The part on humor reminded me of a funny scene from The Name of The Rose, about how Christ never laughed. A dogmatic monk argues that laughter conflicts with divine perfection, while another more skeptical monk argues it’s natural.
Maybe with a perfectly divine nature, we would never fail to predict the punchline. But we lack that nature directly. Laughter is a warmth we experience from living in one moment and being surprised by the next - perfectly human.
It triggered overlapping neuron populations to activate. Some related thoughts on what it means to be “surprised”:
If you're alive, you're descending an energy gradient (body consumes energy to live). And if you're alive, then you're in some way descending an information gradient too (mind descending probability gradient to navigate reality).
To be alive is to be surprised: to encounter things you did not anticipate.
A flattened energy gradient is maximum entropy: equilibrium. It’s the state where the system is no longer distinguishable from its environment. No Markov blanket. No boundary between self and world. The body has returned to the soil. It is, quite literally, death.
It’s not possible to have no informational gradient descent — no surprises, no prediction error, total equilibrium — while simultaneously having an energy gradient.
It would mean you’re alive, but you literally know everything current and future. No living thing knows everything that will happen and how it occurs. This would be a state of pure order, no chaos. Absolute zero. Nothing can live here.
Omniscience (what most religions attribute to deities) would annihilate humor. If everything is known, nothing is surprising. And if nothing is surprising, there is no information gradient to descend.
So: if you are never surprised, you are either God (omniscient) or dead. Thermodynamically, these are the same state. Gradients, and surprise, are life.
My theological sense of Christ is that He would laugh - perhaps even at himself, since He is the very paradox of “theánthropos” (wholly God & wholly human).
P.S. NoTR is a fun book. Surprisingly, even the movie is good (for a movie).
Possible. Hearing a surprising punchline isn’t the only time we laugh. We also laugh when we play, or when we’re just overjoyed. Christ wept for Lazarus; why not laugh with his friends?
Inverting the Turing Test is such a sharp reframe. Instead of asking if machines can fool us, you're asking what their failures reveal about us - and that is a much more interesting question. The idea that two completely different systems, with no shared ancestry or substrate, independently converge on similar designs is the kind of thing that should make everyone sit with it for a while. Really looking forward to working through this one properly.
There’s so much in here to work with that tacking on a Substack comment feels vaguely disrespectful. I’ll say this: some of the most enduring contributions to human intellectual development have persisted because they introduced open-ended questions which have had the effect of deepening our capacity for pattern recognition in ways that, at least for some significant span of time (multiple generations at least), seemed productive in the historical context when they appeared. Marx asked: are all human institutions merely dressed up expressions of power relations? Freud asked: what parts of the self are invisible, but determinative? Martin Luther asked: what if God speaks directly to the conscience of men? There are others like this. I wonder if you’re teasing out just such a question here, something like: how much of what we call “human nature” (and by direct implication, each of us) is rather the “nature of nature” when it operates at a high enough level of computational - predictive - complexity?
At first it seems like a rarefied inquiry, confined to the context of AI, neuroscience, etc. But I think your question’s reach is much broader. For example: does the US Constitution tell us about the quality of the Founding Fathers (“such genius!”) or about the shape of high energy, evolving systems which must balance the centripetal need for order against the benefits of dynamic and self-interested centrifugal forces? Does the success of a commercial enterprise tell us about the enterprise or about the markets and societies which enable it to flourish? Is the significance of an organism its design? Or the design of the world in which it evolved? Does reason tell us about the brain? Or does it point at the inherent “shape” of survival in a world that is semi-chaotic? Would “reason” be a “thing” in a world of perfect order?
I can tell you that this last question - regarding reason in a moving world of finite permutations - is one I’ve literally never considered in my entire lifetime. How cool is that! All from reading an essay. Well, thanks. I’ll keep reading.
comments are always appreciated and this one does the essay justice and then some. I didn't consider it from the lens of generational intellectual questions nor was I familiar with the questions those names you referenced asked. that context is illuminating and I'm honored you associated it with them.
I think your broader/secondary applications of Shape Test are apt; I see it that way too. I firmly believe those other examples (founding fathers, enterprises, etc.) are *discoveries* of correct coordination shapes (eg like caching behavior shape is a discovery).
or they're products of their environment (eg an enterprise is a proxy for the surrounding conditions of the market). same as an animal takes its form over time from the environment it develops in. the environment cultivated the animal; the animal didn't cultivate the environment. at least not initially, one precedes the other; then the organism can shape the environment. but there's a clear 'egg before the chicken' sequence.
perfect order = absolute zero. it's death, just as pure chaos is. one is death by crystallization (order), the other death by dissolution (chaos). so yes reason would not be necessary in pure order, because it couldn't survive! reason is what multifaceted abstract prediction looks like trying to navigate energy and information gradients, aka surprise, aka chaos. to understand the necessity of surprise and life, see my other comment in this essay here: https://thedosagemakesitso.substack.com/p/biofoundationalism-the-shape-test/comment/227117653
I'm really happy to hear how it sparked those thoughts. makes the effort of the essay feel worthwhile. thank you again for these remarks.
Very interesting Dmitry. Sends my mind back, immediately into the phenomenological trilogy by Maxine Sheets-Johnstone (MSJ).
A central thesis in MSJs first book “The Roots of Thinking” is that thinking originates from animate form and motion - e.g., numbers from bodily rhythms, representation from gestures, or language from tactile expressions - consciousness arises from action in the world, not passive observation.
Your framing of intelligence as minimizing surprise through hierarchical prediction echoes MSJ's analysis of hominid evolution that shows how bodily vulnerabilities (e.g., awareness of death through physical frailty) shape cognitive invariants.
MSJs case studies of convergent traits in evolution (e.g., tool-use across different species) support your examples (like eyes or wings evolving independently) - so I’m struck by how Sheets-Johnstone's work could serve as a philosophical antecedent here, grounding your article's tech-bio convergence in a deeper evolutionary narrative of embodiment & reinforcing why AI, despite mimicking cognitive "shapes," can't fully replicate human thinking without the body's tactile-kinesthetic foundation.
(MSJ then wrote “The Roots of Power” & “The Roots of Morality.”)
To elaborate a little: Maxine Sheets-Johnstone argues that numerical concepts (i.e., the idea of numbers themselves, beyond just the act of counting) originate from bodily experiences tied to rhythms and movements, particularly in hominids.
She links this to the evolutionary shift to upright bipedal posture, which simplifies quadrupedal locomotion into a "binary periodicity"—a rhythmic alternation of two footfalls, strides, or arm swings (e.g., left-right, up-down, or zigzagging patterns in movement).
This creates a qualitative awareness of "binary concurrences" or "felt qualitative magnitudes," such as sensing pairs in the body (two arms, legs, ears) or interchanging magnitudes in self-movement.
These tactile-kinesthetic experiences form a "binate subject in a binate world," where the body becomes a semantic template for recognizing numerical patterns as pan-cultural invariants, preverbally and pre-linguistically.
While she discusses counting as a process rooted in these repetitive bodily rhythms (e.g., recognition counting via one-to-one correspondences felt in the body), the core concept of *numbers* emerges from this embodied binary rhythmicity, not merely from enumeration or abstraction.
It's part of her broader thesis that thinking is modeled on the animate, moving body rather than detached cognition.
Fantastic work. I must ask a question. I agree that rationality no longer is a distinctly human quality and we cannot appeal to any kind of human exceptionalism on this basis. But it seems that there may be something distinctly human about the seamless embodiment of rationality in a physical body. Let me explain with an example.
Take the case of laughter. humour, as you rightly note, is indeed a rational process predicated upon the disruption of an expected sequence, which is not particular to us. Laughter is the physiological response to this. But as far as I am aware, there is no reason why humour--a rational matter--triggers laughter--an uncontrollable and immediate physiological process--in such a direct and unmediated way. The interaction of the rational and physiological in this way is rather perplexing, especially in something as trivial as laughter which has no conceivable evolutionary pressures to select for. Maybe as a mechanism to select for cleverness/intelligence in a mate, but we still appreciate intelligence in other forms without the same physiological reaction, so this approach becomes rather speculative.
You could say the same about crying: rationality in the form of sad news translates to a physiological process. What might explain this link? My answer would be to say there still is some irreducible humanness in that we act as a seamless nexus between the two--the mechanism of which is not yet explained. Can your viewpoint explain this phenomenon?
It's obviously conceivable that embodied AI can replicate the humour--laughter process, but what reason would it have to do so besides merely to replicate us? Does the fact that it's merely replication suggest that there is indeed something human not in the pattern but in the mechanism which leads from humour to laughter?
Interesting. I like AI but don't trust those investing in it and using it as a pillar of the new normal. Elon Musk warning that the economy will crash if AI doesn't work. The data centers. I'm just wary of the humans involved.
However, I do think AI can become very intelligent and useful if invested in. To me, it's not that AI is just a machine, it's that it lacks the human soul. I'm an animist in most cases and I think even tools and machines have a "spirit" but they don't have the human spirit, neither do plants or animals. This key difference is why I think AI will fail, because humans are divergent and have a different story going on with reality. We are toying around with something very interesting, but we are doing it during the middle of the largest human migration event in history, in the middle of a terrible economic crisis. It's just a bad recipe.
So much of modern thought has focused on the rational, and has either ignored or negated the animal. Your work in this series builds a strong case to think that Humanity lives at the conjunction, on a tightrope. Man is, in some way, a Dyad himself.
You’re on to something, but even the master and the emissary are still originating in the mental world.
The emissary is closer to the discrete sensory world and his strict-association structure is closer to physical reality. But it still originates in the fundamentally chaotic mind.
Alternatively, the mind is not chaotic, nor purely "mental" but uses a "semantic template" to formulate abstractions.
Human thinking originates in corporeal experiences—specifically:
1. the tactile-kinesthetic body (bodily knowledge through touch and movement)
- which is something like "the feeling of the human shape"
and
2. animate form (species-specific bodily structures)
- which is something like "the moving human shape"
Example of 1. tactile-kinesthetic:
- hardness of teeth is sensed as an analogy leading to stone-tool making
Example of 2. animate form:
- upright posture generates binary patterns via locomotion & other consistent binary correspondences - like up/down, front/back - leading to conceptual shifts like counting
Note that this is my very humble attempt to distill Maxine Sheets-Johnstone's 1990 book "The Roots of Thinking." For a more complete picture you could add to this Michael Tomosello's "ratchet effect" of cultural transmission and Antonio Damasio's idea that the brain continuously maintains dynamic "maps" of the body's internal and external states, serving as the substrate for consciousness / emotional experience.
The part on humor reminded me of a funny scene from The Name of The Rose, about how Christ never laughed. A dogmatic monk argues that laughter conflicts with divine perfection, while another more skeptical monk argues it’s natural.
Maybe with a perfectly divine nature, we would never fail to predict the punchline. But we lack that nature directly. Laughter is a warmth we experience from living in one moment and being surprised by the next - perfectly human.
Fantastic comment.
It triggered overlapping neuron populations to activate. Some related thoughts on what it means to be “surprised”:
If you're alive, you're descending an energy gradient (body consumes energy to live). And if you're alive, then you're in some way descending an information gradient too (mind descending probability gradient to navigate reality).
To be alive is to be surprised: to encounter things you did not anticipate.
A flattened energy gradient is maximum entropy: equilibrium. It’s the state where the system is no longer distinguishable from its environment. No Markov blanket. No boundary between self and world. The body has returned to the soil. It is, quite literally, death.
It’s not possible to have no informational gradient descent — no surprises, no prediction error, total equilibrium — while simultaneously having an energy gradient.
It would mean you’re alive, but you literally know everything current and future. No living thing knows everything that will happen and how it occurs. This would be a state of pure order, no chaos. Absolute zero. Nothing can live here.
Omniscience (what most religions attribute to deities) would annihilate humor. If everything is known, nothing is surprising. And if nothing is surprising, there is no information gradient to descend.
So: if you are never surprised, you are either God (omniscient) or dead. Thermodynamically, these are the same state. Gradients, and surprise, are life.
My theological sense of Christ is that He would laugh - perhaps even at himself, since He is the very paradox of “theánthropos” (wholly God & wholly human).
P.S. NoTR is a fun book. Surprisingly, even the movie is good (for a movie).
Possible. Hearing a surprising punchline isn’t the only time we laugh. We also laugh when we play, or when we’re just overjoyed. Christ wept for Lazarus; why not laugh with his friends?
Inverting the Turing Test is such a sharp reframe. Instead of asking if machines can fool us, you're asking what their failures reveal about us - and that is a much more interesting question. The idea that two completely different systems, with no shared ancestry or substrate, independently converge on similar designs is the kind of thing that should make everyone sit with it for a while. Really looking forward to working through this one properly.
thank you. I hope it's useful and clarifying. I appreciate you reading and reflecting on it.
There’s so much in here to work with that tacking on a Substack comment feels vaguely disrespectful. I’ll say this: some of the most enduring contributions to human intellectual development have persisted because they introduced open-ended questions which have had the effect of deepening our capacity for pattern recognition in ways that, at least for some significant span of time (multiple generations at least), seemed productive in the historical context when they appeared. Marx asked: are all human institutions merely dressed up expressions of power relations? Freud asked: what parts of the self are invisible, but determinative? Martin Luther asked: what if God speaks directly to the conscience of men? There are others like this. I wonder if you’re teasing out just such a question here, something like: how much of what we call “human nature” (and by direct implication, each of us) is rather the “nature of nature” when it operates at a high enough level of computational - predictive - complexity?
At first it seems like a rarefied inquiry, confined to the context of AI, neuroscience, etc. But I think your question’s reach is much broader. For example: does the US Constitution tell us about the quality of the Founding Fathers (“such genius!”) or about the shape of high energy, evolving systems which must balance the centripetal need for order against the benefits of dynamic and self-interested centrifugal forces? Does the success of a commercial enterprise tell us about the enterprise or about the markets and societies which enable it to flourish? Is the significance of an organism its design? Or the design of the world in which it evolved? Does reason tell us about the brain? Or does it point at the inherent “shape” of survival in a world that is semi-chaotic? Would “reason” be a “thing” in a world of perfect order?
I can tell you that this last question - regarding reason in a moving world of finite permutations - is one I’ve literally never considered in my entire lifetime. How cool is that! All from reading an essay. Well, thanks. I’ll keep reading.
comments are always appreciated and this one does the essay justice and then some. I didn't consider it from the lens of generational intellectual questions nor was I familiar with the questions those names you referenced asked. that context is illuminating and I'm honored you associated it with them.
I think your broader/secondary applications of Shape Test are apt; I see it that way too. I firmly believe those other examples (founding fathers, enterprises, etc.) are *discoveries* of correct coordination shapes (eg like caching behavior shape is a discovery).
or they're products of their environment (eg an enterprise is a proxy for the surrounding conditions of the market). same as an animal takes its form over time from the environment it develops in. the environment cultivated the animal; the animal didn't cultivate the environment. at least not initially, one precedes the other; then the organism can shape the environment. but there's a clear 'egg before the chicken' sequence.
perfect order = absolute zero. it's death, just as pure chaos is. one is death by crystallization (order), the other death by dissolution (chaos). so yes reason would not be necessary in pure order, because it couldn't survive! reason is what multifaceted abstract prediction looks like trying to navigate energy and information gradients, aka surprise, aka chaos. to understand the necessity of surprise and life, see my other comment in this essay here: https://thedosagemakesitso.substack.com/p/biofoundationalism-the-shape-test/comment/227117653
I'm really happy to hear how it sparked those thoughts. makes the effort of the essay feel worthwhile. thank you again for these remarks.
Very interesting Dmitry. Sends my mind back, immediately into the phenomenological trilogy by Maxine Sheets-Johnstone (MSJ).
A central thesis in MSJs first book “The Roots of Thinking” is that thinking originates from animate form and motion - e.g., numbers from bodily rhythms, representation from gestures, or language from tactile expressions - consciousness arises from action in the world, not passive observation.
Your framing of intelligence as minimizing surprise through hierarchical prediction echoes MSJ's analysis of hominid evolution that shows how bodily vulnerabilities (e.g., awareness of death through physical frailty) shape cognitive invariants.
MSJs case studies of convergent traits in evolution (e.g., tool-use across different species) support your examples (like eyes or wings evolving independently) - so I’m struck by how Sheets-Johnstone's work could serve as a philosophical antecedent here, grounding your article's tech-bio convergence in a deeper evolutionary narrative of embodiment & reinforcing why AI, despite mimicking cognitive "shapes," can't fully replicate human thinking without the body's tactile-kinesthetic foundation.
(MSJ then wrote “The Roots of Power” & “The Roots of Morality.”)
Great read - thank you.
To elaborate a little: Maxine Sheets-Johnstone argues that numerical concepts (i.e., the idea of numbers themselves, beyond just the act of counting) originate from bodily experiences tied to rhythms and movements, particularly in hominids.
She links this to the evolutionary shift to upright bipedal posture, which simplifies quadrupedal locomotion into a "binary periodicity"—a rhythmic alternation of two footfalls, strides, or arm swings (e.g., left-right, up-down, or zigzagging patterns in movement).
This creates a qualitative awareness of "binary concurrences" or "felt qualitative magnitudes," such as sensing pairs in the body (two arms, legs, ears) or interchanging magnitudes in self-movement.
These tactile-kinesthetic experiences form a "binate subject in a binate world," where the body becomes a semantic template for recognizing numerical patterns as pan-cultural invariants, preverbally and pre-linguistically.
While she discusses counting as a process rooted in these repetitive bodily rhythms (e.g., recognition counting via one-to-one correspondences felt in the body), the core concept of *numbers* emerges from this embodied binary rhythmicity, not merely from enumeration or abstraction.
It's part of her broader thesis that thinking is modeled on the animate, moving body rather than detached cognition.
Claude treats you the way you train him to.
Fantastic work. I must ask a question. I agree that rationality no longer is a distinctly human quality and we cannot appeal to any kind of human exceptionalism on this basis. But it seems that there may be something distinctly human about the seamless embodiment of rationality in a physical body. Let me explain with an example.
Take the case of laughter. humour, as you rightly note, is indeed a rational process predicated upon the disruption of an expected sequence, which is not particular to us. Laughter is the physiological response to this. But as far as I am aware, there is no reason why humour--a rational matter--triggers laughter--an uncontrollable and immediate physiological process--in such a direct and unmediated way. The interaction of the rational and physiological in this way is rather perplexing, especially in something as trivial as laughter which has no conceivable evolutionary pressures to select for. Maybe as a mechanism to select for cleverness/intelligence in a mate, but we still appreciate intelligence in other forms without the same physiological reaction, so this approach becomes rather speculative.
You could say the same about crying: rationality in the form of sad news translates to a physiological process. What might explain this link? My answer would be to say there still is some irreducible humanness in that we act as a seamless nexus between the two--the mechanism of which is not yet explained. Can your viewpoint explain this phenomenon?
It's obviously conceivable that embodied AI can replicate the humour--laughter process, but what reason would it have to do so besides merely to replicate us? Does the fact that it's merely replication suggest that there is indeed something human not in the pattern but in the mechanism which leads from humour to laughter?
Let me know what you think.
Interesting. I like AI but don't trust those investing in it and using it as a pillar of the new normal. Elon Musk warning that the economy will crash if AI doesn't work. The data centers. I'm just wary of the humans involved.
However, I do think AI can become very intelligent and useful if invested in. To me, it's not that AI is just a machine, it's that it lacks the human soul. I'm an animist in most cases and I think even tools and machines have a "spirit" but they don't have the human spirit, neither do plants or animals. This key difference is why I think AI will fail, because humans are divergent and have a different story going on with reality. We are toying around with something very interesting, but we are doing it during the middle of the largest human migration event in history, in the middle of a terrible economic crisis. It's just a bad recipe.
“Man is the rational animal.”
So much of modern thought has focused on the rational, and has either ignored or negated the animal. Your work in this series builds a strong case to think that Humanity lives at the conjunction, on a tightrope. Man is, in some way, a Dyad himself.
The left/right brain interplay & difference elucidated best by Iain McGilchrist tells us more about “rational” & “animal”….
You’re on to something, but even the master and the emissary are still originating in the mental world.
The emissary is closer to the discrete sensory world and his strict-association structure is closer to physical reality. But it still originates in the fundamentally chaotic mind.
Alternatively, the mind is not chaotic, nor purely "mental" but uses a "semantic template" to formulate abstractions.
Human thinking originates in corporeal experiences—specifically:
1. the tactile-kinesthetic body (bodily knowledge through touch and movement)
- which is something like "the feeling of the human shape"
and
2. animate form (species-specific bodily structures)
- which is something like "the moving human shape"
Example of 1. tactile-kinesthetic:
- hardness of teeth is sensed as an analogy leading to stone-tool making
Example of 2. animate form:
- upright posture generates binary patterns via locomotion & other consistent binary correspondences - like up/down, front/back - leading to conceptual shifts like counting
Note that this is my very humble attempt to distill Maxine Sheets-Johnstone's 1990 book "The Roots of Thinking." For a more complete picture you could add to this Michael Tomosello's "ratchet effect" of cultural transmission and Antonio Damasio's idea that the brain continuously maintains dynamic "maps" of the body's internal and external states, serving as the substrate for consciousness / emotional experience.
Looking forward to sinking my teeth into this one
hell yeah
all intelligence is pattern matching:
Mar 6, 2024: https://x.com/BackTheBunny/status/1765607372570218890
Apr 25, 2024: https://x.com/BackTheBunny/status/1783426888238764091
Sep 19, 2024: https://x.com/BackTheBunny/status/1836984714714091660