Language is not essential for the cognitive processes that underlie thought(scientificamerican.com) |
Language is not essential for the cognitive processes that underlie thought(scientificamerican.com) |
who cares right?
Most things we know, we are probably not aware of. And for most of us, direct experience of everything that surrounds us in the world certainly exceeds by several order of magnitude the best bandwidth we can ever dream to achieve through any human language.
Ok, there are no actual data to back this, but authors of the article don't have anything solid either to back such a bold statement, from what is presented in the article.
If most of what we know of the world would mostly be things we were told, it would obviously be mostly a large amount of phatic noises, lies and clueless random assertions that we would have no mean to distinguish from the few stable credible elements inferable by comparing with a far more larger corpus of self experiments with realty.
The claims here are exceptionally limited. You don't need spoken language to do well on cognitive tests, but that has never been a subject of debate. Obviously the deaf get on fine without spoken language. People suffering from aphasia, but still capable of communication via other mechanisms, still do well on cognitive tests. Brain scans show you can do sudoku without increasing bloodflow to language regions.
This kind of stuff has never really been in debate. You can teach plenty of animals to do fine on all sorts of cognitive tasks. There's never been a claim that language holds dominion over all forms of cognition in totality.
But if you want to discuss the themes present in Proust, you're going to be hard pressed to do so without something resembling language. This is self-evident. You cannot ask questions or give answers for subjects you lack the facilities to describe.
tl;dr: Language's purpose is thought, not all thoughts require language
Language's purpose - why it arose - is more likely communication, primarily external communication. The benefit of using language to communicate with yourself via "inner voice" - think in terms of words - seems a secondary benefit, especially considering that less than 50% of people report doing this.
But certainly language, especially when using a large vocabulary of abstract and specialist concepts, does boost cognitive abilities - maybe essentially through "chunking", using words as "thought macros", and boosting what we're able to do with our limited 7+/- item working memory.
For one, how would you know? It left no fossils, nor do we have any other kind of record from that time.
For another, the very question implies a teleological view of evolution, which is arguably wrong.
As for what 50% of people report (where did that number come from?), we have virtually zero intuitive insight into the inner workings of our minds in general, or of the way we process language. All the knowledge that has been obtained about how language works--linguistics--has been obtained by external observation of a black box. (FMRIs and the like provide a little insight inside that black box, but only at the most general level--and again, that's not intuition.)
Why the introduction of "spoken?" Sign languages are just as expressive as spoken language, and could easily be written. Writing is a sign.
> But if you want to discuss the themes present in Proust, you're going to be hard pressed to do so without something resembling language. This is self-evident.
And it's also a bad example. Of course you can't discuss the use of language without the use of language. You can't discuss the backstroke without any awareness of water or swimming, either. You can certainly do it without language though, just by waving your arms and jumping around.
> Language's purpose is thought
Is it, though? Did you make that case in the preceding paragraphs? I'm not going to go out on a limb here and alternatively suggest that language's purpose is communication, just like the purpose of laughing, crying, hugging, or smiling. This is why we normally do it loudly, or write it down where other people can see it.
We'd better hope that is true, because if we didn't have non-linguistic mastery of the cognitive processes underlying thought it's hard to see how we could even acquire language in the first place.
One must ask why this is such a common occurrence on this (and almost all other) social media, and conclude that it is because the structure of social media itself is rotten and imposes selective pressures that only allow certain kinds of content to thrive.
The actual paper itself is not readily accessible, and properly understanding its claims and conclusions would take substantial time and effort - by which point the article has already slid off the front page, and all the low-effort single-sentence karma grabbers who profit off of simplistic takes that appeal to majority groupthink have already occupied all the comment space "above the fold."
I think this may have been partially substantiated through experiments in decoding thoughts with machine sensors.
If this turns out to-not- to be true it would have huge implications for AI research.
Language is a very poor substitute for freely flowing electrical information - it is evolved to compensate for the bottlenecks to external communication - bottlenecks that are lacking an internal analogue.
It's also a highly advanced feature - something as heavily optiimised as evolved life would not allow something as vital as cognition to be hampered by a lack of means for high fidelity external expression.
But the most fundamental boon that it offered was in terms of planning and organisation. Before language we'd point and grunt and go there and do the thing that we were gesturing that we were going to do.
But that's a very crude form of planning - you're pretty much just going all in on Leeroy Jenkins.
But actually (and horrifically) I think it's the gift of Kane that speaking and planning permitted; well organised humans (the smartest things on the planet) have been figuring out increasingly better ways to both kill each other and not to die themselves in a brutal feedback loop for a very long time now.
It's brutal as fuck, but it's Darwinian gold.
Language is designed to be expressible with low fidelity vibrating strings - it is very clear that the available bandwidth is in the order of bytes per second.
Verses a fucking neural network with ~100 billion neurons.
Come on man, seriously - the two communication modalities are completely incomparable.
Also the title is editoralized for no reason. It makes searching, recognizing, citing etc waaay harder, and full of errors. I'll flag it.
No. But I'm going to stop there, because there are pages of comments saying the exact opposite (and of course some agreeing with you) above.
Base consciousness is surely not dependent on language, but I suspect base consciousness may be extremely different from what one might expect, so much that compared to what we perceive as consciousness, might seem something close to death.
And I'll hold to the notion that the complete absence of language (and its underlying structure) would resemble death if death can be resembled. Perhaps death is only the excoriation of thought, cognition and language, with something more fundamental persisting.
Note that human speech ability required more than brain support - it also required changes to the vocal apparatus for pronunciation (which other apes don't have), indicating that communication (vocalization) was either driving the development of language, or remained a very important part of it.
If people had no idea if they think with words or not, presumably they would say so.
It doesn't take words to understand implication of a club in your hand and a body of dead ape. From there it takes either violence or words to defend yourself(rightfully or not). Here, using language to explain the situation is more efficient.
Come on man, seriously - the two communication modalities are completely incomparable.
Clearly the information traveling around on the phone network couldn't possibly be the same as the low bandwidth vibrating strings used in face to face communication. Obviously.
The internal communications of the mind have no need for such constraints (and evolved hundreds of millions of years beforehand).
Anyway, I don't know what you were actually trying to argue here: you just built a simulated brain out of people, and the massively multi-agent distributed nature of the language of that machine is (emergently) incomparable with vocalised language.
This I think is why so much popular psychology is so vacuous - the slogans are merely things that triggered some people to figure out how to improve their mental actions, but there's no strong linkage between the two.
The one thing I wonder is if it's mostly "code duplication": iow, would we be able to develop language by using a different region of the brain, or do we actually do cognitive processes in the language part too?
In other words, is this simply deciding to send language processing to the GPU even if we could do it with the CPU (to illustrate my point)?
How would one even devise an experiment to prove or disprove this?
To me it seems obvious that our language generation and processing regions really involve cognition as well, as languages are very much rule based (even of they came up in reverse: first language then rules): could we get both regions to light up in brain imaging when we get to tricky words that we aren't sure how to spell or adapt to context like declensions of foreign words
> But you can build these models that are trained on only particular kinds of linguistic input or are trained on speech inputs as opposed to textual inputs.
As someone from this side of the "fence" (mathematics and CS, though currently obly a practicing software engineer), I don't think LLMs provide this opportunity that is in any way comparable to human minds.
Comparing performance of small kids developing their language skills (I've only had two, but one is enough to prove by contradiction) to LLMs (in particular for Serbian), LLMs like ChatGPT had a much broader vocabulary, but kids were much better at figuring out complex language rules with very limited number of inputs (noticed by them making mistakes on exceptions by following a "rule" at 2 years of age or younger).
The amount of training input GenAI needs is multiple orders of magnitude larger compared to young kids.
Though it's not a fair comparison: kids learn language by listening, immitation, watching, smelling, hearing and in context (you'll talk about bread at breakfast).
So let's be careful in considering LLMs a model of a human language process.
I don't know why Russell is catching strays. Saying language exists to make possible thoughts which could not exist without it does not in any way imply that you can't think without language.
Then comes the need to transmit/transfer understanding.
From the fine article:
> various properties that human languages have—there are about 7,000 of them spoken and signed across the world—are optimized for efficiently transmitting information, making things easy to perceive, easy to understand, easy to produce and easy to learn for kids.
Somewhat relatedly, I've started suspecting over the past few years that this is why I struggle to multitask or split my attention; while I can ruminate on several things at once, the "output" of my thinking is bottlenecked by a single stream that requires me to focus on exclusively to get a anything useful from it. Realizing this has actually helped me quite a bit in terms of being more productive because I can avoid setting myself up for failure by trying to get too much done at once and failing rather than tackling things one at a time.
IMO this rather reinforce Sapir-Whorf positions than refute, it means more than literal language/grammar influence thoughts. That's directly against UG theory that predetermined rigid grammar is all you need.
"Before my teacher came to me, I did not know that I am. I lived in a world that was a no-world. I cannot hope to describe adequately that unconscious, yet conscious time of nothingness. I did not know that I knew aught, or that I lived or acted or desired. I had neither will nor intellect. I was carried along to objects and acts by a certain blind natural impetus. I had a mind which caused me to feel anger, satisfaction, desire. These two facts led those about me to suppose that I willed and thought. I can remember all this, not because I knew that it was so, but because I have tactual memory. It enables me to remember that I never contracted my forehead in the act of thinking. I never viewed anything beforehand or chose it. I also recall tactually the fact that never in a start of the body or a heart-beat did I feel that I loved or cared for anything. My inner life, then, was a blank without past, present, or future, without hope or anticipation, without wonder or joy or faith.
It was not night—it was not day.
. . . . .
But vacancy absorbing space, And fixedness, without a place; There were no stars—no earth—no time— No check—no change—no good—no crime.
My dormant being had no idea of God or immortality, no fear of death.
I remember, also through touch, that I had a power of association. I felt tactual jars like the stamp of a foot, the opening of a window or its closing, the slam of a door. After repeatedly smelling rain and feeling the discomfort of wetness, I acted like those about me: I ran to shut the window. But that was not thought in any sense. It was the same kind of association that makes animals take shelter from the rain. From the same instinct of aping others, I folded the clothes that came from the laundry, and put mine away, fed the turkeys, sewed bead-eyes on my doll's face, and did many other things of which I have the tactual remembrance. When I wanted anything I liked,—ice-cream, for instance, of which I was very fond,—I had a delicious taste on my tongue (which, by the way, I never have now), and in my hand I felt the turning of the freezer. I made the sign, and my mother knew I wanted ice-cream. I "thought" and desired in my fingers. If I had made a man, I should certainly have put the brain and soul in his finger-tips. From reminiscences like these I conclude that it is the opening of the two faculties, freedom of will, or choice, and rationality, or the power of thinking from one thing to another, which makes it possible to come into being first as a child, afterwards as a man.
Since I had no power of thought, I did not compare one mental state with another. So I was not conscious of any change or process going on in my brain when my teacher began to instruct me. I merely felt keen delight in obtaining more easily what I wanted by means of the finger motions she taught me. I thought only of objects, and only objects I wanted. It was the turning of the freezer on a larger scale. When I learned the meaning of "I" and "me" and found that I was something, I began to think. Then consciousness first existed for me. Thus it was not the sense of touch that brought me knowledge. It was the awakening of my soul that first rendered my senses their value, their cognizance of objects, names, qualities, and properties. Thought made me conscious of love, joy, and all the emotions. I was eager to know, then to understand, afterward to reflect on what I knew and understood, and the blind impetus, which had before driven me hither and thither at the dictates of my sensations, vanished forever.
I cannot represent more clearly than any one else the gradual and subtle changes from first impressions to abstract ideas. But I know that my physical ideas, that is, ideas derived from material objects, appear to me first an idea similar to those of touch. Instantly they pass into intellectual meanings. Afterward the meaning finds expression in what is called "inner speech." When I was a child, my inner speech was inner spelling. Although I am even now frequently caught spelling to myself on my fingers, yet I talk to myself, too, with my lips, and it is true that when I first learned to speak, my mind discarded the finger-symbols and began to articulate. However, when I try to recall what some one has said to me, I am conscious of a hand spelling into mine.
It has often been asked what were my earliest impressions of the world in which I found myself. But one who thinks at all of his first impressions knows what a riddle this is. Our impressions grow and change unnoticed, so that what we suppose we thought as children may be quite different from what we actually experienced in our childhood. I only know that after my education began the world which came within my reach was all alive. I spelled to my blocks and my dogs. I sympathized with plants when the flowers were picked, because I thought it hurt them, and that they grieved for their lost blossoms. It was two years before I could be made to believe that my dogs did not understand what I said, and I always apologized to them when I ran into or stepped on them.
As my experiences broadened and deepened, the indeterminate, poetic feelings of childhood began to fix themselves in definite thoughts. Nature—the world I could touch—was folded and filled with myself. I am inclined to believe those philosophers who declare that we know nothing but our own feelings and ideas. With a little ingenious reasoning one may see in the material world simply a mirror, an image of permanent mental sensations. In either sphere self-knowledge is the condition and the limit of our consciousness. That is why, perhaps, many people know so little about what is beyond their short range of experience. They look within themselves—and find nothing! Therefore they conclude that there is nothing outside themselves, either.
However that may be, I came later to look for an image of my emotions and sensations in others. I had to learn the outward signs of inward feelings. The start of fear, the suppressed, controlled tensity of pain, the beat of happy muscles in others, had to be perceived and compared with my own experiences before I could trace them back to the intangible soul of another. Groping, uncertain, I at last found my identity, and after seeing my thoughts and feelings repeated in others, I gradually constructed my world of men and of God. As I read and study, I find that this is what the rest of the race has done. Man looks within himself and in time finds the measure and the meaning of the universe."
What if the things are part of a set, chosen for uniqueness and distinguishability. Meanings determined by tradition?
There's a lot of territory between the two.
His work explores the neuropsychology of emotions WAIT DON'T GO they are actually the substrate of consciousness, NOT the other way around.
We have 7 primary affective processes (measurable hardware level emotions) and they are not what you think[2]. They are considered primary because they are sublinguistic. For instance, witnessing the color red is a primary experience, you cannot explain in words the color red to someone who has not ever seen it before.
His work is a really fascinating read if you ever want to take a break from puters for a minute and learn how people work.
PS the reason this sort of research isn't more widely known is because the behaviorist school was so incredibly dominant since the 1970s they made it completely taboo to discuss subjective experience in the realm of scientific discourse. In fact the emotions we are usually taught are not based on emotional states but on muscle contractions in the face! Not being allowed to talk about emotions in psychological studies or the inner process of the mind is kinda crazy when you think about it. So only recently with neuroimaging has it suddenly become ok to acknowledge that things happen in the brain independent of externally observable behavior.
[2] - seeking - fear - anxiety and grief - rage - lust - play!!! - caring
[3] if this sounds familiar at all it's because Jordan Peterson cites Jaak Panksep all the time. Well 50% of the time, the other 50% is CG Jung and the final 50% is the book of Exodus for some reason.
Another related tool is religion (for emotions instead of thoughts,) which funnily enough faces the same divergence language does.
Right now society that calls itself "secular" simply does not understand the role of religion, and its importance in society.
To be clear, I don't belong to any religion, I am saying one needs to be invented for people who are currently "secular."
In fact, you have the disorganized aspects of religion already. All one needs to spot these are to look at the aspects that attempt to systematize or control our feelings. Mass media, celebrities for example.
Instead of letting capitalistic forces create a pseudoreligion for society, it's better if people come together and organize something healthier, intentionally.
Thinking some type of materialism is even mostly correct, with the sum over all mostly materialist theories being close to 1, isn't a religion at all.
Serializing much higher dimensional freeform thoughts into language is a very lossy process, and this kinda ensures that mostly only the core bits get translated. Think of times when someone gets an idea you're trying to convey, but you realize they're missing some critical context you forgot to share. It takes some activation energy to add that bit of context, so if it seems like they mostly get what you're saying, you skip it. Over time, transferring ideas from one person to the next, they tend towards a very compressed form because language is expensive.
This process also works on your own thoughts. Thinking out loud performs a similar role, it compresses the hell out of the thought or else it remains inexpressible. Now imagine repeated stages of compressing through language, allowing ideas to form from that compressed form, and then compressing those ideas in turn. It's a bit of a recursive process and language is in the middle of it.
“You cannot ask a question you that you have no words for”
- Judea Pearl
I think that using a LLM as the referred telepathy device to a wolfram-alpha/mathematica like general reasoning module is the way to AGI. The reasoning modules we have today are still much to narrow because of the very broad and deep search trees exploding in complexity. There is the need for a kind of pathfinder which could come from common knowledge already encoded in LLMs, like in o1. An system playing with real factual reasoning but exploring in directions coming from world knowledge.
What is still missing is the dialectic between possible and right, a physics engine, the motivation of analysed agents, the effects of emergent behavior and a lot of other -isms. But they may be encoded in the reasoning-explorer. And of course loops, more loops, refinement, working hypotheses and escaping cul-de-sacs.
There are people with great language skills and next to no reasoning skills. Some of them have general knowledge. If you ever talked to them, for a at least an hour freely meandering topics you will know. They seem intelligent for a couple of minutes but after a while you realise that they can refer fact, even interpret metaphors, but they will not find an elegant one, to navigate abstraction levels, even to differentiate root cause from effect or motivation and culture from cold logic. Some of them even ace IQ or can program but none did math so far. They hate, fear or despise rational results violation their learned rules. Sorry, chances are if you hate reading this, maybe you are one (or my English is annoyingly bad).
I love talking to people outside my bubble. They have an incredible broad diversity in abilities and experiences.
Several factors contribute to the unfalsifiability of this claim:
Subjectivity of Thought: Thought processes are inherently internal and subjective. There is no direct method to observe or measure another being's thoughts without imposing interpretative frameworks influenced by social and material contexts.
Defining Language and Thought: Language is not merely a collection of spoken or written symbols; it is a system of signs embedded within social relations and power structures. If we broaden the definition of language to include any form of symbolic representation or communication—such as gestures, images, or neural patterns—then the notion of thought occurring without language becomes conceptually incoherent. Thought is mediated through these symbols, which are products of historical and material developments.
Animal Cognition and Symbolic Systems: Observations of animals like chimpanzees engaging in strategic gameplay or crows crafting tools demonstrate complex behaviors. Interpreting these actions as evidence of thought devoid of language overlooks the possibility that animals utilize their own symbolic systems. These behaviors reflect interactions with their environment mediated by innate or socially learned symbols—a rudimentary form of language shaped by their material conditions.
Limitations of Empirical Testing: To empirically verify that thought can occur without any form of language would require accessing cognitive processes entirely free from symbolic mediation. Given the current state of scientific methodologies—and considering that all cognitive processes are influenced by material and social factors—this is unattainable.
Because of these factors, Stix's claim cannot be empirically tested in a way that could potentially falsify it. It resides outside the parameters of verifiable inquiry, highlighting the importance of recognizing the interplay between language, thought, and material conditions.
Cognitive processes and language are deeply intertwined. Language arises from collective practice; it both shapes and is shaped by the material conditions of the environment. Thought is mediated through language, carrying the cognitive imprints of the material base. Even in non-human animals, the cognitive abilities we observe may be underpinned by forms of symbolic interaction with their environment—a reflection of their material engagement with the world.
Asserting that language is not essential for thought overlooks the fundamental role that social and material conditions play in shaping both language and cognition. It fails to account for how symbolic systems—integral to language—are embedded in and arise from material realities.
Certain forms of thought might appear to occur without human language, but this perspective neglects the intrinsic connection between cognition, language, and environmental conditiond. Reasoning itself can be viewed as a form of internalized language—a symbolic system rooted in social and material contexts. Recognizing this interdependence is crucial for a comprehensive understanding of the nature of thought and the pivotal role language plays within it.
Or vice versa?
I don't use an inner monologue but my imagination is fairly good at creating new images.
If I had time and free use of an LLM, I'd like to investigate how well it understands constructional synonymy, like "the red car" and "the car that is red" and "John saw a car on the street yesterday. It was red." I guess models that can draw pictures can be used to test this sort of thing--surely someone has looked into this?
> this kinda ensures that mostly only the core bits get translated
The kinda is doing a lot here. Many times the very act of trying to communicate a thought colors/corrupts the main point and gives only one perspective or a snapshot of the overall thought. There's a reason why they say a picture is worth a thousand words. Except the mind can conjure much more than a static picture. The mind can also hold the idea and the exceptions to the idea in one coherent model. For me this can be especially apparent when taking psychedelics and finding that trying to communicate some thoughts with words requires constant babbling to keep refining the last few sentences, ad libidum. There are exceptions of course, like for simple ideas.
Yeah! Sometimes the thought isnt compressible and language doesnt help. But a lot of times it is, and it does
I spent the next few days trying to understand how that process worked. I would force myself to think in words and sentences. It was incredibly limiting! So slow and lacking in images, in abstract relationships between ideas and sensations.
It took me another few years to realise that many people actually depend on those structures in order to produce any thought and idea.
Also, many people simply repeat facts they were told. "We need words to think" is simply a phrase this person learned, a fact to recite in school settings. It doesn't mean they deeply reflected on this statement or compared it with their experience.
Try it now: Tap your hand on the desk randomly. Can you recall how many times you did it without "saying" a sequence in your head like "1, 2, 3" or "A, B, C" etc?
If yes, how far can you count? With a language it's effectively infinite. You could theoretically go up to "1 million 5 hundred 43 thousand, 2 hundred and 10" and effortlessly know what comes next.
For context I have both abstract "multimedia" thought processes and hypervisor-like internal narrative depending on the nature of the experience or task.
If I want to translate this knowledge into a number, I need to count the taps I am seeing in my head. At that point I do need to think of the word for the number.
I could even do computations on these items in my mind, imagine dividing them into two groups for instance, without ever having to link them to words until I am ready to do something with the result, such as write down the number of items in each group.
An example of this would be when I’m lifting weights with a friend and am lost in the set/focusing on mind-muscle connection, and as a result I forget to count my reps. I am usually quite accurate when I verify with my lifting partner the number of reps done/remaining.
As OP mentioned, many people have no internal speech, otherwise known as anendophasia, yet can still do everything anyone with an internal dialogue can do.
Similarly for me, I can do “mental object rotation” tasks even though I have aphantasia.
I would note though I have a really difficult time with arithmetic and mechanical tasks like counting. Mostly I just lose attention. Perhaps an inner voice would help if it became something that kept a continuity of thought.
This is a parallel stream, because if I count with imagined pictures, then I can speak and listen to someone talking without it disturbing the process. If I do it with subvocalization, then doing other speech/language related things would disturb the counting.
Yes. Seriously, these kind of questions are so surprising. It tells you that everyone's experience is just a little different.
It's the equivalent of <thinking> tags for LLM output.
Other animals with at best very limited language, are still highly intelligent and capable of reasoning - apes, dogs, rats, crows, ...
Just as Gödel showed that no formal system can be both complete and consistent, language as a system cannot fully encapsulate the entirety of cognitive processes without relying on foundational assumptions that it cannot internally validate. Attempting to describe thought without acknowledging this limitation is akin to seeking completeness in an inherently incomplete framework. Without language, the discussion becomes impossible, rendering the initial claim fundamentally flawed.
This is known as subitising.
...maybe I do this sometimes myself. Remembering the proper names of things is effort.
X . . X . . X . . . X . X . . .
X . . X . . X . X . . . X . . .
X . . . X . . . . . X . X . . .
X . X X X . X . X X X .
A . . A . . A . . A . .
B . B . B . B . B . B .
X . . X . X X X . X X . X . X X . . X . X X . . X X . X X . X . . X . X X . . X X . X . . X . . X X X X . . X X X X . . X . . X . X X . . X X . X . . X . X X . X X . . X X . X . . X X . X . X X . X X X . X . .
A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . . A . . . .
B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . . B . . . . . .
C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . . C . .
https://www.sciencealert.com/theres-a-big-difference-in-how-...
We recently put the project I've been working on for the last year out into the field for the first time. As was fully expected, some bugs emerged. I needed to solve one of them. I designed a system in my head for spawning off child processes based on the parent process to do certain distinct types of work in a way that gives us access to OS process-level controls over the work, and then got about halfway through implementing it. Little to none of this design involved "words". I can't even say it involved much "visualization" either, except maybe in a very loose sense. It's hard to describe in words how I didn't use words but I've been programming for long enough that I pretty much just directly work in system-architecture space for such designs, especially relatively small ones like that that are just a couple day's work.
Things like pattern language advocates aren't wrong that it can still be useful to put such things into words, especially for communication purposes, but I know through direct personal experience that words are not a necessary component of even quite complicated thought.
"Subjective experience reports are always tricky, jerf. How do you know that you aren't fooling yourself about not using words?" A good and reasonable question, to which my answer is, I don't even have words for the sort of design I was doing. Some, from the aforementioned pattern languages, yes, but not in general. So I don't think I was just fooling myself on the grounds that even if I tried to serialize what I did directly into English, a transliteration rather than a translation, I don't think I could. I don't have one.
I'm also not claiming to be special. I don't know the percentages but I'm sure many people do this too.
If you stand outside under the sun, do you have to be able to write the word "sun" in order to feel warm?
Since we all have language and opinions about it, the risk of genericness is high with a title like this. It's like this with threads about other universal topics too, such as food or health.
The actual paper [1] says that functional MRI (which is measuring which parts of the brain are active by sensing blood flow) indicates that different brain hardware is used for non-language and language functions. This has been suspected for years, but now there's an experimental result.
What this tells us for AI is that we need something else besides LLMs. It's not clear what that something else is. But, as the paper mentions, the low-end mammals and the corvids lack language but have some substantial problem-solving capability. That's seen down at squirrel and crow size, where the brains are tiny. So if someone figures out to do this, it will probably take less hardware than an LLM.
This is the next big piece we need for AI. No idea how to do this, but it's the right question to work on.
[1] https://www.nature.com/articles/s41586-024-07522-w.epdf?shar...
When the first chess engines came out they only employed one of these: calculation. It wasn't until relatively recently that we had computer programs that could perform all of them. But it turns out that if you scale that up with enough compute you can achieve superhuman results with calculation alone.
It's not clear to me that LLMs sufficiently scaled won't achieve superhuman performance on general cognitive tasks even if there are things humans do which they can't.
The other thing I'd point out is that all language is essentially synthetic training data. Humans invented language as a way to transfer their internal thought processes to other humans. It makes sense that the process of thinking and the process of translating those thoughts into and out of language would be distinct.
After all, that's what Artificial General Intelligence would at least in part be about: finding and proving new math theorems, creating new poetry, making new scientific discoveries, etc.
There is even a new challenge that's been proposed: https://arcprize.org/blog/launch
> It makes sense that the process of thinking and the process of translating those thoughts into and out of language would be distinct
Yes, indeed. And LLMs seem to be very good at _simulating_ the translation of thought into language. They don't actually do it, at least not like humans do.
If "general cognitive tasks" means "I give you a prompt in some form, and you give me an incredible response of some form " (forms may differ or be the same) then it is hard to disagree with you.
But if by "general cognitive task" you mean "all the cognitive things that human do", then it is really hard to see why you would have any confidence that LLMs have any hope of achieving superhuman performance at these things.
To some extent this is true.
To calculate A + B you could for example generate A, B for trillions of combinations and encode that within the network. And it would calculate this faster than any human could.
But that's not intelligence. And Apple's research showed that LLMs are simply inferring relationships based on the tokens it has access to. Which you can throw off by adding useless information or trying to abstract A + B.
Solving puzzles is a specific cognitive task, not a general one.
Language is a continuum, not a puzzle. The problem with LLMs is that testing has been reduced to performance on language puzzles, mostly with hard edges - like bar exams, or letter counting - and they're a small subset of general language use.
When it comes to general intelligence, I think we are trying to run before we can walk. We can't even make a computer with a basic, animal level understanding of the world. Yet we are trying to take a tool that was developed on top of system that already had an understanding of the world and use it to work backwards to give computers an understanding of the world.
I'm pretty skeptical that we're going to succeed at this. I think you have to be able to teach a computer to climb a tree or hunt (subhuman AGI) before you can create superhuman AGI.
https://arstechnica.com/ai/2024/10/llms-cant-perform-genuine...
or do you maybe think no logical reasoning is needed to do everything a human can do? Tho humans seem to be able to do logical reasoning
Then LLMs came along, and ML folks got rather too excited that they contain implicit knowledge (which, of course, is required to deal with ambiguity). Then the new aspiration as "all in one" and "bigger is better", not analyzing what components are needed and how to orchestrate their interplay.
From an engineering (rather than science) point of view, the "end-to-end black box" approach is perhaps misguided, because the result will be a non-transparent system by definition. Individual sub-models should be connected in a way that retains control (e.g. in dialog agents, SRI's Open Agent Architecture was a random example of such "glue" to tie components together, to name but one).
Regarding the science, I do believe language adds to the power of thinking; while (other) animals can of course solve simple problems without language, language permits us to define layers of abstractions (by defining and sharing new concepts) that goes beyond simple, non-linguistic thoughts. Programming languages (created by us humans somewhat in the image of human language) and the language of mathematics are two examples where we push this even further (beyond the definition of new named concepts, to also define new "DSL" syntax) - but all of these could not come into beying without human language: all formal specs and all axioms are ultimately and can only be formulated in human language. So without language, we would likely be stuck at a very simple point of development, individually and collectively.
EDIT: 2 typos fixed
Based on my experience with toddlers, a rather smart dog, and my own thought processes, I disagree that language is a fundamental component of abstraction. Of sharing abstractions, sure, but not developing them.
When I'm designing a software system I will have a mental conception of the system as layered abstractions before I have a name for any component. I invent names for these components in order to define them in the code or communicate them to other engineers, but the intuition for the abstraction comes first. This is why "naming things" is one of the hard problems in computer science—because the name comes second as a usually-inadequate attempt to capture the abstraction in language.
In my personal learning journey I have been exploring the space of intuitive learning which is dominant in physical skills. Singing requires extremely precise control of actions we can't fully articulate or even rationalise. Teaching those skills requires metaphors and visualising and a whole lot of feedback + trial & error.
I believe that this kind of learning is fundamentally non verbal and we can achieve abstraction of these skills without language. Walking is the most universal of these skills and we learn it before we can speak but if you study it (or better try to program a robot to walk with as many degrees of freedom as the human musculoskeletal system) you will discover that almost all of us don't understand what all the things that go into the "simple" task of walking!
My understanding is that people who are gifted at sports or other physical skills like musical instruments have developed the ability to discover and embed these non verbal abstractions quickly. When I practise the piano and am working on something fast, playing semiquavers at anything above 120bpm is not really conscious anymore in the sense of "press this key then that key"
The concept of arpeggio is verbal but the action is non verbal. In human thought where does verbal and non-verbal start and end? Its probably a continuum
A black box that works in human language and can be investigated with perturbations, embedding visualizations and probes. It explains itself as much ore more than we can.
Not to over-hype LLMs, but I don't see why this results says this. AI doesn't need to do things the same way as evolved intelligence has.
Open AI O1 seems to be trained on mostly synthetic data, but it makes intuitive sense that LLMs work so well because we had the data lying around already.
Similar reason we look for markers of Earth-based life on alien planets: it's the only example we've got of it existing.
An Ab Initio AGI would maybe be free of our legacy, but LLMs certainly are not.
I would expect a ship-like intelligence a la the Culture novels to have non-English based cognition. As far as we can tell, our own language generation is post-hoc explanation for thought more so than the embodiment of thought.
for more, see "Assembly Theory"
LLMs basically become practical when you simply scale compute up, and maybe both regions are "general compute", but language ends up on the "GPU" out of pure necessity.
So to me, these are entirely distinct questions: is the language region able to do general cognitive operations? What happens when you need to spell out "ubiquitous" or declense a foreign word in a language with declension (which you don't have memory patterns for)?
I agree it seems obvious that for better efficiency (size of training data, parameter count, compute ability), human brains use different approach than LLMs today (in a sibling comment, I bring up an example of my kids at 2yo having a better grasp of language rules than ChatGPT with 100x more training data).
But let's dive deeper in understanding what each of these regions can do before we decide to compare to or apply stuff from AI/CS.
No this is not true. For two reasons.
1. We call these things LLMs and we train it with language but we can also train it with images.
2. We also know LLMs develop a sort of understanding that goes beyond language EVEN when the medium used for training is exclusively language.
The naming of LLMs is throwing you off. You can call it a Large Language Model but this does not mean that everything about LLMs are exclusively tied only to language.
Additionally we don't even know if the LLM is even remotely similar to the way human brains process language.
No such conclusion can be drawn from this experiment.
A crow has a small brain, but also has very small neurons, so ends up having 1.5B neurons, similar to a dog or some monkeys.
The absence of both of these things is an incredible crippler for technological development. It doesn't matter how intelligent you are, you're never going to achieve much technologically without these.
I don't think brain size correlations is as straightforward as 'bigger = better' every time but we simply don't know how intelligent most of these species are. Land and Water are completely different beasts.
And it turns out that human brain volume and intelligence are moderately-highly correlated [1][2]!
[1]: https://pmc.ncbi.nlm.nih.gov/articles/PMC7440690/ [2]: https://www.sciencedirect.com/science/article/abs/pii/S01602...
https://www.scientificamerican.com/article/gut-second-brain/
There are 100 million in my gut, but it doesn't solve any problems that aren't about poop, as far as I know.
https://en.wikipedia.org/wiki/List_of_animals_by_number_of_n...
If the suspiciously round number is accurate, this puts the human gut somewhere between a golden hamster and ansell's mole-rat, and about level with a short-palated fruit bat.
On the other hand, further understanding how to engage complex cognitive processes in nonverbal individuals is extremely useful and difficult to accomplish.
All intelligence is the mitigation of uncertainty (the potential distributed problem.) if it does not mitigate uncertainty it is not intelligence, it is something else.
Intelligence is a technology. For all life intelligence and the infrastructure of performing work efficiently (that whole entropy thing again) is a technology. Life is an arms race to maintain continuity (identity, and the very capacity of existential being.)
The modern problem is achieving reliable behavioral intelligence (constrained to a specific problem domain.) AGI is a phantasm. What manifestation of intelligence appears whole and complete and is always right? These are the sorts of lies you tell yourself, the ones that get you into trouble. They distract from tangible real world problems, perhaps causing some of them. True intelligence is a well calibrated “scalar” domain specific problem (uncertainty) reducer. There are few pressing idempotent obstructions in the real world.
Intelligence is the mitigation of uncertainty.
Uncertainty is the domain of negative potential (what,where,why,how?)
Mitigation is the determinant resolve of any constructive or destructive interference affecting (terminal resolve within) the problem domain.
Examples of this may be piled together mountains high, and you may call that functional AGI, though you would be self deceiving. At some point “good enough” may be declared for anything so passing as yourselves.
Basically we need Multimodal LLM's (terrible naming as it's not an LLM then but still).
There's been progress. Look at this 2020 work on neural net controlled drone acrobatics.[1] That's going in the right direction.
I’d be extremely surprised if AI recapitulates the same developmental path as humans did; evolution vs. next-token prediction on an existing corpus are completely different objective functions and loss landscapes.
I then looked it up and they had each copy/pasted the same Stack overflow answer.
Furthermore, the answer was extremely wrong, the language I used was superficially similar to the source material, but the programming concepts were entirely different.
What this tells me is there is clearly no “reasoning” happening whatsoever with either model, despite marketing claiming as such.
Stepping back a level, it may only actually tell us that MRIs measure blood flow.
> Recent work has revealed that the neural activity patterns correlated with sensation, cognition, and action often are not stable and instead undergo large scale changes over days and weeks—a phenomenon called representational drift.
[...]
So, I'm not sure how conclusive this fmri activation study is either.
Though, is there a proto language that's not even necessary for the given measured aspects of condition?
Which artificial network architecture best approximates which functionally specialized biological neutral networks?
OpenCogPrime:KnowledgeRepresentation > Four Types of Knowledge: https://wiki.opencog.org/w/OpenCogPrime:KnowledgeRepresentat... :
> Sensory, Procedural, Episodic, Declarative
From https://news.ycombinator.com/item?id=40105068#40107537 re: cognitive hierarchy and specialization :
> But FWIU none of these models of cognitive hierarchy or instruction are informed by newer developments in topological study of neural connectivity;
ultimately, there's no reason that a general algorithm couldn't do the job of a specific one, just less efficiently.
An easy conclusion to jump to but I believe we need to be more careful. Nothing in these findings proves conclusively that non-verbal reasoning mechanism equivalent to humans couldn't evolve in some part of a sufficiently large ANN trained on text and math. Even though verbal and non-verbal reasoning occurs in two distinct parts of the brain, it doesn't mean they're not related.
You mean besides a few layers of LLMs near input and output that deal with tokens? We have the rest of the layers.
1. Syntax
2. Semantics
3. Pragmatics
4. Semiotics
These are the layers you need to solve.
Saussure already pointed out these issues over a century ago, and Linguists turned ML Researchers like Stuart Russell and Paul Smolensky tried in vain to resolve this.
It basically took 60 years just to crack syntax at scale, and the other layers are still fairly far away.
Furthermore, Syntax is not a solved problem yet in most languages.
Try communicating with GPT-4o in colloquial Bhojpuri, Koshur, or Dogri, let alone much less represented languages and dialects.
Higher order faculties aside, animals seem like us, just simpler.
The higher functioning ones appear to have this missing thing too. We can see it in action. Perhaps all of them do and it is just harder for us when the animal thinks very differently or maybe does not think as much, feeling more, for example.
----
Now, about that thing... and the controversy:
Given an organism, or machine for this discussion, is of sufficiently robust design and complexity that it can precisely differentiate itself from everything else, it is a being.
This thing we are missing is an emergent property, or artifact that can or maybe always does present when a state of being also presents.
We have not created a machine of this degree yet.
Mother nature has.
The reason for emergence is a being can differentiate sensory input as being from within, such as pain, or touch, and from without, such as light or motion.
Another way to express this is closed loop vs open loop.
A being is a closed loop system. It can experience cause and effect. It can be the cause. It can be the effect.
A lot comes from this closed loop.
There can be the concept of the self and it has real meaning due to the being knowing what is of itself or something, everything else.
This may be what forms consciousness. Consciousness may require a closed loop, and organism of sufficient complexity to be able to perceive itself.
That is the gist of it.
These systems we make are fantastic pieces. They can pattern match and identify relationships between the data given in amazing ways.
But they are open loop. They are not beings. They cannot determine what is part of them, what they even are,or anything really.
I am both consistently amazed and dismayed at what we can get LLM systems to do.
They are tantalizingly close!
We found a piece of how all this works and we are exploiting the cral out of it. Ok fine. Humans are really good at that.
But it will all taper off. There are real limits because we will eventually find the end goal will be to map out the whole problem space.
Who has tried computing that? It is basically all possible human thought. Not going to happen.
More is needed.
And that "more" can arrive at thoughts without having first seen a few bazillion to choose from.
An example was the problem of memory shared between systems. ML people started doing LLM’s with RAG. I looked into neuroscience which suggested we need a hippocampus model. I found several papers with hippocampus-like models. Combining LLM’s, vision, etc with hippocampus-like model might get better results. Rinse repeat for these other brain areas wherever we can understand them.
I also agree on testing the architectures with small, animal brains. Many do impressive behaviors that we should be able to recreate in simulators or with robotics. Some are useful, too, like how geese are good at security. Maybe embed a trained, goose brain into a camera system.
I am not convinced it follows. Sure LLMs don’t seem complete however there’s a lot of unspoken inference going on in LLMs that don’t map into a language directly already - the inner layers of the deep neural net that operates on abstract neurons.
Perhaps, but the relative success of trained LLMs acting with apparent generalised understanding may indicate that it is simply the interface that is really an LLM post training;
That the deeper into the network you go (the further from the linguistic context), the less things become about words and linguist structure specifically and the more it becomes about things and relations in general.
(This also means that multiple interfaces can be integrated, sometimes making translation possible, e.g.: image <=> tree<string>)
We have, it's called DreamCoder. There's a paper and everything.
Everything needed for AGI exists today, people simply have (incorrect) legacy beliefs about cognition that are holding them back (e.g. "humans are rational").
Despite being an LLM skeptic of sorts, I don’t think that necessarily follows. The LLM matrix multiplication machinery may well be implementing an equivalent of the human non-language cognitive processing as a side effect of the training. Meaning, what is separated in the human brain may be mixed together in an LLM.
You would think the whole "split-brain" thing would have been the first clue; apparently not.
We need to add the 5 senses, of which we have now image, audio and video understanding in LLMs. And for agentic behavior they need environments and social exposure.
I think this is completely wrong-headed. It's like saying that until cars came about we just didn't have anything other than animals that could move around under its own power, therefore in order to understand how animals move around we should go and study cars. There is a great gulf of unsubstantiated assumptions between observing the behaviour of a technological artifact, like a car or a statistical language model, and thinking we can learn something useful from it about human or more generally animal faculties.
I am really taken aback that this is a serious suggestion: study large language models as in-silico models of human linguistic ability. Just putting it down in writing like that rings alarm bells all over the place.
I could enter what we all here call the "Zone" quite often when i was young (once while doing math :D). I still can, but rarely on purpose, and rarely while coding. I have a lot of experience in this state, and i can clearly say that a marker of entering the zone is that your thoughts are not "limited" by language anymore and the impression of clarity and really fast thinking. This is why i never thought that language was required for thinking.
Now the question: would it be possible to scan the brain of people while they enter the zone? I know it isn't a state you can reach on command, but isn't it worth to try? understand the mechanism of this state? And maybe understand where our thought start?
That is, until the code refuses to work. Then the code is a bitch and I need a break.
> Language serves not only to express thoughts, but to make possible thoughts which could not exist without it. It is sometimes maintained that there can be no thought without language, but to this view I cannot assent: I hold that there can be thought, and even true and false belief, without language. But however that may be, it cannot be denied that all fairly elaborate thoughts require words.
> Human Knowledge: Its Scope and Limits by Bertrand Russell, Section: Part II: Language, Chapter I: The Uses of Language Quote Page 60, Simon and Schuster, New York.
Of course, that would contravene the popular narrative that philosophers are pompous idiots incapable of subtlety.
Practically, I think the origins of fire-making abilities in humans tend to undermine that viewpoint. No other species is capable of reliably starting a fire with a few simple tools, yet the earliest archaeological evidence for fire (1 mya) could mean the ability predated complex linguistic capabilities. Observation and imitation could be enough for transmitting the skill from the first proto-human who successfully accomplished the task to others.
P.S. This is also why Homo sapiens should be renamed Homo ignis IMO.
It’s doubtless to me that thinking happens without intermediary symbols; but it’s also obvious that I can’t think deeply without the waypoints and context symbols provide. I think it is a common sense opinion.
Just a few days ago was "What do you visualize while programming?", and there's a few of us in the comments that, when programming, think symbolically without language: https://news.ycombinator.com/item?id=41869237
> You can ask whether people who have these severe language impairments can perform tasks that require thinking. You can ask them to solve some math problems or to perform a social reasoning test, and all of the instructions, of course, have to be nonverbal because they can’t understand linguistic information anymore.
Surely these "non-verbal instructions" are some kind of language. Maybe all human action can be considered language.
A contrarian example to this research might be feral children, i.e people who have been raised away from humans.[0] In most cases they are mentally impaired; as in not having human-like intelligence. I don't think there is a good explanation why this happens to humans. And why it doesn't happen to other animals, which develop normally in species-typical way whether they are in the wild or in human captivity. It seems that most human behavior (even high-level intelligence) is learned / copied from other humans, and maybe this copied behavior can be considered language.
If humans are "copy machines", there's also a risk of completely losing the "what's it like to be a human" behavior if children of the future are raised by AI and algorithmic feeds.
> DA was impaired in solving simple addition, subtraction, division or multiplication problems, but could correctly simplify abstract expressions such as (b×a)÷(a×b) or (a+b)+(b+a) and make correct judgements whether abstract algebraic equations like b − a = a − b or (d÷c)+a=(d+a)÷(c+a) were true or false.
> Sensitivity to the structural properties of numerical expressions was also evaluated with bracket problems, some requiring the computation of a set of expressions with embedded brackets: for example, 90 [(3 17) 3].
Discussions of whether or not these sorts of algebraic or numerical expressions constitute a "language of mathematics" aside (despite them not engaging the same brain regions and structures associated with the word "language"); it may be the case that these sorts of word sequences and symbols processed by structures in the brain's left hemisphere are not essential for thought, but can still serve as a useful psychotechnology or "bicycle of the mind" to accelerate and leverage its innate capabilities. In a similar fashion to how this sort of mathematical notation allows for more concise and precise expression of mathematical objects (contrast "the number that is thrice of three and seventeen less of ninety") and serves to amplify our mathematical capacities, language can perhaps be seen as a force multiplier; I have doubts whether those suffering from aphasia or an agrammatic condition would be able to rise to the heights of cognitive performance.
If you look up 'mentalese' you can find a bunch written about it. There's an in-depth article by Daniel Gregory and Peter Langland-Hassan, in the incredible Stanford Encyclopedia of Philosophy, on Inner Speech (admittedly, I'm taking a leap to think they mean precisely the same thing). [2]
[0] Steven Pinker, The Blank Slate: The Modern Denial of Human Nature (2002)
[1] Oxford English Dictionary
> In the stage of Cause and Effect, the relationships between mental and physical phenomena become very clear and sometimes ratchet-like. There is a cause, such as intention, and then an effect, such as movement. There is a cause, such as a sensation, and there is an effect, namely a mental impression.
Trying to increase the frequency at which you oscillate between physical sensations and mental sensations is a fascinating exercise.
[0] https://www.mctb.org/mctb2/table-of-contents/part-iv-insight...
They should start with what is their definition of language. To me it's any mean you can use to communicate some information to someone else and they generally get a correct inference of what kind of representations and responses are expected is the definition of a language. Whether it's uttered words, a series of gestures, subtle pheromones or a slap in your face, that's all languages.
For the same reason I find extremely odd that the hypothesis that animals don't have any form of language is even considered as a serious claim in introduction.
Anyone can prove anything and its contrary about language if the term is given whatever meaning is needed for premises to match with the conclusion.
Since it controls my limbs, I consider it to be the real me. My inner monologue is a bit frustrated that it can't control my limbs, and it can't really communicate with whoever controls my limbs.
Then there is my inner monologue, which does my thinking almost always, in an auditory way: imagine the sound of spoken words in an ~5 sec long duration, and let the answer appear. I consider it as an auditory deducing thingy, and also an intelligence on its own.
I am mostly fine with this, tho I am curious about my non-verbal me, and I wish I'd know more about it.
[0] https://archive.org/details/a-a-the-mystical-and-magical-sys...
https://www.cbc.ca/news/canada/saskatchewan/inner-monologue-...
Take riding a bike: I presume even people with an overactive inner monologue aren't constantly planning their actions (brakes, steering, turns) in words. Then just extend that out to most other stuff.
I want to remind everyone that your experiences are unique and do not necessarily translate to all other people.
There could be functional redundancies or alternative systems at play that we haven't identified, systems that allow thought to access linguistic capabilities even when the specialized language areas are offline or unnecessary. The question of what "language in thought" looks like remains open, particularly in tasks requiring comprehension. This underscores the need for further exploration into how thought operates and what role, if any, latent or alternative linguistic functionalities play when conventional language regions aren't active.
In short, we may have a good understanding of language in isolation, but not necessarily in its broader role within the cognitive architecture that governs thought, comprehension, and meaning-making.
Much later, I did begin to think mostly in words, and (perhaps for unrelated reasons?) my thinking became much less efficient.
Also related, I experienced temporarily enhanced cognition while under the influence of entheogens. My thoughts, which normally fade within seconds, became stretched out, so that I could stack up to 7 layers of thought on top of each other and examine them simultaneously.
I remember feeling greatly diminished, mentally, once that ability went away.
What it seemed like subjectively though is that my thoughts themselves became "longer", imagine planks of wood. You can stack them (slightly offset, like a video timeline with layers), and the wider they are, the more ideas you can stack before it topples over.
I have unfortunately been unable to replicate the experience. There were after-effects for a few weeks where my senses and cognition were markedly enhanced, but this faded after a few weeks.
My main take-away here is "why are we trying to make machines smarter than humans, we should try to make humans smarter"! (I guess Neuralink kinda does that, but it doesn't actually make the human part smarter...)
I doubt that this is different for other people. I believe that those people who claim that they never think using language are never thinking about the abstract or remote things about which I think using language.
For instance, I can think about a model of CPU without naming it, if it has been included in some of the many computers that I have used during the years, by recalling an image of the computer, or of its motherboard, or of the CPU package, or recalling some experiences when running programs on that computer, how slow or how responsive that felt, and so on.
I cannot think about a CPU that I have never used, e.g. Intel 11900K, without naming it.
Similarly, I can think without language about the planet Jupiter, which I have seen directly many times, or even about the planet Neptune, which I have never seen with my eyes, but I have seen in photographs, but I cannot think otherwise than with words about some celestial bodies that I have never seen.
The same for verbs, some verbs name actions about which I can think by recalling images or sounds or smells or tactile feelings that correspond with typical results of those actions. Other verbs are too abstract, so I can think about the corresponding action only using the word that names it.
For some abstract concepts, one could imagine a sequence of images, sounds etc. that would suggest them, but that would be like a pantomime puzzle and it would be a too slow way of thinking.
I can look at a wood plank thrown over a precipice and I can conclude that it may be safe to walk on it without language, but if I were to design a bridge guaranteed to resist to the weight of some trucks passing on it, I could not do that design without thinking with language.
Therefore I believe that language is absolutely essential for complex abstract thinking, even if there are alternative ways of thinking that may be sufficient even most of the time for some people.
This makes me think of the Tao Te Ching, which opens with (translation dependent, of course)
The Tao that can be spoken is not the eternal Tao
The name that can be named is not the eternal nameWhy I mention this is that I see both language and reasoning as rooted in this more fundamental cognitive ability of "coherent sequencing". This sits behind all kinds of planning and puzzling tasks where you have to project forward a sequence of theoretical actions and abstractly evaluate the outcome.
Which is all to say, I don't think language and reasoning are the same, but I do think it is likely they stem from the same underlying fundamental mechanisms in our brain. And as a consequence, it's actually quite plausible that LLMs can reconstruct mechanisms of reasoning from language, in a regressive model kind of fashion. ie: just because their are other ways to reason doesn't exclude language as a path to it.
In my mind there should be some kind of parallel/hierarchical model that comes after language layers and then optionally can be converted back to a series of tokens. The middle layers are trained on world models such as from videos, intermediary layers on mapping, and other layers on text, including quite a lot of transcripts etc. to make sure the middle layers fully ground the outer layers.
I don't really understand transformers and diffusion transformers etc., but I am optimistic that as we increase the compute and memory capacity over the next few years it will allow more video data to be integrated with language data. That will result in fully grounded multimodal models that are even more robust and more general purpose.
I keep waiting to hear about some kind of manufacturing/design breakthroughs with memristors or some kind of memory-centric computing that gives another 100 X boost in model sizes and/or efficiency. Because it does seem that the major functionality gains have been unlocked through scaling hardware which allowed the development of models that took advantage of the new scale. For me large multimodal video datasets with transcripts and more efficient hardware to compress and host them are going to make AI more robust.
I do wish I understood transformers better though because it seems like somehow they are more general-purpose. Is there something about them that is not dependant on the serialization or tokenization that can be extracted to make other types of models more general? Maybe they are tokens that have scalars attached which are still fully contextualized but are computed as many parallel groups for each step.
But I thought in images and I still do in part. so I don’t think you need words to think.
I thought the people who did were overly computerized, maybe thinking in an over defined way.
We only consciously "know" something when we represent it with symbols. There are also unconscious processes that some would consider "thought", like driving a car safely without thinking about what you're doing, but I wouldn't consider those thoughts.
I find an interesting parallel to Chain of Thought techniques with LLMs. I personally don't (consciously) know what I think until I articulate it.
To me this is similar to giving an LLM space to print out intermediary thoughts, like a JSON array of strings. Language is our programming language, in a sense. Without representing something in a word/concept, it doesn't exist.
"Ich vermute, dass wir nur sehen, was wir kennen." - Nietzsche, where "know" refers to labeling something by projecting a concept/word onto it.
My take away is that language is secondary to thinking - aka intuitive pattern detection. Language is the Watson to Sherlock.
The corollary is that treating language as primary in decision making is (sometimes) not as effective as treating it as secondary. At this point in my life (I'm old) I seem to have spent much of my life attempting to understand why my pattern matching/intuition made a choice that turned out to be so superior to my verbal language process.
I guess this was the experiment the proved the point.
Humans not taking this approach doesn’t mean that AI cannot.
And yeah it seems that core primitives of intelligence exist very low in our brains. And with people like Michael Levin, there may even be a root beside nervous systems.
Spoiler alert: brains require a lot of blood, constantly, just to not die. Looking at blood flow on an MRI to determine neural circuitry has to deal with the double whammy of both an extremely crude tool and a correlation/causation fallacy.
This article and the study are arguably useless.
I used to rationalize to myself along similar lines for a long time, then I realized that I'm just not as smart as I thought I was.
I'm brilliant - I've read volumes of encyclopedias, my hobbies include comparative theology, etymology, quantum mechanics and predicting the future with high accuracy (I only mention stuff I'm certain of tho ;) but so much so it disturbs my friends and family.
The highest I scored was in the 160s as a teenager but I truly believe they were over compensating for my age - only as an adult have I learned most children are stupid and they maybe in fact didn't over compensate. I am different than anyone else I've ever personally met - I fundamentally see the world different.
All of that is true but that's a rather flawed way of assessing intelligence - fr. I'm being serious. The things we know can free us as much as they can trap us - knowledge alone doesn't make a man successful, wealthy, happy or even healthy - I'm living evidence of this. That doesn't cut it as a metric for prediction of much. There are other qualities that are far more valuable in the societal sense.
Every Boss I've ever worked for has been dumber than me - each one I've learned invaluable stuff from. I was a boss once - in my day I owned and self taught/created an entire social network much like FB was a few years ago, mine obviously didn't take off and now I'm a very capable bum. Maybe someday something I'm tinkering with will make me millions but prolly not, for many reasons, I could write books if I wanted ;)
At the end of the day, the facts are what they are - there is an optimal level of intelligence that is obviously higher than the bottom but is nowhere near the top tier, very likely near that 100 IQ baseline. What separates us all is our capabilities - mostly stuff we can directly control, like learning a trade.
A Master Plumber is a genius plumber by another name and that can and obviously is most often, learned genius. What you sus about yourself is truth - don't doubt that. No IQ test ever told me I lacked the tenacity of the C average student that would employ me someday - they can't actually measure the extent of our dedicated capacity.
I kno more than most people ever have before or rn presently - I don't know as much about plumbing as an apprentice with 2 years of a trade school dedicated to plumbing and a year or two of experience in the field, that's the reality of it. I could learn the trade - I could learn most every trade, but I won't. That's life. I can tell you how you the ancients plumbed bc that piqued my curiosity and I kno far more about Roman plumbing than I do how a modern city sewer system works. That's also life.
It isn't what we kno or how fast we can learn it - it's what we do that defines us.
Become more capable if you feel looked down on - this is the way bc even if what you hone your capabilities of can be replicated by others most won't even try.
That's my rant about this whole intelligence perception we currently have as a society. Having 100 IQ is nowhere near the barrier that having 150 IQ is.
Rant aside, to the article - how isn't this obvious? I mean feelings literally exist - not just the warm fuzzy ones, like the literal feeling of existence. Does a monkey's mind require words to interpret pain or pleasure for example. Do I need to know what "fire" or "hot" is in a verbal context to sufficiently understand "burn" - words exists to convey to to others what doesn't need to be conveyed to us. That's their function. Communication. To facilitate communication with our social brethren we adopt them fundamentally as our Lego blocks for understanding the world - we pretend that words comprising language are the ideas themselves. A banana is a - the word is the fruit, they are the same in our minds but if I erase the word banana and all it's meaning of the fruit and I randomly encounter a banana - I still can taste it. No words necessary.
Also, you can think without words, deliberately and consciously - even absentmindedly.
And LLMs can't reason ;)
Truthfully, the reality is that a 100 IQ normal human is far more capable than any AI I've been given access to - in almost every metric I attempted to asses I ultimately didn't even bother as it was so obvious that humans are functionally superior.
When AI can reason - you, and everyone else, will kno it. It will be self evident.
Anyways, tldr: ppl are smarter than given credit for, smarter and much more capable - IQ is real and matters but far less than we are led to believe. People are awesome - the epitome of biological life on Earth and we do a lot of amazing things and anyone can be amazing.
I hate it when the Hacker News collective belittles itself - don't do that. I rant here bc it's one of the most interesting places I've found and I care about what all of you think far more than I care about your IQ scores.
The abstract visualizations I could build in my mind where comparable to semi-transparent buildings that I could freely spin, navigate and bend to connect relations.
In my mid-twenties, someone introduced me to the concept of people using words for mental processes, which was completely foreign to me up to this point.
For some reason, this made my brain move more and more towards this language-based model and at the same time, I felt like I was losing the capacity for complex abstract thoughts.
Still to this day I (unsuccessfully) try to revive this and unlearn the language in my head, which feels like it imposes a huge barrier and limits my mental capacity to the capabilities of what the language my brain uses at the given time (mostly EN, partially DE) allows to express.
I think that I ultimately developed an obsessive need to cite all my ideas against the literature and formulate natural language arguments for my claims to avoid being bludgeoned over the head with wordcelry and being seen as inferior for my lesser verbal fluency despite having written software for years at that point, since early childhood, and even studied computer science.
Basically what to most people is so obvious that it becomes transparent ("air") isn't to us, which apparently is an incredible gift for becoming a language researcher. Or a programmer.
It seems more like a complement to it: the idea arises, and then I have this compulsion to verbalise it, which gets quite frustrating as it takes several iterations. Clearly words do matter to me as a way to structure and record my ideas but there is something that pre-empts verbalisation and to some extent resists it.
I cannot provide insight on how I arrive at ideas. Even when I did literary criticism, the best I can say is that I absorbed lots of text and then suddenly a pattern would spring out. But the same things would happen for me studying maths or the hard sciences.
Software engineering is actually a bit different for me because I am not naturally a good algorithmic problem solver. Really I am somebody very passionate about computing who has a near-compulsion to see and collect more and more technology. So for me it is as simple as saying "this resembles a reader monad" or "this puns on the active record pattern". Less impressive than my humanities intelligence but worth maybe 10x the amount in the labour market :-)
This begs a question though: Since programming is mostly done with language - admittedly primitive/pidgin ones - why isn't that a struggle? Not sure if you're a programmer yourself, but if so do you prefer certain programming languages for some sense of "less-verbalness" or does it even matter?
Just wondering, not attacking your claim per se.
Parent isn't saying they can't handle language (and we wouldn't have this discussion in the first place), just that they better handle complexity and structure in non verbal ways.
To get back to programming, I think this do apply to most of us. Most of us probably don't think in ruby or JS, we have a higher vision of what we want to build and "flatten" it into words that can be parsed and executed. It's of course more obvious for people writing in say basic or assembly, some conversion has to happen at some point.
> The dog's owner's house's roof's angle's similarity to an equilateral triangle is remarkable.
I very strongly suspect that you're overestimating yourself.
It's hard for me to understand where my peers are coming from on the other side of this argument and respond without being dismissive, so I'll do my best to steelman the argument later.
Machine learning models are function approximators and by definition do not have an internal experience distinct from the training data any more than the plus operator does. I agree with the sentiment that even putting it in writing gives more weight to the position than it should, bordering on absurdity.
I suppose this is like the ELIZA phenomena on steroids, is the only thing I can think of for why such notions are being entertained.
However, to be generous, lets do some vigorous hand waving and say we could find a way to have an embodied learning agent gather sublinguistic perceptual data in an online reinforcement learning process, and furthermore that the (by definition) non-quantifiable subjective experience data could somehow be extracted, made into a training set, and fit to a nicely parametric loss function.
The idea then is that could find some architecture that would allow you to fit a model to the data.
And voila, machine consciousness, right? A perfect model for sentience.
Except for the fact that you would need to ignore that in the RL model gathering the data and the NN distilled from it, even with all of our vigorous hand waving, you are once again developing function approximators that have no subjective internal experience distinct from the training data.
Let's take it one step further. The absolute simplest form of learning comes in the form of habituation and sensitization to stimuli. Even microbes have the ability to do this.
LLMs and other static networks do not. You can attempt to attack this point by fiatting online reinforcement learning or dismissing it as unnecessary, but I should again point out that you would be attacking or dismissing the bare minimum requirement for learning, let alone a higher order subjective internal experience.
So then the argument, proceeding from false premises, would claim that the compressed experience in the NN could contain mechanical equivalents of higher order internal subjective experiences.
So even with all the might vigorous hand waving we have allowed, you have at best found a way to convert internal subjective processes to external mechanical processes fit to a dataset.
The argument would then follow, well, what's the difference? And I could point back to the microbe, but if the argument hasn't connected by this point, we will be chasing our tails forever.
A good book on the topic that examines this in much greater depth is "The Self Assembling Brain".
That being said, I am hella jealous of the VC money that the grifters will get for advancing the other side of this argument.
For enough money I'd probably change my tune too. I can't by a loaf of bread with a good argument lol
Should we expect experts in cognitive science exposing their view in a scientific publication to stick to the narrowest median view of language though? All the more when in the same article you quote people like Russell who certainly didn't have a naïve definition of language when expressing a point of view on the matter.
And slapping in general can definitely communicate far more than a single thing depending on many parameters. See https://www.33rdsquare.com/is-a-slap-disrespectful-a-nuanced... for a text exploring some of nuances of the meaning it can encompasse. But even a kid can get that slap could perfectly have all the potential to create a fully doubly articulated language, as The Croods 2 creators funnily have put in scene. :D
Even tools present us a certain 'language', talking to us via beeps, blinks and buzzes, and are having increasingly interesting discussions amongst themselves (e.g. subreddit simulator, agent based modeling). Recent philosophers of technology as Mark Coeckelbergh present a comprehensive argument for why we need to move away from the tool/language barrier [0], and has been part in informing the EC Expert Group on AI [1].
[0]: https://www.taylorfrancis.com/books/mono/10.4324/97813155285...
[1]: https://philtech.univie.ac.at/news/news-about-publicatons-et...
> Do any forms of thought—our knowledge of the world and ability to reason over these knowledge representations—require language (that is, representations and computations that sup-port our ability to generate and interpret meaningfully structured word sequences)?
Emphasis on "word sequences," to the exclusion of, e.g. body language or sign language. They go on to discuss some of the brain structures involved in the production and interpretation of these word sequences:
> Language production and language understanding are sup-ported by an interconnected set of brain areas in the left hemisphere, often referred to as the ‘language network'.
It is these brain areas that form the basis of their testable claims regarding language.
> Anyone can prove anything and its contrary about language if the term is given whatever meaning is needed for premises to match with the conclusion.
This is why "coming to terms" on the definitions of words and what you mean by them should be the first step in any serious discussion if you aim to have any hope in hell of communicating precisely; it is also why you should be skeptical of political actors that insist on redefining the meanings of (especially well-known) terms in order to push an agenda. Confusing a term with its actual referent is exceedingly commonplace in modern day.
Think about it: almost every nontrivial conversation you’ve had or comment/blog/article/book you’ve read constituted an entirely new (to you) utterance which you understood and which enabled you to acquire new ideas and information you had previously lacked. No non-human animals have demonstrated this ability. At best they are able to perform single-symbol utterances to communicate previously-understood concepts (hungry, sad, scared, tired) but are unable to combine them to produce a novel utterance, the way a child could tell you about her day:
“Today the teacher asked me to multiply 3 times 7 and I got the answer right away! Then Bobby farted and the whole class was laughing. At lunch I bit my apple and my tooth felt funny. I think it’s starting to wiggle! Sally asked me if I could go to her house for a sleepover but I said I had to ask mom and dad first.”
We maybe disagree, in the sense that it seems to be mixing indefinitely bounded expressiveness with actual unlimited expression production that could potentially be in a bijective relationship with the an infinite set of expression.
We human are mortals and even at the whole humankind scale, we will produce a finite set of utterances.
The main thing bringing so much flexibility to languages, is our ability to reuse, fit and evolve them as we go through indefinitely many inedit experiences of the world. So something like context change tolerance. But if we want to be fair with crediting admirable unknowingly extensive creativeness, we should first consider the universe as a whole, with its permanent flow of novel context, which also include all interpretations of itself through mere mortals as ourself.
I guess I've always just assumed it refers to some feature that's uniquely human—notably, recursive grammars.
And recursion as the unique trait for human language differentiation is not necessarily completely consensual https://omseeth.github.io/blog/2024/recursive_language/
Also, let's recall that in its broader meaning, the scientific consensus is that humans are animals and they evolved through the same basic mechanism as all other life forms that is evolution. So even assuming that evolution made some unique language hability emerge in humans, it's most likely that they share most language traits with other species and that there is more things to learn from them that what would be possible if it's assumed they can't have a language and thoughts.
The parent article is mostly about thinking without "words", not necessary without a "language".
Some thoughts might be completely different from sentences in a language, probably when they have a non-sequential nature, but other thoughts are exactly equivalent to a sentence in a language, except that they do not use the words.
I can look and see to things that I recognize, e.g. A and B, and I can see that one is bigger than the other and I can think "A is bigger than B" without thinking at the words used in the spoken language, but nonetheless associating some internal concepts of "A", "B" and "is greater than", exactly like when formulating a spoken sentence.
I do not believe that such a thought can be considered as an example of thinking without language, but just as an example that for a subset of the words used in a spoken language there is an internal representation that is independent of the sequence of sounds or letters that compose a spoken or written language.
All other things being equal, its is a reason to provisionally reject the hypothesis that those kinds of thought use language as introducing entities (the ties between those kinds of thought and language) into the model of reality being generated that are not needed to explain any observed phenomenon.
Music or sports are more interesting to investigate (in my opinion) since those specific actions won’t be preprogrammed and must be learned independently.
The same way we build abstractions for language in order to perform “telepathy” it seems like for music or sports we build body-specific abstractions. They work similar to words within our own brain but are not something easily communicated since they’re not tied to any language, it’s just a feeling.
I think it’s an interesting point that quite often the best athletes or musicians are terrible coaches. They probably have a much more innate internal language for their body that cannot be communicated easily. Partially, I think, that their body is more different than others which helps them be exceptional. Or that weaker athletes or musicians need to focus much more on lessons from others, so their body language gets tied much closer to human language and that makes it much easier for them to then communicate the lessons they learn to others.
Once you've figured out how to use language, explain why this is important and to who. Then maybe what the upshot will be. The fact that someone has proven something to be true doesn't make it important.
The comment I replied to made it sound like it's important to the field of AI. It is not. Almost zero serious researchers think LLMs all by themselves are "enough". People are working on all manner of models and systems incorporating all kinds of things "not LLM". Practically no one who actually works in AI reads this paper and changes anything, because it only proves something they already believed to be true and act accordingly.
I don't feel like this is a very meaningful argument because if you can do that generation then you must already have a superhuman machine for that task.
I'm not saying we cannot create a self conscious entity - I'm saying that none of the stuff we've made so far can become self aware or conscious as we are bc we haven't made it correctly. Nobody worried about AGI has anything to worry about rn - at best the models we have now may someday be able to trick us into perceiving them as aware but fundamentally they cannot attain that state out of what they are now, so it will be bullshit if one "wakes up" soon.
This bias is real. Current gen ai works proportionally well the more known it is. The more training data, the better the performance. When we ask something very specific, we have the impression that it’s niche. But there is tons of training data also on many niche topics, which essentially enhances the magic trick – it looks like sophisticated reasoning. Whenever you truly go “off the beaten path”, you get responses that are (a) nonsensical (illogical) and (b) “pulls” you back towards a “mainstream center point” so to say. Anecdotally of course..
I’ve noticed this with software architecture discussions. I would have some pretty standard thing (like session-based auth) but I have some specific and unusual requirement (like hybrid device- and user identity) and it happily spits out good sounding but nonsensical ideas. Combining and interpolating entirely in the the linguistic domain is clearly powerful, but ultimately not enough.
I'm not sure why everyone assumes an AGI would just automatically do creativity considering most people are not very creative, despite them quite literally being capable, most people can't create anything. Why wouldn't an AGI have the same issues with being "awake" that we do? Being capable of knowing stuff - as you pointed out, far more facts than a person ever could, I think an awake AGI may even have more "issues" with the human condition than us.
Also - say an AGI comes into existence that is awake, happy and capable of truly original creativity - why tf does it write us poetry? Why solve world hunger - it doesn't hunger. Why cure cancer - what can cancer do to to it?
AGI as currently envisioned is a mythos of fantasy and science fiction.
I suppose one could build an LLM around a lora that's being continuously trained to attempt to get it to adapt to new scenarios.
I was just pointing out that a crow's brain is built on a more advanced process node than our own. Smaller transistors.
I like to think that it is my gut brain that is telling me that it's okay to have that ice cream...
(also important to note that NNs/LLMs operate on... abstract vectors, not "language" -- not relevant as a response to your post though).
Needs to be a closed loop, running on its own.
We get its attention, and it responds, or frankly if we did manage any sort of sentience, even a simulation of it, then the fact is it may not respond.
To me, that is the real test.
It seems that the second link may indicate otherwise but I'm still pretty skeptical. This requires extraordinary evidence. Furthermore there may be a more practical limit of "stack size" or "context size" that effectively exceptionalizes humans (especially considering the size and proportional energy consumption of our brains).
Other animals have cognitive processes, and languages, or at least it seems to be something scientifically consensual. Thus the surprise reading the kind of statement given in introduction.
Whether humans have exceptional language habilites or even "just" a biggest board to play on with the same basic facilities seems to be a completely different matter.
Your responding to a claim that was never made. The claim was don't assume humans are smarter than whales. Nobody said whales are more intelligent than humans. He just said don't assume.
Anyway, this is just like solipsism, you won't find a sincere one outside the asylum. Every Reddit intellectual writing such tired drivel as "who's to say humans are more intelligent than beasts?" deep down knows the score.
(/s)
Intelligence is the ability to use experience to predict your environment and the outcomes of your own actions. It's a tool for survival.
However, I do think that a meaningful intelligence comparison between humans and dolphins, etc, would conclude that we are more intelligent, especially based on our reasoning/planning (= multi-step prediction) abilities, which allows us not only to predict our environment but also to modify it to our desires in very complex ways.
Remember in CS theory, a language is just a set of strings. If you think in pictures that is STILL a language if your pictures are structured.
So I'm really handwaving the above just to suggest that it all depends on the assumptions that each expert is making in elucidating this debate which has a long history.
Unless we're getting metaphysical to the point of describing quantum systems as possessig a language, there are various continuous analog systems that can compute without a formal grammar. The language system could be the one that thinks in discrete 'tokens'; the conscious system something more complex.
Siegmund, J., Kästner, C., Apel, S., Parnin, C., Bethmann, A., Leich, T. & Brechmann, A. (2014). Understanding understanding source code with functional magnetic resonance imaging. In Proceedings of the 36th International Conference on Software Engineering (pp. 378-389).
Peitek, N., Siegmund, J., Apel, S., Kästner, C., Parnin, C., Bethmann, A. & Brechmann, A. (2018). A look into programmers’ heads. IEEE Transactions on Software Engineering, 46(4), 442-462.
Krueger, R., Huang, Y., Liu, X., Santander, T., Weimer, W., & Leach, K. (2020). Neurological divide: An fMRI study of prose and code writing. In Proceedings of the ACM/IEEE 42nd International Conference on Software Engineering (pp. 678-690).
Peitek, N., Apel, S., Parnin, C., Brechmann, A. & Siegmund, J. (2021). Program comprehension and code complexity metrics: An fmri study. In 2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE) (pp. 524-536). IEEE.
[0]: https://www.frontiersin.org/10.3389/conf.fninf.2014.18.00040...
[1]: https://ieeexplore.ieee.org/abstract/document/8425769
Anecdotally, when I write code, I don’t “talk in my head”. The structures that I have in my brain are in fact difficult to put into words, and I can only vaguely describe them as interconnected 3D shapes evolving over time, or even just “feelings” and “instincts” in some cases.
The code that comes out of that process does not, in fact, describe the process fully, even though it describes exactly what the computer should do. That’s why reading someone else’s code can be so difficult - you are accessing just the end product of their thinking process, without seeing the process itself.
There is no reason to assume consciousness is Turing computable [1].
[1] https://en.m.wikipedia.org/wiki/Church%E2%80%93Turing_thesis
[1] Mark Bickford, Liron Cohen, Robert L. Constable, and Vincent Rahli. 2018. Computability Beyond Church-Turing via Choice Sequences. In Proceedings of the 33rd Annual ACM/IEEE Symposium on Logic in Computer Science (LICS '18). Association for Computing Machinery, New York, NY, USA, 245–254. https://doi.org/10.1145/3209108.3209200
Proper multimodal models natively consider whatever input you give them, store the useful information in an abstracted form (i.e not just text), building it's world model, and then output in whatever format you want it to. It's no different to a mammals, just the inputs are perhaps different. Instead of relying on senses, they rely on text, video, images and sound.
In theory you could connect it to a robot and it could gather real world data much like a human, but would potentially be limited to the number of sensors/nerves it has. (on the plus side it has access to all recorded data and much faster read/write than a human).
This (awesome!) researcher would likely disagree with what I’ve just said based on this early reference:
In the early 2000s I really was drawn to the hypothesis that maybe humans have some special machinery that is especially well suited for computing hierarchical structures.
Also, calling "generative grammar" productive seems wrong to me. It's been around for half a century -- what tools has it produced? At some point theory needs to come into contact with empirical reality. As far as I know, generative grammar has just never gotten to this point.
The only reason humans have that "communication model" is because that's how you model other humans you speak to. It's a faculty for rehearsing what you're going to say to other people, and how they'll respond to it. If you have any profound thoughts at all, you find that your spoken language is deficient to even transcribe your thoughts, some "mental tokens" have no short phrases that even describe them.
The only real thoughts you have are non-verbal. You can see this sometimes in stupid schoolchildren who have learned all the correct words to regurgitate, but those never really clicked for them. The mildly clever teachers always assume that if they thoroughly practice the terminology, it will eventually be linked with the concepts themselves and they'll have fully learned it. What's really happening is that there's not enough mental machinery underneath for those words to ever be anything to link up with.
I am a sensoral thinker, I often think and internally express myself in purely images or sounds. There are, however, some kinds of thoughts I've learned I can only fully engage with if I speak to myself out loud or at least inside of my head.
The most appropriate mode of thought depends upon the task at hand. People don't typically brag about having internal monologues. They're just sharing their own subjective internal experience, which is no less valid than a chiefly nonverbal one.
e.g. the neural electrochemical output has a specific sequence that triggers the production of a certain hormone in your pituitary gland for e.g. and the hormone travels to the relevant body function activating/stopping it.
Probably by putting simulated animals into simulated environments where they have to survive and thrive.
Working at animal level is uncool, but necessary for progress. I had this argument with Rod Brooks a few decades back. He had some good artificial insects, and wanted to immediately jump to human level, with a project called Cog.[1] I asked him why he didn't go for mouse level AI next. He said "Because I don't want to go down in history as the inventor of the world's greatest artificial mouse."
Cog was a dud, and Brooks goes down in history as the inventor of the world's first good robotic vacuum cleaner.
Just a personal opinion, but in my shitty When H.A.R.L.I.E. Was One (and others) unpublished fiction pastiche (ripoff, really), I had the nascent AI stumble upon Cyc as its base for the world and "thinking about how to think."
I never thought that Cyc was enough, but I do think that something Cyc-like is necessary as a component, a seed for growth, until the AI begins to make the transition from the formally defined, vastly interrelated frames and facts in Cyc to being able to growth further and understand the much less formal knowledgebase you might find in, say Wikipedia.
Full agreement with your animal model is only sensible. If you think about macaques, they have a limited range of vocalization once they hit adulthood. Noe that the mothers almost never make a noise at their babies. Lacking language, when a mother wants to train an infant, she hurts it. (Shades of Blindsight there) She picks up the infant, grasps it firmly, and nips at it. The baby tries to get away, but the mother holds it and keeps at it. Their communication is pain. Many animals do this. But they also learn threat displays, the promise of pain, which goes beyond mere carrot and stick.
The more sophisticated multicellular animals (let us say birds, reptiles, mammals) have to learn to model the behavior of other animals in their environment: to prey on them, to avoid being prey. A pond is here. Other animals will also come to drink. I could attack them and eat them. And with the macaques, "I must scare the baby and pain it a bit because I no longer want to breastfeed it."
Somewhere along the line, modeling other animals (in-species or out-species) hits some sort of self-reflection and the recursion begins. That, I think, is a crucial loop to create the kind of intelligence we seek. Here I nod to Egan's Diaspora.
Looping back to your original point about the training data, I don't think that loop is sufficient for an AGI to do anything but think about itself, and that's where something like Cyc would serve as a framework for it to enter into the knowledge that it isn't merely cogito ergo summing in a void, but that it is part of a world with rules stable enough that it might reason, rather than "merely" statistically infer. And as part of the world (or your simulated environment), it can engage in new loops, feedback between its actions and results.
Is that the dominant chain, or is the simpler “I’ve seen animals here before that I have eaten” or “I’ve seen animals I have eaten in a place that smelled/looked/sounded/felt like this” sufficient to explain the behavior?
Warning, watch out for waving hands: The way I see it is that cognition involves forming an abstract representation of the world and then reasoning about that representation. It seems obvious that non-human animals do this without language. So it seems likely that humans do too and then language is layered on top as a turbo boost. However, it also seems plausible that you could build an abstract representation of the world through studying a vast amount of human language and that'll be a good approximation of the real-world too and furthermore it seems possible that reasoning about that abstract representation can take place in the depths of the layers of a large transformer. So it's not clear to me that we're limited by the data we have or necessarily need a different type of data to build a general AI although that'll likely help build a better world model. It's also not clear that an LLM is incapable of the type of reasoning that animals apply to their abstract world representations.
OTOH, humans (and animals) do get other data feeds (visual, context, touch/pain, smell, internal balance "sensors"...) that we develop as we grow and tie that to learning about language.
Obviously, LLMs won't replicate that since even adults struggle to describe these verbally.
While I agree this is possible, I don't see why you'd think it's likely. I would instead say that I think it's unlikely.
Human communication relies on many assumptions of a shared model of the world that are rarely if ever discussed explicitly, and without which certain concepts or at least phrases become ambiguous or hard to understand.
Feed it all the video ever recorded, hook it up to web cams, telescopes, etc. This says a lot about how the universe works, without using a single word.
The data itself will be most senses collecting raw data about the world most of the day for 18 years. It might require a camera on the kid’s head which I don’t like. I think people letting a team record their life is more likely. Split the project up among many families running in parallel, 1-4 per grade/year. It would probably cost a few million a year.
(Note: Parent changes might require an integration step during AI training or showing different ones in the early years.)
The training system would rapidly scan this information in. It might not be faster than human brains. If it is, we can create them quickly. That’s the passive learning part, though.
Human training involves asking lots of questions based on internal data, random exploration (esp play) with reinforcement, introspection/meditation, and so on. Self-driven, generative activities whose outputs become inputs into the brain system. This training regiment will probably need periodic breaks from passive learning to ask questions or play which requires human supervision.
Enough of this will probably produce… disobedient, unpredictable children. ;) Eventually, we’ll learn how to do AI parenting where the offspring are well-behaved, effective servants. Those will be fine-tuned for practical applications. Later, many more will come online which are trained by different streams of life experience, schooling methods, etc.
That was my theory. I still don’t like recording people’s lives to train AI’s. I just thought it was the only way to build brain-like AI’s and likely to happen (see Twitch).
My LLM concept was to do the same thing with K-12 education resources, stories, kids games, etc. Parents already could tell us exactly what to use to gradually build them up since they did that for their kids year by year. Then, several career tracts layering different college books and skill areas. I think it would be cheaper than GPT-4 with good performance.
Imagine trying to limit, control, or explain a being without familiar cognitive structures.
Is there a reason to care about such unfamiliar modalities of cognition?
Anything that doesn't have a spine, I'm pretty sure.
Also if we look at just auditory, tons of creatures are deaf and don't need that.
> Imagine trying to limit, control, or explain a being without familiar cognitive structures.
I don't see why any of that that affects whether it's intelligent.
Presumably they have some sort biological input processing or sensory inputs. They don't eat data.
Because whales or dolphins didn’t evolve hands. Hands are a foundational prerequisite for building technology. So if whales or dolphins had hands we don’t know if they would develop technology that can rival us.
Because we don’t know, that’s why he says don’t assume. This isn’t a “deep down we know” thing like your more irrational form of reasoning. It is a logical conclusion: we don’t know. So don’t assume.
The "they MIGHT be as intelligent, just lacking hands" theory can't have the same weight as "nah" in an honest mind seeing the overwhelming clues (yes, not proof, if that's what you want) against it. Again, same way that you can't disprove solipsism.
I'm not sure how you would make meaningful comparisons here. We can't communicate to them as they communicate and we live in almost completely different environments. Any such comparison would be extremely biased to us.
>which allows us not only to predict our environment but also to modify it to our desires in very complex ways.
We modify our environment mostly through technology. Intelligence is a big part of technology sure but it's not the only part of it and without the other parts (hands with opposable thumbs, fire etc), technology as we know it wouldn't exist and our ability to modify the environment would seem crippled to any outside observer regardless of how intelligent we may be.
It's not enough to think that the earth revolves around the sun, we need to build the telescopes (with hands and materials melted down and forged with fire) to confirm it.
It's not enough to dream and devise of flight, we need the fire to create the materials that we dug with our hands and the hands to build them.
It's not enough to think that Oral communication is insufficient for transmitting information through generations. What else will you do without opposable thumbs or an equivalent ?
Fire is so important for so many reasons but one of the biggest is that it was an easy source of large amounts of energy that allowed us to bootstrap technology. Where's that easy source of energy underwater ?
Without all the other aspects necessary for technology, we are relegated to hunter/gatherer levels of influencing the environment at best. Even then, we still crafted tools that creatures without opposable thumbs would never be able to craft.
At least to our own perception, and degree of understanding, it would appear that the ocean habitat(s) of dolphins are far less diverse and demanding. Evidentially complex enough to drive their intelligence though, so perhaps we just don't understand the complexity of what they've evolved to do.
That said, i definitely would not say the Ocean is particularly less diverse or demanding.
Even with our limited understanding, there must be adaptations for Pressure, Salinity, light, Energy, Buoyancy, Underwater Current etc that all vary significantly by depth and location.
And the bottlenose dolphin for instance lives in every ocean of the world except the Arctic and the Antarctic oceans.
I’m dumb - I’ve skimmed through a few Wikipedia articles, my hobbies include forgetting passwords, misunderstanding instructions, and being confused by even simple math. It disturbs my friends how little I grasp most things.
One time I scored so low on a test that they didn’t even bother to tell me the number. I’ve always known I’m just not sharp – I see the world through this fog that most people seem to walk right through.
And honestly, all of that is true, but it’s a pretty bad way of measuring stupidity. The more you don’t know, the more free you are, but also trapped. Ignorance alone doesn’t make a person unsuccessful, broke, or confused, though I’m living proof that it sure doesn’t help. That’s not a good indicator of much, though. There are plenty of other things that are way more useful in life.
Every boss I’ve ever had? All smarter than me, without a doubt. They’ve taught me invaluable lessons, like how not to trip over my own feet. I’ve tried and failed at many things, and now I’m just an especially capable fool. Maybe something I’m messing up now will lead to something someday, but probably not. For many reasons. I could write books about how to fail spectacularly, though ;)
At the end of the day, the facts are what they are - being smarter than me doesn’t seem hard, but that’s not the point. What really matters is learning practical things, like how to do simple tasks without messing them up.
A Master Plumber is a genius plumber, and that’s genius you can learn. You could probably learn anything – heck, I could too, but I probably won’t. That’s life. I know more about ancient Roman sewers than I do about unclogging my own toilet, and that’s also life.
It’s not about what you know or how fast you can figure it out - it’s about what you actually manage to do. Even if everyone can do what you’re trying to do better than you, most won’t even try.
That’s my rant about how dumb I am.
You can't do this. It's not a matter of IQ, it's a matter of math. Higher order effects are essentially impossible to predict because the level of detail you need to know the initial conditions in is not possible. Even in simple systems where all the rules are known like a billiards table. Furthermore, if you could do this, you would be a billionaire by now just from trading the stock market. This claim alone makes me doubt the rest of your comment.
It's an intuitive process. Almost always the most likely things that can will be what happens - the top 3 most likely outcomes of whatever will almost always contain the thing that does happen but that list must be generated adequately, factoring in the system, players and rules of whatever it is - for example: "at work" , "co-workers" , "who gets a promotion" - the most deserving person only might get the promotion, what the Boss wants is the actual key factor for predicting that outcome.
I rarely reply to replies - to make myself even more conceited, I'm not suggesting that you should feel special or anything, I'm noting this bc you've hit a button of mine - as knowing what will happen before it does is both one of my favorite things to do and an almost natural function of my experience at this point. Not everything can be predicted and I'm not talking like "on x date x will happen exactly" - not typically, there are exceptions. I've never been able to adequately explain this but I will attempt bc I think everyone can do this to some extent.
IQ factors in bc I'm able to do this bc I have an encyclopedia in my head that I am constantly updating as much as I am able - all the time, everyday constantly adding data to a "hard drive" that has so many files I honestly don't even kno what's all there at this point - I don't even try.
Almost everything I've ever read, almost every concept I've actually thought about (highdeas or altered state ponderings included) and everything I've written out by hand is still inside my head rn and I can retrieve if I need it - I unlock it with passion of all things, for example were we to have a heated debate about the Roman Empire during the course of our conversation everything I've ever learned about Rome would come back to me - to the point I could quote professor's lectures verbatim or quote off encyclopedia entries that support my argument exactly from the copy that exists inside my head. I have to be into it for it to work.
Anyways, back to the future ;)
If your correctly identify the parameters of any given situation and you can account for what people want - the intent behind their actions, what motivates them, you will see that their desires, the chain of causation and what is actually possible incredibly narrows down the actual possibilities from the "anything can happen" point of view to one where the next event in the chain is rather obvious.
So, your right - I can't do it Fr and I literally make assumptions, inferences and operate off of hypothetical data often - and more often than not I am able to predict what will happen with high accuracy.
That works bc all people are essentially the same and we all have the same underlying motivations regardless of all demographic factors.
Meh, this could be a book.
To another point of yours - I've made others an incredible amount of money in the stock market and cryptocurrency. I've never been very motivated by money. I think your Buffets, Musks and Bezos have mental disorders and I don't envy their obsession in the slightest.
This was fun :)
Have an awesome day!
If the category you are working with is the kind of thing that you have to construct such nuance, and circles, and "yes but also..."s around, perhaps you might question your category outright?
Just to say, have you ever maybe thought that what we call "intelligence" is somewhat determined more by time and place than it is by our collective answers to multiple choice questions? Just maybe something to think about.
I am not sure what sort of LLM-powered bot is behind them, or whether it's one person with some sort of schizophrenia, but once you notice it you will see at least one of these per popular post.
Fixations around "intelligence"/IQ is huge these days, I have found, among young men, not just because of the AI stuff.
And humans in general can still, for now, write and be passionate and maybe have some misplaced enthusiasm on internet forums!
What you are describing is the difference between a Nicolai Tesla and a Thomas Edison - Tesla was far more brilliant than Edison and what he's was doing was superior to what Edison was doing in the same field of study but Edison won and Tesla died poor and most of his greatest discoveries died with him.
The world is not made for smart people - like I said, an IQ of 150 is an incredible obstacle to a normal life. That is relevant. IQ alone is not enough but of someone spends a lifetime living with a high IQ and paying my sort of attention they will see connections that others don't, stuff will just be obvious to them that others cannot see at all, stuff that others lose sleep over won't bother them and what causes them to lose sleep others won't understand.
I think you need to think further in your thought process if you think something so substantial is merely a consequence of time and place - I assure you there is more to it.
Now a little bit cheekier advice. Let me help you make an argument for your position against someone like me, because I really didn't see how I was to be convinced that this thing we are talking about is "substantial" enough to reduce any which way. The only way you can really argue with my position here is to take the point of view that "intelligence" is something like a metaphysical category. It doesn't have to be exactly that, but the important thing is that intelligence is more something "justice" than it is like "phenotype" or "language" or "neuron". You simply cannot be a materialist and also hold the fundamental nature of something like "intelligence," I wont insult you by filling in the dots there, but just know you are committed to a bit of woo-woo when you want to go all the way like you are (which is fine, I like woo-woo, just not in this case). You want, at the end of the day, for the brain to be more than an organ, but a truly teleological entity, which by chance has now touched upon something "substantial". Its a rough argument to make these days, you would of done great in the Enlightenment though :).
Also, what I believe you were trying to get at with the Tesla/Edison thing is this idea that Edison manifests my point of view because he was more successful in like the capitalist sense than Tesla. But that is just surface level here, and not at all what I am saying. I would come back and say "no, those are simply too people we are now calling intelligent, for different reasons." Intelligence isnt about results, its about certain things that we value (at a given time). And I am not quite sure even how you want to make your point here, we all generally consider Tesla as a very intelligent man these days, people even did back then!
Right, but big brains do actively impede you - they require a lot of energy, so there needs to be some offsetting benefit.
LLMs absolutely 100% can reason, if we take the dictionary definition; it’s trivial to show their ability to answer non-memorized questions, and the only way to do that is some sort of reasoning. I personally don’t think they’re the most efficient tool for deliberative derivation of concepts, but I also think any sort of categorical prohibition is anti-scientific. What is the brain other than a neural network?
Even if we accept the most fringe, anthropocentric theories like Penrose & Hammerhoff’s quantum tubules, that’s just a neural network with fancy weights. How could we possibly hope to forbid digital recreations of our brains from “truly” or “really” mimicking them?
Basically, if humans have had meaningful discussions about it, the product of their reasoning is there for the LLM, right?
Seems to me, the "how many R's are there in the word "strawberry" problem is very suggestive of the idea LLM systems cannot reason. If they could, the question is not difficult.
The fact is humans may never have actually discussed that topic in any meaningful way captured in the training data.
And because of that and how specific the question is, the LLM has no clear relationships to map into a response. It just does best case, whatever the math deemed best.
Seems plausible enough to support the opinion LLM'S cannot reason.
What we do know is LLMs can work with anything expressed in terms of relationships between words.
There is a ton of reasoning templates contained in that data.
Put another way:
Maybe LLM systems are poor at deduction, save for examples contained in the data. But there are a ton of examples!
So this is hard to notice.
Maybe LLM systems are fantastic at inference! And so those many examples get mapped to the prompt at hand very well.
And we do notice that and see it like real thinking, not just some horribly complex surface containing a bazillion relationships...
First, while it is a fringe idea with little backing it, it's far from the most fringe.
Secondly, it is not at all known that animal brains are accurately modeled as an ANN, any more so than any other Turing-compatible system can be modeled as an ANN. Biological neurons are themselves small computers, like all living cells in general, with not fully understood capabilities. The way biological neurons are connected is far more complex than a weight in an ANN. And I'm not talking about fantasy quantum effects in microtubules, I'm talking about well-established biology, with many kinds of synapses, some of which are "multicast" in a spatially distinct area instead of connected to specific neurons. And about the non-neuronal glands which are known to change neuron behavior and so on.
How critical any of these differences are to cognition is anyone's guess at this time. But dismissing them and reducing the brain to a bigger NN is not wise.
Um... What? That is a huge leap to make.
'Reasoning' is a specific type of thought process and humans regularly make complicated decisions without doing it. We uses hunches and intuition and gut feelings. We make all kinds of snap assessments that we don't have time to reason through. As such, answering novel questions doesn't necessarily show a system is capable of reasoning.
I see absolutely nothing resumbling an argument for humans having an "ineffable calculator soul", I think that might be you projecting. There is no 'categorical prohibition', only an analysis of the current flaws of specific models.
Personally, my skepticism about imminent AGI has to do believing we may be underestimating the complexity of the software running on our brain. We've reached the point where we can create digital "brains", or atleast portions of them. We may be missing some other pieces of a digital brain, or we may just not have the right software to run on it yet. I suspect it is both but that we'll have fully functional digital brains well before we figure out the software to run on them.
'Reasoning' is a specific type of thought process
I personally have lots of thoughts on the topic and look to Kant and Hegel for their definitions of reason as the final faculty of human cognition (after sensibility, understanding, and judgement), and I even think there’s good reason (heh) to think that LLMs are not a great tool for that on their own. But my point is that none of the LLM critics have a definition anywhere close to that level of specificity.
Usually, “reason” is used to mean “good cognition”, so “LLMs can’t reason” is just a variety of cope/setting up new goalposts. We all know LLMs aren’t flawless or infinite in their capabilities, but I just don’t find this kind of critique specific enough to have any sort of scientific validity. IMHO
The whole issue with "reasoning" is that is an incompletely defined concept. Over what domain, what problem space, and what kind of experimental access do we define "reasoning"? Search is better as a concept because it comes packed with all these things, and without conceptual murkiness. Search is scientifically studied to a greater extent.
I don't think we doubt LLMs can learn given training data, we already accuse them of being mere interpolators or parrots. And we can agree to some extent the LLMs can recombine concepts correctly. So they got down the learning part.
And for the searching part, we can probably agree its a matter of access to the search space not AI. It's an environment problem, and even a social one. Search is usually more extended than the lifetime of any agent, so it has to be a cultural process, where language plays a central role.
When you break reasoning/progress/intelligence into "search and learn" it becomes much more tractable and useful. We can also make more grounded predictions on AI, considering the needs for search that are implied, not just the needs for learning.
How much search did AlphaZero need to beat us at go? How much search did humans pack in our 200K years history over 10,000 generations? What was the cost of that journey of search? That kind of questions. In my napkin estimations we solved 1:10000 of the problem by learning, search is 10000x to a million times harder.
I think the paper should've included controls, because we don't know how strong the result is. They certainly may have proven that humans can't reason either.
Some people will use any limitation of LLMs to deny there is anything to see here, while others will call this ‘moving the goalposts’, but the most interesting questions, I believe, involve figuring out what the differences are, putting aside the question of whether LLMs are or are not AGIs.
While I generally do suspect that we need to invent some new technique in the realm of AI in order for software to do everything a human can do, I use analogies like chess engines to caution myself from certainty.
The most interesting thing about LLMs is probably how much relational information turns out to be encoded in large bodies of our writing, in ways that fancy statistical methods can access. LLMs aren’t thinking, or even in the same ballpark as thinking.
Not true. You yourself have failed at reasoning here.
The problem with your logic is that you failed to identify the instances where LLMs have succeeded with reasoning. So if LLMs both fail and succeed it just means that LLMs are capable of reasoning and capable of being utterly wrong.
It's almost cliche at this point. Tons of people see the LLM fail and ignore the successes then they openly claim from a couple anecdotal examples that LLMs can't reason period.
Like how is that even logical? You have contradictory evidence therefore the LLM must be capable of BOTH failing and succeeding in reason. That's the most logical answer.
Apple’s recent research summarized here [0] is worth a read. In short, they argue that what LLMs are doing is more akin to advanced pattern recognition than reasoning in the way we typically understand reasoning.
By way of analogy, memorizing mathematical facts and then correctly recalling these facts does not imply that the person actually understands how to arrive at the answer. This is why “show your work” is a critical aspect of proving competence in an education environment.
An LLM providing useful/correct results only proves that it’s good at surfacing relevant information based on a given prompt. That fact that it’s trivial to cause bad results by making minor but irrelevant changes to a prompt points to something other than a truly reasoned response, i.e. a reasoning machine would not get tripped up so easily.
It’s bloody obvious that when I classify success I mean that the llm is delivering a correct and unique answer for a novel prompt that doesn’t exist in the original training set. No need to go over the same tired analogies that have been regurgitated over and over again that you believe LLMs are reusing memorized answers. It’s a stale point of view. The overall argument has progressed further then that and we now need more complicated analysis of what’s going on with LLMs
Sources: https://typeset.io/papers/llmsense-harnessing-llms-for-high-...
https://typeset.io/papers/call-me-when-necessary-llms-can-ef...
And these two are just from a random google search.
I can find dozens and dozens of papers illustrating failures and successes of LLMs which further nails my original point. LLMs both succeed and fail at reasoning.
The main problem right now is that we don’t really understand how LLMs work internally. Everyone who claims they know LLMs can’t reason are just making huge leaps of irrational conclusions because not only does their conclusion contradict actual evidence but they don’t even know how LLMs work because nobody knows.
We only know how LLMs work at a high level and we only understand these things via the analogy of a best fit curve in a series of data points. Below this abstraction we don’t understand what’s going on.
The evidence is using one instance of the LLM parroting training data while completely ignoring contradicting evidence where the LLM created novel answers to novel prompts out of thin air.
>Observations trump claims.
No. The same irrational hallucinations that plague LLMs are plaguing human reasoning and trumping rational thinking.
The condition of “some people are bad at thing” does not equal “computer better at thing than people”, but I see this argument all the time in LLM/AI discourse.
It could be said not as well as the ones that don't need SO.
Anyway, it seems to me we are generally all in agreement (in this thread, at least), but are now being really picky about... language :)
Generally, I absolutely agree that he is not humble in the sense of expressing doubt about his strongly held beliefs. He’s been saying pretty much the same things for decades, and does not give much room for disagreement (and ofc this is all ratcheted up in intensity in his political stances). I’m using humble in a slightly different way, tho: he insists on qualifying basically all of his statements about archaeological anthropology with “we don’t have proof yet” and “this seems likely”, because of his fundamental belief that we’re in a “pre-Galilean” (read: shitty) era of cognitive science.
In other words: he’s absolutely arrogant about his core structural findings and the utility of his program, but he’s humble about the final application of those findings to humanity.
Contrast to the statistical approach. It's easy to point to something like Google translate. If Chomsky's approach gave us a tool like that, I'd have no complaint. But my sense is that it just hasn't panned out.
Right, and this is why claims that models are “reasoning” can’t be taken at face value. This space is filled with overloaded terms and anthropomorphic language that describes some behavior of the LLM but this doesn’t justify a leap to the belief that these terms actually represent the underlying functionality of the model, e.g. when terms like “hallucinate”, “understand”, etc. are used, they do not represent the biological processes these ideas stem from or carry the implications of a system that mimics those processes.
> Everyone who claims they know LLMs can’t reason are just making huge leaps of irrational conclusions because not only does their conclusion contradict actual evidence but they don’t even know how LLMs work because nobody knows.
If you believe this to be true, you must then also accept that it’s equally irrational to claim these models are actually “reasoning”. The point of citing the Apple paper was that there’s currently a lack of consensus and in some cases major disagreement about what is actually occurring behind the scenes.
Everything you’ve written to justify the idea that reasoning is occurring can be used against the idea that reasoning is occurring. This will continue to be true until we gain a better understanding of how these models work.
The reason the Apple paper is interesting is because it’s some of the latest writing on this subject, and points at inconvenient truths about the operation of these models that at the very least would indicate that if reasoning is occurring, it’s extremely inconsistent and unreliable.
No need to be combative here - aside from being against HN guidelines, there just isn’t enough understanding yet for anyone to be making absolute claims, and the point of my comment was to add counterpoints to a conversation, not make some claim about the absolute nature of things.
If a novel low probability conclusion that is correct was arrived at from a novel prompt where neither the prompt nor the conclusion existed in the training set, THEN by logic the ONLY possible way the conclusion was derived was through reasoning. We know this, but we don't know HOW the model is reasoning.
The only other possible way that an LLM can arrive at low probability conclusions is via random chance.
>The point of citing the Apple paper was that there’s currently a lack of consensus and in some cases major disagreement about what is actually occurring behind the scenes.
This isn't true. I quote the parent comment:
"What this tells me is there is clearly no “reasoning” happening whatsoever with either model, despite marketing claiming as such."
>Everything you’ve written to justify the idea that reasoning is occurring can be used against the idea that reasoning is occurring. This will continue to be true until we gain a better understanding of how these models work.
Right and I took BOTH pieces of contradictory evidence into account and I ended up with the most logical conclusion. I quote myself:
"You have contradictory evidence therefore the LLM must be capable of BOTH failing and succeeding in reason. That's the most logical answer."
Right. And this, again, was my conclusion. But I took it a bit further. Read again what I said in the first paragraph of this very response.
>No need to be combative here - aside from being against HN guidelines, there just isn’t enough understanding yet for anyone to be making absolute claims, and the point of my comment was to add counterpoints to a conversation, not make some claim about the absolute nature of things.
You're not combative and neither am I. I respect your analysis here even though you dismissed a lot of what I said (see quotations) and even though I completely disagree and I believe you are wrong.
I think there's a further logical argument you're not realizing and I pointed it out in the first paragraph. LLMs are arriving at novel answers from novel prompts that don't exist in the data set. These novel answers have such low probability of existing via random chance that the ONLY other explanation for it is covered by the broadly defined word: "reasoning".
Again, there is also evidence of prompts that aren't arrived at via reasoning, but that doesn't negate the existence of answers to prompts that can only be arrived via reasoning.
Unfortunately, you won't get one. We simply don't know enough about cognition to create rigourous definitions of the type you are looking for.
Instead, this paper, and the community in general are trying to perform practical capability assessments. The claim that the GSM8k measures "mathematical reasoning" or "logical reasoning" didn't come from the skeptics.
Alan Turring didn't try to define intelligence, he created a practical test that he thought would be a good benchmark. These days we believe we have better ones.
> I just don’t find this kind of critique specific enough to have any sort of scientific validity. IMHO
"Good cognition" seems like dismisal of a definition, but this is exactly the definition that the people working on this care about. They are not philosphers, they are engineers who are trying to make a system "better" so "good cognition" is exactly what they want.
The paper digs into finding out more about what types of changes impacts peformance on established metrics. The "noop" result is pretty interesting since "relevancy detection" isn't something we commonly think of as key to "good cognition", but a consequence of it.
As humanity has struggled to understand the world, it has frequently given names to concepts that seem to matter, well before it is capable of explaining with any sort of precision what these things are, and what makes them matter - take the word 'energy', for example.
It seems clear to me that one must have these vague concepts before one can begin to to understand them, and also that it would be bizarre not to give them a name at that point - and so, at that point, we have a word without a locked-down definition. To insist that we should have the definition locked down before we begin to investigate the phenomenon or concept is precisely the wrong way to go about understanding it: we refine and rewrite the definitions as a consequence of what our investigations have discovered. Again, 'energy' provides a useful case study for how this happens.
A third point about the word 'energy' is that it has become well-defined within physics, and yet retains much of its original vagueness in everyday usage, where, in addition, it is often used metaphorically. This is not a problem, except when someone makes the lexicographical fallacy of thinking that one can freely substitute the physics definition into everyday speech (or vice-versa) without changing the meaning.
With many concepts about the mental, including 'reasoning', we are still in the learning-and-writing-the-definition stage. For example, let's take the definition you bring up: reasoning as good cognition. This just moves us on to the questions of what 'cognition' means, and what distinguishes good cognition from bad cognition (for example, is a valid logical argument predicated on what turns out to be a false assumption an example of reasoning-as-good-cognition?) We are not going to settle the matter by leafing through a dictionary, any more than Pedro Carolino could write a phrase book just from a Portugese-English dictionary (and you are probably aware that looking up definitions-of-definitions recursively in a dictionary often ends up in a loop.)
A lot of people want to jump the gun on this, and say definitively either that LLMs have achieved reasoning (or general intelligence or a theory of mind or even consciousness, for that matter) or that they have not (or cannot.) What we should be doing, IMHO, is to put aside these questions until we have learned enough to say more precisely what these terms denote, by studying humans, other animals, and what I consider to be the surprising effectiveness of LLMs - and that is what the interviewee in the article we are nominally discussing here is doing.
You entered this thread by saying (about the paper underlying an article in Ars Tech [1]) I’ll pop in with a friendly “that research is definitely wrong”. If they want to prove that LLMs can’t reason..., but I do not think there is anything like that claim in the paper itself (one should not simply trust what some person on HN says about a paper. That, of course, goes as much for what I say about it as what the original poster said.) To me, this looks like the sort of careful, specific and objective work that will lead to us a better understanding of our concepts of the mental.
The first three paragraphs you wrote very succinctly and obviously summarize the fundamental flaw of our modern science - that it can't make leaps, at all.
There is no leap of faith in science but there is science that requires such leaps.
We are stuck bc those most capable of comprehending concepts they don't understand and are unexplainable - they won't allow themselves to even develop a vague understanding of such concepts. The scientific method is their trusty hammer and their faith in it renders all that isn't a nail unscientific.
Admitting that they don't kno enough would be akin to societal suicide of their current position - the deciders of what is or isn't true, so I don't expect them to withhold their conclusions til they are more able to.
They are the "priest class" now ;)
I agree with your humble opinion - there is much more we could learn if that was our intent and considering the potential of this, I think we absolutely ought to make certain that we do everything in our power to attain the best possible outcomes of these current and future developments.
Transparent and honest collaboration for the betterment of humanity is the only right path to an AGI god - to oversimplify a lil bit.
Very astute, well formulated position, presented in accessible language and with humility even!
Well done.
I think it's something like the counting parts of problems that current models are shaky with, and I imagine it's a training data problem.
Other examples exist.
[0]That example is due to tokenization. DoH! I knew better too.
Ah well.
For that not to be the case, you'd have to take the position that humans experience consciousness and they talk about consciousness but that there is no causal link between the two! It's just a coincidence that the things you find yourself saying about consciousness line up with your internal experience?
https://www.lesswrong.com/posts/fdEWWr8St59bXLbQr/zombies-zo...
To deal with the awkwardly apparent fact that consciousness certainly seems to have physical effects, zombiephiles challenge the notion that physics is causally closed, so that it is conceivable that something non-physical can cause physical effects. Their approach is to say that the causal closure of physics is not provable, but at this point, the argument has become a lexicographical one, about the definition of the words 'physics' and 'physical' (if one insists that 'physical' does not refer to a causally-closed concept, then we still need a word for the causal closure within which the physical is embedded - but that's just what a lot of people take 'physical' to mean in the first place.) None of the anti-physicalists have been able, so far, to shed any light on how the mind is causally effective in the physical world.
You might be interested in the late Daniel Dennett's "The Unimagined Preposterousness of Zombies": https://dl.tufts.edu/concern/pdfs/6m312182x
As many people have pointed out, Searle's argument begs the question by tacitly assuming that if anything about the room understands Chinese, it can only be the person within it.