Re: Roles, again

"Nicola Guarino" <>
Message-id: <guarino.1161224661A@>
Date: Mon, 11 Sep 95 17:30:21 +0100
From: "Nicola Guarino" <>
Subject: Re: Roles, again
To: "Pat Hayes" <>,,
Cc: "Pierdaniele Giaretta" <>,
        "Massimiliano Carrara" <>
X-Mailer: VersaTerm Link v1.1
Precedence: bulk
On Sep 8, 1995, Pat Hayes wrote (on Sowa's discussion about "Firstness",
"Secondness" and "Thirdness"):

>At  1:28 PM 9/8/95 +0100, Nicola Guarino wrote:
>>What is needed
>>is an effort for translating the relevant ideas (if any) into ordinary
>>logic, I agree with Pat on this point. 
>Not quite. My suggestion is simpler: that these words be simply ignored and
>ordinary logic be used to do the job at hand. 

Part of my "job at hand" is to make explicit, by means of ordinary logic,
the ontological commitments usually hidden within "ad-hoc" (task-dependent)
logical theories in order to make them understandable (if not reusable) by
people different from those who originally wrote them. To understand such
commitments, something different from ordinary logic is necessary: logic can
be (with some difficulties) a good instrument to describe reality, but
reality itself needs to be studied as such, at least in its very general,
"qualitative" aspects. That's why ontology (as a philosophyical discipline)
is important, and why I find stimulating the work by those people (including
Aristotle, Husserl, and Peirce) who have addressed that task.

>>However, I'm convinced that there *are* relevant ideas hidden behind this
>>mysticism. The notion which lies behind, as far as I have understood, is
>>that of *dependence* (to be more precise, "conceptual" or "notional"
>>dependence, as Peter Simons calls it in his book on parts [p. 297]). A
>>predicate like Mother, for instance, is dependent on Child if there cannot
>>be a mother without a child ("Secondness").
>>(x)((Mother x) implies (exist y)(Child y) & (x is-mother-of y)))
>I know that this means that Foot depends on Toe, just as Person depends on
>date-of-birth. If you really, really want nonextensionality, then use a
>(x)((Mother x) implies Neccessary((exist y)(Child y) & (x is-mother-of y))))

I avoided here modality just for the sake of simplicity: in the papers I
have cited I use modality, plus some other restrictions to avoid trivial
cases (for instance, Mother should not be necessarily false, and Child
should not be necessarily true). By the way, the modal operator should be
put outside:

Necessarily, (x)((Mother x) implies (exist y)(Child y) & (x is-mother-of y)))

details depend however on which kind of modal semantics you use, of course.

>> A predicate like Person, on the
>>other hand, can be conceived as indepedent if the fact that X is a Person
>>does not necessarily imply that somebody else is an instance of another
>>predicate ("Firstness").
>Ah, careful. It does, I suggest, imply that some THING must be an instance
>of another prediacte and have a relation to X. For example a Person must
>have their time of birth. Every concept we have gets its meaning from the
>network of relations that bind it to other concepts. There arent any
>'primitive' ones. 

In fact we must be *very* careful: there are many forms of dependence, and
their formalization still presents various technical difficulties. One major
problem is exactly what you have addressed: every concept is bound to a
network of relations, but the point to be captured is that these relations
are not all the same (as you seem to suggest): some of them are - I would
say - "internal" to the concept, while others are "external", in the sense
that they bound together different concepts. For instance, database people
distinguish between attributes and relations, while from the logical point
of view both of them are just binary relations. 

Now, the idea of conceptual dependency should somehow account for this:
internal relations (like the one between a person and his time of birth)
don't contribute to dependency, while external relations (like the one
between a mother and her child) do. There are various technical ways to
obtain this (again, I point to you Simons' book), but the problem is still
open. However, the fact that database (and object-oriented) people use such
a distinction (whithout any formalization) should suggest us that it may be

>Exercise: consider, if you can, an orange which is shiny,
>purple, oval, used to make savory dinners, and tastes like an eggplant. Its
>just a mistake to call it an 'orange', right? But if Orange is
>'independent' and had Firstness, we can just assert that this thing is an
>orange (a very unusual orange) in spite of any other properties or
>relations it might have to anything else. This is just Aristotelian
>essentialism rearing its seductive old head again.

The only way to recognize such a mistake is by detecting an inconsistency
with the "meaning postulates" which characterize the predicate "Orange".
This is not related to the fact that "Orange" is dependent or not. Knowing
that a predicate is dependent, on the other hand, can help for instance in
maintaining and organization purposes (if X depends on Y then I should
consider to delete X when deleting Y; X and Y may be allocated in contiguous
locations, and so on...)

> Finally, if we have reified situations (or states
>>of affairs, eventualities, or whatever) in our domain, we can think that the
>>fact that X is a Mother implies the existence of a "Motherhood" situation
>Why isnt motherhood(ness) simply the relation of being-a mother-of? Why
>should we reify such murky entities as these 'situations'? (Incidentally,
>these arent Barwisean 'situations': in that situation theory, a situation
>has a spatiotemporal location. What is the motherhood 'state of affairs'?
>All the moments of birth? 

I agree that, for this particular example, reifying Motherhood doesn't help
too much (that's wy I used "if"). In general, however, I would admit
Barwisean "situations" in my ontology.

Dow Dwiggins wrote on Sep 11 1995:

>I don't believe a
>role belongs anywhere in the type hierarchy of an ontology (in OO terms, I
>don't believe that a role can adequately be represented by a single class).
>It'd be closer to say that a role is an axiom, or collection of them,
>that's part of the representation of a relation.

The utility of having a role in an ontology is just to be able to establish
subsumption relationships with other roles or types. In general, it can be
very useful (but it may be not). I  agree however that placing a role in a
taxonomy doesn't "represent" it completely, of course: when we "mark" a
predicate as a role and establish some subsumption relationships we are just
characterizing in a very rough way its meaning. Only suitable further
postulates (axioms) can do a better job (maybe via a systematic reification
of relations).

Even if we don't place the role in the taxonomy, however, it's important to
"mark" the corresponding unary predicate as a role, just to distinguish it
from a property (like "Red"), whose corresponding unary predicate is usually
kept out of the taxonomy, too.


Nicola Guarino
National Research Council     phone: +39 49 8295751
LADSEB-CNR                    fax:   +39 49 8295778
Corso Stati Uniti, 4          email:
I-35127 Padova                WWW: