Re: SUO: RE: Defining words
Ian,
To clarify the point with regard to SUMO, I was making the
following claim:
1. Any language of the Common Logic family (KIF, CGs, OWL,
predicate calculus, etc.) has a minimal amount of
syntax and logical expressive power.
2. Every feature in the CL family is also present in the
"core vocabulary" every natural language that linguists
have ever analyzed and described on planet Earth.
3. By "core vocabulary", I mean the collection of words and
syntactic features that include the ability to express
the following:
a) Conjunction, disjunction, conditional ("if-then"),
negation, existence, and universal (all or every)
together with ways of showing scope (such as clause
or phrase boundaries and subordination).
b) Classes of words that represent people, things, events,
attributes, manners, times, places, purposes, and the
copulative ("is" or something like it), and relations
among these things. (Only the very general types at
the top are assumed -- no specific subtypes.)
c) Pronouns and other indexicals for showing cross
references.
d) Metalevel mechanisms for talking about words and
defining new ones in terms of paraphrases that may
take one or more sentences.
4. Given these features, anything that can be defined in any
CL language can also be defined by the core features of
any natural language.
5. As evidence, there are large numbers of NLs on earth today,
which have been expanded from their agrarian roots to the
needs of fully industrial societies with no change to their
core vocabularies. As an example, Japan developed from an
iron-age agrarian society in the late 19th century to a
modern industrial society in less than a century. In the
process, their language borrowed large numbers of new words,
but the core syntax and vocabulary remains the same.
6. The example of the difficulty of teaching stone-age
people the meaning of something they have never
encountered has nothing to do with the capability of
the language. But it has everything to do with the fact
that people cannot learn new words successfully when they
have no experience with the things they represent.
IN>... What we're arguing about is whether any natural language
> can express any concept whatsoever. This is not a definitional
> claim about natural languages; it is an empirical claim which
> can be falsified by considering actual languages, e.g.
> languages of stone-age cultures versus languages of
> industrialized cultures.
As I pointed out, that has nothing to do with the nature of
the language -- it is purely a matter of how people learn
new words.
IN>OK, this is where we disagree. You're arguing that all natural languages
>are equally expressive. One consequence of this is that it should be
>possible to express any English concept in any stone-age language without
>using any extra-linguistic means, e.g. without pointing or other gestures.
More precisely, I claim that you would teach stone-age people
that concept in the same way that you would teach an English
speaking child that concept. You can test that point by going
to some Amish community, where the use of modern machinery,
radio, and TV is not allowed.
IN>... my point that all natural languages are not fully expressive.
The only evidence you have given is about somebody who
tried to convey the concept of "book" to people who could
not read. That only proves that if concept B depends on
concept A, you have to teach A before you can teach B.
IN> If the training can only be accomplished extra-
> linguistically, then your claim is false.
The example of Helen Keller demonstrates that people are
capable of acquiring a very sophisticated education without
being able to see or hear. It may take a long time, and it
may require someone with high intelligence and motivation,
but it is possible. Helen K. learned a lot by reading books
in Braille without being able to see the pictures. After she
learned the core vocabulary and a lot of concepts by touch,
the rest was acquired by linguistic means.
John