[a_gray]

> Abstract conceptual thinking cannot be separated from our language ability

1. Clearly false (what words do you use when you plan to throw a ball to a target?)

2. There is no reason to believe other animals are incapable of abstract thinking (although we can clearly see they are far less intelligent)

3. I don't know what your long screed about bigotry and religion has to do with language ability

Just imagine how slow university-level maths would be if you had to do all the thinking in words in your head.

Sorry if I sound rude, but this is such a common misconception people believe in, and it's so blatantly not true so I'm baffled people keep on claiming it is 'scientific' to believe that language is identical to intelligence. No wonder some people insist LLMs are just weeks away from becoming AGIs.

[visarga]

> Clearly false (what words do you use when you plan to throw a ball to a target?)

If a monkey can do it, it's not very abstract.

[Tor3]

When I drive to and from work I drive through several places where I have to mentally imagine what several other drivers plan to do, and, depending on what they actually do, (and what problems they may run into due to the slippery road) what the traffic picture will look like in a few moments. Which will let me decide if I stop right here, for example, so that the car which will otherwise be blocked by me and itself will block the traffic coming the other way can pass in front of me and resolve the (future) situation.. things like that. I have to do the whole thought experiment in my brain in a second or two. There are no words at all involved in that. And LLMs can't do such things as of now. I doubt a monkey could either.

At work I do a lot of design thinking. I don't use words for that either, that comes later when I document the thing.

[MrMcCall]

Spacial dynamics don't need abstract thinking because they're concrete. For example: throwing a ball, or catching a ball.

Understanding the physics or logic behind complex moving pieces does require abstract thinking, but that's not what your brain does to perform physical feats.

We saw them, practiced dealing with them, then became proficient, or not.

[Tor3]

I'm talking about the time analyzing, not the time executing. And that's definitely abstract. And, as I mentioned, this applies to my design work too. And when programming. No words involved, until I write the documentation.

[MrMcCall]

> I'm talking about the time analyzing, not the time executing.

I see, and I agree.

> And when programming. No words involved, until I write the documentation.

Your variable names must be horrible ;-)

I do get your point about visualizing software relationships in the design phase, though. I'd still say, as a 40+ year programmer, that we can never escape the concept-words that are the foundation of programming, e.g. types, variables, functions, classes, encapsulation, pipeline, executable, etc.

Even if we are not consciously thinking of the terms, my guess is that, at a certain level of proficiency, we are using them at a kindof subconscious level.

It's an interesting meta-topic, and I won't say you're wrong, but we've certainly stumbled into a mostly unexplored land where no map yet exists. That's why programming is so challenging, difficult, and fascinating.

Peace be with you, friend.

[a_gray]

But it is abstract, without requiring language. Driving would be a better example, the other guy put it better.

Look at my other example - maths. How do you reason about solving mathematical problems rapidly with words? Many mathematicians visualise abstract problems without using words, only using words later to attempt to write more thorough proofs.

[MrMcCall]

> 1. Clearly false (what words do you use when you plan to throw a ball to a target?)

That's physical, not abstract. Our entire body's neural net is used to develop those physical skills. When it comes time to catch or throw a ball, no abstract thinking is involved, just doing.

If you want to study the physics of throwing a ball, then you need abstract thought, but not for just practicing it.

> 2. There is no reason to believe other animals are incapable of abstract thinking (although we can clearly see they are far less intelligent)

And there is reason to believe they're capable of abstract thinking?

> 3. I don't know what your long screed about bigotry and religion has to do with language ability

Because we communicate abstract ideas to one another via language. Once we acquire a mind-topic, we can then focus on it, as per our desire and mental focusability. That's why programming is so difficult.

But we can also contemplate our moral basis with respect to another mind-topic that teaches us how to abstractly evaluate it. It can be called religion or ethics or whatever. But we navigate our choices against whatever we deem permissible or impermissible, desirable or repellant. And we can self-evolve that basis against which we make our choices. It's an essential aspect of human nature, and it requires abstract concepts such as compassion, harm, happiness, sadness, kindness and anger, to name but a few.

> Just imagine how slow university-level maths would be if you had to do all the thinking in words in your head.

Just imagine how uni-level maths would be if there were no abstract concepts being taught with actual words.

Are you saying that not only did your maths education only use diagrams, but that no verbal explanation was involved?

> Sorry if I sound rude, but this is such a common misconception people believe in, and it's so blatantly not true so I'm baffled people keep on claiming it is 'scientific' to believe that language is identical to intelligence.

It's not rude, but it is just wrong. For one, I never said it was identical to intelligence. Intelligence is evaluating an abstract concept net to attempt to evaluate how a particular concept relates to it.

Abstract concepts are built on the same logical network that facilitates our ability to process and produce language. How we build and utilize our ever-evolving network results in abilities that we call intelligence.

> No wonder some people insist LLMs are just weeks away from becoming AGIs.

That's not me, brother. I'm not even on the LLM hype train, much less think they're going to facilitate AGI. They probably have a few niche uses, but that's my take on all the LLM stuff.

[a_gray]

> That's physical, not abstract. Our entire body's neural net is used to develop those physical skills. When it comes time to catch or throw a ball, no abstract thinking is involved, just doing.

Abstraction is about generalising beyond the physical objects we experience. The ability to understand the physics behind how objects work is abstract, but it's a low enough level of abstraction that even simple animals are capable of it.

> Just imagine how uni-level maths would be if there were no abstract concepts being taught with actual words.

Language is pretty good for communication, obviously.

> Are you saying that not only did your maths education only use diagrams, but that no verbal explanation was involved?

My driving test also used language to convey the lesson. But I don't talk to myself before I decide which direction to turn the wheel...

> Abstract concepts are built on the same logical network that facilitates our ability to process and produce language.

Overlap, certainly. Anything else is conjecture.

[MrMcCall]

> The ability to understand the physics behind how objects work is abstract, but it's a low enough level of abstraction that even simple animals are capable of it.

No, it's a physical coordination of our body's neural net that connects our senses, brain, nerves, and musculature to perform coordinated feats. It's the direct opposite of abstract: it's concretely physical. And we share this neural net wetware wiring with our cousins, the animals.

There is nothing abstract about pulling your hand off the stove, but describing anything requires abstract concept-nets.

> Language is pretty good for communication, obviously.

It's essential for communicating abstract concepts, as well as contemplating them.

> My driving test also used language to convey the lesson. But I don't talk to myself before I decide which direction to turn the wheel...

That's because you have taught your wetware how to physicalize the concept net's decision tree by learning how to recognize the inputs and coordinate the proper responses. Lather, rinse, repeat.

> Overlap, certainly. Anything else is conjecture.

Anything that can be named exists as an abstract mental concept; that's what a concept is. The map is not the territory, because it is abstract, because the map exists only in our brain, even if we learned its contours from a paper document. That's what makes it abstract, because we have abstracted it to a mental concept, where its name refers to a network of other names in a describable way.

That is how langauge is built up from literally nothing but our elders' repetitive descriptions of objects and their interactions, kinda like typed values and functions where the typing describes their relationships to other typed values and functions. And the relationships are abstract concepts, too.

Thanks for this interesting philosophical discussion. Peace be with you.