[Top] [All Lists]

Re: [ontolog-forum] Ontology and kantian propositions

To: ontolog-forum@xxxxxxxxxxxxxxxx
From: "John F. Sowa" <sowa@xxxxxxxxxxx>
Date: Sat, 06 Aug 2011 23:02:16 -0400
Message-id: <4E3E0038.9070600@xxxxxxxxxxx>
On 8/6/2011 11:21 AM, Marcelino Sente wrote:
> Regarding the distinction between analytic and synthetic propositions,
> it seems that my impressions to fit you. I saw this relationship with
> the area of ontologies,but also do not know if it is correct. I searched
> the literature and found no work that makes this relationship with
> ontologies. I am interested in finding out ifthere is something to
> explore in ontologies (ontology engineering, computer science, etc.)
> using these Kantian distinctions.    (01)

This issue is definitely relevant to ontology.  The big 20th century
debate about analytic and synthetic propositions was between two
prominent philosophers, Rudolf Carnap and William Van Orman Quine.    (02)

For a good, readable summary of the issues from Carnap's point of view,
I recommend his _Introduction to the Philosophy of Science_, which
Amazon sells for $9.23.    (03)

The whole book is well worth reading, but you can skip to Chapter 18,
pp. 177-183 for a brief discussion of Kant's distinctions.  Then
skip to Chapters 27 and 28, pp. 257-274, for further discussion.
It also has pointers to Quine's objections for further reading.    (04)

To relate that to ontology:    (05)

  1. A statement (in logic or in a controlled natural language) is
     logically true iff its truth follows from the logical operators
     independently of the meaning of the terms in it.    (06)

     Examples:  "For all x, P(x) implies P(x)."
                "Everything red is red."    (07)

  2. A statement is analytically true iff its truth follows from
     the meaning of the terms in it.    (08)

     Ex:  "Everything red is colored."
          "If a cat is on a mat, then an animal is on something."
          "If Benedict is a bachelor, then Benedict is unmarried."    (09)

Many people are trying to use ontology to define the background
knowledge that would be sufficient to determine the analytically
true statements about the terms in their domain of interest.    (010)

However, many people (not just Quine) have observed that there
are many problematical issues.  For example, Pope Benedict is
unmarried, but most people would not call him a bachelor.    (011)

Since priests and other clergy in the Catholic Church take
a vow of celibacy, that makes them ineligible to get married
without breaking their vows.  Some do.  But if the Pope did,
he would create a huge controversy.  (I wish he would.)    (012)

Quine and others have pointed out that the number of such
problematical issues is enormous for any significant body
of knowledge.  That is an issue that makes many people
(including me) highly skeptical about the goal of creating
a single universal ontology that has detailed definitions
and axioms for specifying all analytical truths.    (013)

Instead, my recommendation is to use an underspecified
collection of terms with very few axioms.  For detailed
reasoning about specific problems, use those terms in an
open-ended collection of small theories (microtheories),
each of which is specialized for a narrow domain.    (014)

John    (015)

Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/  
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/  
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/ 
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J    (016)

<Prev in Thread] Current Thread [Next in Thread>