Formalizing Natural Language Semantics

What did you say? And what did you mean by it?

Moderators: AMod, iMod

PeteOlcott
Posts: 1514
Joined: Mon Jul 25, 2016 6:55 pm

Formalizing Natural Language Semantics

Post by PeteOlcott »

(a) John owns a brick house
(b) John owns a house
(c) John owns a counterfeit Monet
(d) John owns a Monet

∃!JOHN ∈ Male_Humans
∃!y ∈ Houses
(Owns(JOHN, y) & Construction_Material(y, BRICK))

∃!JOHN ∈ Male_Humans
∃!y ∈ Houses
(Owns(JOHN, y)

∃!JOHN ∈ Male_Humans
∃!y ∈ Work-of-Art
∃!MONET ∈ Famous_Artists
∃!MONET_ART ∈ Artwork(MONET)
(Owns(JOHN, y) & ~Authentic(y, MONET_ART))

∃!JOHN ∈ Male_Humans
∃!y ∈ Work-of-Art
∃!MONET ∈ Famous_Artists
∃!MONET_ART ∈ Artwork(MONET)
(Owns(JOHN, y) & Authentic(y, MONET_ART))

Copyright 2020 Pete Olcott

https://www.researchgate.net/publicatio ... y_YACC_BNF
PeteOlcott
Posts: 1514
Joined: Mon Jul 25, 2016 6:55 pm

Re: Formalizing Natural Language Semantics

Post by PeteOlcott »

https://en.wikipedia.org/wiki/Colorless ... _furiously
The only way that we can know that the above sentence is a truth bearer with the semantic value of Boolean false is a type mismatch error between its tokens.

Copyright 2020 Pete Olcott
Scott Mayers
Posts: 2446
Joined: Wed Jul 08, 2015 1:53 am

Re: Formalizing Natural Language Semantics

Post by Scott Mayers »

PeteOlcott wrote: Sat Feb 29, 2020 4:55 pm (a) John owns a brick house
(b) John owns a house
(c) John owns a counterfeit Monet
(d) John owns a Monet

∃!JOHN ∈ Male_Humans
∃!y ∈ Houses
(Owns(JOHN, y) & Construction_Material(y, BRICK))

∃!JOHN ∈ Male_Humans
∃!y ∈ Houses
(Owns(JOHN, y)

∃!JOHN ∈ Male_Humans
∃!y ∈ Work-of-Art
∃!MONET ∈ Famous_Artists
∃!MONET_ART ∈ Artwork(MONET)
(Owns(JOHN, y) & ~Authentic(y, MONET_ART))

∃!JOHN ∈ Male_Humans
∃!y ∈ Work-of-Art
∃!MONET ∈ Famous_Artists
∃!MONET_ART ∈ Artwork(MONET)
(Owns(JOHN, y) & Authentic(y, MONET_ART))

Copyright 2020 Pete Olcott

https://www.researchgate.net/publicatio ... y_YACC_BNF
Is your "∃!" stand for "Some particular" versus "some" unspecified?

As to semantics, while Predicate Calculus adds quantification and predicates, what are you defining as the 'semantic meaning' fundamentally? Is your system postulating variables (then proving constants) or constants (then proving variables)?
PeteOlcott
Posts: 1514
Joined: Mon Jul 25, 2016 6:55 pm

Re: Formalizing Natural Language Semantics

Post by PeteOlcott »

Scott Mayers wrote: Sat Feb 29, 2020 10:23 pm Is your "∃!" stand for "Some particular" versus "some" unspecified?

As to semantics, while Predicate Calculus adds quantification and predicates, what are you defining as the 'semantic meaning' fundamentally? Is your system postulating variables (then proving constants) or constants (then proving variables)?
The same way that it works in the human mind it can be expressed in Higher Order Logic:
All [Conceptual knowledge]** is merely the connections between concepts that can be
expressed as connections between finite strings.

**Conceptual knowledge: The set of knowledge that can be completely expressed using language.
Skepdick
Posts: 14364
Joined: Fri Jun 14, 2019 11:16 am

Re: Formalizing Natural Language Semantics

Post by Skepdick »

Pete, you don't even know what Higher Order logic is. All of your expressions are 1st/2nd order expressions.

Whatever you think is a "high order logic" there is a higher order logic than that. The tower of abstraction is infinitely high.
Rather than wasting another 12000 hours of your life, I can recommend a $120 book for you.

https://www.amazon.com/Higher-Order-Com ... 3662479915

While you are waiting for your book to arrive, read this paper. In the first 5 pages it offers you a model of semantics that is 4-layers nested - way more sophisticated, and more "accurate" than anything you are currently proposing.
PeteOlcott wrote: Sat Feb 29, 2020 10:43 pm **Conceptual knowledge: The set of knowledge that can be completely expressed using language.
So that's excludes all tacit knowledge then?
That's ironic, because tacit knowledge is algorithmic. Know-how.
PeteOlcott
Posts: 1514
Joined: Mon Jul 25, 2016 6:55 pm

Re: Formalizing Natural Language Semantics

Post by PeteOlcott »

Skepdick wrote: Sun Mar 01, 2020 10:30 am Pete, you don't even know what Higher Order logic is. All of your expressions are 1st/2nd order expressions.

Whatever you think is a "high order logic" there is a higher order logic than that. The tower of abstraction is infinitely high.
Rather than wasting another 12000 hours of your life, I can recommend a $120 book for you.

https://www.amazon.com/Higher-Order-Com ... 3662479915

While you are waiting for your book to arrive, read this paper. In the first 5 pages it offers you a model of semantics that is 4-layers nested - way more sophisticated, and more "accurate" than anything you are currently proposing.
PeteOlcott wrote: Sat Feb 29, 2020 10:43 pm **Conceptual knowledge: The set of knowledge that can be completely expressed using language.
So that's excludes all tacit knowledge then?
That's ironic, because tacit knowledge is algorithmic. Know-how.
I know that I know that I know that I am correct.
No one has been able to even begin to form any counter-example showing that I am not correct.
Everyone merely presumes that I am incorrect and then proceeds on that basis.
My correct refutation of the Halting Problem proofs will change all that.

Minimal Type Theory has an [assign alias] operator that can assign a whole expression to a variable.
This variable can be quantified over in another expression, ad infinitum.

When I use the term conceptual knowledge I am referring to all knowledge that can be expressed using language.
You read what I was saying backwards.
Scott Mayers
Posts: 2446
Joined: Wed Jul 08, 2015 1:53 am

Re: Formalizing Natural Language Semantics

Post by Scott Mayers »

PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm I know that I know that I know that I am correct.
No one has been able to even begin to form any counter-example showing that I am not correct.
Everyone merely presumes that I am incorrect and then proceeds on that basis.
My correct refutation of the Halting Problem proofs will change all that.

Minimal Type Theory has an [assign alias] operator that can assign a whole expression to a variable.
This variable can be quantified over in another expression, ad infinitum.

When I use the term conceptual knowledge I am referring to all knowledge that can be expressed using language.
You read what I was saying backwards.
I can't even figure out what you are saying. One has to know what you are saying first in order to "presume [you] incorrrect" BEFORE preceding.

When I asked you about how you interpret your rationale as "semantic", I am confused because you are trying to use variables as constants but cannot do this by assigning meaning to its contents, just like "propositional calculus" cannot do. "Predicate logics" only add relations of propositions with quantified generalities. And though these can be 'true' of reality regarding logic itself, it doesn't speak of the literal meaning of the propositions conceived.

What is your goal here? What is your view say about the meaning of class terms and meaning that isn't already understood by the symbolic logics already developed?
Last edited by Scott Mayers on Mon Mar 02, 2020 4:31 pm, edited 1 time in total.
Skepdick
Posts: 14364
Joined: Fri Jun 14, 2019 11:16 am

Re: Formalizing Natural Language Semantics

Post by Skepdick »

PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm I know that I know that I know that I am correct.
No one has been able to even begin to form any counter-example showing that I am not correct.
It's not our job to give you a counter-example, Pete. That's your job.

If you can't prove yourself wrong then your idea is not even wrong.

If you don't take that as an example of fallacious reasoning, then I have nothing for you. You aren't a scientist. You are a dumb philosopher.
PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm When I use the term conceptual knowledge I am referring to all knowledge that can be expressed using language.
How would you determine whether any particular linguistic expression expresses "conceptual knowledge" or not?

Here is a linguistic expression: I know that apples are delicious.

It is clearly expressed in language. Is that "conceptual knowledge" or not?
Last edited by Skepdick on Mon Mar 02, 2020 4:37 pm, edited 2 times in total.
Scott Mayers
Posts: 2446
Joined: Wed Jul 08, 2015 1:53 am

Re: Formalizing Natural Language Semantics

Post by Scott Mayers »

Skepdick wrote: Mon Mar 02, 2020 4:30 pm
PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm I know that I know that I know that I am correct.
No one has been able to even begin to form any counter-example showing that I am not correct.
It's not our job to give you a counter-example, Pete. That's your job.

If you can't prove yourself wrong then your idea is not even wrong.

If you don't take that as an example of fallacious reasoning, then I have nothing for you. You aren't a scientist. You are a dumb philosopher.
I think you're kinda being a "skep-dick" here by calling him 'dumb'.
Skepdick
Posts: 14364
Joined: Fri Jun 14, 2019 11:16 am

Re: Formalizing Natural Language Semantics

Post by Skepdick »

Scott Mayers wrote: Mon Mar 02, 2020 4:33 pm I think you're kinda being a "skep-dick" here by calling him 'dumb'.
It's an assertion. A statement of fact. Not an insult.

The decision-procedure I used to assert his "dumbness" was explicitly stated.

Leave your feelings at the door.
Scott Mayers
Posts: 2446
Joined: Wed Jul 08, 2015 1:53 am

Re: Formalizing Natural Language Semantics

Post by Scott Mayers »

Skepdick wrote: Mon Mar 02, 2020 4:35 pm
Scott Mayers wrote: Mon Mar 02, 2020 4:33 pm I think you're kinda being a "skep-dick" here by calling him 'dumb'.
It's an assertion. A statement of fact. Not an insult.

The decision-procedure I used to assert his "dumbness" was explicitly stated.

Leave your feelings at the door.
But to insult his point as 'dumb', you require being certain to understand what the confusion is and know why his view is not being 'smart'. If you just don't understand him, then just say that. Own your own role in understanding/misunderstanding what he stated without implying yourself that you know better. Because then you are reversing the onus on you to prove WHY you know he is not being smart.

[I'm turning your 'dumb' into 'not smart' here to accord to your claim of it not being an insult. Then you MEAN that you do not think he is being practically correct in his approach where 'smart' means "practically correct or effective behavior". THAT would not come across insulting and NOT steal his his onus to explain more clearly for you.]
PeteOlcott
Posts: 1514
Joined: Mon Jul 25, 2016 6:55 pm

Re: Formalizing Natural Language Semantics

Post by PeteOlcott »

Skepdick wrote: Mon Mar 02, 2020 4:30 pm
PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm I know that I know that I know that I am correct.
No one has been able to even begin to form any counter-example showing that I am not correct.
It's not our job to give you a counter-example, Pete. That's your job.

If you can't prove yourself wrong then your idea is not even wrong.

If you don't take that as an example of fallacious reasoning, then I have nothing for you. You aren't a scientist. You are a dumb philosopher.
PeteOlcott wrote: Mon Mar 02, 2020 4:07 pm When I use the term conceptual knowledge I am referring to all knowledge that can be expressed using language.
How would you determine whether any particular linguistic expression expresses "conceptual knowledge" or not?

Here is a linguistic expression: I know that apples are delicious.

It is clearly expressed in language. Is that "conceptual knowledge" or not?
Within the context of your expression: "know" means subjective assessment.
If we use the most literal meaning of {know} meaning that anyone not liking apples
is logically incorrect in their dislike of apples then your sentence becomes false.
Skepdick
Posts: 14364
Joined: Fri Jun 14, 2019 11:16 am

Re: Formalizing Natural Language Semantics

Post by Skepdick »

Scott Mayers wrote: Mon Mar 02, 2020 4:49 pm
Skepdick wrote: Mon Mar 02, 2020 4:35 pm
Scott Mayers wrote: Mon Mar 02, 2020 4:33 pm I think you're kinda being a "skep-dick" here by calling him 'dumb'.
It's an assertion. A statement of fact. Not an insult.

The decision-procedure I used to assert his "dumbness" was explicitly stated.

Leave your feelings at the door.
But to insult his point as 'dumb'
Pay attention..... IT IS NOT AN INSULT.

It is an assertion.
Scott Mayers wrote: Mon Mar 02, 2020 4:49 pm you require being certain to understand what the confusion is and know why his view is not being 'smart'.
Because it is not even wrong. It's not only untestable, but it is also unfalsifiable.

I said that. How did you miss it?
Scott Mayers
Posts: 2446
Joined: Wed Jul 08, 2015 1:53 am

Re: Formalizing Natural Language Semantics

Post by Scott Mayers »

Skepdick wrote: Mon Mar 02, 2020 4:55 pm
Scott Mayers wrote: Mon Mar 02, 2020 4:49 pm
Skepdick wrote: Mon Mar 02, 2020 4:35 pm
It's an assertion. A statement of fact. Not an insult.

The decision-procedure I used to assert his "dumbness" was explicitly stated.

Leave your feelings at the door.
But to insult his point as 'dumb'
Pay attention..... IT IS NOT AN INSULT.

It is an assertion.
READ and interpret the rest in context. You only save grace of it NOT being an insult if you mean that "YOU do not understand what he is saying."

The term, "not smart" may be appropriate. But then you still have the onus to prove WHY for ASSERTING such a conclusive belief unexpressed.
Skepdick
Posts: 14364
Joined: Fri Jun 14, 2019 11:16 am

Re: Formalizing Natural Language Semantics

Post by Skepdick »

PeteOlcott wrote: Mon Mar 02, 2020 4:52 pm Within the context of your expression: "know" means subjective assessment.
Yes Pete. All knowledge is subjective because all assertions are subjective.
PeteOlcott wrote: Mon Mar 02, 2020 4:52 pm If we use the most literal meaning
You lack a formal definition of "literal meaning" - that's the very thing you are claiming to be formalizing here, right!?!?
PeteOlcott wrote: Mon Mar 02, 2020 4:52 pm is logically incorrect in their dislike of apples then your sentence becomes false.
Please don't bullshit me. It is logically, syntactically, grammatically and semantically correct.

Empirically demonstrable.

https://repl.it/repls/PinkClientsideBlock

Code: Select all

from universe import *

assert Skepdick.likes(Apples)
assert Skepdick.likes(Peaches)
assert not Skepdick.likes(Oranges)

print('All assertions are TRUE')
Last edited by Skepdick on Mon Mar 02, 2020 5:12 pm, edited 3 times in total.
Post Reply