Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: SUO: RE: The Story So Far




Ian Niles <iniles@teknowledge.com>, to Matthew West:

>.... The formal ontologies
>relating to numbers, measures, organisms, agents, etc. that have been
>incorporated into the merged ontology do not, from what I can see, hinge on
>debates about nominalism versus realism or about the source and
>justification of our concepts or etc.

Maybe not, but they *do* encode (and, in the way they  are written 
presume) specific ontological stances on a number of issues, and 
thereby implicitly rejecting other positions.

> >
> > I, Pat Hayes, John Sowa, Chris Partridge, Nicola Guarino
> > (please correct me
> > if I am mistaken) and perhaps others would argue that we
> > would be better
> > served by understanding the existing World Viewpoints and
> > relating them to
> > each other.
>
>I think everything here hinges on your notion of "World Viewpoints".  If, by
>this, you mean formal ontologies relating to mereology, biological
>categories, set theory, etc., then I agree that we are well served by trying
>to understand them and relate them to each other (and incorporate them into
>a merged ontology).

Take an example, the axioms for 'part' in the merged ontology. These 
axioms can be thought of as applying to 'solid everyday things' in 
the 3-d sense, which would be continuants in an endurantist ontology; 
or they can be thought of as applying to pieces of 3-d space (what 
might be termed merological locations); or they can be applied to 4-d 
histories; or they can be applied to occurrents in an endurantist 
ontology, but not at the same time as to occurrents, because they 
would use a different notion of 'part' (episode rather than 
continuant-part). They could be applied to a much broader universe of 
mereosums which would include things like 'the accumulated stuff in 
any set of pieces of space', or they could be restricted to sets 
consisting of countable sums of spheres in n-space. If collections 
are identified with mereosums, as Chris P and Nicola urge, they could 
be applied to collections (with some care), but if not then not. They 
might or might not be taken to include surfaces as parts (I earlier 
thought they excluded them, but it is possible to reinterpret 'part' 
so that surfaces are parts), but one cannot be agnostic about it as 
the truth of other axioms turns on the decision.  As they stand they 
are the ontological equivalent of a toga hanging on a hook; they 
could be worn a lot of different ways, but - and this is the crucial 
point - *not all at once*. If two people try to wear the toga in 
different ways at the same time, it will break.

>Furthermore, I agree that there are some deep
>theoretical issues that we have to work through, e.g. the representation of
>time.  If these latter issues prevent us from developing a single,
>monolithic ontology, then I think we should get as clear as possible about
>the the different viewpoints in the debate and then formalize and package
>these different viewpoints in independent modules.  It would be unfortunate
>if we didn't end up with a single, overarching, and consistent conceptual
>schema, but perhaps this goal is unattainable.  If it is unattainable, then
>I think the best we can do is to make clear the various representational
>choices and bundle them up in consistent and independent packages.

What we are most in need of is a methodology for stating such 
packages and describing the relations between them. Until we have 
this we are bound to constantly descend into disagreements. I think 
that John Sowa probably has a head start on this business, and I 
concur with his earlier message.

> >
> > Discussion
> >
> > As far as I can see there are four possible outcomes to Ian's
> > approach.
> >
> > 1. Ian solves the problem that previous philosophers have
> > failed to solve
> > and creates a single ontology that everyone agrees is how the
> > world is.
> >
> > 2. Ian recreates one of the possibilities that are already
> > known about.
> >
> > 3. Ian creates a new ontology with a different world
> > viewpoint than those
> > already existing, adding one more to the list that
> > philosophers don't agree
> > about.
> >
> > 4. Ian fails to create a consistent ontology.
>
>First of all, the ontology isn't going to be created single-handedly by me.
>The merged ontology is the product of a range of formal ontologies, and it
>will be refined on the basis of comments and criticisms from the SUO WG.
>
>Second, I don't agree that this four-fold set of choices exhausts the range
>of possibilities, because I don't see that we're trying to get straight
>about "how the world is".  As I see it, we're trying to create a conceptual
>schema that should, insofar as possible, support every conceptual model that
>anyone would ever want to develop.

Doesn't that seem almost obviously impossible, just on the face of 
it, if you agree (as you seem to have done) that these various models 
have incompatible conceptual schemes?

> In other words, we're not trying to
>stake claims on the nature of the world;

Everyone agrees with this, Ian, so please stop repeating it.

>we're essentially trying to develop
>the semantic content of a language which can express such claims (and many
>other claims as well).  To ask after the "World Viewpoint" of an ontology
>construed in this way would be analogous to asking after the "World
>Viewpoint" that informs the English language or the Library of Congress
>system of subject categories.  A natural language and subject categorization
>scheme don't embody a single view about the way the world is; they provide a
>representational framework for stating and locating views about the world.

Well, any thing that can be said can be said in English, of course, 
with long enough sentences (though most of science would be hard to 
do that way, not to mention engineering and business: ever tried to 
say an Excel spreadsheet in English?) but it seems impossibly naive 
to say that NL doesn't embody a world-view.  Of course every natural 
language does, in its very grammar. A language that uses tenses 
incorporates an ontology of time into its grammar. Indoeuropean 
languages seem to have an ontology of 'events' built into their 
grammar ( I use scare quotes since the category is rather looser than 
physical event.) There *is* an ontological viewpoint which informs 
English (arguably, informs all Indoeuropean langauges: hence the 
earlier interest in Sanskrit, by the way.) I suspect there is one 
(perhaps several?) that underlies every library classification scheme 
as well, though I know little about their rationale.

>I would argue that we should regard our upper-level ontology in a similar
>manner.

I think this is a recipe for confusion.

Pat Hayes

---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes