the noun `plate' (as in dinner plate) no
action comes to mind.
TOU QEOU is genitive. That's syntax and the fact that it is
collocated with PROSEUCHi is also syntactic.
There's also features of the words which allow them to be collocated
the way they are. At least one part of language which allows such
efficient use is the branch of semantics called lexical semantics.
There is a vocabulary developing which helps us describe what is going
on.
As a contrary example, I can't say
EN THi PROSEUCHi TOU PAROYIDOS
I might add, though, that if the author had built a fair amount of
context around this phrase, it may, in fact, be understood. "I pray
[to] God" is quite easily understood on its own; "I pray [to] plate"
needs more textual context to understand what that might mean. The
point, though, is that PROSEUCH and QEOS can be quite efficiently
collocated since there is considerable extra-linguistic information
available to disambiguate the genitive for the native, fluent, Greek
speaker/hearer. In other words, there exists information utilized to
determine meaning which is NOT in the text.[1] This information can, I
believe, be semantically described. That is, we've got some
vocabulary to describe what is going on.
So, when one sees a genitive collocated with its head noun there are a
number of textual and extra-textual data which come to play. This is
really my grumble with the grammars and those lists--I don't have a
description of the semantic features associated with a given syntactic
construction which determine the meaning.[2] They give me a list of
the semantically loaded syntactic features, but do not give me a
semantic map to help me walk through them.
This is why I appreciated Kenneth's "I raise you and call your poker
hand." When ever someone says, "Well, you know, experience is the
best teacher," which, interpreted is a lot like, "Well, just read more
Greek," I think of that Ben Franklin quote (or misquote) and recall
what he REALLY said: "Experience is a dear teacher, a fool will learn
by no other." The point being that what I need is supervised and
somewhat structured (guided) experience. Just throwing me at the wall
of experience helps me stick, but it also hurts.
In that regard I want to thank this list. Part of that role is filled
by the many who help out on this list. THANK YOU! But, it seems to
me, that the grammars can do better. I can't have a dedicated teacher
kindly watching over my shoulder until the language clicks in place
(or there's enough dents in the wall for me to grab a hold of). And
(with no offense intended) I also need some scientific rigor since I'm
not peacefully convinced that the Big Greek's gut feel in a particular
context is something we should just happily hope in.
So, what is it I need? Well, I've got a list. O! goodie. I've got
examples under each list item. That's helpful, too. But what I don't
have is a description USING SEMANTIC TERMS which tell me how to take
that syntax, located as it is in a particular context, and from that
data determine what has or needs to change in the cognitive
environment inside my head[3]. In that regard, grammars also don't
seem to address how to exegete the semantic features from the text.
(I'm using `exegete' rather loosely to emphasize that what I'm looking
for comes FROM the text, but I'm also referring to the semantic
features such as "O! That's a verbal noun".)
Now, I feel--I really do--that I'm asking for the moon. So, let me
ask a somewhat gentler question: what are some of the pieces we would
have to have in order for a book to look like that? What should that
book look like?
For example, let me ask this question of those far more expert in the
Greek language than me. If the head noun to a genitive is a verbal
noun (in the semantic sense), then MUST the genitive be either
objective or subjective? That is, will the verbal noun "expect" the
collocated genitive to be either the actor or the thing acted upon.
If not, what other factor(s) come into play? And how are those
factors semantically interacting?
Thank you all!
Mike Sangrey
---
[1] This is why lawyers will ALWAYS have jobs. It is impossible to code
ALL information relevant to interpretation into a given text (eg contract).
An author only tries to code SUFFICIENT information for interpretation.
Bias seeks to manipulate this extra-linguistic information.
[2] I will quite readily admit that `determine' is too strict a term
to use here. Communication is far too highly complex for me to hope
for a deterministic model. However, I am hoping for something better
than `gut feel'. `Gut feel' is nothing more than a bias engendered
best fit algorithm.
[3] In my mother tongue, as I process each word, various changes occur
inside my head. Certain things are turned on, turned off, or readied
on the side lines (eg. if I say `restaurant' things like `menu' will
be more readily accessed. Like `atmosphere' takes on a specific meaning
in the context of `restaurant'). This, at least to my way of
thinking, essentially defines `fluency' to me.
--
Mike Sangrey
msangrey@BlueFeltHat.org
Landisburg, Pa.
"The first one last wins."
"A net of highly cohesive details reveals the truth."
---
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 - 15:37:02 EDT