Re: The sign relation
fritz@rodin.wustl.edu (Fritz Lehmann)
Date: Tue, 6 Jun 95 15:23:15 CDT
From: fritz@rodin.wustl.edu (Fritz Lehmann)
Message-id: <9506062023.AA05677@rodin.wustl.edu>
To: cg@cs.umn.edu
Subject: Re: The sign relation
Cc: fritz@rodin.wustl.edu, pdoudna@aol.com, srkb@cs.umbc.edu
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
Pat Cassidy responded on defining the "sign-relation".
Among other things, he said:
----begin quote---
... neither
the sign relation nor identity can be completely defined formally in a
way that all "interpretants" will interpret in an identical fashion,
as long as the set of interpretants includes humans. If we confine
ourselves to mechanical interpretants, perhaps it will be possible.
Clearly, we can't completely control the variability in the way humans
interpret signs, since human understanding depends upon individual
experience as well as the better-defined experiences from formal
teaching methods.
----end quote----
Agreed, that the sign relation can't be "completely defined"
and that "we can't completely control" human interpretations. But
there are broad areas of intersubjective agreement. Nobody considers
the Eiffel Tower a kumquat. As we've discussed earlier, the fact that
ontological classes have disputed definitions at the borderline
doesn't mean there can't be full agreement on core meanings. That
is, necessary (maybe not sufficient) axioms for membership in a
class. This applies as much to the sign-relation as to anything else,
as far as I can see.
[discussion of various technical triad-to-dyad tricks omitted]
-----begin quote---
So it seems that an attempt to break
a triadic relation into binary relations tends to present unnnecessary
complications. Other examples may be worse, for example
value_ratio(dollar,cent,100)
My attempts to represent this in binary relations were too clumsy
to repeat here. But more important than the difficulty of finding
a truly identical relation based on binary relations is the likelihood
that such relations will be less perspicuous to the humans who have to
build and use these systems. Unless there is some very compelling
need, due to computational efficiency or accuracy of representation,
I hope we can all agree to keep intuitively triadic relations as
formally triadic.
----end quote----
Agreed. Maybe even Pat Hayes will acquiesce in this.
---begin quote---
With respect to the sign relation, Fritz says that the
interpretant is a necessary part of the relation, with which I agree,
but he also says that time is not part of the relation, and this
doesn't seem right. He himself gives an example, later on, where
the name of an organization (or a country, or a person) may change.
This seems to be a powerful argument that if this relation is to be
represented as triadic, the "interpretant" must include the time at
which the relation is defined.
---end quote---
This is case (very typical) in which the relation as usually used
(the representation-relation) is actually a complex of individual
events. So the fact that "Dusseldorf" is used to refer to Dusseldorf
is grounded on a lot of little events: a person intending Dusseldorf
utters (says or writes) "Dusseldorf" and a person who reads or hears
"Dusseldorf" forms a mental concept which through further, internal
sign-relations represents the actual city in Germany. When naming
conventions change (and this often involves authorities and laws,
or just code authorities e.g. in the case of DUNS numbers or part numbers)
this gets transmitted to the participants and the representation-
events conform to the new convention. All this representation/
semiotic stuff is complicated; for most practical ontologies it's
a job the ontologist would like somebody else to have tackled. I
expect the REPRESENTATION/SEMIOTIC ontology to be one that practical
users will simply get "off the shelf" without worrying too much about
the details.
------begin quote----
My vote is for a triadic relation,
of "sign" A, "object" B, and "representation system" C,
in which we have buried the complexity of the representation
system (time of specification of the sign, human interpretant(s),
context, purpose, computer program using this sign, etc.) in
a single entity.
----end quote----
OK, as long as the more primitive underlying representation events
still involve the interpretant (utterer and receiving interpretant).
----begin quote---
I also think that "referent" is a better sign for B than "object" --
is "object" a standard term now?
Pat Cassidy
[alias "PatC" to avoid confusion with the eminent Pat (Hayes)]
---end quote----
"Referent" is fine, but it may cause some confusion among
Conceptual Graphs people, who are often discussing the CG "referent
field". "Object" has too many meanings as well, I guess.
By the way, Pat, I noticed that your revision of Roget's
Thesaurus is now the official version in the Gutenberg Project
and is referenced by thousands of people's and institutions'
World Wide Web homepages around the world. Is there an article
which describes your revisions and further proposed improvements?
Yours truly, Fritz Lehmann
GRANDAI Software, 4282 Sandburg Way, Irvine, CA 92715, U.S.A.
Tel:(714)-733-0566 Fax:(714)-733-0506 fritz@rodin.wustl.edu
=============================================================