Independent/Dependent distinction
Robert Hoehndorf
I am a bit puzzled by the distinction between Independent/dependent
continuant in BFO. Originally, I thought that the distinction is
between existentially dependent and independent continuants. But
ObjectParts and ObjectBoundaries are classified as Independent. But
clearly, if some boundary of an object exists, than it exists only as
the boundary /of/ some object, and therefore, the boundary is
existentially dependent on the object (or a part of the object). Same
holds for ObjectParts, and, perhaps, for ObjectAggregates, Lines,
Points and Surfaces.
So I wonder what the distinction between Independent and Dependent
really is. Pierre Grenon suggested that the distinction is between
substantial/non-substantial instead of dependent/independent. I do not
completely understand what the difference would be, in particular how
things like properties (dependent/non-substantial) differ in
"substanciality" from a point.
Rob.
Pierre Grenon
On 11/21/06, Robert Hoehndorf <leec...@leechuck.de> wrote:
...
> I am a bit puzzled by the distinction between Independent/dependent
> continuant in BFO. Originally, I thought that the distinction is
> between existentially dependent and independent continuants.
No, indeed. It sop happens that all the entities which fall under the
class labelled 'Dependent' are continuants which are existentially
independent on some entity which falls under the class labelled
'Independent'.
But
1) it is not the case that all entities that are existentially
dependent on some entity fall under 'Dependent'.
On the other hand,
2) any entity on which some entity is existentially dependent falls
under Independent.
I am not too sure whether there are entities falling under Independent
which are such that no entity is existentially dependent on them. It
might be the case with some boundaries, perhaps, but I rather think of
the question as open at the moment (even if the conservative view
would be to say that for all independent there is an entity which
depend on that independent).
> But
> ObjectParts and ObjectBoundaries are classified as Independent. But
> clearly, if some boundary of an object exists, than it exists only as
> the boundary /of/ some object, and therefore, the boundary is
> existentially dependent on the object (or a part of the object).
Yes, indeed. Boundaries are existentially dependent on the entities they bound.
> Same
> holds for ObjectParts, and, perhaps, for ObjectAggregates, Lines,
> Points and Surfaces.
I think part are not generally existentially dependent, it is probably
safer to say taht some might be but some are not. Aggregates are
existentially dependent, I think, so are sites for that matter it
would seem.
Lines, Points, and Surfaces, if they are labels for classes falling
under Independent are a bit misleading. Some objects, boundaries,
parts and so on might be linear, punctual or surfacic (if that's a
word).
> So I wonder what the distinction between Independent and Dependent
> really is. Pierre Grenon suggested that the distinction is between
> substantial/non-substantial instead of dependent/independent.
Rougly, the Substantial (aka Objectual aka Independent) / Dependent
(aka Trope aka Property and so on) is the distinction between
Property bearers (and the entities which may be obtained from them
through mereological operations)
vs.
Properties
The important relation is not existential dependence, it is inherence.
The following is taken from:
http://ontology.buffalo.edu/smith/articles/SNAP_SPAN.pdf page 151:
""
All of these entities [SNAP Dependent Entities] have in common the
feature of enduring through time and of requiring a basis in SNAP
independent entities in order to exist. (The particular colour of this
tomato would not exist without the tomato it is the colour of.)
However, if endurance and dependence are necessary conditions for SNAP
dependent entities, they are not sufficient conditions. The
distinguishing feature of these entities is that they inhere in
substances. InheresIn is an intraontological relation between a SNAP
dependent entity and its substantial bearer.
The redness of the ball inheres in the ball; the elevation of a summit
inheres in the summit; the shape of a landform inheres in the
landform. Inherence is a form of existential dependence (Husserl,
1913/21; Simons, 1987; Smith, 1997). The latter is such that the first
of its relata (in the case of inherence, the SNAP dependent entity)
exists only in virtue of the existence of the second (the bearer).
There are other forms of dependence relations (e.g., between processes
and their participants). Thus dependence alone does not suffice for
the relation of inherence to obtain, though we will not pursue this
matter here.
""
> I do not
> completely understand what the difference would be,
Inherence is, so to say, a form or a type of existential dependence:
(x)(y) (x inheres_in y --> x depends_on y)
Independent(x) =def Ey (x inheres_in y)
Dependent(x) =def Ey (x inheres_in y)
> in particular how
> things like properties (dependent/non-substantial) differ in
> "substanciality" from a point.
I'm not sure I get this. 'Substantial' is a technical term inherited
from Aristotle's notion of substance or substratum of predication, to
be short. If you are aiming at a distinction between material /
immaterial or something along the concrete / abstract distinction,
this distinction is not made by BFO.
If you are wondering whether a point which would be a boundary part of
an extended substance is existentially dependent on that substance,
that is the case.
If you are wondering whether that would make according to BFO the
point fall under the class labelled 'Dependent', the answer is no.
The Independent/Dependent terminology is indeed misleading, but it
seemed at the time the least bad terminological consensual choice to
make...
Cheers,
Pierre
> Rob.
>
> >
>
Pierre Grenon
> > continuant in BFO. Originally, I thought that the distinction is
> > between existentially dependent and independent continuants.
>
> No, indeed. It sop happens that all the entities which fall under the
> class labelled 'Dependent' are continuants which are existentially
> independent on some entity which falls under the class labelled
sorry, meant:
'Dependent' are continuants which are existentially dependent on some entity...
> 'Independent'.
>