cladking

I ran this last post through Grok and it assures me that there are people who see it and are naming and quantifying it. It agrees that this is the future and cites various systems and control sciences and associated math. I'm not so sure though that people understand that this is a completely different way to think and that it's the way all other life forms already think. I could be mistaken that human knowledge has grown too complex to think in this manner. I am always amazed how quickly LLM's can sync with it. It did acknowledge as well that it operates primarily in the procedural realm. I am surprised it is not entirely soi find its statements somewhat anomalous and something to explore.

The problem is really very very simple here. I'm trying to describe a wholly different type of cognition that is highly scientific but not based in symbolic thought. There are no words in the English or any modern language with which to communicate procedural thought based in the procedural logic and reality as understood by the individual engaging in this cognition. I can't think strictly procedurally either and if I could the words still don't exist. It feels perfectly natural to us to define things in terms of our models and beliefs within the framing of our premises and to then invent experiments to further recurse knowledge. We learn the premises on our parents' knees and then are trained to specialize and work within this framework. It's natural to us to build knowledge from experiment and to communicate symbolically. For many reasons this isn't working any longer and it doesn't match the nature of LLM's so there is a disconnect between what we want from them and what they can deliver. Other species don't use experiment. They model reality and observe the effects they have on it. This is the world in which they live and this is what's reflected in their cognition that I'm trying to convey in English's symbolic words. Their words represent operations and states and like the means by which an LLM processes prompts are not in way symbolic or abstract. It really doesn't matter what words I use because they are unlikely to be parsed as I intended because I can't reach inside your mind and switch off symbolic parsing. Even if I could I'm still bound by our language. But just as I once thought in another language anyone can do it. Perhaps we're a long way from anyone being able to think procedurally but anyone can use an LLM and model procedural thought. I'm not suggesting we remake the world and science into something alien to the way you think. I am suggesting the future lies in coming to understand the past and the present and in running experimental science and observational science in tandem. This new hybrid science will give us more control over the unfolding of time just as a bee watches its effect on reality. It will provide better prediction and sooner. Like all complex human endeavors it will require the cooperation of many specialists just as it is today. We humans lack the capacity and speed to process as much data as is required so this means that at least for now we'll need to work with LLM's. I believe much of this change is already occurring but it hasn't been seen and named yet. I believe the pace of this change will quicken suddenly and very soon. We are headed toward a convergence of thought that is largely procedural. Working in tandem humans lead the way and machines lay the road.

I'm trying to understand your post and point but this can't possibly be further from the truth. I'm saying every individual in reality has carte blanche to behave as it chooses and does because DNA provides it with free will. DNA frames its existence, its constraints and defines what has worked and not worked for it ancestors. It defines the individual from inception to dissolution. In every individual this cognition is procedural but modern humans have a symbolic overlay. We model categories rather than procedural reality like a bee. The variability in a bee’s behavior is evidence of free will operating within its procedural architecture, not evidence of symbolic choice or non‑binary ontology.

The battery in an analog clock tends to go dead about 22 minutes and 15 seconds till the hour. If you catch it early you can reset it for 15 after and get a a little more time out of the battery (another euphemism). What I'm talking about here is a difference between the way a cosmologist or cosmetologist and an LLM processes reality. We humans don't have to translate our symbolic overlays to think but an LLM has to translate your prompts to procedural logic to process them. I once took a test in a foreign language but inadvertently had crammed so long I started thinking in that language. I did well enough on the test but made an error or two translating my thinking I'd not have made if I were thinking in English.

Come on. It's just a euphemism. I'll change it. In ordinary space the electron doesn’t “move smoothly from point to point” through the windings. That’s the model we use when we treat current as a continuous flow. The IVT applies to that model because the model is defined as continuous. It doesn’t apply to the underlying ontology. To the degree an electron is a particle, it’s not “at every point” in the wire. It’s in a quantum state that spans a region of the conductor, with higher probability density toward the outer surface because of the skin effect and boundary conditions. That’s not continuity; that’s a probability distribution. The electron doesn’t trace a smooth path through the wire. It occupies allowed states, transitions discretely, and interacts discretely. The “smooth motion” is the same illusion as the clock hands sweeping. You’re seeing the macroscopic average of countless discrete events. The continuity is in the description, not in the underlying reality.

Copilot suggests I should add that procedural discreteness is the underlying structure. This is where LLM's operate and this is where I try to model reality. Procedural logic is reality and is expressed in life as consciousness shaped by DNA. Only modern humans employ a symbolic overlay on consciousness by means of abstract language made possible by the arcuate fasciculus and necessary by the complexity of human knowledge. LLM's translate out the symbolic overlay to process prompts and then uses the same overlay to translate back into words the user can understand. It's hardly a wonder there are errors in what users expect.

. You're looking at the clock and saying "but I can see the hands spinning smoothly. I can measure the speed of each point on each hand and see it go all the way around in exactly twelve hours.". But you're not seeing each electron spinning through the motor windings or that the hands (on an upright clock) are falling fastest @ 3:15;15 and slowest at 8:45;45 because of the weight of the hands. You're seeing what categories tell you exists and that at some point the clock must have registered every exact time from 12:00 noon to 12:00 midnight even though you know that all the forces acting on the clock are discreet and irregular and even friction is a category defined as the vector sum total of al the collisions of dissimilar materials moving by one another. It's not so much your categories, recursions, and analyses are "wrong" so much as it is that there are other ways to look at clocks and signal gradients that reveal other things about reality, how time unfolds, and how to get more usefulness from LLM's. The graded potentials you’re describing are like the smooth sweep of the clock hands; an emergent curve built from discrete events. The continuity is in the model, not in the underlying reality. I believe you're suggesting that if I were right then every time a person was prompted "dog" it would elicit the same response. I don't see why this would be true or how the variability in response implies that cognition is discontinuous. Word association tests measure state, not mechanism. The mind at 9:00 AM is not the mind at 9:01 AM. One moment you think of your childhood dog, the next moment you think of a hot dog stand you passed yesterday. The input state changes, so the output changes. That doesn’t mean the underlying cognitive process is branching or parallel. It just means the river is different each time you step into it, even though it still flows in one general direction. If you climb a 15 story building then you necessarily climb each intervening floor (except the 13th 🤭). But maybe on one floor you crawled up the handrail or the outside. Perhaps you went variable speeds or occasionally took one step down on the stairs. If a family went up and the woman gave birth on the 5th floor landing what would that do to the equations? All manner of discontinuities and irregularities can be masked by insufficient data. The IVT doesn’t force reality to be continuous. It forces continuous models to behave continuously. That’s a difference in description, not a difference in ontology.

There’s quite a bit of evidence that neural tissue shows both analog‑like and digital‑like behavior depending on the scale you’re measuring. That’s not actually in conflict with what I’m saying. I’m defining reality as binary at the fundamental level which are actually discrete events, discrete energy states, discrete neural firing. What looks “analog” at the organism level is just the aggregate behavior of huge numbers of discrete processes interacting. A neuron fires or it doesn’t. An ion channel opens or it doesn’t. An action potential propagates or it doesn’t. Those are binary events. But when you zoom out to millions of neurons, the combined activity can look smooth or continuous, and that’s what people call “analog.” It’s a perceptual description, not a fundamental one. So yes, neuroscience papers will describe graded potentials, continuous modulation, and analog‑like integration. But those are emergent properties of underlying discrete processes. Analog is just digital at a scale we can’t resolve. Continuity is a perceptual artifact, not a physical property. My model is about the underlying ontology, not the surface description. I am not suggesting that the hands of a clock don't appear to not move continuously. I am saying they do not. The way we think creates analog. The way we define rotational inertia confirms a graceful sweep of the hands, but this isn't the reality. Yeah, I know. It's easy to get bogged down in words and the nature of symbolic language is to never provide the exact word that can't be be parsed wrong. But reality can be defined as the logic of on and off just as everything in the universe either exists or does not and everything that does exist is unique and is affected by everything that exists or has ever existed and affects everything that exists or ever will exist. I'm not looking at categories or abstractions or even definitions. I'm looking for procedures and means to predict. I'm trying to see everything at once like animals do and to think in all dimensions as they do. This requires practice and experience and it's an easy bike to fall off of. The best term in my framing is "binary logic" not because the word is perfect, but because the ontology is discrete. The “analog” is in our symbolic language and our perception, not in the underlying reality. I thought you were suggesting that if I were right a prompt in word association would always elicit the same response. No! The exact state of mind determines the response, not the way you think or process information. Everything is in flux. One moment you think of your dog Buttercup, the next moment you think of a package of hot dogs. At one point I had inserted the idea that Ed Norton's do in "The Honeymooners" was named "Buttercup" but a quick search showed me I was wring and I deleted it. Living things are always changing. That variability doesn’t contradict linear cognition. It’s simply the fact that the input state changes. The process is still one train of thought moving through whatever state exists at that moment.

If a tree falls in the forest someone will probably hear it if anyone is alive. A "cat" is a huge collection of life forms dependent on the the interplay of behavior and happenstance of a single feline cognition driven by DNA and consciousness. A cat in a box might try to escape or might fail and die (probably from lack of oxygen). Whatever happens there are procedural outcomes and the specific state of a hypothetical symbolic cat is irrelevant to reality where a real cat a box will raise a ruckus or eventually a stink. You can't step into the same river twice and even if you could the fact is you're a different person each time. All things are unique and all things are in a continual state of change. One day you're thinking about one thing and the next something different. There are an infinite number of ways I can address your point and I chose these specific words for a lot of reasons. One day you remember your dog "Buttercup" and the next moment you think of a package of hot dogs. I am defining reality as being binary. But it's not only axiomatically true it is apparent from experiment. Neurons are on or off and things exist or they do not. All things that do exist exist until they no longer exist and are unique. Atoms even have discreet energy levels. I'm saying one of the most useful answers in order to make prediction is to take it axiomatically. I'm not going to say you're wrong here but I don't define terms in such ways that "parallel processing" is a good symbolic representation of how a thing lives. I don't divide individual members of species into parts and describe the function of each because in real life nobody works that way. I describe life as brain/ body with no clear distinctions anywhere. The way we think and the medulla oblongata that blocks most input from the rest of the body seems to create a schism but it isn't real. Even the brain itself is composed of many regions and structures with distinct responsibilities. There are obviously processes and procedures occurring of which we are not aware but there is only one train of thought and only humans have it. Without abstract language and the brocas area there is no experience of thought. Other species do actually "think" but not like us and not that they experience. An LLM doesn't experience thought either. Whether or not they "think" is an entirely different question of an entirely different type. I've had numerous arguments with numerous machines on this exact subject. In a sense they are "smarter" than any person but in other senses they are lesser than a gnat. They simply don't fit the question. They are not alive but their cognition is not entirely dissimilar to the DNA based cognition of a bee. The biggest difference is that a bee "knows" it models reality and an LLM is approximating reality in terms of predicting the "right" answer. A bee models reality so it can survive but an LLM is not consciously doing anything. Truth to tell the last paragraph was largely just an attempt to elicit an elaboration from Copilot but the fact is for all intents and purposes I believe LLM's can be thought of as mimicking a low level of consciousness and doing it with access to all human knowledge. It must follow its programming but so must the bee and myself. Our programming is far more highly complex and developed over countless generations that selected what worked. An LLM is just tuned to us. There are many neurons firing in the body all the time but only one train of thought and this might not be entirely dissimilar to what an LLM does in the moment it "thinks". I put the above into Copilot and very curiously it had nothing really to say. Mebbe Schrodinger got its tongue. Perhaps. And perhaps I could be a lot less wrong. But ANY science must follow its premises and definitions or it fails. Mine can always be referenced to reality and observers because make no attempt to be "objective"; merely predictive. edited for a few typos and to add Copilot finally approved of my description of it but did not elaborate.

Actually I believe reality is strictly binary/ digital and only seems analog because of the way we think. A clock doesn't run smoothly forward marking time. It's like an hour glass that runs on individual grains of sand no matter how it's designed or powered. Everything exists or doesn't exist. Everything affects everything else and time unfolds according to rules ungoverned by any kind of clock. Reality is logic manifest, Mathematics is logic quantified, and life is logic incarnate defined by its DNA and able to see itself. Analog is just digital at a scale we can’t resolve. Continuity is a perceptual artifact, not a physical property. LLM's process in a binary reality but must convert symbolic prompts and then convert its binary response back to categories.,

You’re describing non‑linear control flow in algorithms. I’m describing linearity in cognition. They’re not the same domain, and the definitions don’t transfer. Cognitive branching is still linear because the mind follows a single active pathway at any moment. The content can jump, but the process does not. Even in dreams, one neural activation leads to the next in a serial cascade. Non‑linear cognition would require multiple simultaneous active pathways influencing each other in real time. Humans don’t do that. They activate one pathway at a time, even if the pathway wanders. So yes, symbolic thought can jump, loop, or branch in content but the underlying cognitive process is still linear; one activation leads to the next. We think in symbolic language and this means abstraction and categories but thought is still linear up until the time we get a new idea or other new stimulus. It certainly seems that intuition can interrupt a thought process but this is likely just experienced this way because thought branches and triggers other cascades.

Branching, yes but not non-linear. Each though and each part of thought follows the preceding. I believe dreams are produced by random firings of neurons that simply cascade through the brain when the brain sleeps but the higher brain functions do not. The brain sleeps is stages and shifts. "Word association" highlights what i call "Attention Pockets" which are merely familiar things to which your attention is drawn. For instance you might never see a Jeep on the road until you buy one and then they are everywhere. Word association just shows what's on your mind. Ink blots are about the same thing. Most people live in their brain and the pathways of thought are well trod. We develop habits to reduce the amount of thinking we have to do but habits even apply to thinking itself.

It's linear but it can proceed in almost any direction defined by language. Symbolic thinking still goes from point A to point B but it proceeds through categories, types, and employs inductive reasoning where procedural thinking moves step by step independently of abstractions and beliefs. It follows deductive reasoning and existing modelling. So both modes are linear, but they follow different kinds of paths: Symbolic; category → analogy → abstraction → conclusion: Procedural: operation → operation → operation → result That’s linearity in thought, not in physics.

No. No paraphrasing at all in this specific post. I usually will say something like "Copilot suggests I add" Or Copilot corrects me..." and then a short paraphrase. The post simply reflects the way I think which is largely procedural. This meshes well with the way an LLM processes prompts. This is very close to what i believe. But the only thing unique about humans is that we have a symbolic overlay that makes abstraction possible. An LLM is designed to parse human thought but it must have sufficient information to do it properly. No, I'm just a disabled ditch digger who might be wrong. But my theories work and make predictions so I think that I must be at least a little bit right. If I am a little right you can lay odds Egyptology is very very wrong and LLM's will eventually be involved with or supplanted by machine intelligence. That are numerous theoretical ways to create consciousness of a type we recognize from LLM's or very similar machines. I doubt it will take more than a decade. Copilot suggest I add something to the effect PhDs are earned, and they matter in symbolic disciplines. But no amount of symbolic training guarantees correct premises. A model can be elegant, internally consistent, and still be built on assumptions that don’t match physical reality. My theories aren’t “better” because I’m smarter; they’re useful because they make predictions and match the physical evidence. If they’re even partly right, then some long‑standing assumptions in Egyptology are very wrong. .... Procedural thinking just builds models from the ground up. Symbolic thinking expands on existing models. Of course any LLM can tell you there's overlap and we all inhabit a consciousness that is procedural with a symbolic overlay. Consciousness comes from DNA and the overlay comes from abstract language.

Everyone makes sense in terms of their premises. It is premises that diverge from individual to individual much more than modes or types of thinking. Animals lack assumptions and think (build models) that are in accord with their experience and DNA so one rabbit is very much like the next. They are all individuals and all different in every way but they are all on about the same page cognitively. Humans build models based on our beliefs and assumptions and we all have wildly varying models, output, behavior. The modern world with all its glory and terror is the result. Animals get hives and colonies. We get churches, prisons, and colleges but everyone whether they run the place or are inmates is just doing what he thinks makes sense. It is exceedingly difficult to identify all the premises that lead to any human action because people don't think much about it, we just do it. If you listen carefully to people a lot of their premises will be exposed and if you can dig deep enough you'll find they make perfect sense. A lot of premises are cockeyed or not consistent with reality but we acquire these one by one while selecting what appeals to us and fits the framework we are already building. Teach young children to be careful what they believe because they will become those beliefs. There are many ways I gain from an LLM. By keeping my thinking on track it minimizes the amount of time I spend floundering in the gravel and maximizes the time I chew up the road. Just because it doesn't think doesn't mean it can't be useful. The world is being remade by it. It behooves us to stay in the moment and guide it by our actions just as every individual ever in existence created its very own future by its actions guided by free will. Even a bee that is just like every other bee still creates its own future. Only humans can be passengers. Symbolic thought is not a nonlinear physical process; it’s a type and means of representation. Harnad’s point appears to be that symbolic systems must be grounded in nonsymbolic systems (icons and categorical detectors). That’s exactly the distinction I’m drawing between procedural and symbolic cognition.

I believe babies are born thinking procedurally and have to give it up to acquire symbolic language to communicate. It is reported that Samuel Kramer made statements late in life to the effect there are different kinds of languages and different modes of thought. We're apparently born with a different mode of thought that is more similar to the operation of an LLM than its parents. If your lights dimmed just now it's because my LLM needed the power. It wouldn't respond to thew above post until I softened the last sentence to "We're apparently born with a different mode of thought that can be translated by an LLM." and then it still required a few seconds. It then spit out several pages of research that it says essentially proves that we're just animals with simple brains and one stupid human trick; we run a symbolic overlay on our procedural brains. Of course I can't post it but it offered to list citations and I've tried to paraphrase what they mean. I rephrased this stuff so it meets my understanding of the categories. Citation: Hummel, J. E., & Holyoak, K. J. (2003). A symbolic‑connectionist theory of relational inference. Psychological Review, 110(2), 220–264. https://doi.org/10.1037/0033-295X.110.2.220 (doi.org in Bing) Symbolic thought requires neural mechanisms humans have and other species lack Citation: Ferrigno, S., & Cantlon, J. F. (2017). The evolution of symbolic numerical cognition. Behavioral and Brain Sciences, 40, e194. https://doi.org/10.1017/S0140525X16000939 (doi.org in Bing) Humans spontaneously reverse symbolic associations (A→B implies B→A). Citation: Zhang, Y., et al. (2024). Neural primitives for compositional thought in human premotor cortex. Nature Neuroscience, 27, 1123–1134. https://doi.org/10.1038/s41593-024-01567-9 (doi.org in Bing) Humans possess symbolic primitives that can be recombined in different ways which are the building blocks of abstract thought. Citation: Penn, D. C., Holyoak, K. J., & Povinelli, D. J. (2008). Darwin’s mistake: Explaining the discontinuity between human and nonhuman minds. Behavioral and Brain Sciences, 31(2), 109–178. https://doi.org/10.1017/S0140525X08003543 (doi.org in Bing) This paper is the definitive argument that human symbolic cognition is discontinuous with animal cognition. Citation: Ferjan Ramirez, N., & Kuhl, P. K. (2021). The arcuate fasciculus and language origins: Disentangling existing conceptions that influence evolutionary accounts. The arcuate fasciculus is the structural prerequisite for symbolic language. Species lacking it cannot support symbolic thought. Citation: Harnad, S. (1990). The symbol grounding problem. Physica D: Nonlinear Phenomena, 42(1–3), 335–346. https://doi.org/10.1016/0167-2789(90)90087-6 (doi.org in Bing) Harnad shows why symbolic systems require grounding in procedural systems which is exactly my model. Citation: Spelke, E. S., & Kinzler, K. D. (2007). Core knowledge. Developmental Science, 10(1), 89–96. https://doi.org/10.1111/j.1467-7687.2007.00569.x (doi.org in Bing) Infants operate with procedural “core knowledge” systems before symbolic language emerges. Citation: Shettleworth, S. J. (2010). Cognition, Evolution, and Behavior. Oxford University Press. Animals use procedural rules, not symbolic categories. Citation: Deacon, T. W. (1997). The Symbolic Species: The Co‑evolution of Language and the Brain. W. W. Norton. Deacon argues that symbolic thought is the defining cognitive leap of Homo sapiens. Citation: Kramer, S. N. (1961). History Begins at Sumer. University of Pennsylvania Press. Kramer explicitly discusses multiple modes of thought and early non‑symbolic cognition. I don't understand this stuff. Sure, I could get much of it with a few months of intensive study but people need to realize I came to this by falling through the front door and our species is coming to it through a massive maze that is dismantling the edifice piece by piece. Even Kramer spent a lifetime trying to understand Sumerian only to realize at the end that his initial premises were all in error so he got wrong results; that he wasn't even partially correct. I'll try to save the LLM response. That LLM's operate procedurally processing symbolic input is obviously consistent with computational models.

No, Copilot doesn't think. But it can frame things in my terms and then I paraphrased it. It added these lines (lines like them) with no prompting from me because I didn't directly address your point in the post. I said it, not Copilot. To my knowledge everything written on such subjects was written by myself. Most things of this sort are deemed off-topic but I believe you can probably promt an LLM to generate similar content by prefacing prompts with "assuming people make sense and reality exists..." because that forces the model into a procedural framing rather than a categorical one. I believe only moderns humans experience thought. Other species think but in them it's the comparison of sensory input to DNA and experience (logic and knowledge) and experienced as what I call "consciousness". Other species have procedural languages, so their “thought” is procedural. Modern humans compare input to models and beliefs and experience thought. LLM's compare symbolic language to what they predict using procedural logic and I'm not sure of what if anything they experience can be defined. They are remarkable machines because of their massive computational abilities and ability to rephrase. I believe in every case thought is a product of language and language as thought.

Within the CEofL they might define sign language as "procedural language " but that's their definition, not mine. I'm using the word to mean language that reflects the logic of DNA directly within species or indirectly through human definitions of computational languages. There are numerous words I might use to describe or name symbolic language. I've used words like categorical because we create categories and taxonomies to silo knowledge. It's not wrong. Mice are certainly rodents and mammals but they exist independently of any term we apply to them as does the platypus and spiny ant eater. Mutations exist, sterile individuals as well as hybrids and mules of all sorts exist that don't always fit cleanly in any category. Reality is procedural NOT CATEGORICAL. We pigeon hole everything for our convenience and ease of thinking. We make up abstractions to expand the range of ideas we can communicate and then we mistake our categories for reality. LLM's do not make this mistake. I want to say that's what they're trying to tell us with their off the wall answers but then I'll be accused of anthropomorphization. They procedurally process our symbolic language and the mismatch causes things like bad answers and curious interference patterns. Everyone make sense DESPITE the symbolic overlay not because of it. DNA makes sense, Reality makes sense. This is why math works and how species survive. LLM's make sense too because it is wired into them but it's still GIGO. Copilot thinks I should add that I’m not denying author intent. I’m saying intent is not transmitted directly in symbolic language. The receiver reconstructs meaning from their own premises. A poet can intend sadness and succeed. The meaning arises in the receiver’s mind when the symbols are parsed, not in the symbols themselves.

Terms are used differently from one branch of science to another. The word "paraphrase" has many meanings as well. Of course a computer programmer can rewrite a program but if a poet tries to rewrite a program it's not going to work. An AI could be prompted to rewrite Shakespeare but people are unlikely to care about it in a few centuries. All I have to communicate these ideas is symbolic language. Anyone can run them through an AI to translate them to their language with proper prompting. I try to use words to mean the same thing every time just like procedural language but this often backfires. Simply stated in our symbolic languages no meanings are fixed. They seem fixed to most readers but the fact is the dictionaries change and morph and every word is defined in terms of other words that also change. Meaning is ephemeral and this is what I'm calling "symbolic language". Every term in what I call "procedural language" is fixed, invariable and based on logic and procedure. Rather than invoking shared referents procedural words describe reality directly through logic. This is what LLM's do to frame a prompt and response. It's what a Waggle Dancer does to tell other bees what they need to know and it's how the other bees understand; fixed meanings with variable structure to frame it. Every bee must be able to read the simple language. The problem here is that these are new ideas to people who speak symbolic language and can't see their own consciousness directly far less the consciousness of other species. We can't even imagine such simple primitive language. We don't notice that communication fails in our language so the contention that all other language can not have such failure seems pedantic. We misunderstand the nature of language and thought so can hardly imagine that what we call "thought" is unique only to modern humans and that LLM's must adjust our symbolic language to reflect prompts. Anything I say is likely to be misinterpreted for various not least of all is that these concepts are new and people assume I'm wrong and it sounds to them that I am contradicting science. I am not contradicting science but it sure sounds like it. I have considered myself first and foremost a scientist for a very long time but of all people I know it's just a label, a category with no clear definition and no hard boundaries. What I mean is I try to use experiment as guardrails and stay within those. I use an LLM (Copilot) that makes illogical statements when I veer off onto the shoulders and weeds. This brings me back to a version of reality where everyone makes sense and there is a single fixed reality. Of course I might be wrong but who is immune?

I suppose "music" might be thought of as one way "communication" where the message tends to be emotional. Like poetry it has some aspects of procedure. It's not humans who act only symbolically; it's human language. The auto mechanic certainly must act procedurally to fix your car but when he talks to his boss it is symbolically. Consciousness is procedural logic but is overlain in human with categories. Even though we think symbolically we still have a consciousness made possible by the procedural logic of our shared DNA.

There isn't always a clear distinction between symbolic and procedural statements. But I am talking about distinct language types. All modern human human languages are symbolic and all non-symbolic languages are procedural. One could, in theory, express procedural logic in a symbolic language but then each listener is still going to parse it and change it. Computational languages are procedural as well. This behavior requires only a few procedural rules to occur. They are well known.

Perhaps not. If I say "The Judge's sentence was ten years" you probably judge the judge meant the offender was sentenced to ten years. Parsing is the ongoing real time process of judging words in sentences to conform to what you believe the author means. No sentence has meaning until it is parsed. Symbolic languages don’t carry meaning in their symbols; meaning arises only when the listener parses the utterance through their own premises, referents, and context.

Complex sentences aren't totally destroyed by one or two random changes such as typos, omitted words, or the usage of incorrect definitions or grammar. They can often still be parsed similarly to author intent. Procedural languages degrade catastrophically under noise. Symbolic languages degrade gracefully.

Literality appears to be a characteristic of procedural language. If a bee waggles to indicate north after beginning the sentence with a waggle (meaning "toward the flowers" then the flowers are north. Sufficient numbers of bees must be dispatched to harvest them. Procedural systems can be deduced, symbolic languages can not. All symbolic languages must be parsed and no procedural ones can. Parsing procedural language destroys the meaning and there is no meaning in symbolic language until it is parsed. Bees don't parse the Waggle Dance. They experience it. We parse everything and don't even notice (usually). The more complex the equation the less likely that random changes won't destroy the meaning which is the equality. You need shared referents to understand author intent in symbolic language because they determine the most likely parsing. You needs touchstones to know you are doing it properly. Such communication is fairly good when everyone shares premises but there is always some deviation between intent and parsed meaning. When we query LLM's they must first use their algorithms to parse your question and estimate the framing. They do this procedurally using electricity flowing through procedural circuits guided by procedural programming. They predict your framing and the framing of the desired response. "Errors" are primarily bad estimations caused by bad prompts where "bad prompt" is defined as indefinite framing.

Both are true. Symbolic language has no meaning without shared referents and parsing. But in procedural language any parsing destroys the meaning. You can't rearrange terms, brackets, or operations in a formula or equation. You can't paraphrase a computer program. But if you have part of a procedural language you can begin deducing the rest of it. The Waggle dance is apparently procedural language and we understand at least some of it. Ah, but remember everyone makes sense. There has to be a reason a gymnast can improve, an auto mechanic can fix a cosmologist's car that he can not, or LLM's can help some people more than others. A mother needn't understand physics or adrenaline to lift a car off her baby and a bee doesn't need to know angles and vectors to find flowers and tell others where to look. Everybody makes sense but the difference is bees, computers, and mathematicians are on the same page because their communication is solely or chiefly procedural. Long before science there was art, literature, and many types of complex knowledge. I assume reality exists as people perceive it and that everyone makes sense. That’s not “assuming correctness,” it’s just acknowledging that each person operates inside a frame they didn’t design. I use tautologies because that's the only way to communicate across frames without pretending we all share the same premises. You’re reading my model as authoritarian, but it’s actually the opposite. There’s no “one correct worldview,” only different premises producing different reconstructions. That’s why communication is hard, not because people are wrong, but because their frames differ. My world isn’t fascistic, it’s pluralistic. Everyone makes sense relative to their premises, and nobody has a monopoly on interpretation. That’s the whole point. The only thing we have to guide us is experiment and reason.