Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: SUO: On the supreme supertype




>Aldo Gangemi wrote:
>>  >On further thought, I'd like to suggest a different possible name for the
>>  >supreme supertype, other than "thing" or "entity" which seem to be
>>  >traditional...
>>  >
>>  >Perhaps a better word for the supreme supertype could be 
>>"referent", meaning
>>  >that which can be referenced.
>>  >
>>  >I'm thinking along the following lines: Whatever cannot be 
>>referenced cannot
>>  >be described and hence cannot be categorized or placed in an ontology. Of
>>  >that which cannot be referenced, we must be silent...

for the sake of correct referencing ;-):

I did NOT write the above quoted, but Philip did it. I commented that 
'referent'
is not appropriate for its theory-ladenness in philosophy and 
generally for the reason indicated by John (Sowa) in another reply.

>This doesn't seem like a good idea to me.  In mathematics, at least,
>there are lots of things that can't be referenced but about which we
>are not silent.  For instance, any property that we prove holds of all
>real numbers says something about all the reals, even those for which
>we could never have sort any procedure that would enable us to pick
>them out and refer to them explicitly.

I agree

>If you respond that such numbers are still, in some abstract sense,
>capable of being referenced, then it seems to me that your notion of
>"referent" has lost any bite that distinguishes it from "thing" and
>"entity".

I only suggested that 'reference' (NOT 'referent') would be a 
possible choice: for example it is used in semiotics for any sign you 
can talk about. But you are right: one or another, that's not all 
that different.

All the best

Aldo
-- 



*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*;*
Aldo Gangemi
Ontology and Conceptual Modeling Group
ITBM-CNR (National Research Council)
Viale Marx 15, 00137
Roma Italy
+3906.86090249
mailto://gangemi@saussure.irmkant.rm.cnr.it
http://saussure.irmkant.rm.cnr.it