Re:Language is The Scaffold of The Mind (2019)
by motohagiography on Thursday November 04, 2021 @09:25PM
(#29114470)
I get the impression that philosophy of mind people neither have dogs nor ride horses seriously, as both are supremely complex relationships and conversations without verbal language - as defined by being a way of communicating abstractions, counterfactuals, and non-present things.
The dogs I've had all understood nonverbal and almost invisible cues, but the horse case would probably blow their minds. A ride is a conversation with another being that asks it to suspend its instincts as a literal prey animal, let a predator climb on its back, maintain this suspension of instinct, and then contact muscles, move each limb and bend its body in a precise way in response to microexpressions of the body that originate from intentions in the riders mind, both in and outside its natural habitat, and even trust you enough to blindly jump obstacles without any idea of what's on the other side of them if they are doing XC.
Serious riders cultivate a sophisticated and rich sub-language for communicating with completely different beings - and one that transfers between them! Putting on shows is just a way to make money and support the habit, but it has almost nothing to do with the human art of communicating with other beings. When you reframe riding a horse as riding an alien, suddenly it's insanely cool that we can do these incredibly sophisticated things without language.
I wish I could source it but there was an AI talk posted here where the speaker said, "language is a tool we evolved to act through others," and I think it was in the context of getting ML agents to develop ways to co-ordinate and scale. It's a rudimentary tool, but it is not the substrate or the real for anything other than perhaps the ego, which I believe is also an artifact of language.
Language is not so much the scaffolding as it is a keyhole that provides a very narrow view of the world, and perhaps of the self, which I suspect is also an artifact of language. The notion of "I," is above the fold of language, whereas when you are working with another being without language, the "I," disappears, much like when we have other intense experiences like speeding on a motorcycle or tripping balls on a mismeasured dose of hallucinogens, not often at the same time. BJJ practitioners talk about similar ways of relating as well.
Maybe it's just me, but as someone who spends much of their time cultivating skills outside the domain of language, I'd wonder if philosophers of mind need to get out more?
Re:Language is The Scaffold of The Mind (2019)
by analog31 on Thursday November 04, 2021 @10:04PM
(#29114666)
I know it's qualitatively different, and not to make light of your comment... But I couldn't help amusing myself with the thought, that I have the same kind of relationship with my bicycle.
A difference is that if I close my eyes while riding a horse, the horse probably won't kill me.
Re:Language is The Scaffold of The Mind (2019)
by marcosdumay on Thursday November 04, 2021 @10:34PM
(#29114871)
Riding a horse is a conversation of sorts. The horse has its own opinions on how fast it wants to go and what path it wants to take. It doesn't just blindly obey you, although you can make yourself "loud enough" to override it.
Riding a donkey, on the other hand is completely different, as you just can't make yourself "loud enough" and must always settle for a compromise.
Re:Language is The Scaffold of The Mind (2019)
by joe_the_user on Thursday November 04, 2021 @11:36PM
(#29115175)
Is it possible to say that dogs and horses are creatures beyond any machine and to say that language grants human beings unique advantages and a unique flexibility?
Re:Language is The Scaffold of The Mind (2019)
by motohagiography on Friday November 05, 2021 @09:52AM
(#29119218)
It is, and I think there was previously a general view that only humans have language that handles abstractions like counterfactuals and non-present objects, which was disproven by border collies searching for named objects.
The "riding an alien," frame opens up the idea of other intelligences to ask the question of, is there an analogy where if let's say most humans and our language lacked the tools to communicate or reason together about things that were not there in the room with us, how would you communicate with those people?
It's a bit like Plato's Cave, but imo, the answer is math. Language is not the substrate for cognition and the mind, math is.
The way we communicate concepts to horses, interestingly, is via geometry. A horse has no concept of lines and circles, but by establishing a foundation of lines, circles and other figures, we build language with them by associating movements that express our intentions with those figures and points on them. The reason it's not just "training," and I would argue it's language, is because once you have established this geometric languge, a horse can determine your intentions from the ground, meters away, and not on its back. Modern olympic dressage in this sense is a kind of cargo cult that preserves some of these ideas and techniques, but in a weird ritual pagentry form, where the deeper art is something each individual practices. I'd speculate that humans have been doing math as long as we've been riding horses isn't a total coincidence.
It makes one wonder what else geometry can be a vehicle for communicating, and what it might mean to encode intentions via geometry for us. It's worth re-watching the Ted Chiang / Denis Villeneuve movie Arrival and Amy Adams' characters' flashbacks in it for a really beautiful illustration of this.
Re:Language is The Scaffold of The Mind (2019)
by mrjangles on Thursday November 04, 2021 @10:15PM
(#29114733)
>Research on adult individuals with aphasia has demonstrated that math, theory of mind, and many other cognitive abilities are independent from language.4 Patients with severe language impairments perform comparably to the rest of us when *asked* to complete arithmetic tasks, reason about people’s intentions, determine physical causes of actions, or decide whether a drawing depicts a real-life event.
Ok something is weird here. How were they *asked* to do these things if they have no language skills. Is the author brushing over the fact that they still had language skills, but they were not as good as they used to be? In which case all her conclusions about what we can do without any language skills are irrelevant as they don't apply, which would explain why she brushes over this point, I guess.
Re:Language is The Scaffold of The Mind (2019)
by fourtrees on Saturday November 06, 2021 @08:05AM
(#29129502)
I'm not sure exactly what constitutes a 'severe language impairment'; there are lots of aphasias out there that only affect very specific parts of the whole language/writing system in the brain. They could be comprehend normally, but have trouble producing (on some level) speech or vice versa. If could comprehend spoken language without difficult, they could 'easily' complete tasks that didn't require those neural connections. Without knowing what the specific symptoms were, this is kind of a moot point.
Re:Language is The Scaffold of The Mind (2019)
by inter_netuser on Friday November 05, 2021 @03:05AM
(#29116216)
Nearly all of such studies are on stroke/trauma victims.
What weirds me out is that they completely disregard that no stroke/trauma victims are identical, but I guess they try their best to derive some insights on this rather heterogenous population.
These are hard experiments to run. I fear that as soon as we figure out how dolphins speak we are going to run such gruesome experiments, only intentional and controlled, on intelligent life.
Re:Language is The Scaffold of The Mind (2019)
by tim333 on Friday November 05, 2021 @09:48AM
(#29119165)
>Our language is but a scaffold for our minds: indispensable during construction
I'm not convinced it's indispensable. Kids have minds before they get talking and some famous thinkers like Einstein "didn't speak full sentences until he was 5 years old". He was famous for his visual imagination and I think that's more indispensable. Language is a nice add on.
Re:Language is The Scaffold of The Mind (2019)
by kazinator on Thursday November 04, 2021 @09:56PM
(#29114634)
> More specifically, can you imagine your mind without language? Can you think, plan, or relate to other people if you lack words to help structure your experiences?
Yes; for instance, when you use your mind to come up with how to rotate a sofa to get it around the bend in a stairwell, that does not really involve verbal/syntactic constructions, except for expressing the requirement at the highest level.
If you have to communicate the solution to your moving helpers, there will be a lot of hand gesturing, and use of pronouns. From an audio-only recording of the scene, capturing the verbalizations, nobody can reconstruct the manipulation of the sofa in even scant amounts of detail.
If people had to accurately document everything they do, every decision they make, in words, they wouldn't get anything done; they would be too busy editing words.
Re:Language is The Scaffold of The Mind (2019)
by fsckboy on Thursday November 04, 2021 @07:25PM
(#29113685)
language is much less the scaffold of the mind than people think it is. You know that feeling when you can't remember somebody's name, or you can't remember the word for something? You have no trouble thinking of the thing, you just can't come up with the word? That's the scaffolding of the brain, not the words.
Re:Language is The Scaffold of The Mind (2019)
by gumby on Thursday November 04, 2021 @07:41PM
(#29113792)
The scaffold does not remain once the building is complete. Once you have learnt an abstract idea like “justice” (to use an example from the article) you can still reason about it even if you lose the word.
Re:Language is The Scaffold of The Mind (2019)
by tim333 on Friday November 05, 2021 @09:36AM
(#29119001)
Though you can have an idea about things like justice without any language scaffolding. I got a quite a few views on an edit I did of Frans de Waal's talk where monkeys get annoyed by injustice https://youtu.be/-dMoK48QGL8
Re:Language is The Scaffold of The Mind (2019)
by unsui on Thursday November 04, 2021 @07:46PM
(#29113830)
Those are simply incidents where the scaffold foothold temporarily slips.
A true counterexample is pure interaction without words, akin to what dogs and other animals do (excepting things like recognizing their name), or meditative state where the conscious mind is no longer driving.
Words provide an information-space regularity mapping to similar constructs in other minds, a shared "game" with shared (crowdsourced) rules.
This, in turns, allows the cognitive chunking of words/ideas mappings so that they can in turn be used as constructs within larger intentional chunks, largely automated by your brain machinery, so that you don't even have to be aware of the syntax, only the semantics.
A world without words would be like navigating in an ocean without lighthouses and stars: anything is possible, but you never really know where you are and where anything else is (including others' minds)
Re:Language is The Scaffold of The Mind (2019)
by hyperpallium2 on Thursday November 04, 2021 @08:06PM
(#29113972)
Better: "Scaffold of the Neocortex". Repiles and mammals know what things are, and can reason about them, a little.
Human-level reasoning requires something like language, for generality. Particularly, to reason about reasoning - to argue, to see the argument of another, to see their point of view, to trade disimilar things, to "see where you went wrong", to deceive.
It's analogous to: animals are FSA; we are TMs.
You can argue for the intelligence of other animals, but they can't argue for yours.
Re:Language is The Scaffold of The Mind (2019)
by V-2 on Thursday November 04, 2021 @07:42PM
(#29113805)
If you can't remember the word, you must have nevertheless have learned it at some point in the past (and then forget). By learning it, you've imprinted the concept in your mind. Afterwards you just lost the identifier. But still, language was how you were able to acquire the idea to begin with. Remembering a person is different, but a person isn't an abstract entity.
Re:Language is The Scaffold of The Mind (2019)
by mannykannot on Thursday November 04, 2021 @09:51PM
(#29114612)
Some philosophers hold that learning about something by acquaintance or experience (including the experience of trying to do something, which, when we are successful, gives us know-how) is different from learning, and coming to believe in, a collection of true propositions. Others, such as Stanley and Williamson in their influential paper "Knowing How", argue that all knowledge is, fundamentally, propositional knowledge - a position known as intellectualism.
I have always had strong doubts about the latter position, and I do not think it can survive the challenge presented by the sort of cases presented in this article.
With knowledge by acquaintance or experience (including of pain), we grasp the concept before we have a name for it, and the name is an optional relationship. I am well aware of this because I have a small problem remembering names, but when it happens, it does not interfere in any other way in my knowledge of that person. When it happens, I have, indeed, temporarily forgotten a name that I once knew (or at least lost conscious access to it), but it seems plausible that most of what I still know about them could have been learned without ever knowing their name.
Language can supplement our experience in at least a couple of ways: we can learn some concepts that might, in principle, be learned by experience, but we have not actually done so (this is not quite the same, however, which might explain my yearning to visit places with an interesting history, even if everything from that time have crumbled to dust, and so, intellectually, I realize that such a visit will not lead to significant new insights.) Secondly, it is not clear to me that one could gain abstract concepts such as the square root of -1 or entropy by language-free experience alone, which I think is the thesis expressed in the article's title.
[1] Stanley, Jason and Timothy Williamson, 2001, “Knowing How”, Journal of Philosophy, 98(8): 411–444. doi:10.2307/2678403
Re:Language is The Scaffold of The Mind (2019)
by kaba0 on Friday November 05, 2021 @12:32PM
(#29121576)
I’m fairly sure that the few n years of our lives are full of examples of very abstract things we didn’t learn through language.
Like, the whole of 3D space, physics, object permanence, etc are most definitely not learnt through language at all (though arguably, we might have some innate knowledge on some of these categories? I believe 3D objects reliably break down into the same basic 3D shapes in all humans)
Re:Language is The Scaffold of The Mind (2019)
by V-2 on Friday November 05, 2021 @02:56PM
(#29123369)
The idea of object permanence is not abstract in the sense that I mean. You get to experience it. But could anyone grasp eg. the idea of Einstenian time dilation without language? Or 4+ dimensions?
Re:Language is The Scaffold of The Mind (2019)
by fsckboy on Thursday November 04, 2021 @07:49PM
(#29113859)
No. You can meet people and get to know them without learning their names. You can learn to shoe horses without knowing the word horseshoe, and you can even get a very clear grasp of what a horseshoe does, as a deaf person would.
Re:Language is The Scaffold of The Mind (2019)
by V-2 on Friday November 05, 2021 @02:42PM
(#29123224)
This doesn't address (much less invalidate) my point. Pain is not an abstract concept.
This distinction (between non-abstract and abstract entities) already becomes blurred when it comes to things like colors. In their case your quip starts to break down - contrary to our intuition, it turns out that you do need the word in order to experience the concept. To some extent at least.
It's been proven that - even though colors exist on physical level - languages (vocabulary) do have an impact on how humans perceive them.
Eg.
"Researchers found that Greek and Russian speakers, who have dedicated words for light and dark blue, were more likely to see a light blue triangle against a dark blue background (and vice versa), than they were to identify green triangles against green backgrounds. Speakers of German, which has no such distinction, were no better at seeing shades of blue triangles than green."
Re:Language is The Scaffold of The Mind (2019)
by remir on Thursday November 04, 2021 @08:45PM
(#29114227)
Langage is built on memory, and without memories, how can there be thoughts?
If you are from some uncontacted tribe in the Amazon that were never exposed to any modern technology, you cannot think about computers, skyscrapers or microwave ovens because you would not have the memories of these things.
So it's seems to me that memory is the scaffold of the mind.
Re:Language is The Scaffold of The Mind (2019)
by HKH2 on Tuesday November 09, 2021 @11:56PM
(#29171220)
I can't imagine any teacher would say that language precedes memory. When teaching, the first thing you should do is set a context to help students recall relevant memories to work with.
I think words themselves are kept somewhat stable due to being in overlapping memories, and thus protected from the brain cleanup process.
Re:Language is The Scaffold of The Mind (2019)
by remir on Friday November 05, 2021 @08:54PM
(#29126418)
New ideas are always built on existing notions. Same with science fiction or any inventions. Nothing gets created out of nothing. Yes it is simple, because that's reality.
Imagine you are completely blind. The concept of colours would exist in your mind because you have access to literature mentioning it. You would know it's a thing, it exist, and yet, because you haven't been exposed to colors, you would have no memories of them of your own.
Now imagine if everybody in society was blind. Because nobody would have been exposed to colours, there would be no memories of them. It wouldn't be a thing as an idea in people's mind. As a result, there would be no way of thinking about the color red, or blue. It wouldn't be part of language either. Why describe something that we don't know that exist?
Re:Language is The Scaffold of The Mind (2019)
by heresie-dabord on Thursday November 04, 2021 @06:25PM
(#29113219)
For language to work, there must be reasoning: "Given the available tooling, how do I communicate this thing or idea?" and "How can I understand what the other person just said or wrote?"
For broken communication to be fixed, there must be debugging. Discussion and laughter are essential.
Human reason is as ancient a capacity as language itself.
The dogs I've had all understood nonverbal and almost invisible cues, but the horse case would probably blow their minds. A ride is a conversation with another being that asks it to suspend its instincts as a literal prey animal, let a predator climb on its back, maintain this suspension of instinct, and then contact muscles, move each limb and bend its body in a precise way in response to microexpressions of the body that originate from intentions in the riders mind, both in and outside its natural habitat, and even trust you enough to blindly jump obstacles without any idea of what's on the other side of them if they are doing XC.
Serious riders cultivate a sophisticated and rich sub-language for communicating with completely different beings - and one that transfers between them! Putting on shows is just a way to make money and support the habit, but it has almost nothing to do with the human art of communicating with other beings. When you reframe riding a horse as riding an alien, suddenly it's insanely cool that we can do these incredibly sophisticated things without language.
I wish I could source it but there was an AI talk posted here where the speaker said, "language is a tool we evolved to act through others," and I think it was in the context of getting ML agents to develop ways to co-ordinate and scale. It's a rudimentary tool, but it is not the substrate or the real for anything other than perhaps the ego, which I believe is also an artifact of language.
Language is not so much the scaffolding as it is a keyhole that provides a very narrow view of the world, and perhaps of the self, which I suspect is also an artifact of language. The notion of "I," is above the fold of language, whereas when you are working with another being without language, the "I," disappears, much like when we have other intense experiences like speeding on a motorcycle or tripping balls on a mismeasured dose of hallucinogens, not often at the same time. BJJ practitioners talk about similar ways of relating as well.
Maybe it's just me, but as someone who spends much of their time cultivating skills outside the domain of language, I'd wonder if philosophers of mind need to get out more?
A difference is that if I close my eyes while riding a horse, the horse probably won't kill me.
Riding a donkey, on the other hand is completely different, as you just can't make yourself "loud enough" and must always settle for a compromise.
You don't get that from a bicycle.
The "riding an alien," frame opens up the idea of other intelligences to ask the question of, is there an analogy where if let's say most humans and our language lacked the tools to communicate or reason together about things that were not there in the room with us, how would you communicate with those people?
It's a bit like Plato's Cave, but imo, the answer is math. Language is not the substrate for cognition and the mind, math is.
The way we communicate concepts to horses, interestingly, is via geometry. A horse has no concept of lines and circles, but by establishing a foundation of lines, circles and other figures, we build language with them by associating movements that express our intentions with those figures and points on them. The reason it's not just "training," and I would argue it's language, is because once you have established this geometric languge, a horse can determine your intentions from the ground, meters away, and not on its back. Modern olympic dressage in this sense is a kind of cargo cult that preserves some of these ideas and techniques, but in a weird ritual pagentry form, where the deeper art is something each individual practices. I'd speculate that humans have been doing math as long as we've been riding horses isn't a total coincidence.
It makes one wonder what else geometry can be a vehicle for communicating, and what it might mean to encode intentions via geometry for us. It's worth re-watching the Ted Chiang / Denis Villeneuve movie Arrival and Amy Adams' characters' flashbacks in it for a really beautiful illustration of this.
Ok something is weird here. How were they *asked* to do these things if they have no language skills. Is the author brushing over the fact that they still had language skills, but they were not as good as they used to be? In which case all her conclusions about what we can do without any language skills are irrelevant as they don't apply, which would explain why she brushes over this point, I guess.
What weirds me out is that they completely disregard that no stroke/trauma victims are identical, but I guess they try their best to derive some insights on this rather heterogenous population.
These are hard experiments to run. I fear that as soon as we figure out how dolphins speak we are going to run such gruesome experiments, only intentional and controlled, on intelligent life.
Language Is the Scaffold of the Mind - https://news.ycombinator.com/item?id=21085946 - Sept 2019 (37 comments)
I'm not convinced it's indispensable. Kids have minds before they get talking and some famous thinkers like Einstein "didn't speak full sentences until he was 5 years old". He was famous for his visual imagination and I think that's more indispensable. Language is a nice add on.
Yes; for instance, when you use your mind to come up with how to rotate a sofa to get it around the bend in a stairwell, that does not really involve verbal/syntactic constructions, except for expressing the requirement at the highest level.
If you have to communicate the solution to your moving helpers, there will be a lot of hand gesturing, and use of pronouns. From an audio-only recording of the scene, capturing the verbalizations, nobody can reconstruct the manipulation of the sofa in even scant amounts of detail.
If people had to accurately document everything they do, every decision they make, in words, they wouldn't get anything done; they would be too busy editing words.
A true counterexample is pure interaction without words, akin to what dogs and other animals do (excepting things like recognizing their name), or meditative state where the conscious mind is no longer driving.
Words provide an information-space regularity mapping to similar constructs in other minds, a shared "game" with shared (crowdsourced) rules.
This, in turns, allows the cognitive chunking of words/ideas mappings so that they can in turn be used as constructs within larger intentional chunks, largely automated by your brain machinery, so that you don't even have to be aware of the syntax, only the semantics.
A world without words would be like navigating in an ocean without lighthouses and stars: anything is possible, but you never really know where you are and where anything else is (including others' minds)
Human-level reasoning requires something like language, for generality. Particularly, to reason about reasoning - to argue, to see the argument of another, to see their point of view, to trade disimilar things, to "see where you went wrong", to deceive.
It's analogous to: animals are FSA; we are TMs.
You can argue for the intelligence of other animals, but they can't argue for yours.
I have always had strong doubts about the latter position, and I do not think it can survive the challenge presented by the sort of cases presented in this article.
With knowledge by acquaintance or experience (including of pain), we grasp the concept before we have a name for it, and the name is an optional relationship. I am well aware of this because I have a small problem remembering names, but when it happens, it does not interfere in any other way in my knowledge of that person. When it happens, I have, indeed, temporarily forgotten a name that I once knew (or at least lost conscious access to it), but it seems plausible that most of what I still know about them could have been learned without ever knowing their name.
Language can supplement our experience in at least a couple of ways: we can learn some concepts that might, in principle, be learned by experience, but we have not actually done so (this is not quite the same, however, which might explain my yearning to visit places with an interesting history, even if everything from that time have crumbled to dust, and so, intellectually, I realize that such a visit will not lead to significant new insights.) Secondly, it is not clear to me that one could gain abstract concepts such as the square root of -1 or entropy by language-free experience alone, which I think is the thesis expressed in the article's title.
[1] Stanley, Jason and Timothy Williamson, 2001, “Knowing How”, Journal of Philosophy, 98(8): 411–444. doi:10.2307/2678403
Like, the whole of 3D space, physics, object permanence, etc are most definitely not learnt through language at all (though arguably, we might have some innate knowledge on some of these categories? I believe 3D objects reliably break down into the same basic 3D shapes in all humans)
You: "No. You can meet people and get to know them without learning their names."
I'm confused, as I can't see how your observation - insightful as it is - is supposed to contradict what I wrote.
Did you not read my comment in full, or is there something I am missing?
This distinction (between non-abstract and abstract entities) already becomes blurred when it comes to things like colors. In their case your quip starts to break down - contrary to our intuition, it turns out that you do need the word in order to experience the concept. To some extent at least.
It's been proven that - even though colors exist on physical level - languages (vocabulary) do have an impact on how humans perceive them.
Eg. "Researchers found that Greek and Russian speakers, who have dedicated words for light and dark blue, were more likely to see a light blue triangle against a dark blue background (and vice versa), than they were to identify green triangles against green backgrounds. Speakers of German, which has no such distinction, were no better at seeing shades of blue triangles than green."
I put this question to various cats, dogs, rats and rooks of my acquaintance. The answer was a unanimous though wordless yes.
If you are from some uncontacted tribe in the Amazon that were never exposed to any modern technology, you cannot think about computers, skyscrapers or microwave ovens because you would not have the memories of these things.
So it's seems to me that memory is the scaffold of the mind.
I think words themselves are kept somewhat stable due to being in overlapping memories, and thus protected from the brain cleanup process.
That seems too simple to be correct. How would you explain inventions? New ideas? Science fiction?
Imagine you are completely blind. The concept of colours would exist in your mind because you have access to literature mentioning it. You would know it's a thing, it exist, and yet, because you haven't been exposed to colors, you would have no memories of them of your own.
Now imagine if everybody in society was blind. Because nobody would have been exposed to colours, there would be no memories of them. It wouldn't be a thing as an idea in people's mind. As a result, there would be no way of thinking about the color red, or blue. It wouldn't be part of language either. Why describe something that we don't know that exist?
For broken communication to be fixed, there must be debugging. Discussion and laughter are essential.
Human reason is as ancient a capacity as language itself.
https://en.wikipedia.org/wiki/Linguistic_relativity