And Rosta wrote:
> Mark:
>> Richard Hudson wrote:
>>>
>>> Thanks for this interesting exchange. My view of Dick is that there is
>>> something permanent running through my various appearances on the world
>>> stage (though I have no idea how to push this idea back to me when I
>>> was
>>> much younger).
>>
>> What is permanent is your identity. Identity is not something that
>> inheres
>> in any complex event, it's something that is created by the mind when
>> the
>> event is reified as an object.
>
> And how should identity be modelled? On the face of things, it looks like
> a relation between token and type but that would lead to weird
> inheritance paradoxes, especially if one tried to state generalizations
> about identities.

This is a little circuitous, but bear with me.

In open-world cognitive systems, I see identity (of some X) as a
relationship between X and the person for whom X has an identity.

Since we might as well be talking about this person's cognitive network,
or model of the universe, we might as well understand this relationship to
be between X and a cognitive network M in which it is given identity.

Notice that a similar situation arises (in cognitive networks) in the case
of belief (which is as close as you'll ever get to the concept of
knowledge, much less truth). To wit, belief in a property P is a
relationship between P and the person who believes that P. Since we might
as well be talking about this person's cognitive network M, we might as
well understand this relationship to be between P and M.

Now, when we're modelling beliefs in a cognitive network that's supposed
to model Mark's mind, we don't have any relations from something to P1, P2
... Pn that establish that Mark believes P1, P2, ... Pn. That's not how it
works. We establish that Mark believes P1 by inserting P1 into the
cognitive network: Belief in a property is established by presence of that
property, not in any other way.[1]

I think that the phenomenon of identity should be treated in just the same
way as belief. We don't need any relations from something to X, Y and Z in
Mark's mind to establish that Mark has given identity to X, Y and Z. The
fact that X, Y and Z are identifiable (in terms of whatever processing
goes on) in Mark's mind already does the job.


-- Mark

[1] Note that this does not preclude representation of other people's
beliefs in your own mind using relations, that the ramifications are
entirely different, and that they may not actually coincide with an
omniscient model of the other people's minds: people don't always believe
what you think they believe. It also does not preclude metacognitive
properties about your own beliefs, noting that these metacognitions may or
may not coincide with your actual beliefs as represented in an omniscient
model: you don't always believe what you think you believe.


Mark P. Line
Polymathix
San Antonio, TX