RE: Improbable Probability Statements

From: Jim Beale (
Date: Sat May 24 1997 - 05:22:06 EDT

At 12:50 PM +0000 5/23/97, Clayton Bartholomew wrote:

>Jim wrote:
>Surface irregularities do not negate the existence of a perfectly
>orderly, rational, deep structure.
>Are natural languages *perfectly orderly, rational?* This sounds to me
>like 19th century science, which was convinced that under all the
>apparent complexity is a *perfectly orderly, rational* world which
>requires nothing but cautious observation to explain. I have
>reservations about this.

Hi Clayton;-

This sure is an interesting topic!! I think we should change
the Subject header though. :-)

I didn't say that natural languages are perfectly orderly and
rational. I wouldn't deny it, but then I would suggest that
a distinction be made between the language *in essentia* and the
language *per accidens*; and in an analysis of the language per
accidens there are instances of competent usage and instances
of incompetent usage. I wouldn't think it reasonable to infer
anything about the English language from a witness in a courtroom
who says, "I ain't seen nothing, nohow. I don't gotta tell yous
nuttin anyways (then spits;-)" or an example of the violation of
a selectional rule that Chomsky gives, "colorless green ideas
sleep furiously." Such speech acts do not display any genuine
irregularities in the structure of the language, but rather the
linguistic incompetence of the speaker. And we do not admit such
sentences as evidence into the grammatical courtroom, because,
having an innate conception of the essentials of the universal
grammar, we know (_partly_ a priori) that they are not genuine.

And my distinction was between the universal syntactical and
semantic structure of *language-in-general* and any specific
language with its specific idiosyncracies. The universal and
the particular are related by means of a generative grammar;
which is what we are (supposed to be;) attempting to formulate
inductively for Greek. The rules (inductive generalizations) are
formed in an attempt to explain what we already know; that is,
how it is that language is a bearer of meaning. We are merely
trying to codify what we already understand quite well - though
we can't quite put our finger on how we know (perhaps John 1:9?).
These rules are extraordinarily complex because of the ways in
which the universal grammar must be mapped to express the thoughts
of free agents in the midst of complex intellectual surroundings.

>I am not arguing for chaos. In fact I think the term *Chaos* as it is
>currently used by the Chaos Cult is a loaded word. The class of
>phenomena which are being called *Chaos* are really *complex systems.*
>There is order there, but order of a complexity that it does not
>appear to the casual empiricist.

Substitute 'non-linear' for 'complex'. There are complex linear
systems which never exhibit chaotic behavior, and the simplest
of non-linear systems (e.g. x[n+1] = 2*x[n]^2-1; where x[0]=.75)
can generate chaotic results. The point of the "chaos cult" is
that these things which are apparently random have quite simple
explanations. The results appear indeterminate, but the equation
which produced them is deterministic, and extraordinarily simple.

>I am suggesting that NT Greek and all natural languages are like
>*complex systems.* Complex systems are difficult to model. They behave
>according to rules, but all the rules are not currently understood.
>*Complex Systems* are modeled with high level mathematics (certainly
>not my forte). But for those who cannot do this kind of work, models
>can be built using metaphors to help us guess more accurately at what
>is taking place.

The rules of the surface syntax are established by inductive
generalization of actual instances of competent usage. The rules
are extremely complex. But the extraordinary thing is that we
understand the rules, because we can make sense of completely
novel sentences. We know the rules, but we have a most difficult
time writing them down!

>Now I am assuming that when Jim says * a perfectly orderly, rational,
>deep structure* this is not the kind of thing he is talking about. I
>am assuming that Jim thinks that the *perfectly orderly, rational*
>aspect of language is observable using traditional methods of
>empiricism. This notion is what I think has serious problems.

Indeed, I am saying nothing of the sort. Deep structure is not
immediately observable (but it is logically necessary to presuppose
nonetheless!) Perhaps a quote will help to clarify my position:

   ... the deep structures for which universality is claimed
   may be quite distinct from the surface structures of
   sentences as they actually appear. Consequently, there
   is no reason to expect uniformity of surface structures...
   (_Aspects of the Theory of Syntax_, 118)

Anyone who argues for innate knowledge of universals is clearly not
an empiricist. Empiricists, such as Quine, are behaviorists when it
comes to language learning. In my opinion, behaviorism is a most
ridiculous theory. The inability of the behaviorist to give an account
of how we can use language POLUTROPWS - in ways that are creative and
outside the scope of previous experience - scuttles his theory every
time. We all understand radically novel sentences, and this is just
what the behaviorist _cannot_ explain.

>For the purpose of translation and exegesis of biblical Greek, I think
>that a certain tolerance of indeterminacy needs to be cultivated. One
>analogy I use to help me with this is borrowed from electronics.
>Analog circuits have signals and they have noise. If the signal to
>noise ratio is high enough there is no problem in understanding the
>signal. Indeterminacy in language is *like* (not equal to, but like)
>noise. We can read and understand the NT and LXX because the signal is
>strong and the noise is weak.

It is interesting that you say this. What is necessary here is to
presuppose an innate possession of linguistic universals. If this
was not the case, not only would the child not be able to learn a
language in the first place, but, one would not be able to properly
distinguish signal from noise (that is, as I take it, competent from
incompetent linguistic acts.) (see Chomsky, _Aspects of the Theory
of Syntax_, pp. 27, 47-59.)

>Now the *Complex Systems/Chaos* people have come along and told us
>that noise has structure, meaningful structure, if we can only decode
>it. This kind of thinking may be fruitful in building a better
>language model but it is going to be a technical business and few of
>us liberal arts types will be competent to participate.

Oh well! It's a sad fact that not everyone is gifted in every area!
It's a painful reality -- I'd love to be a great basketball player,
but I fear the worst!! It's just not going to happen.


In Christ,
Jim Beale

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:38:17 EDT