Conceptual Truth can be understood as math

[PeteOlcott]

(1) I made my point.
(2) Prolog proves my point.
(3) Done

Liar!

When you produce the algorithm which takes your own statement as an input-string, and it returns True as a result, THEN you have proven Tarski wrong.

I've done the ground-work for you: https://repl.it/repls/FuchsiaVivaciousDemo

I await your implementation.

Conceptual truth is ONLY mutually interlocking semantic tautologies that can ALWAYS
be represented as the satisfaction of stipulated relations between finite strings.


That no counter-examples exist refuting the above statement proves that it is true.
That Prolog fully implements the above for first order logic proves that it is true for first order logic.
That Tarski is refuted by the above statement is logically entailed by the above statement.
That Prolog cannot prove that the above statement is true is a limitation of first order logic.

[Skepdick]

That no counter-examples exist refuting the above statement proves that it is true.

https://en.wikipedia.org/wiki/Argument_from_ignorance

The statement above is either true, false or undecidable.

That Prolog fully implements the above for first order logic proves that it is true for first order logic.

Pete, if you are a liar, your claim is false. If you are a truth-teller your claim is true.

I don't know whether you are a liar or a truth-teller.
Perhaps you don't know whether you are a liar or a truth-teller either.

If you write the algorithm to determine the correct answer, then it's settled.

That Prolog cannot prove that the above statement is true is a limitation of first order logic.

Bullshit, Pete! Prolog is Turing-complete - it's N-th order logic.
The only limitation is your ability to implement the algorithm.

I feel like you are just looking for excuses now.

[Eodnhoj7]

(1) You are getting the analytic versus synthetic distinction incorrectly.
(2) The truth teller paradox is not Boolean.

I am not arguing boolean, and you are analyzing in making that distinction. Analysis breaks a phenomenon into parts...period.

I will refrain from responding until after you show that you know these things well enough.

You are assuming boolean thinking as a foundation. Second the distinction between analytic/synthetic is subject to the context. What context determines it as true or false? In one context I will alway be right, and in another false.

But these seperation of contexts always requires an inherent analysis.

Analysis always requires, at minimum, the observation of phenomenom through the breaking apart or atomization of it...thus the progression of one axiom to many axioms which compose it.

This applies to any subject ranging from math, to biology to the pscyhe.

For example in the analysis of an animal:

The whole animal is observed.

This whole animal is dissected into various parts (heart, lungs, etc.)

The parts are observed in how they are connected (heart pumps blood to lungs, lungs absorb air I might into blood for heart) and maintained by cycles.

[PeteOlcott]

Analysis always requires, at minimum, the observation of phenomenom

An analytic sentence is any sentence that can be verified as
completely true entirely on the basis of the meaning of its words.
no phenomenom no context
Example: “Ophthalmologists are doctors”

[PeteOlcott]

That no counter-examples exist refuting the above statement proves that it is true.

https://en.wikipedia.org/wiki/Argument_from_ignorance

The statement above is either true, false or undecidable.

You got confused about the meaning of ad ignorantiam.
(1) If I consider proof to be based on the fact that I am not aware of counter-examples this is ad ignorantiam.
(2) If no counter-examples exist, what-so-ever this proves NOT FALSE.

[PeteOlcott]

That no counter-examples exist refuting the above statement proves that it is true.

That Prolog cannot prove that the above statement is true is a limitation of first order logic.

Bullshit, Pete! Prolog is Turing-complete - it's N-th order logic.
The only limitation is your ability to implement the algorithm.

Unless the proof is directly written in higher order logic and this higher order logic
is directly processed by a HOL inference engine humans will not be able to determine
the validity of the proof because the proof steps will be hidden in a mountain of detail.

[Skepdick]

You got confused about the meaning of ad ignorantiam.
(1) If I consider proof to be based on the fact that I am not aware of counter-examples this is ad ignorantiam.
(2) If no counter-examples exist, what-so-ever this proves NOT FALSE.

Negative. You have equated True and Provable in your system.
This is the same as saying that all unproven propositions are false-by-default until proven True.

But then in point (2) above you are declaring the exact opposite! You are insisting that your claim should be exempt from this rule and assumed true UNTIL proven false. You are breaking the inference rules of your own system.

So I shall hold you accountable to True == Provable and I expect you to prove this claim:

Conceptual truth is ONLY mutually interlocking semantic tautologies that can ALWAYS
be represented as the satisfaction of stipulated relations between finite strings.

[Skepdick]

Unless the proof is directly written in higher order logic and this higher order logic
is directly processed by a HOL inference engine humans will not be able to determine
the validity of the proof because the proof steps will be hidden in a mountain of detail.

Pete, this is total nonsense.

Who is going to write the inference engine? A human, right? It's obviously not going to write itself.
I can only imagine that such a complex system isn't going to be written in Assembler. In fact, it's probably going to be written in some high-level language with Natural Language Processing functionality and human-friendly abstractions.

The human that writes the inference engine understands EXACTLY what the inference engine needs to do in order to classify sentences as "true" or "false".Have you ever written an algorithm that emulates complex human judgment and it works by accident?

Behind every object/abstraction/function, behind every line of code there is human intent - a clear strategy and tactics towards achieving some immediate or overall goal.

And so you can't just arbitrarily say that the function takes a string and returns a boolean. All you've done is defined the interface.
Show me the implementation. Or even some high-level pseudo-code. How would you approach the parsing of your very own claim?

What information would you extract from the string and how in order to assert its truth-value.

I imagine step1 would be to split up the sentence into words. Trivial.
What's step 2?

By Curry-Howard the implementation is the actual proof.

[Eodnhoj7]

An analytic sentence is any sentence that can be verified as
completely true entirely on the basis of the meaning of its words.
no phenomenom no context
Example: “Ophthalmologists are doctors”

Yes and the meaning is subject to context.

"Opthalmologists are doctors" is a true and false statment.

It is true because ophthalmologist are doctors.

It is false because ophthalmologists are not doctors but eye doctors. They are also humans. They are also....x. It is false because it is incomplete.

Doctor may also mean general practitioner, thus the statement is also false because doctor is a meaningless word.

[PeteOlcott]

You got confused about the meaning of ad ignorantiam.
(1) If I consider proof to be based on the fact that I am not aware of counter-examples this is ad ignorantiam.
(2) If no counter-examples exist, what-so-ever this proves NOT FALSE.

Negative. You have equated True and Provable in your system.
This is the same as saying that all unproven propositions are false-by-default until proven True.

But then in point (2) above you are declaring the exact opposite! You are insisting that your claim should be exempt from this rule and assumed true UNTIL proven false. You are breaking the inference rules of your own system.

So I shall hold you accountable to True == Provable and I expect you to prove this claim:

Conceptual truth is ONLY mutually interlocking semantic tautologies that can ALWAYS
be represented as the satisfaction of stipulated relations between finite strings.

For the body of all knowledge that can be verified as completely true
entirely on the basis of the meaning of its words:
True(x) = Provable(x)
False(x) = Provable(~x)
There are only two ways that this can be proven:
(1) Write out every element of the set of all conceptual knowledge and find there are no exceptions to the rule.
(2) Prove that exceptions to the rule are categorically impossible.

I know that exceptions to the rules are categorically impossible yet do not
have an effective way to show this without more feedback from others.

Give me one exception to the rule and I will show you where you erred.

[PeteOlcott]

Unless the proof is directly written in higher order logic and this higher order logic
is directly processed by a HOL inference engine humans will not be able to determine
the validity of the proof because the proof steps will be hidden in a mountain of detail.

Pete, this is total nonsense.

Who is going to write the inference engine? A human, right? It's obviously not going to write itself.
I can only imagine that such a complex system isn't going to be written in Assembler. In fact, it's probably going to be written in some high-level language with Natural Language Processing functionality and human-friendly abstractions.

The human that writes the inference engine understands EXACTLY what the inference engine needs to do in order to classify sentences as "true" or "false".Have you ever written an algorithm that emulates complex human judgment and it works by accident?

Behind every object/abstraction/function, behind every line of code there is human intent - a clear strategy and tactics towards achieving some immediate or overall goal.

And so you can't just arbitrarily say that the function takes a string and returns a boolean. All you've done is defined the interface.
Show me the implementation. Or even some high-level pseudo-code. How would you approach the parsing of your very own claim?

What information would you extract from the string and how in order to assert its truth-value.

I imagine step1 would be to split up the sentence into words. Trivial.
What's step 2?

By Curry-Howard the implementation is the actual proof.

Prolog fully implements this proof for the scope of first order logic.
A query returns Yes indicates true.
The negation of a query returns Yes indicates false.
Neither the query nor its negation returns Yes indicates semantically malformed.

Some kinds of semantically malformed are this kind:
https://www.swi-prolog.org/pldoc/man?pr ... rs_check/2

[PeteOlcott]

Yes and the meaning is subject to context.

"Opthalmologists are doctors" is a true and false statment.

It is true because ophthalmologist are doctors.

It is false because ophthalmologists are not doctors but eye doctors. They are also humans. They are also....x. It is false because it is incomplete.

Doctor may also mean general practitioner, thus the statement is also false because doctor is a meaningless word.

All words are defined in a knowledge ontology acyclic graph, thus no word is in the ontology is meaningless.

[Skepdick]

(1) Write out every element of the set of all conceptual knowledge and find there are no exceptions to the rule.
(2) Prove that exceptions to the rule are categorically impossible.

Great. Apply this to your claim using some formal system.

I know that exceptions to the rules are categorically impossible yet do not
have an effective way to show this without more feedback from others.

Well, isn't that the crux of it? I know how to tell the difference between the photo of horse and a photo of a donkey, but I have no effective way to show it formally.

In the universe of formalisms an unexpressed proof is not a proof.

Give me one exception to the rule and I will show you where you erred.

That's not how this works. You haven't defined any rules. You have defined functions. The structure of your system.
It doesn't do anything.


https://repl.it/repls/GiftedPassionateSlash

Code: Select all

def provable(x):
  pass

def true(x):
  return provable(x)

def false(x):
  return provable(not x)

claim='''Conceptual truth is ONLY mutually interlocking semantic tautologies that can ALWAYS
be represented as the satisfaction of stipulated relations between finite strings.'''

if true(claim):
  print('Tarski is wrong!')
elif false(claim):
  print('Tarski is right!')
else:
  print('We have a tri-valued logic!')

[Skepdick]

Prolog fully implements this proof for the scope of first order logic.
A query returns Yes indicates true.
The negation of a query returns Yes indicates false.
Neither the query nor its negation returns Yes indicates semantically malformed.

Yes. That's exactly the structure I have put in place for you!

https://repl.it/repls/GiftedPassionateSlash
Now - show me the real magic. implement the provable() function!

[PeteOlcott]

Well, isn't that the crux of it? I know how to tell the difference between the photo of horse and a photo of a donkey, but I have no effective way to show it formally.

In the universe of formalisms an unexpressed proof is not a proof.

Give me one exception to the rule and I will show you where you erred.

That's not how this works. You haven't defined any rules. You have defined functions. The structure of your system.
It doesn't do anything.

"how to tell the difference between the photo of horse and a photo of a donkey"
There is some feedback, you are not making the analytic versus synthetic distinction correctly.
I am only referring to the body of knowledge that can be determined to be true or false entirely on the basis of the meaning of its words.

Try and find a counter-example of any analytic sentence that is neither provable nor refutable on the basis of the meaning of its words,
Once you understand that this is impossible my proof is done.