Re: On the definition of "ontology"
phayes@ai.uiuc.edu (Pat Hayes)
Message-id: <199510041548.AA00899@a.cs.uiuc.edu>
X-Sender: phayes@tubman.cs.uiuc.edu
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 4 Oct 1995 11:04:21 -0600
To: vet@cs.utwente.nl (Paul van der Vet), guarino@ladseb.pd.cnr.it,
srkb@cs.umbc.edu
From: phayes@ai.uiuc.edu (Pat Hayes)
Subject: Re: On the definition of "ontology"
Cc: piet-hein.speel@2488taux.urlnl.sprint.com
Sender: owner-srkb@cs.umbc.edu
Precedence: bulk
At 2:30 PM 10/4/95 +0100, Paul van der Vet wrote:
>Nicola Guarino writes:
>
>> I would now propose YET ANOTHER definition of an ontology, slightly
>> different (in the form, but not in the content) from that discussed in the
>> paper "Ontologies and Knowledge Bases: Towards a Terminological
>> Clarification", accessible on the web site reported below:
>>
>> "An ontology is a specification of the indended models of a logical
>>language".
>
>I'm somewhat surprised by the plural - when you develop a
>knowledge-based system you'll have one intended model in mind.
This raises an interesting issue (which may be getting too philosphical for
skrb, but here goes anyway.) What one 'has in mind' is presumably expressed
there in some mental representation. Now, how do we know that this
mentalese representation in fact has unique models? All the lessons of
logic would suggest that it usually doesn't; that usually there will be
nonstandard models of our thoughts. (This will almost certainly be true,
for example, if theoremhood in our mentalese is RE and it is given a
semantics which reflects this accurately.)
Now, if this is the case, then what warrant can we have for insisting that
there is "a" model that we have in mind? We have no special access to
models of our mental language: we can only think of things IN that
language.
Comments,anyone? (David McAllester and I have been arguing offline about
this recently. David's view is that we must simply accept our own
mentalese, that his is at least second-order, and that he can then *prove*
that (for example) the integers are unique; my response is that this
conclusion holds only relative to a particular semantic interpretation of
that logic.)
Pat Hayes
------------------------------------------------------------------------------
Beckman Institute (217)244 1616 office
405 North Mathews Avenue (415)855 9043 or (217)328 3947 home
Urbana, Il. 61801 (217)244 8371 fax
Phayes@cs.uiuc.edu