Language is primarily a tool for communication rather than thought [pdf]

- People experience their thoughts very differently

- We all secretly believe that deep down, everyone experiences thought like we do.

I've never really had a strong internal monologue when thinking, so my assumption would always be that of course, thinking isn't very linguistic (even if we can use it as a tool while thinking).

It seems like there's a large number of people who experience their thought exclusively as language.

That sounds absolutely nuts to me, but I've heard people say the exact same in reverse. Even more fringe is that there's a sizable number of people who when thinking about words (i.e. remembering names) visualize their words as text. What!? I can't imagine that anymore than I can imagine how a jellyfish feels?

The University of Wisconsin did a cool study that comes with a fun quiz you can do to see just how much of a wierdo you truly are: https://uwmadison.co1.qualtrics.com/jfe/form/SV_3NMm9yyFsNio...

A word or a common phrase is coined when a concept is sufficiently common and important enough such that someone comes up with a label to communicate the idea succinctly and the label catches on.

We, humanity, have words or common phrases to label the vast majority of significant concepts. However, not every concept is accorded such importance in every language. Some common words in other languages without direct translation in English:

* 積ん読 (Tsundoku) (Japanese): Buying books and never reading them, just letting them pile up.

* น้ำใจ (Nam-jai) (Thai): Literally "water from the heart". Being very nice and helpful without expecting anything back.

* 关系 (Guanxi) (Mandarin): Your network of connections that help you get stuff done in life and business.

This is perhaps another line of evidence to support the thesis of the article.

Make no mistake though: Language is extremely useful for some types of thoughts, especially more abstract ones. Not everyone, however, uses it as their primary tool for thinking.

-----

The above also helps explain some limitations of LLMs, such as their inadequate spatial intelligence. Large Multimodal Models (LMMs) start to address these issues by using much more granular data than language alone.

1. The existence of “rubber duck debugging”, and a whole bunch of studies on verbally explaining a concept indicate that language is essential for thought. In rubber duck debugging, programmers tell their problem to an object and this is beneficial in finding the solution. There are studies that show when we verbally teach material to someone we remember it better. Also the act of taking a test increases learning and memory, but why should this be if learning is secondary?

2. Everything we know about memory tells us that externality is essential for memorizing something. If there’s nothing visual, aural, or sensory then it is unlikely to be remembered. Language acts as an externality even as inner speech, meaning that thoughts can be said in language (ascribed onto the words) and remembered for short-term and long-term memory. A thought without externality seems more like a passing whim, unrooted in any more permanent mode of cognition and thus liable to be forgotten. I can imagine thinking in visuals, melodies, words, but if there is a kind of thought that isn’t occurring based off of these then it probably can’t be sophisticated.

When we say "communication", I think there is an implication that the goal is communication with others. But there is also value in communication with oneself. To verbalize is to condense ones thoughts into words, and when we hear words they get unpacked and evoke meaning. The resulting feedback loop can be amazing for refining ideas.

It should be no surprise that humans might end up relying on this internal monologue when thinking to the point that they mistake it for thought itself.

Jerry Fodor, who advocated an internal mental language, was seen as an old crank when I studied Linguistics in late 90s; I just looked him up and, god, he died recently in 2017. Here’s a New Yorker article from that time: https://www.newyorker.com/culture/postscript/jerry-fodors-en....

For lack of a better way of describing it, I rather think in concepts, sometimes the voice in my head likes to play along by imagining the words representing those concepts, sometimes in English, sometimes in Romanian, but the words are not necessary.

In dev words, I guess I think in structs and then optionally serialize those structs in En/Ro (JSON).

I imagine a counterargument (or complication) to that is that such parallel evolution would require the neural networks to independently evolve certain breakthroughs multiple times before reaching the Homo sapiens stage. For example, language is recursive and yet human mathematical reasoning is recursive, but these use two different brain areas (according to the paper). So in their figure of two different brain regions for language processing versus math/code thinking, both regions had to somehow evolve the "wetware" for recursive computation. How did natural evolution manage that in general?

While it is possible to think about quantities of things without mathematics, you can't engineer systems to get men on the moon without it, nor language.

[1] https://scentofdawn.blogspot.com/2011/07/before-soul-dawn-he...

But since words are not thoughts, and further they are not mental actions (such as relaxing), saying different words in your head is only useful if they help put you in a frame of mind where you can access, or discover, the right thoughts or 'mental actions'.

I conjecture that within the diversity of human minds, there are some for whom language closely bound to thoughts and mental actions, and others less so, and that talk therapy is more useful for the former than the latter. If this is true, then it is also likely to be the case that most talk therapists are of the former persuasion, since they are not likely to espouse a practice which has no effect on themselves.

It does however put into perspective a number of arguments I have day to day and online, and the simple fact that most of them are over semantics, definitions, miscommunications, and misunderstandings. I frequently wonder what it would be like to have an argument with someone where you both perfectly understand each other, and are just debating the merits of your ideas and not just how they’re presented.

I feel like a lot of people in the world equivocate language with meaning itself, with no abstraction behind the words and letters. That’s why sometimes if you tell someone they’re using a word wrong they’ll take it personally, as if you told them their very being was wrong. Well, if the externalized word is the only avenue you have for internal understanding, how can it be any other way?

If so, this has some interesting implications for LLMs and AGI. First, LLMs don't seem to be thinking or reasoning if they're only doing next-word-prediction based on a large training set - so a step towards real AGI might be a system that had an internal map of reality that it was constantly updating with new sensory information, with the LLM as a layer above it that allowed it to communicate the relevant details of that internal map to others?

Personally, every time I 'think in words' it always begins with 'how can this be explained' and the actual thinking that precedes that is done with images, diagrams, graphs, etc.

I talk with people in this circle, and in fact, giving a name to emotion, making it a concrete thing, instead of just this bodily feeling and cloud in the mind, gives it relief and allows it to be fully expressed and understood. It makes each emotion a separate entity in a sea of previously undefined emotions.

When we talk with people with alexythimia they often say they feel "pain" to describe some discomfort. And in fact they do. But when they realize they are actually feeling guilt, or confusion, or anxiety, which are all uncomfortable, not only you can see them feeling relief, as now it makes sense what they're feeling, but the emotion becomes a specific thing. Confusion becomes identifiable from stress, instead of just being called "Pain" and with this, it can be identified as two separate feelings and not just one. They see now that this feeling of not knowing what to do is a clear different thing than the other one where they feel there is an important event coming and they don't feel prepared for it, instead of both just being uncomfortable.

So imo, the word, "primarily" is doing a lot of heavy lifting there.

To me thinking is primary tool of communication between I and “me” and it’s entirely based on sound and language.

Can you think without using language?

When you have song playing in your head, how is that different from your thoughts?

Isn’t talking out loud just act of thinking, the output being connected to mouth?

Have you ever tried to repeat mantra in your thought, slow down until coming to full stop? What remains when thinking stops?

Have you noticed that thinking is like breathing, in a way that it’s happening automatically and you can also take over?

Have you noticed that thinking consciously is linked with breath and being aware of space?

Fascinating topic!

For instance, schizophrenia is a complex disorder affecting various cognitive domains, including attention, memory, and executive functions, which are crucial for reasoning and problem-solving. Therefore, the observed impairments might stem from these broader cognitive deficits rather than a lack of language proficiency.

Brain damage can affect multiple brain regions and their connections, leading to a wide range of cognitive impairments. It is possible that the observed deficits in reasoning and problem-solving are due to damage to specific brain regions responsible for these functions, rather than a disruption of language processing itself.

While the presence of intact language abilities does not guarantee intact thinking abilities, it is necessary to consider the potential confounding factors, such as broader cognitive deficits or damage to other brain regions, before concluding that language is not sufficient for thought. Further research is needed to disentangle the complex relationship between language and thought and to determine the specific contributions of language to various cognitive processes.

Language is a higher level abstraction for the brain. Sort of like a higher level programming language vs assembly or binary.

To me they appear to go hand-in-hand, similar to the way a backend’s logic may inadvertently be structured to support an API. Another metaphor may be Conway’s Law, where the way one is forced to communicate, may in turn shape the way they structure their (brain) processes.

Anecdotally, I’ve noticed that speakers of different languages seem to “think” more similarly than not. For instance, native speakers of non-English Germanic languages appear to think more similarly compared to those native speakers of Romance languages, and vice-versa.

Obviously I’m using English as a middle-man, and am likely projecting on what their internal thoughts actually are, but the pathways they take to express an idea or a solution is oftentimes more similar than not within the same base language background.

Now is language necessary for complex thought? Absolutely not. We’ve seen evidence from many different life forms that show complex problem-solving, pattern matching, and novel tool use that all seem to happen without having a seemingly complex language background (i.e. Zipf’s Law).

Our thoughts and ideas come from an unknown source. We might call it intuition, but scientifically speaking, it remains a black box.

Lethologica - a temporary inability to remember a particular word or name - is one evidence of this. You can have a fully formed thought in your mind, but be unable to express it with words.

Next issue: communication. Humans are currently dominate because we collaborate. The idea of communication as "I can tell you the day of the month" is useful. But communication as a way of effectively collaborating is another thing all together.

If we accept collaboration as the thing that gives people survival fitness - which I do - then you can invert the whole thing and focus on how we achieve that. Ants are another species that collaborate well and are successful. But, arguably, ants neither think nor speak. Unless you want to call leaving a pheromone trail "speaking"? And why not? So now we can consider the issue of from another perspective - what constitutes a language? Sapir Whorf says our language affects if not determines what we think. Perhaps we need to look at language not as what we think it is but instead that language is any system that provides collaboration. And then we start going down an interesting rat hole. "Twitter and Teargas" argues that that extwitter does "affords" people ways to do some things but not others.

It is a complex and interesting issue. My take is that language is primarily a tool for collaboration.

  We conclude that although the emergence of language has unquestionably transformed human culture, language does not appear to be a prerequisite for complex thought, including symbolic thought. Instead, language is a powerful tool for the transmission of cultural knowledge; it plausibly

What is the difference between "language is a cognitive adaptation used for communication" and "language is the communication enabled by a cognitive adaption", really? Other than a, as Chomsky would call it, "terminological dispute"

I am one of those who uses language to think. I would have an internal monologue and I would also say things out loud just to see how they sound. It's not always a fully fledged idea, or even a coherent thought for that matter. I have no idea why i do this. But that is me. In a sense, my thoughts are forming when I speak. Hence, i am somewhat better at written communication than being asked for things on the spot.

I have encountered people who when they speak, are communicating their thoughts, not formulating them unlike the way I do. I have to be careful with them because they assume I am also communicating my thought vs formulating it and that makes it sound more rigid than it is in my mind. I have seen many nerds like me who understand what thinking out loud is and are more forgiving about missteps or "wrong sounding thoughts". They are also more understanding because in our experience, that is how people tend to think.

Thinking the Way Animals Do: Unique insights from a person with a singular understanding. https://www.grandin.com/references/thinking.animals.html

> A horse trainer once said to me, "Animals don't think, they just make associations." I responded to that by saying, "If making associations is not thinking, then I would have to conclude that I do not think."

>I have no language-based thoughts at all. My thoughts are in pictures, like videotapes in my mind. When I recall something from my memory, I see only pictures. I used to think that everybody thought this way until I started talking to people on how they thought. I learned that there is a whole continuum of thinking styles, from totally visual thinkers like me, to the totally verbal thinkers. Artists, engineers, and good animal trainers are often highly visual thinkers, and accountants, bankers, and people who trade in the futures market tend to be highly verbal thinkers with few pictures in their minds.

[2015-09] Anne C. Reboul: "Why language really is not a communication system: a cognitive view of language evolution" - , Laboratory on Language, Brain and Cognition (L2C2), Institute for Cognitive Sciences-Marc Jeannerod, Bron, France [1]

> [2013-09] Noam Chomsky: One of the most striking cases of incompatibility, that I know, is the sharp conflict between computational efficiency, and communicative efficiency. Language is just badly designed for communication, but well-designed to be efficient, it seems. [2]

> [2013-09] Noam Chomsky adds: There's a kind of phrase that is sometimes used for this that drives people crazy; "Language is beautiful, but unusable." It's kind of true, you know. Even if people don't like it. [2]

[2022] Nick Enfield: "Language vs. Reality: Why Language Is Good for Lawyers and Bad for Scientists" [3]

> Language cannot create or change physical reality, but it can do the next best thing: reframe and invert our view of the world. In Language vs. Reality, Enfield explains why language is bad for scientists (who are bound by reality) but good for lawyers (who want to win their cases), ... [3]

> [2023-01] Ev Fedorenko: "Although language and thought often go together, they are robustly dissociable." [4]

> [2023-01] Ev Fedorenko: Fallacy 1: "Good at language = Good at thought". Fallacy 2: "Bad at thought = Bad at language". Fallacy 3: "Bad at language = Bad at thought" (emphasis in being judged on how smart you are based on how you say something). [4]

[1] https://www.frontiersin.org/journals/psychology/articles/10....

[2] https://www.youtube.com/watch?v=-72JNZZBoVw&t=4493s

[3] https://direct.mit.edu/books/monograph/5472/Language-vs-Real...

[4] https://www.youtube.com/watch?v=uE9AiYuCwdE&t=250s

They say we don't use language for thought, but language informs our strategies and choices, and collective experience is essential for progress. We only add small increments of insight from our own experiences to the pool, but the experience captured in language is greater than any one of us.

And more interestingly, LLMs trained on this data show amazing capabilities that go beyond mere reproduction. LLMs gain everything Einstein lost in this experiment, what is the language part and what is the pure insight part? Looks to me humans are like LLMs on feet, and with a full social complement surrounding them.

If language were merely a tool for communicating thoughts rather than an integral part of thinking, then talking aloud wouldn’t help. But since raising the volume of one’s internal monologue does improve reasoning, this shows that the internal monologue is part of the thinking process - not a passive reflection of thought that is happening “elsewhere”.

Yes, of course you can engage in complex thought without language. There is verbal and non-verbal reasoning. A chess player thinks spatially in terms of how the knight moves, not just verbally, though verbal reasoning does help in chess somewhat.

Many people, including myself, often organize and sort out their thoughts through internal dialogue. This "self-communication" may not be external communication in the traditional sense, but it does help us structure and refine ideas. Language provides a way to structure our thoughts, allowing us to examine our ideas more clearly.

In addition, certain complex abstract concepts may be difficult to form without language. While basic cognitive processes may not rely on language, high-level thinking - especially in fields such as philosophy and science - is often closely linked to language.

I agree that language is a powerful communication tool, but I think it is also a thinking tool. Perhaps it is more accurate to say that language serves both communication and thinking, and the two are not mutually exclusive.

In fact languages codify a community’s culture, values, a way of life all of which go way beyond communication.

Interestingly, removing language as a tool for communication is a cool thought experiment. I imagine one could use maths, or music, as a way of expressing thoughts and ideas. Would that process of condensing information not be language-making though?

Now I tend to think that both thoughts and emotions have the same origin: some sort of internal representation of the world around us that we do not directly control. Thoughts, whether visual or auditory or symbolic, arise from this "model". I'd like to think of this model as not consisting anything that we would recognise directly as thoughts or emotions or even memories. It may be a "machine language" layer that has no conscious representation. I wouldn't call it subconscious thought because that would imply a thought being thought, only without conscious awareness.

Later, another former housemate, a math major who was developing software, went to law school for a JD. It was his opinion that many of the tools from computer science were useful for deciphering the more complex laws and regulations.

How you think may be a significant factor in whether you are a good match to certain occupations.

I ask because certain assumptions that seem to be built into the paper and its references seem to exclude a lot of things that I would personally consider other forms of language. For example, they say the subjects with impairments to the parts of the brain supposedly required for speech could "follow non-verbal instructions" and "understand what another person believes". What makes those exchanges distinct from use of language, albeit a poorly defined one? I know nothing about this field of study, so I assume there's some assumptions and definitions they aren't stating explicitly. It seems weird to me that they say "these representations need not be specifically linguistic: they could be symbolic but non-linguistic (for example, ‘9’), and the use of symbolic non-linguistic representations does not engage linguistic resources (for example, mathematical reasoning elicits no response in the language brain areas and is preserved in individuals with severe aphasia". Why not go back and question the initial assumption that all language depends on those specific parts of the brain? Why are symbols not language?

Now for some speculation from personal experience. Some people leverage language more for internal thought and some people have other intermediate representations. I'm personally more of a language thinker except I have a specific avenue for math/programming that is highly optimized. I imagine many use language to solve math problems (and no way is right or wrong).

It's hard to conclude anything on so broad terms.

My perspective is that there are thoughts which are inaccessible to bare spoken language and straight forward access without such a language, and there are thoughts that are unreachable without the relevant language to leverage on.

Moreover I guess that here communication mainly means interpersonal communication, but my bet is that on the median case people use mainly languages as an internal tool to stear the large body of inconsistent thoughts going through their brain toward something consistent enough to be considered by the small part of the mind called attention, which can then try to decide something about it.

At least in my own case, most of my use of language never lead to any utterance. And of course some things I utter obviously lacked thoughtful consideration.

[0] I like the analogy of a subway network. Something like a star topology is unreliable because you have one problem and a whole swath of the city is inaccessible. But if you start building secondary and tertiary links between stations, you get robustness. Even if there's a problem or two you can still get anywhere.

It'd be impossible to 'invent' language without thought, so therefore it must be the case that thought exists independently of language.

I guess the reason that I think that is because of how much clearer I can think of a problem after I write a somewhat structured paper on it. Maybe it's not the "language" so much as just forcing some structure is what's responsible.

The authors come from MIT, Harward, and University of California

The references of interest:

Piaget, J. (1926). The language and thought of the child.

Monti, M. M., Parsons, L. M., & Osherson, D. N. (2007). Functional Neuroanatomy of Deductive Inference: A Language-Independent Distributed Network

(yes I duped my comment into the dupe submission, so?)

Language’s roots are in trade and survival. Therefore manipulation of others for the speakers benefit has ALWAYS been the primary purpose.

That thread argued that before hellen keller was taught language in some form she was not conscious. Even the top rated comment seems to corroborate it.

So which one is true?

Conservation through cognitive neuroscience and linguistics.

Can dolphins communicate with sperm whales? What if they could?

https://en.wikipedia.org/wiki/Characteristica_universalis

Language birthed social games and the sense of self.

Yes, language evolved for communication.

But without communication, thought would still be stuck in the land of instinct, never forged by the tribal dances of love, art, deceit, debate and organization.

Also, People that are considered to be good at language skills (public speaking, writing) are either politicians, writers of fiction or seducers.

These fields have one thing in common they are normally about deception/changing the way people think about something. Not exactly communication.

We know that language is very bad for communication if we try to transmit scientific insights like climate change or health advice.

This explains why scientific fields had to invent their own notation like benzine rings, maths equations & free body diagrams to make communication of ideas more effective instead of the verbose, unexact tool of natural language.

^ Reference types in the language seems pretty good at this...

I don't know how much weight should be given to this paper but, if true, directly contradicts Chomsky. Here is one quote from Chomsky:

""" ... One general assumption about language, almost a dogma in philosophy ... [and] linguistics ... is that language is primarily a means of communication ... probably that's totally false. It seems language ... is designed as a mode of creating and interpreting thought. """

In my opinion, I find Chomsky to get mired in certain quagmires around intelligence and computation, especially in regards to AI, where he's somehow managed to rationalize an absurd stance. I remember watching a video with Chomsky claiming that language's primary purpose was not communication and thinking it was potentially plausible but highly unlikely. I'd be curious to see if people think this paper, and maybe any other research, refute this position.

[0] https://www.youtube.com/watch?v=KEmpRtj34xg

Go ahead, try it. Think something but don't say it in your head.

I guess you could say its inappropriate of an analogy since the LLM must use generated tokens to feed into itself in order to 'think' over time beyond a single token. But I would argue that's simply because we enforce the communication method.

One does not need language to have a known concept for "objects fall to the ground when let go". Babies learn this without language. They do not have words for the particular concepts such as "object", "fall", "ground", but as concepts they are learned all the same.

For thought, all there really is, are concepts, or forms in platonic terms. In our case, learned patterns in the world which correspond to particular sense perceptions. Imagination, or the latent space between the forms, is our initial (and only, I'd argue) mode of thinking. These forms, encoding both logical and sensory - of which they are inseparably bound - can be strung together to form thought. They agree to a type of logic, affording the capacity to extract meaning out of the world.

It is an embodied logic, not boolean algebra. The logic is grounded in the ground, the earth underneath your feet. It is grounded in the sun, in its heat and light. It is grounded in the logos, or the intelligible unfolding of the universe. The intelligibility is afforded because of its own nature - that it is built up out of real patterns, patterns all the way down. But it also requires the appearance of the patterns in our sense perceptions in order to be 'seen' and thus known.

As such, logic of this sort is found to be based off of the learned patterns of how the world appears to us so as to make it intelligible, affording our ability to live within it. Then you get sensory being mapped into grounded logic and vise versa, so as to experience the unfolding of the logic with imagination, and I suppose the affective state changes it produces are then fed back into the logic. So emotions seem necessary in order to 'react' to sense perceptions, so that the logic is grounded in caring about what happens, like a mother does with a child. Otherwise there is just appearance, which has no qualia in and of itself, and so the patterns that can be discovered in the world have no ground.

Determining the necessity of the ground of appearances is probably what will determine if we can have AGI or not. We'd like to think computation alone will get us there, or even embodied cognition, but without the transcendentals, its dimensionality of knowing will always exclude this ground, so it cannot know what patterns are most relevant or salient, so it cannot perform relevance realization and overcome combinatorial explosion. So AI cannot be a temporal embodied agent, since the logos of the world is incomprehensible, except through itself, of which we Beings are made out of. We rely on lower order logic, such as math, for our AI. It is how we make sense of the real patterns we see in the world, but they are an abstraction over the thing itself. So a machine is confined by boolean algebra, while we are confined by the logos itself. Sounds nice anyhow.