How Best to Define BFO Universals That Are Not BFO Objects?

[ANTHONY PETOSA]

(This post is directed to Dr. Barry Smith, but I welcome any and all
feedback).

If one bases specialization across BFO Objects on the physical trait
differentia when creating Aristotelian definitions, then what are the
criteria and/or best practices for establishing the differentia across
the remaining BFO 2.0 categories?

Suppose we establish a quick and dirty definition for MusicianRole.

"MusicianRole = Def. MusicianRole is a BFO Role whereby its bearer
serves to perform music."

...and for ProfessionalMusicianRole...

"ProfessionalMusicianRole = Def. ProfessionalMusicianRole is a
MusicianRole whereby its bearer has acquired expertise in playing some
musical instrument." (I make no distinction here between an external
musical instrument and one's voice.)

...and lastly...

"StudioMusicianRole = Def. StudioMusicianRole is a
ProfessionalMusicianRole whereby its bearer performs contractually in a
professional audio recording environment on a regular basis."

The challenge here is whether this is a viable taxonomic hieararchy in
terms of ontological realism. Is ProfessionalMusicianRole truly a
specialization of MusicianRole or does ProfessionalMusicianRole amount
to an attributive class? Is ProfessionalMusicianRole a conjunctive form
of Professionalism and Musician, or does this specialization stand? The
answer does not seem as obvious as when applying Aristotelian
definitions to BFO Objects. In another post on this list, I was properly
corrected in my inaccurate attempt to define "FurniturePiece" as a BFO
Object. It was pointed out that I conflated BFO Object with BFO Role.

Do any specific criteria exist or, at the least, are there established
best practices for expressing the differentia within Aristotelian
definitions when applied across the various BFO universals? If not, will
these criteria and/or best practices be developed and documented as a
supplement to BFO 2.0? I think this goes a very long way toward enabling
end users to implement BFO 2.0 in their ontology work in a manner that
holds true to ontological realism.

I realize the near-term goal is to develop a First-Order Logic version
of BFO 2.0, and this is critical knowledge for engineering BFO 2.0 in
some set-theoretic machine-consumable language (e.g., OWL). However, I
personally view the aforementioned criteria and/or best practices to be
equally important as a next step toward enabling better BFO 2.0
implementation. I understand this in itself does not in any way replace
an Ontologist's years of knowlege acquisition relating to ontological
realism and to its practical application, but, nonetheless, these
criteria and/or best practices would greatly enhance BFO 2.0's
implementability for the end user. At the very least, it would enable
the end user to interact with an Ontologist in a more meaningful way (no
pun intended).


Regards,

Anthony Petosa