W3C home > Mailing lists > Public > www-rdf-logic@w3.org > July 2002

Re: questions on assertion

From: Giles Hogben <giles.hogben@jrc.it>
Date: Wed, 10 Jul 2002 10:21:12 +0200
Message-ID: <008001c227ea$c0b5b6c0$162abf8b@pcdsa22>
To: "pat hayes" <phayes@ai.uwf.edu>, <www-rdf-logic@w3.org>, <www-rdf-interest@w3.org>

> >But what about this sentence in the RDF model spec (also in the latest
> >draft)?
> >
> >"This document describes a model theory for RDF(S) which treats the
language
> >as simple assertional language, in which each triple makes a distinct
> >assertion and the meaning of any triple is not changed by adding other
> >triples"
> >
> >An assertion is traditionally a statement that you believe something to
be
> >true.
>
> NO, I think that is a slight misstatement. An assertion is a
> statement that something is true; it need not make any reference to
> belief. Usually, when a rational agent asserts something, we assume
> that they believe it to be true; but that is a pragmatic inference,
> not part of the very notion of asserting. Its easy to think of cases
> where it breaks down, eg telling lies.
>
> The notion of assertion is much simpler and more fundamental than
> that of belief. In the context of software agents, one can argue
> whether there are any beliefs involved at all, but making an
> assertion by publishing a sentence is a clear, unambiguous,
> measurable kind of event.
>
I don't agree here. I think what you are saying is circular.
What is a "statement that something is true" if it does not contain some
notion of belief.
I am not getting philosophical here, I am talking about the practical matter
of how such statements can be used in environments where trust is critical.
In the area of trust, I would take a particular meaning of belief, which is
that socially, I commit myself to undergo a sanction if the statement turns
out to be false. For example if an agent makes an assertion that a can of
baked beans contains baked beans, and it does not, the law is able to use
that statement to sue the company who manages the agent for breach of the
trade description act. In trust critical environments, it is crucial that I
am able to express a semantic which expresses my legal/social commitment to
a statement.Of course people are able to make contextual inferences about
such things, but there seems to be no reason why the SW could not be built
from the bottom up to support this kind of commitment. At the moment, it
seems that the semantics is too confused to do this.
Received on Wednesday, 10 July 2002 04:16:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:42 GMT