Re: [ontolog-forum] Ontologies and languages

[Thomas Johnston <tmj44p@xxxxxxx>]

It seems to me that John's four points reduce to the recommendation in his #1. 

And I would like to point out that, in the process of doing #2, a lot more than terminological simplification is taking place. As #2 is being done, an immense number of ontological choices are being made. I would recommend that, to the best of the developers' abilities, those choices be explicitly documented. 

For example, ontology can hardly be done without deciding on a set of basic categories. Different sets of categories have been defined by Aristotle, various medieval philosophers such as Scotus and Ockham, Kant, Husserl, Roderick Chisholm, Reinhardt Grossman, Hoffman and Rosenkrantz, E. J. Lowe and others. 

Aside: it just occurred to me that some might wonder what more there is to ontology than deciding on a set of categories, and working out the consequences of that decision. Basically, what more there is is describing how change happens. This corresponds very nicely, in Aristotle, to his major works (i) Categories and (ii) Metaphysics, in the latter of which he develops a theory of essential change vs. accidental change, and of what it is that endures through change.

In this connection, I recommend an excellent article, "Categories", from the Stanford Encyclopedia of Philosophy. References to the above-mentioned can be found there, along with several different sets of categories, laid out in outline form. The article is far from a complete enumeration (skipping over a lot of medieval work, for example). But it's strong on recent (20th century onwards) work, and its discussion of methodological issues is excellent.

Then there is the question of methodology, for example whether we base our categories on linguistic distinctions, or on phenomenological insight. Gilbert Ryle did a lot of work on category mistakes. Frege based his ontology on the distinction between names and predicates. In distinguishing kinds of things, Michael Dummett emphasizes the importance of identity criteria strong enough to support, not just identification, but also re-identification. My own approach, in Bitemporal Data: Theory and Practice, is to supplement an Ockhamist reduction of Aristotelian categories with the additional category of events, and to insist on the importance, with Dummett, of re-identification (This thing here is the same thing I encountered yesterday, over there) of individuals within any category type.

So in developing categories for a simple way of expressing properties and relationships of objects (and where are events and processes in PROSE, by the way?), we are not winnowing philosophical chaff to get to the wheat kernels. We are making choices all along the way. What the history of Philosophy, and especially of ontology, shows us is that important philosophers have not failed at the winnowing task, but simply worked hard and carefully to -- with apologies for the shift of metaphor -- slice the ontological pie in different ways. (Or is it to slice a particular ontological pie, chosen from God's bakery, in a particular way?)

Ultimately, we simply have to choose the best set of categories we can come up with, and then proceed from there. But simple personal intuition, unsupported by an awarness of what the great minds of the past, working on the same issues, have come up with, is not a good way to start. 

What this relatively new field of formal ontology / ontology engineering does, that classical work in ontology did not do nearly so clearly (and, except for Ockham really didn't do at all), was to use formalizations of some of the patterns of inference and reasoning apparent in ordinary discourse to draw out the deductive consequences of initial/earlier ontological claims. This suffers from the fact that any no formal deductive system expresses all the inferential patterns found in ordinary discourse. But I think it is a price well worth paying. I think it is a major advance in ontology to have started down this path. 

One more book on these issues that has been important to me is Ernest Davis, Representations of Commonsense Knowledge (Morgan-Kaufmann, 1990), especially his chapters on Quantity, Time, Space, Physics, Minds, Plans and Goals, and Society.

Regards,

Tom

On Friday, June 19, 2015 9:48 AM, John F Sowa <sowa@xxxxxxxxxxx> wrote:

Ed, William, and Pat,

EL
> Students everywhere are taught how to arrange pieces of information by
> educators who are unaware of elementary pieces of information that are
> well designed to be easily arranged. To me, few educational practices
> seem more obviously unreasonable.

I agree.  Ontology has been debated in philosophical circles for
at least 2500 years, and many versions of formal logic have been
used to define and reason about ontology for about 2300 years.

The result is an immense literature with a huge number of terms
that have been used (formally and informally) in a very large
number of ways.

As I have said for years, I like the ideas behind your original
system called PROSE (Properties and Relations of Objects Simply
Expressed).  As a general strategy, I would recommend:

  1. A formal logic with the barest *minimum* amount of terminology.
    It must at least contain FOL + the option of quantifying over
    functions and relations + the option of using metalanguage for
    talking about whatever languages are being defined.

  2. A huge *purging* of the immense philosophical terminology
    to a minimal set that is formally defined in the logic of #1.

  3. The option of designing an open-ended family of formal notations,
    linear and/or graphic, that have a precise mapping to #1 and #2.

  4. There may be huge debates about how to map NL terminology
    (including any and all terms in philosophy, science, business,
    the arts, etc., to the terms in point #2).  But any proposed
    solution must be defined in the logic and minimal terminology of
    points #1 and #2 (or #3, which is defined in terms of #1 and #2).

WF
>> ....sigh...

PH
> This is why I decline to take part in these, um, debates.

I sympathize.  I have been tied up with some deadlines for the
past week.  But I'll make a few more comments about the debates
in this thread and the "qua" thread -- perhaps over the weekend.

John

_________________________________________________________________
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

_________________________________________________________________
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    (01)