BFO-ontology / BFO

BFO repository including source code and latest documents
Creative Commons Attribution 4.0 International
266 stars 44 forks source link

Bear vs. Inhere #227

Open schneidert opened 3 years ago

schneidert commented 3 years ago

The uses of the terms 'bear' and 'inhere' and their synonymity in BFO are not consistent with their natural language definitions.

The term 'bear' has a common definition of 'carry' or 'support', not a permanent situation. While the common definition of inhere is "exist essentially or permanently in", a permanent situation.

A 'Quality' inheres in an entity, a permanent situation. A entity bears a 'Role', a temporary situation.

This synonymity introduces confusion to users, fails to provide needed ontological distinctions, and should be corrected.

Skreen5hot commented 3 years ago

Hello Schneidert, sorry this page never got the traction I was hoping for, so I missed your comment. It looks like you answered your own question. You see that bear is not permanent and neither is a role and inhere is implies a permanent situation like the relation between quality and an entity. What exactly do you find confusing about them. Do you have a suggestion?

In natural language, the two are often conflated. (I have a son. I have a BFA. I have 2 arms. I am a father. I am an artist. I am a human.) Naturally, we know that you were not born with a son or a BFA but you were born with 2 arms. It seems appropriate to make a clear distinction for something we all assume in a formal system.

jonathanvajda commented 6 months ago

The uses of the terms 'bear' and 'inhere' and their synonymity in BFO are not consistent with their natural language definitions.

The term 'bear' has a common definition of 'carry' or 'support', not a permanent situation. While the common definition of inhere is "exist essentially or permanently in", a permanent situation.

A 'Quality' inheres in an entity, a permanent situation. A entity bears a 'Role', a temporary situation.

This synonymity introduces confusion to users, fails to provide needed ontological distinctions, and should be corrected.

Could you motivate why qualities would inhere permanently? I think there's also a direction of dependence in view here: my qualities cannot migrate over to you, since whenever they exist they inhere in me. The reverse is not always the case.

I think none of the specifically dependent continuants (quality, disposition, role) would be permanent by definition, even if some of the subclasses are. Like @Skreen5hot, I think what we'd say for roles being temporary would also be true for dispositions and qualities.

Examples: a) This sheet of paper has a quality of white color at time t1, but after adding some dye at time t2 the sheet of paper now bears a quality of red color and no longer bears the quality of white color. b) This anatomical structure (UBERON:0008230) has a healthy quality (obo:SIO_001012) at time t1, but after suffering a traumatic impact at time t2, the anatomical structure has a disordered quality of Rhabdomyolysis (obo:OMIT_0013120), and some later time t3, the structure healed and thus has a healthy quality (obo:SIO_001012) again. These are all qualities. c) This person has a disposition to play piano. After a severe car accident, she no longer has a disposition to play piano. d) That aggregate of rubber does not have a disposition to shatter at time t1, but at time t2 it has a disposition to shatter (it has become brittle).

Perhaps you might be thinking that roles can 'migrate' in that a role might belong temporarily to one entity but at another time belong to another. For example, we might talk like this:

This is definitely a way people talk. But BFO models the way the world is, and it doesn't match the way people talk: my IT admin role is not token identical to your IT admin role, even though the roles are type identical. So, strictly speaking, my IT admin role ceased to exist and a similar one came into existence when you take the job.

Did you have in mind some other qualities, dispositions, or roles that are permanent, in that once you have it, you will always have it? Perhaps once one has suffered a traumatic brain injury, that quality (or disposition?) is irreversible? Or perhaps I am always a father, for example, even if I lost my kids? Maybe this is true of some qualities or roles that they are permanently inhere in the one who bears the dependent-continuant, but I am not sure. But it seems clear it isn't true of the genus.