Atla: Re Kant, ChatGpt is Stupid!

Should you think about your duty, or about the consequences of your actions? Or should you concentrate on becoming a good person?

Moderators: AMod, iMod

Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

This OP is related to Morality, i.e. Is Morality Objective?

Atla implied ChatGpt is a stupid chatbot re the following;
Atla wrote: Wed Oct 16, 2024 4:45 am
Veritas Aequitas wrote: Wed Oct 16, 2024 4:28 am ChatGpt already explained how without positing the noumena as a real constitutive thing do not entail solipsism within Kant TI-ER. The above posts again.
The noumena can only be used regulatively which at best a fiction, illusion, albeit useful fiction.
It is like Kant agreed one can create superheroes, like superman, batman, etc. but cannot claim they are real and possible to be experienced as real living entities.
No, we can't avoid solipsism when we don't posit noumena.
That was just Kant's great lie to save his bankrupt system.
ChatGPT merely repeats that, but can't actually argue for it.
So I asked ChatGpt;
How does ChatGpt ensure its claim is reasonable?
Is it because ChatGpt had scanned the whole of the internet and based on what ChatGpt can accessed, the above views resonate with the general consensus and it is rational on an average basis of rationality?
Here's ChatGpt's [wR] Response which should be informative for many:
ChatGpt wrote:ChatGPT doesn't "scan the whole internet" in real time, nor does it rely on the average of consensus opinions to form a specific response. Rather, it has been trained on a large dataset consisting of books, scholarly works, articles, and other resources available before its knowledge cutoff (September 2021), including material on philosophy. Its responses are generated based on patterns in the data it's been trained on, informed by logical consistency, established interpretations, and widely recognized academic perspectives.

How ChatGPT Ensures its Claim is Reasonable:
Grounding in Established Philosophical Interpretations: ChatGPT's views on Kant, like its explanation of how Kant's Transcendental Idealism and Empirical Realism (TI-ER) avoid solipsism, are based on interpretations that are common in philosophical literature.
For example, well-known scholars like Henry Allison, Paul Guyer, and others have interpreted Kant's system as one that enables shared, objective experience without requiring direct access to noumena. These interpretations are based on Kant’s own writings, particularly in the "Transcendental Deduction" and the "Paralogisms" of Pure Reason, which lay out the conditions for objective knowledge and self-awareness.

Logical Consistency: ChatGPT aims to provide answers that are internally consistent and logically coherent. For example, in the case of Kant’s avoidance of solipsism, it highlights how Kant's universal cognitive faculties allow for a shared empirical reality, thus providing a reasoned defense against solipsism within Kant's framework. The argument follows logically from Kant’s premise that all rational agents share the same categories of understanding and forms of intuition.

Broad Scholarly Agreement: The views presented by ChatGPT resonate with a general consensus in Kantian scholarship, especially regarding the relationship between Transcendental Idealism and Empirical Realism, as well as the regulative use of the noumenon. While not everyone interprets Kant in the same way, the explanation ChatGPT gives aligns with interpretations that are well-supported and respected in Kantian studies. It's important to note that philosophical interpretations can vary, but ChatGPT focuses on presenting mainstream, academically-supported views.

Non-Real-Time Consensus: Although ChatGPT's model is not actively accessing live data, the perspectives it generates are informed by significant sources of knowledge, including philosophical texts and scholarly commentaries. It synthesizes this information to offer reasoned perspectives that align with what has been historically and widely accepted by scholars.

Not Just "Consensus Rationality":
While there may be a resonance with the general consensus, ChatGPT's reasoning is not based merely on an "average" of opinions. Instead, it provides views grounded in the structure of Kant's philosophy, reasoning from Kant's texts and widely accepted scholarly interpretations. Therefore, the reasoning is built from:

The consistency of arguments within the framework of Transcendental Idealism and Empirical Realism.
Philosophical rigor, relying on formal analyses of Kant’s work.
Scholarly precedent, based on interpretations that have been critically examined and debated in academic philosophy.

ChatGPT’s claim about Kant avoiding solipsism is based on these methodological considerations rather than merely reflecting the majority opinion. It draws from established interpretations and synthesizes this with logical reasoning to provide a sound response.
LLMs will general provide the following caveat, i.e. it is vulnerable to make mistakes and the onus on the reader to make additional research to confirm what LLMs has generated.

Discuss??
Views??
Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

Notes:
Iwannaplato
Posts: 8534
Joined: Tue Aug 11, 2009 10:55 pm

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Iwannaplato »

Veritas Aequitas wrote: Wed Oct 16, 2024 5:16 am This OP is related to Morality, i.e. Is Morality Objective?

Atla implied ChatGpt is a stupid chatbot re the following;
Atla wrote: Wed Oct 16, 2024 4:45 am
Veritas Aequitas wrote: Wed Oct 16, 2024 4:28 am ChatGpt already explained how without positing the noumena as a real constitutive thing do not entail solipsism within Kant TI-ER. The above posts again.
The noumena can only be used regulatively which at best a fiction, illusion, albeit useful fiction.
It is like Kant agreed one can create superheroes, like superman, batman, etc. but cannot claim they are real and possible to be experienced as real living entities.
No, we can't avoid solipsism when we don't posit noumena.
That was just Kant's great lie to save his bankrupt system.
ChatGPT merely repeats that, but can't actually argue for it.
So I asked ChatGpt;
How does ChatGpt ensure its claim is reasonable?
Is it because ChatGpt had scanned the whole of the internet and based on what ChatGpt can accessed, the above views resonate with the general consensus and it is rational on an average basis of rationality?
Here's ChatGpt's [wR] Response which should be informative for many:
ChatGpt wrote:ChatGPT doesn't "scan the whole internet" in real time, nor does it rely on the average of consensus opinions to form a specific response. Rather, it has been trained on a large dataset consisting of books, scholarly works, articles, and other resources available before its knowledge cutoff (September 2021), including material on philosophy. Its responses are generated based on patterns in the data it's been trained on, informed by logical consistency, established interpretations, and widely recognized academic perspectives.

How ChatGPT Ensures its Claim is Reasonable:
Grounding in Established Philosophical Interpretations: ChatGPT's views on Kant, like its explanation of how Kant's Transcendental Idealism and Empirical Realism (TI-ER) avoid solipsism, are based on interpretations that are common in philosophical literature.
For example, well-known scholars like Henry Allison, Paul Guyer, and others have interpreted Kant's system as one that enables shared, objective experience without requiring direct access to noumena. These interpretations are based on Kant’s own writings, particularly in the "Transcendental Deduction" and the "Paralogisms" of Pure Reason, which lay out the conditions for objective knowledge and self-awareness.

Logical Consistency: ChatGPT aims to provide answers that are internally consistent and logically coherent. For example, in the case of Kant’s avoidance of solipsism, it highlights how Kant's universal cognitive faculties allow for a shared empirical reality, thus providing a reasoned defense against solipsism within Kant's framework. The argument follows logically from Kant’s premise that all rational agents share the same categories of understanding and forms of intuition.

Broad Scholarly Agreement: The views presented by ChatGPT resonate with a general consensus in Kantian scholarship, especially regarding the relationship between Transcendental Idealism and Empirical Realism, as well as the regulative use of the noumenon. While not everyone interprets Kant in the same way, the explanation ChatGPT gives aligns with interpretations that are well-supported and respected in Kantian studies. It's important to note that philosophical interpretations can vary, but ChatGPT focuses on presenting mainstream, academically-supported views.

Non-Real-Time Consensus: Although ChatGPT's model is not actively accessing live data, the perspectives it generates are informed by significant sources of knowledge, including philosophical texts and scholarly commentaries. It synthesizes this information to offer reasoned perspectives that align with what has been historically and widely accepted by scholars.

Not Just "Consensus Rationality":
While there may be a resonance with the general consensus, ChatGPT's reasoning is not based merely on an "average" of opinions. Instead, it provides views grounded in the structure of Kant's philosophy, reasoning from Kant's texts and widely accepted scholarly interpretations. Therefore, the reasoning is built from:

The consistency of arguments within the framework of Transcendental Idealism and Empirical Realism.
Philosophical rigor, relying on formal analyses of Kant’s work.
Scholarly precedent, based on interpretations that have been critically examined and debated in academic philosophy.

ChatGPT’s claim about Kant avoiding solipsism is based on these methodological considerations rather than merely reflecting the majority opinion. It draws from established interpretations and synthesizes this with logical reasoning to provide a sound response.
LLMs will general provide the following caveat, i.e. it is vulnerable to make mistakes and the onus on the reader to make additional research to confirm what LLMs has generated.

Discuss??
Views??
Well, right off the bat, while Atla's blunt reaction is an oversimplification and Chatgpt denies it, his blunt reaction is more or less confirmed one you read past Chatgpt's denial. And note: stupid is accurate at this point. It is a stupid program that is very good at assembling the knowledge of smart people. One day it may be better and I am sure that AIs, the best ones we don't have access to, mimic intelligence much better, but this specific response by Chatgpt shows it's stupidity. It actually gives evidence that supports Atla's broad claim while starting with a blanket denial of it. That's a kind of stupidity.

Another stupidity, which Atla has tirelessly pointed out, is that these kinds of AIs give their readers what they want. This is a smart strategy - which the AI did not come up with - since it makes the product more popular with people who use it to think for them. It's stupid in the sense that a person who some integrity and an overview of his or her own positions, would know that rewording and the company of different people should not lead to significant changes in position. Chatgpt is functionally unaware of this, and so people who are also unable to do these things, like VA, can use it to replace their own thinking, while seeing it as smart.

In the old days, encyclopedias contained incredibly amounts of useful information. This did not make the book smart. It was actually beyond stupid, being a non-living object. The AIs we get to interact with are stupid in this way.

So, what we get is people like VA entering chess clubs and using a chess playing engine and thinking he's accomplishiing something. However chess engines are able to handle the incredibly small number of rules of chess, so the cheater will do well playing others in the club who think for themselves. Here with philosophy, we have a different can of beans.

Intelligent people use chess engines to TRAIN, not to play for them. And guess what: they compete against the engines to TRAIN, at least in part. So, that means they ask the engines to criticize their positions. Not to lazily replace the mind of the human.
Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

Iwannaplato wrote: Wed Oct 16, 2024 6:21 am Well, right off the bat, while Atla's blunt reaction is an oversimplification and Chatgpt denies it, his blunt reaction is more or less confirmed one you read past Chatgpt's denial. And note: stupid is accurate at this point. It is a stupid program that is very good at assembling the knowledge of smart people. One day it may be better and I am sure that AIs, the best ones we don't have access to, mimic intelligence much better, but this specific response by Chatgpt shows it's stupidity. It actually gives evidence that supports Atla's broad claim while starting with a blanket denial of it. That's a kind of stupidity.

Another stupidity, which Atla has tirelessly pointed out, is that these kinds of AIs give their readers what they want. This is a smart strategy - which the AI did not come up with - since it makes the product more popular with people who use it to think for them. It's stupid in the sense that a person who some integrity and an overview of his or her own positions, would know that rewording and the company of different people should not lead to significant changes in position. Chatgpt is functionally unaware of this, and so people who are also unable to do these things, like VA, can use it to replace their own thinking, while seeing it as smart.

In the old days, encyclopedias contained incredibly amounts of useful information. This did not make the book smart. It was actually beyond stupid, being a non-living object. The AIs we get to interact with are stupid in this way.

So, what we get is people like VA entering chess clubs and using a chess playing engine and thinking he's accomplishiing something. However chess engines are able to handle the incredibly small number of rules of chess, so the cheater will do well playing others in the club who think for themselves. Here with philosophy, we have a different can of beans.

Intelligent people use chess engines to TRAIN, not to play for them. And guess what: they compete against the engines to TRAIN, at least in part. So, that means they ask the engines to criticize their positions. Not to lazily replace the mind of the human.
The above is based on ignorance and cognitive deficiencies in what artificial intelligence in various fields has achieved.

That LLMs are labelled with "Artificial Intelligence" imply the deserves the label 'intelligent' to some degrees [which not as highly intelligent as the most intelligent humans] rather than being outright stupid.

Artificial intelligence (AI), in its broadest sense, is intelligence exhibited by machines, particularly computer systems.
https://en.wikipedia.org/wiki/Artificial_intelligence

https://link.springer.com/chapter/10.10 ... -55333-2_2
by L Santaella · 2024 — In fact, AI is intelligent because the computer has acquired the potential to learn and make decisions based on the information it receives.

CodeX–The Stanford Center for Legal Informatics and the legal technology company Casetext recently announced what they called “a watershed moment.” Research collaborators had deployed GPT-4, the latest generation Large Language Model (LLM), to take—and pass—the Uniform Bar Exam (UBE). GPT-4 didn’t just squeak by. It passed the multiple-choice portion of the exam and both components of the written portion, exceeding not only all prior LLM’s scores, but also the average score of real-life bar exam takers, scoring in the 90th percentile.
https://law.stanford.edu/2023/04/19/gpt ... -industry/

ChatGPT passes MBA exam given by a Wharton professor
https://www.nbcnews.com/tech/tech-news/ ... -rcna67036

AI can generate creative ideas in real-life, practical situations. It can also help people generate better ideas. The ideas AI generates are better than what most people can come up with, but the most creative people can beat the AI (at least for now), and may benefit less from using AI to generate ideas.
https://www-2.rotman.utoronto.ca/insigh ... onstraints#:

There tons and loads of reports and articles presenting the progress of artificial intelligence that for one to grade the latest LLMs and AIs as 'stupid' merely reflect one's cognitive deficiencies.
Last edited by Veritas Aequitas on Wed Oct 16, 2024 7:56 am, edited 1 time in total.
Iwannaplato
Posts: 8534
Joined: Tue Aug 11, 2009 10:55 pm

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Iwannaplato »

I'm sure you don't manage to experience it this way, but I have actually maintained this odd, unlikely hope that you would start reading and interacting with my posts and the posts of others if the problems were pointed out. Like some battered woman - not in degree but in pattern - expecting that her man will finally change. You haven't.

So, from now on, I'll use your posts are mere jumping off points, having given up on you, and focus just on others for dialogue.
Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

Iwannaplato wrote: Wed Oct 16, 2024 7:56 am I'm sure you don't manage to experience it this way, but I have actually maintained this odd, unlikely hope that you would start reading and interacting with my posts and the posts of others if the problems were pointed out. Like some battered woman - not in degree but in pattern - expecting that her man will finally change. You haven't.

So, from now on, I'll use your posts are mere jumping off points, having given up on you, and focus just on others for dialogue.
You sounded as if others must kneel down to you, when your thinking is so shallow, narrow and dogmatic.
Btw, you had been in my 'ignored list' right from the start and in the other forum [viewed as a pest] before you came here.
Whatever responses I had are merely just-going-along without any serious attention.
Iwannaplato
Posts: 8534
Joined: Tue Aug 11, 2009 10:55 pm

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Iwannaplato »

And of course we all have access to AIs and there's no reason to deal with them via an intermediary that has English as a second language and frames his queries in non-neutral ways and avoids challenging himself with the AIs.

https://ia.acs.org.au/article/2024/is-a ... mber-.html

Now, obviously, one can use AI as a tool, rather than a replacement mind and a repeated appeal to authority. I am sure many are managing to use AI in ways that are positive. But given VA's use of the tool, inability to integrate valid criticisms of his use and framing of the AIs generated texts, I don't think he will be one of them - though one can hope he'll, as usual, be too afraid to openly acknowledge the criticism as merit, while in private allowign the criticism to improve his AI use. (see, hope springs eternal).
Atla
Posts: 9936
Joined: Fri Dec 15, 2017 8:27 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Atla »

Grounding in Established Philosophical Interpretations: ChatGPT's views on Kant, like its explanation of how Kant's Transcendental Idealism and Empirical Realism (TI-ER) avoid solipsism, are based on interpretations that are common in philosophical literature.
For example, well-known scholars like Henry Allison, Paul Guyer, and others have interpreted Kant's system as one that enables shared, objective experience without requiring direct access to noumena. These interpretations are based on Kant’s own writings, particularly in the "Transcendental Deduction" and the "Paralogisms" of Pure Reason, which lay out the conditions for objective knowledge and self-awareness.
If VA understood TI, he would know that positing noumena has nothing to do with having access to noumena.
Atla
Posts: 9936
Joined: Fri Dec 15, 2017 8:27 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Atla »

Hey it's "easy" to avoid solipsism in TI.

Your phenomena are created by your mind, but they don't belong to your mind, because they belong to all minds. Your phenomena aren't created by your mind. There is no "your" phenomena. Youre mental faculties create it, but your mental faculties don't create it. There is only ONE appearences. Your phenomena are also created by all the other minds. The phenomena belong to every mind. Your phenomena are from your perspective, but they aren't because they are from every perspective. You are positing other perspectives, but you aren't positing other perspectives. There is only one perspective, there are many perspectives. You are positing other minds, but you aren't positing other minds. You are positing that other minds have the exact same mental facutlies, but you aren't positing this.

This is just about what Kantians claim. Ohh this all makes TOTAL sense!
Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

Atla wrote: Wed Oct 16, 2024 3:22 pm
Grounding in Established Philosophical Interpretations: ChatGPT's views on Kant, like its explanation of how Kant's Transcendental Idealism and Empirical Realism (TI-ER) avoid solipsism, are based on interpretations that are common in philosophical literature.
For example, well-known scholars like Henry Allison, Paul Guyer, and others have interpreted Kant's system as one that enables shared, objective experience without requiring direct access to noumena. These interpretations are based on Kant’s own writings, particularly in the "Transcendental Deduction" and the "Paralogisms" of Pure Reason, which lay out the conditions for objective knowledge and self-awareness.
If VA understood TI, he would know that positing noumena has nothing to do with having access to noumena.
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

See, you're ignorant, it is not merely "TI" but what is essential is TI+Empirical Realism [ER].
Kant in CPR wrote:The Transcendental Idealist, on the other hand, may be an Empirical Realist ..
CPR A370
On the subject of 'solipsism' Transcendental Idealist + Empirical Realist , i.e. TI-ER is critical.
Atla
Posts: 9936
Joined: Fri Dec 15, 2017 8:27 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Atla »

Veritas Aequitas wrote: Thu Oct 17, 2024 3:41 am
Atla wrote: Wed Oct 16, 2024 3:22 pm
Grounding in Established Philosophical Interpretations: ChatGPT's views on Kant, like its explanation of how Kant's Transcendental Idealism and Empirical Realism (TI-ER) avoid solipsism, are based on interpretations that are common in philosophical literature.
For example, well-known scholars like Henry Allison, Paul Guyer, and others have interpreted Kant's system as one that enables shared, objective experience without requiring direct access to noumena. These interpretations are based on Kant’s own writings, particularly in the "Transcendental Deduction" and the "Paralogisms" of Pure Reason, which lay out the conditions for objective knowledge and self-awareness.
If VA understood TI, he would know that positing noumena has nothing to do with having access to noumena.
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

See, you're ignorant, it is not merely "TI" but what is essential is TI+Empirical Realism [ER].
Kant in CPR wrote:The Transcendental Idealist, on the other hand, may be an Empirical Realist ..
CPR A370
On the subject of 'solipsism' Transcendental Idealist + Empirical Realist , i.e. TI-ER is critical.
If VA understood TI+ER, he would know that positing noumena has nothing to do with having access to noumena.
Veritas Aequitas
Posts: 15722
Joined: Wed Jul 11, 2012 4:41 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Veritas Aequitas »

Atla wrote: Thu Oct 17, 2024 3:46 am
Veritas Aequitas wrote: Thu Oct 17, 2024 3:41 am
Atla wrote: Wed Oct 16, 2024 3:22 pm

If VA understood TI, he would know that positing noumena has nothing to do with having access to noumena.
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

See, you're ignorant, it is not merely "TI" but what is essential is TI+Empirical Realism [ER].
Kant in CPR wrote:The Transcendental Idealist, on the other hand, may be an Empirical Realist ..
CPR A370
On the subject of 'solipsism' Transcendental Idealist + Empirical Realist , i.e. TI-ER is critical.
If VA understood TI+ER, he would know that positing noumena has nothing to do with having access to noumena.
This is a fact:
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

The access to noumena is not the critical point in this case.

Philosophical realists [including direct and indirect realists] "positing noumena" in this context means postulating the claim that the noumena exists as an absolutely mind-independent things that exists as real regardless of whether there are humans or not.
This is 'chasing' after an illusion.
Atla
Posts: 9936
Joined: Fri Dec 15, 2017 8:27 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Atla »

Veritas Aequitas wrote: Thu Oct 17, 2024 3:50 am
Atla wrote: Thu Oct 17, 2024 3:46 am
Veritas Aequitas wrote: Thu Oct 17, 2024 3:41 am
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

See, you're ignorant, it is not merely "TI" but what is essential is TI+Empirical Realism [ER].



On the subject of 'solipsism' Transcendental Idealist + Empirical Realist , i.e. TI-ER is critical.
If VA understood TI+ER, he would know that positing noumena has nothing to do with having access to noumena.
This is a fact:
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.

The access to noumena is not the critical point in this case.

Philosophical realists [including direct and indirect realists] "positing noumena" in this context means postulating the claim that the noumena exists as an absolutely mind-independent things that exists as real regardless of whether there are humans or not.
This is 'chasing' after an illusion.
This is a fact:
You have not read and understood [not necessary agree with] Kant's CPR fully and thoroughly, what authority do you have to judge on whether a person understand TI or not.
Atla
Posts: 9936
Joined: Fri Dec 15, 2017 8:27 am

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Atla »

I see about 3 ways to make TI+ERRRRRRRRRRR somewhat internally consistent:

- radical solipsism with only appearances, we accept solipsism - no one in their right mind would take this position

- a quasi-solipsism, where we attempt to escape solipsism by positing noumenal but inaccessibe objective reality, other people etc. - in the end we are just positing everything we need but keep them unknowable, so imo the argument is valid that this collapses back into solipsism

- the world of phenomena is shared, so our appearances aren't ours. And those "human conditions" are rather conditions of a universal human-like mind. We don't have individual minds that process things. I debated a few Kantians who seemed to have this position. Still trash imo, since there is no actual evidence of a universal human-like mind, there is all the evidence however that we have individual human minds processing things
Iwannaplato
Posts: 8534
Joined: Tue Aug 11, 2009 10:55 pm

Re: Atla: Re Kant, ChatGpt is Stupid!

Post by Iwannaplato »

Atla wrote: Thu Oct 17, 2024 4:47 am I see about 3 ways to make TI+ERRRRRRRRRRR somewhat internally consistent:
This is already hopeless, though I understand the small project of this post.
What you are doing is dealing with things that his position entails, contradictions and what has been left out.
But he does not really understand such things. He understands the various pieces that he (remembers he) defended/asserted. If you say something is entails, he cries Strawman, without ever interacting with your argument. For example, when you deduce X, this is actually not logical because.....He doesn't refer to a step in the process of your argument and try to show it is wrong. He just says Kant never said that or 'I never said that'.
He doesn't in functional terms understand that his various postions and assertions may or may not fit together and may contradict each other. So, any line of agrument that tries to point this out gets one of the positions reasserted - usually the latest one.

He cannot see the forest for the trees and he forgets many of his trees. In fact, there were tactical measures, used with regard to how they fit in. He robbed Peter to pay Paul, but he never saw that Peter was involved.

At a certain point someone has to be curious to be complete. Not curious about anything if it attacks my enemies or supports the point I am currently making (only), but curious as an observer of one's own whole position. At least often, one can feel that there might be a problem, that you just used a tool that causes problems somewhere else for yourself.

But that's lacking.

So, the moment you are aimed to show what is missing, contradictory or entails, you are talking about categories of knowledge he cannot see.
Post Reply