Re: H(D,D) cannot even be asked about the behavior of D(D) --- Truth Itself is not Broken.

On 6/15/2024 10:18 AM, Richard Damon wrote:

On 6/15/24 11:03 AM, olcott wrote:

On 6/15/2024 9:46 AM, Richard Damon wrote:

On 6/15/24 10:23 AM, olcott wrote:

>
No you are wrong about this. The first thing that I discovered
about this at least twenty years ago is that it is always an
acyclic graph.

>
Which means there is always a set of root nodes that do not have a truth-maker coming into them.
>

>
When we do this that way that the Cyc project does it {thing} is
the ultimate root node. {thing} is divided up into types of things.
>
By the theory of simple types I mean the doctrine which says that the
objects of thought ... are divided into types, namely: individuals,
properties of individuals, relations between individuals, properties of
such relations, etc.
https://en.wikipedia.org/wiki/History_of_type_theory#G%C3%B6del_1944
>

>
When you try to come up with a concrete counter-example I will
point out your specific mistake.

>
But I have conceptually.
>
Show me a root concept, that has a truth-maker but doesn't depend on anything else. If you use words to describe it, how do those words have meaning without being defined by other words.
>

>
{thing} is the root of the whole knowledge tree.

 And what DEFINES {thing}?
 

Its constituents.

and what distingueshes the things derived from {thing}
 

Their placement in the inheritance hierarchy.

All these need definitions (what are part of truth=making) from OUTSIDE the system.
 

The accurate verbal model of the actual world encoded in
something like simple type theory.

>

There is a fundamental problem of first principles that need to stand on their own without support from anything in the system.
>

>
The definition of the meaning of a term is the truthmaker
for this term. The terms that this definition is composed
of have their own definitions. This is recursively quite
deep yet zero actual cycles.

>
And what makes that definition true?
>

>
What makes puppies not a type of fifteen story office building?

 Because we have defined the terms that way.
 

Yes, you are starting to get this.
Fifteen story office buildings never wag their tail.

>
The correct verbal model of the actual world encodes relations
between types of things as stipulated relations between finite
strings.

 And stipulations don't have truth makers in the system.
 

The verified fact that puppies really do wag their tails
is the truthmaker for {puppies wag their tails}.

>
That we have many human languages that encode the same relations
between types of things in the world and each one does it using
different finite strings proves the stipulated aspect of this.

 And Human Languages have circular definitions for words,

No you are wrong.
Provide a counter-example.
it is always a type hierarchy.

thus you can not trace them to a "root". We need to start with a set of first concepts that we agree OUTSIDE OF LANGUAGE what they mean, and express these definitions as loops within the language.
 

By the theory of simple types I mean the doctrine which says that the objects of thought ... are divided into types, namely: individuals, properties of individuals, relations between individuals, properties of such relations, etc.
https://en.wikipedia.org/wiki/History_of_type_theory#G%C3%B6del_1944
This can be further simplified to types of relations between
finite strings.

These words have no "truth-makers"
 

Incorrect.

>

How can you write a "defintion" for the first term of your system?
>

>
It is the same sort of knowledge tree that the Cyc project uses
to encode an accurate verbal model of the actual world.

 and, as I asked, how do they actually DEFINE {thing} or diferentiate between the sub-concepts off of {thing}
 

{Thing} is the root and is defined by itself constituents types of things.

Only by using information from OUTSIDE the system.
 

Not at all.

>

You ALWAYS need to reference something outside your system, and when you then include that source, you need to find the root of THAT system, and your problem continues.
>

>
Not really, even the root of the knowledge tree {thing}
is defined in terms of its constituents.
>
>

 So, if those constituents are in the system, we have a circular definition, and if outside, it isn't a self-sufficient system.

A type hierarchy definitely does not have any cycles.
Most all of human knowledge that can be expressed using
language can be fully encoded as relations between finite
strings in a type hierarchy.
Some knowledge requires specifying a sequence of procedural
steps. This requires sequence, selection and iteration.
The Cyc project uses GUIDs in place of finite strings.
This makes their system generic to all human knowledge.
I am sorry to say that Doug Lenat died last year.
--
Copyright 2024 Olcott "Talent hits a target no one else can hit; Genius
hits a target no one else can see." Arthur Schopenhauer