participation and inherence

[Stefan Schulz]

Hello,

what are the most recent names and definitions of the BFO relations of
participation and inherence ?
what's the name of the the inverse relation of "inheres-in" ?

Thanks in advance

Stefan

--
PD Dr. Stefan SCHULZ [stsc...@uni-freiburg.de]
Universitätsklinikum - Abt.Medizinische Informatik
Stefan-Meier-Strasse 26 D-79104 Freiburg
[home: Eschholzstr. 70 D-79115 Freiburg]
+49 (0)761 2033252, 2036702, 2049089, FAX 2033251
http://www.imbi.uni-freiburg.de/medinf/~schulz.htm
Skype: stschulz

[Fabian Neuhaus]

-- Stefan,

When we wrote the relations ontology paper we decided not to make the
distinction between substance universals and quality universals. Hence
according to the relation ontology the this apple is an instance of the
universal redness. This is at odds with the four categorical view
according to which there is an entity color_of_this_apple' which is
an instance of red and which inheres in the apple. The reason why we
embraced this approach is that we wanted to focus on distinctions that
are directly relevant for biologists.

I attached a paper where we analyze the inherence relation formally.
According to my knowledge the definition of the participation relation
has not changed since the RO paper.

Best
Fabian

[Chris Mungall]

Hi Fabian

Can you point to the part of the relations ontology where it dictates
an instantiation relationship between continuant instances and
quality types? The ontology is simply silent on this matter AFAIK.

The OBO Relations ontology and definitions can be found at http://
obofoundry.org/ro

> > <UPtheory_revisited.pdf>
>

[Fabian Neuhaus]

Hi Chris,

Of course we never said explicitly: "this continuant particular is an
instant of that quality type", since we were aware that we were
simplifying matters. What I meant is: within the framework of the RO
paper there is no other way to describe the relation between an apple
and its color then to say that this apple is an instance of the color
red (at a given time) -- which is the way anybody who knows FOL or OWL
or some formal semantics would expect it to work.

We did not mention the inherence relation, but we did not write
anything in the RO which directly contradicts the 4-categorical view --
at least I hope so!

Fabian

[Smith, Barry]

Stefan,

This is what I think we need:

inheres-in between a dependent continuant and an independent continuant

do we really need an inverse relation for this?

Participates_in is still defined as in the paper.

Regarding the relation between an apple and its
redness, we can designate this as exemplification, defined as follows:

independent continuant instance c exemplifies
dependent continuant type C1 =def there is some
instance c1 of C1 and c1 inheres_in c.

BS

[Smith, Barry]

Stefan,

This is what I think we need:

inheres-in between a dependent continuant and an independent continuant

do we really need an inverse relation for this?

Participates_in is still defined as in the paper.

Regarding the relation between an apple and its
redness, we can designate this as exemplification, defined as follows:

independent continuant instance c exemplifies
dependent continuant type C1 =def there is some
instance c1 of C1 and c1 inheres_in c.

BS

At 12:04 PM 12/5/2006, Stefan Schulz wrote:

[Cristian Cocos]

> what are the most recent names and definitions of the BFO relations of
> participation and inherence ?
> what's the name of the the inverse relation of "inheres-in" ?

See below how inherence was discussed on the RO list a while ago. (This
pretty much reflects the definition of inherence in the BFO manual.)

"...if one allows for multiple inherence, the situation boils down to:

i. At the instance level one has the primitive "c1 inheres_in c2 at t"
relation, satisfying the principle of non-migration (a inheres_in b at t & a
exists_at t' -> a inheres_in b at t') and possibly other axioms ("There are
no bare particulars" etc.)

ii. At the universal level we have the two definitions:

C1 inheres_in C2 =df. (c1)(t) [c1 instance_of C1 at t -> (Ec2) (c2
instance_of C2 at t & c1 inheres_in c2 at t)]

C2 is_bearer_of C1 =df. (c2)(t) [c2 instance_of C2 at t -> (Ec1) (c1
instance_of C1 at t & c1 inheres_in c2 at t)]

No other clauses/qualifications necessary: non-migration takes care of
everything."

Cristian