ONT Re: New List & Classification of Signs
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
What are some of the things that we would
like to be able to do in the real world
with our data and our logic, with our
DB's and our KB's?
Well, for instance, one might take a notion
to wonder about the truth of the statement:
If Perth is 400 miles from Omaha then Perth is in America.
First of all, a certain amount of logical and semiotic analysis
would be necessary to decide whether and under what conditions
any such statement can even possess a determinate truth value.
We all know the familiar conversational implicatures and
the polite pragmatic presuppositions that would attend such
a statement if we met with it in the context of a logic text,
as a participant in a classroom discussion, or as a contestant
on a logic exam, but you are falling into forgetfulness -- I said
the "real world", right up front, and never yet the twain have met,
so you can start by forgetting those convenient, all too convenient
assumptions, the rules of engagement in classroom arenas and theatres.
Here's a sample, just in case you've forgotten what it's time to forget:
There's no such thing as a proper name, a sign so magically sympathetic with
its one and only intended object that it can pick it out of a crowded cosmos
and will this one thing with purity of heart as its uniquely sole denotation.
Strings of char like "Omaha" and "Perth" can only do so much, nothing at all
by themselves, and serving as signs that are capable of determining singular
existents just isn't one of them. And if you're determined to find the rest
of that misplaced determination in the context, countryside, county, environ,
neighborhood, province, state, surround, or vicinity then you are barking up
the wrong bailiwick. All of which the pragmatic thinker says by saying that
words from "Time" to "Timbuktu" are properly read as "symbols", and if they
mean, then they mean what they mean just because some interpretant says so.
And before we get done doing that, as the requisite analysis of the question
may itself be a matter of experimental trial and persistent error that will
have to be carried on concurrently with our tries at answering it, we will
want to check the coverage of the statement against a real world database,
or many, just to see if our mini-theory has models or counter-models there,
in a representation where models or counter-models really count for something.
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o