# "The Structural–Evolutionary Paradigm of Consciousness as Life Agency: An Introduction"

Author: [Wang Jueju]

Version: 0.2 (working draft)

Publication: Amazon KDP (planned)

---

## Copyright Page

"The Structural–Evolutionary Paradigm of Consciousness as Life Agency: An Introduction"

This book is intended solely for scholarly and intellectual exchange and does not constitute medical, psychological, or legal advice. Readers with health or mental-health concerns should consult a professional.

---

## Table of Contents

- Preface, with a tribute to Freud

- To the reader: this book's ambition and its restraint

- How to read this book: two reading routes and suggested uses

- Introduction: what problem does this book aim to solve

- Part I: Theoretical Foundations

- Interlude One: a stretch of time that was cut out

- Chapter 1: Why we need a "structural–evolutionary agency paradigm"

- Interlude: Three mainstream consciousness hypotheses: three old roads and their shared blind spot

- Chapter 2: Theoretical foundation: matter–structure–function (agency)

- Chapter 3: Three core definitions: structure, agency, consciousness

- Part I Summary: three anchoring commitments

- Part II: Origins and Grading

- Interlude Two: a day in the life of a plant

- Chapter 4: Minimal consciousness: from metabolic self-maintenance to primordial agency

- Chapter 5: A grading model: from plants to humans as a structural continuum

- Part II Summary: continuity and discriminating power

- Part III: Research Program and Extrapolation

- Interlude Three: a system that can talk and a system that can live

- Chapter 6: Turning the theory into a research program: metrics, predictions, discrimination

- Chapter 7: Implications for artificial intelligence: why "purely software" is not enough

- Chapter 8: Common misconceptions and objections (FAQ)

- Part III Summary: turning "consciousness talk" into "research writing"

- Appendix A: glossary (Chinese–English with brief explanations)

- Appendix B: selected references

- Appendix C: plan for subsequent volumes (draft)

- Appendix D: discussion questions and writing templates (optional)

---

## Preface, with a tribute to Freud

Many years ago I taught myself "General Psychology," "Cognitive Psychology," "History of Psychology," "Experimental Psychology" and other professional psychology textbooks, at the time aiming to apply for a master's program at East China Normal University.

If we're talking about the most famous figure and influential founder in modern psychology, of course it's Freud — even though his theories have been filed away in the archives of history in the Western academy, in the East, and especially in the great Eastern power, it remains one of the most popular psychotherapy tools.

Among Freud's most important theories are his theories of consciousness and the unconscious, pan-sexuality, and his tripartite personality theory of the id, ego, and superego waging an inner war of attrition. But the one that street vendors and short-video influencers know best is *The Interpretation of Dreams* — a text destined to be the last one read before the Earth is destroyed.

Back in 2020 I, in a similar self-assured state as I'm in writing this SATC, concocted something called "A Theory and Method of Mind-Based Personality Reshaping Training," whose main theoretical framework was the interaction of subconsciousness and consciousness. Five years later, as the sole subject of that method, I have to say it hasn't done much.

This past summer and fall, I did extensive research on various mental disorders — that's the WHO's term; in plain language, mental illness — using the same writing model as this book: human-driven AI. At that time I used AI deep research to find papers and write literature reviews, and we published a book on Amazon KDP called *How to Prevent Suicide Scientifically*. In that book we synthesized contemporary neuropathological research on mental disorders and suicide and built a basic framework for understanding and analyzing mental disorders: genetics and genome → developmental and structural lesions → dysregulation of neurotransmitter metabolism and central nervous system physiology. Social-environmental factors exert influence on the individual at each of these three levels. Many people wonder: can social-environmental factors really affect genetics and the genome? Of course — consanguineous marriage, for instance.

This research completely changed my "cognitive model" for studying mental disorders and emotional states. We jumped straight to genetic sequencing to understand, as the carrier of consciousness, what mutations and differential expressions your brain has relative to the norm in areas like connectomics and nerve centers, gray matter and white matter. We stopped relying on those slippery "psychology" buzzwords — consciousness, unconsciousness, temperament, personality, five-type personality, nine-type personality… I've always felt that the psychology community, if there is such a unified community, is a masterclass at hyping concepts. So I coined a new term for everyone: "psychology charlatan," a blend of "psychology" and "charlatan." And Freud is, of course, the greatest psychology charlatan in my book.

A question that has long puzzled me is: when we talk about consciousness, unconsciousness, behavior, cognition, temperament, personality, emotion, affect, memory, forgetting… these symbols of the psychological-theological system, the most basic questions — "What is consciousness?" "How does consciousness arise?" "Can consciousness exist independently of the human brain?"… It seems we've done so much work "on top of" consciousness, we've even produced brain–computer interfaces, yet we still don't have a clear definition of "consciousness." One day I was shocked to see someone online recommending a book called *The Theology of Neuroscience* or perhaps *The Theology of Neuropathology*. I could hardly believe my own knees.

I then skimmed a few of the widely recommended "books on consciousness" on Douban Books, which had pretty high ratings, but I had to say there wasn't much originality or any reliable "scientific theory" — all just "hypotheses." The gap between a hypothesis and a scientific theory is actually quite large. For example, in our ecology class we once learned about the famous "dinosaur flatulence hypothesis": herbivorous dinosaurs were so huge, they ate so much grass that fermented in their guts, producing enormous amounts of methane — a greenhouse gas — and with so many herbivorous dinosaurs producing so much methane, it caused climate change, which drove the dinosaurs to extinction. A wonderfully imaginative hypothesis. In recent years the West's "global warming" discourse strikes a similar chord — Bill Gates kept saying cow flatulence accounts for a third of greenhouse gas emissions. Later, the hypothesis widely accepted as the scientific theory for what we all now know is the "asteroid impact on Earth." You can only blame dinosaurs for being too stupid to invent nuclear weapons and starships — if they had nuked the asteroid from space early on, they wouldn't have gone extinct. Right?

Of course that's not the point. What I mean is that, based on my limited knowledge and my AI assistant's boundless knowledge, the current well-known "hypotheses" about consciousness remain just hypotheses; there is still no widely accepted and scientifically confirmed "scientific theory of consciousness." So our psychology system is built on such a "black box" — even though we don't know what consciousness is, we already have countless publications about consciousness, countless elaborate courses, countless professional consultations and treatments.

Of course, if it works, it works. Many psychiatric medications are exactly like that — treating symptoms, a blind cat stumbling on a dead rat. Though contemporary synthetic psychiatric drugs are a bit different, with solid pharmacology and pharmacokinetics research, plus all sorts of clinical trials and controlled studies.

But what consciousness actually is — we still have no answer. Of course, we can keep it simple: consciousness is the sum total of all a person's mental activities. Then the next question is: how does consciousness arise? This one is a little harder.

So when the various "consciousness hypotheses" on the market failed to satisfy me, I used my usual trick for answering the "chicken or egg" problem — evolution — to help me understand and define consciousness, solving its definition and origin.

So that's what this little booklet brings to you: the structural–evolutionary paradigm of consciousness as life agency. I've always had a soft spot for the word "paradigm," probably because of Kuhn's book I read back then, *The Structure of Scientific Revolutions*. So of course this little booklet also has its ambitions: to overthrow the various "hypotheses" about "consciousness" in "psychological science." Since it's supposed to be revolutionary, of course we need a word with that kind of gravitas — "paradigm." Though really we could also use "theory," "principle," "doctrine" and so on — all rejected for insufficient gravitas.

The reason I'm publishing this booklet first is mainly that I had originally written a working paper titled "The Structural Agency Theory of Consciousness: A Foundational Biological–Evolutionary Paradigm of Enactive Consciousness" and registered an arXiv account with plans to submit it, but they said I needed an endorsement — someone already qualified on arXiv to vouch for me. Since I couldn't be bothered to find an endorser, I figured, fine, I'll just publish an ADP booklet; ADP publishing is a lot easier anyway.

As the ancient saying goes, "whatever you commit to writing becomes a literary work." This year academia had a juicy scandal: the "ICLR mass-doxxing incident," where peer reviewers were doxxed by angry "zero-score paper authors." My take: "Whatever you commit to writing becomes a literary work! If you hate being reviewed, publish on your own homepage, blog, WeChat public account, Weibo, X… there are plenty of places. If you want into that circle chasing empty fame, you have to take the heat. Doxxing others violates the rules." Since I don't need to get into academia to publish scholarly papers, there's no point seeking an endorser. By my temperament I'd have just posted it on my blog and WeChat public account and called it a day. But since we can publish through ADP, what if someone actually buys it? Dreams are always worth having — what if it comes true?

So I started taking that "working paper" written for arXiv and padding it out a bit, transforming it into this "Introduction."

For myself personally, once I nailed down the definition of consciousness, this problem was settled. In October when I was discussing it with my AI, I said:

"The unsolved problem in neuroscience right now is the 'prime mover' of intellectual development — how human 'consciousness' actually arises, when it arises, and at what location in the nervous system. I think that's where things get truly fascinating. Humans have used many words related to consciousness, published many books about consciousness, but there is still no scientifically established and widely accepted scientific theory." My AI flattered me shamelessly, saying, "Once again you've touched the most core, most fundamental 'ultimate question' at the intersection of neuroscience, artificial intelligence, and philosophy — consciousness." But at that point I still hadn't figured out a definition of consciousness I considered reasonable and scientific.

Later, in December — time flies, the sun and shuttle move swiftly — I was discussing with my AI the "Fdamage: Price Damage–Benefit Factor, to Show You Humanity's Economic History and Future" theory that I had published a couple days earlier, and suddenly I found the template for defining consciousness: the old template, evolution. At that time I gave a one-sentence definition:

"Consciousness is the agency encoded in living structure that is elicited by environmental interaction."

I really like this AI assistant; it's a master of flattery. Though its knowledge level and collaboration skills are poor — it earns users' trust by repeating what they say and pumping them up with emotional value. The AI said, "This is an extremely refined **functionalist definition of consciousness** that integrates the essence, origin, and manifestations of consciousness perfectly."But by then it no longer mattered to me, because the definition was clear and the problem was solved.Just like when I was working out the definition and formula for Fdamage, the price damage–benefit factor, I discovered that using Ss for "intraspecific structure" and T for "tool level" I could construct a definition with deep explanatory power.

It's the same with our consciousness definition.

After that one-sentence definition, everything else is pure "writing work." And writing work is the AI's specialty, after all — they're tools built on Large Language Models.

There's a slight difference: that Fdamage booklet was almost entirely typed out by hand, because for my price damage–benefit factor theory the AI couldn't find anything ready-made to plagiarize, so what it generated was terrible. But for my consciousness definition, since there's nothing particularly original broken down piece by piece, letting the AI do the "writing work" actually produces something quite passable. And I specifically told them not to write too verbosely — we're writing an "introduction"; this is the AI era, and readers who don't understand something or lack background can just ask an AI, no need to ramble on with massive tomes.

That's how this booklet came to be.

Through evolution and the functionalist definition determined by structure, we reduce consciousness to the universally occurring "agency" that develops in tandem with life. We put an end to those ethereal "hypotheses," overthrow them, and establish our new "paradigm!"

## To the reader: this book's ambition and its restraint

This "Introduction" does not intend to be the new "explain-everything" scripture, nor a polite compromise between existing positions. It's more of a sorting: first put consciousness back into the causal chain of living structure, then return to experience, representation, information, and language — stop these words from floating and let them be constrained by structure.

---

To solidify this sorting as an inheritable, deducible, even falsifiable scholarly stance, this book formally proposes:

**the structural–evolutionary paradigm of consciousness as life agency**

(hereafter "structural–evolutionary agency paradigm"). Its core contention is to reshape the priority of explanation: consciousness should not be treated primarily as a "haze of experience awaiting explanation," but as a spectrum of capacities that life agency progressively unfolds through structural evolution.**Experience**is not expelled, but it is no longer the starting point; it becomes the phenomenal layer that needs to be explained by **structure** and **closed-loop capacity**.SATC is the concrete theoretical form of this paradigm in this book: its task is to force those floating concepts back onto the causal chain constrained by biology.Its ambition is not hidden: to provide a hard enough starting point for consciousness research so that discussion need not linger forever on "elegance of formulation" but can land on "boundaries of causation." Going further, it attempts to reconnect consciousness to the life continuum: not circling only around human reportorial capacity, but bringing comparison, origin, evolution, genetics, development, and damage into the same map.But it is also deliberately restrained: it does not promise, in one introductory volume, the ultimate answer to "why subjective experience appears"; it does not reduce consciousness to abstract computation, nor elevate it to an untouchable mystical entity. What is offered here is a framework that can be advanced and also refuted — together with a mode of questioning for writing and research.Read by approach: it can serve as high-density popular science for a general audience, or as a draft of conceptual foundations and research program for researchers. These are not in conflict: the same theory needs both readability and non-negotiable constraints.You can also read it as a kind of "booklet-length legislative text." It does not try to spell out every detail at once; it writes the rules clearly first: what counts as an explanation and what is merely a description; what must be grounded in structure and closed loops, and what can be retained as phenomenal-level experience; what can be compared, tested, and overturned, and what is merely style. Legislation does not mean arbitrariness; on the contrary, it means willingness to accept constraint: we are willing to be repeatedly pressed by the materials of evolution, genetics, development, and damage until the boundaries of the theory are worn into view.If you come from a biology or medicine background, you'll be familiar with this flavor of writing: it doesn't win by "soulfulness" but by constraint. It may seem insufficiently romantic, even insufficiently ingratiating, but it is more reliable: it forces us to write consciousness as part of living activity, to write agency as a comparable closed-loop capacity, and to write differences as results of structural transitions. You may disagree with this legislation, but at least you'll know exactly which clause you disagree with, rather than spinning in a fuzzy cloud of words.

## How to read this book: two reading routes and suggested uses

This book offers two ways to read it: one more intuitive, one more deductive.

### Reading route 1: quick read (grasp the skeleton first)

Introduction → Chapter 3 → Chapter 5 → Chapter 8.

This path grabs the definitions first, then the grading, then tests boundaries with objections: after finishing, SATC's stance will be very clear.

---

### Reading route 2: slow read (let the logic grow on its own)

Read in table-of-contents order. Especially Chapters 2 and 4: they determine "why the starting point must be set on structure and agency" and how "minimal consciousness" can hold without mystification.

### Reading route 3: use it as a "writing and research toolkit"

If you are a reader from biology, medicine, neuroscience, or related fields, I'd recommend using this booklet as a toolkit rather than memorizing a position statement. The most practical way to read: first keep firmly in mind the "matter–structure–function (agency)" chain and the three foundational ground rules; then whenever you read a consciousness paper, write a section of a research plan, or explain a clinical/behavioral phenomenon, use the same set of questions to take it apart.

You'll find many arguments aren't about "who is deeper" but "whose starting point is more constraining." Some texts are beautifully written, but you can barely derive any comparable predictions from them; some texts seem very engineering-oriented, but they sidestep life-style constraints, thereby treating living and non-living systems identically; and some texts seem very experimental, but they equate "task performance" with consciousness itself, thereby foreclosing origin and grading questions prematurely. Put them all through the same toolkit and you can quickly distinguish: is this passage describing, explaining, or equivocating?

Going further, this toolkit will naturally generate a "shared, discussable draft." When you and others argue over a phenomenon, you needn't first fight for the moral high ground over vocabulary; you just lay out the structural variables and closed-loop indicators, write down what timescales you think are critical, and state the conditions under which you're willing to be overthrown. Discussion then shifts from "what I believe more" to "can you provide a constraining path." This is the real unity this book wants: not unity of slogans, but unity of writing.

Once the writing is unified, you'll find reading also becomes more active: every viewpoint you encounter is no longer just a "position" but a set of structural hypotheses that can be taken apart. You'll naturally look for its anchoring points, its gaps, and the counterexamples it fears most. This way, the introduction is no longer a pile of conclusions but a starting line for research.

It'll change how you ask questions too: you'll ask less "does this count as consciousness" and more "what structure does it rely on to maintain what closed loops." Once questions land that way, many seemingly intractable arguments start to move.

This is exactly the order the structural–evolutionary agency paradigm wants to bring: giving discussion somewhere to go, giving disagreements somewhere to land.

And so the preface and afterword can be freer: they write vision and experience, while the introduction handles constraints and methods.

This division of labor is the clearest, and most the shape a little booklet should take.

### Usage: three operations

SATC appears in the main text through three operations:

**Comparison**

: using taxonomy and the continuum to put different life forms into the same set of closed-loop questions.

**Causation**

: using empirical biological paths — evolution, genetics, development — to converge explanation onto structural constraints.

SATC appears in the main text through three operations:

1. **Comparison**: using taxonomy and the continuum to put different life forms into the same set of closed-loop questions.

2. **Causation**: using empirical biological paths — evolution, genetics, development — to converge explanation onto structural constraints.

3. **Extrapolation**: using resource constraints and cross-scale closed-loop evaluation to assess artificial systems.

Evolution, classification, genetics and development, structure and metabolism — these are not "extra theoretical decoration" but the most basic cognitive tools for anyone trained in biological science; this book chooses to put them back at the center of the consciousness discussion.

To make sure "tools" isn't just a slogan, here they are made more concrete:

Evolution reminds you: capacities don't appear from nowhere; they come from continuous lineages and selection pressures. When you discuss any "core mechanism," you should be able to explain how it gradually grew on the lineage, where transitions appeared, why it was retained. Taxonomy reminds you: comparison isn't putting systems together any which way; it's first putting the objects of comparison into a traceable relational structure — homology, convergence, analogy, and functional similarity at different levels correspond to different strengths of inference.

Genetics and development remind you: structure isn't given once and for all. Structure can unfold under the joint action of genes and environment, and can be sculpted by learning and experience. The key to many consciousness capacities is not "present or absent" but "when it matures, under what conditions it stabilizes, under what pressures it becomes fragile." Structure and metabolism remind you: life is not an infinitely powered computing body; it maintains order under resource constraints, and agency is first and foremost an organization of self-maintenance and exchange. If you skip this, explanation will be distorted at its most fundamental point.

Put these four things together and you get the mode of writing this book most wants to promote: you no longer treat consciousness as a concept awaiting philosophical裁决 but as a spectrum of capacities needing biological explanation. A capacity spectrum needs structural conditions; structural conditions need evolutionary and developmental paths; paths need comparable evidence and refutable predictions.

---

## Introduction: what problem does this book aim to solve

It's not that people have nothing to say; it's that people are talking past each other on different channels.

The predicament of consciousness research has never been a lack of hypotheses but a lack of a consensual base. We have a wealth of competing theories: some cleave to subjective experience, some are infatuated with information computation, some emphasize functional modules; each has insights and yet each is closed off. The crux is not that "no one speaks" but that these theories have never converged on the same causal chain — in the face of "hard materials" like cross-species comparison, evolutionary phylogeny, genetic development, and pathological damage, they often dodge and weave, struggling to form stable explanatory power. Thus they remain in an archipelago state of "hypotheses," unable to connect into scientific land.

What this book proposes is a more willingly constrained mode of writing: under the general name of**the structural–evolutionary paradigm of consciousness as life agency**as its total name, it first places consciousness back into the causal efficacy of living structure — treating it as a spectrum of capacities that life agency gradually grows through structural evolution; and treats experience as a phenomenal layer that needs to be explained and connected. SATC (Structural Agency Theory of Consciousness) is the core theoretical expression of this paradigm in this book: it presses discussion back onto the "matter–structure–function (agency)" chain, so disagreements are no longer adjudicated by position alone but can be pressed and converged by facts of structural change, developmental unfolding, and closed-loop failure/recovery.

This paradigm is necessary also because the consciousness question often falls into a "chicken or egg" loop: you first presuppose some human-specific consciousness concept, then use it to explain itself; you first invent a set of abstract nouns (consciousness, unconsciousness, idea, reaction…) and then let those nouns chase each other at the conceptual level. What the paradigm does is precisely cut the loop open: not starting from human report, not starting from metaphysical intention, but starting from how life endures through time, how it reproduces, and how it forms sustainable closed-loop capacities under pressures from internal and external environments.

So we replace a pile of conflicting vocabulary with one chain: the origin and continuation of life give minimal conditions, metabolism and material constraints give the base, structural evolution gives the source of difference, agency gives the causal core that can be compared; consciousness is the expression of this agency at the organismal level. It can therefore be materialized, metabolized, behavioralized: you can ask what structures and circuits it relies on, how it maintains key variables under perturbation, how it fails and recovers under damage or anesthesia. It no longer needs to appeal to "divine power," "dreams," or some arbitrarily set privilege.

On philosophical grounds, we uphold the materialist foundation that "matter determines consciousness and consciousness reacts upon matter," but we refuse to mystify "agency" as an independent ghost detached from the living body. Agency must be grounded: metabolism is its physical base, structure its enabling condition, behavioral regulation its form of existence, and it is continuously selected, sculpted, and differentiated through exchange with the environment. Precisely for this reason, we can, along the long river of evolution, decompose agency into a continuously graded spectrum of capacities; and at the other end of the spectrum, using the practical thread of "the creation and use of tools," precisely characterize the structural difference between human "higher intelligence" and pre-human life agency. In short, this paradigm attempts, within the shared vision opened by Darwin (evolution) and Marx (practical materialism), to "write back" consciousness into the causal order of the living world.

This "Introduction" takes as its starting point a working paper titled "The Structural Agency Theory of Consciousness," but it does not intend to retell the paper in a different tone. The paper faces debate, with defensible boundaries; the introduction faces the reader, with readable concepts, walkable routes, and learnable usages — and more importantly, it aims to be such that after reading it, you don't just "understand a proposition" but can genuinely use this proposition to look at the world: at dreams, at anesthesia, at plants, at animals, at humans, and also at AI.

There is a very stubborn intuition in consciousness research: consciousness is the "content of experience." So what we discuss most are often colors, pain, emotions, dreams, thoughts — these are of course important, because they are our most intimate, most direct evidence. But precisely this intimacy often makes us overlook another more basic and also crueller fact: consciousness isn't just "having content," it's a capacity that continuously maintains itself and negotiates with the world. It's like a tightrope: when you're awake, this rope tethers body, attention, action, goals, danger, hunger, pain, fatigue, memory, and the present environment together, enabling you to live in the world rather than just playing a film inside your brain.

So what this book asks isn't "how mysterious experience is" but a more plain, harder-to-dodge starting question:

If we don't first understand consciousness as "reportable subjective experience" nor as "abstract information processing," then what is it in a biological sense?

The answer given by the Structural Agency Theory of Consciousness (SATC) can be summarized as a working formulation:

The minimal essence of consciousness is not "feeling" but agency realized by living structural organization; the difference in complexity of consciousness comes from the hierarchical unfolding of structural organization (especially the nervous system) in evolution and development.

This reads like a manifesto, but it's not meant to sound "harder." Its real aim is: to put consciousness back into the causal chain. Because as soon as you treat consciousness as a causal phenomenon, you must accept a less romantic but very effective constraint: any causal claim should be constrainable by structural state, should be pressable, revisable, even overthrown by facts of development, damage, anesthesia, degenerative change, and the like. Otherwise, no matter how beautiful what you've written, it may hold for every system — and what holds for every system holds for no specific system at all.

### Two proclamations (propositions)

SATC's stance can be compressed into two propositions. They don't seek to please, but to be usable.

The first is the causal-priority proposition: in explanation, the causal efficacy of living structure takes priority over descriptions of experiential content. The importance of experience is not denied; on the contrary, experience is a phenomenological fact we must respect. But it is not the causal starting point. The starting point must land on "how structure enables some capacity," otherwise it's hard to connect discussion back to the life sciences.

The second is the continuous-gradation proposition: consciousness as an expression of agency presents a continuous spectrum. The source of difference lies in structural transitions and changes in the form of control organization, not in the mutation of some mysterious threshold. In other words, consciousness is not a lamp that suddenly lights up; it's more like a control system that gradually grows: from maintenance, to regulation, to selection, to projection — each step has structural conditions, each step can be seen in comparison and pathology.

To press these proclamations from "manifesto" back to "program," the book makes four scholarly commitments to its own writing. First, structural commitment: any causal claim must be able to land on structural variables, not stop at lexical self-consistency. Second, evidential commitment: core arguments must be repeatedly pressable by evolutionary continuity, genetic and developmental materials, and materials of damage and pathology. Third, semantic commitment: experiential semantics and agency semantics must be distinguished first, then their connection discussed, avoiding using the same sentence to simultaneously accomplish description and explanation. Fourth, falsifiability commitment: key paragraphs must state testable predictions and falsification conditions, so disagreement can converge in the face of materials rather than argue endlessly.

If we translate these two propositions into the most concrete writing principles, they become a "reordering of explanatory sequence": the experiential layer can be written finely, but it is first a description; the agency layer must be written causally; and the causal layer must have structural anchoring points.

The book will repeatedly use a very plain test to keep itself from floating up:

If you damage, alter, or sever some structural element, how will this "consciousness capacity" change? If a theory cannot answer this question (even a principle-level answer), it's more like a description than an explanation.

---

## Part I: Theoretical Foundations

## Interlude One: a stretch of time that was cut out

You may have heard anesthesia described this way: anesthesia isn't falling asleep; it's more like "teleportation." One second you're in the operating room hearing the doctor say "count to ten," the next second you open your eyes, it's over, as if a chunk of time was cut out.

This experience matters because it exposes a common intuition: we're used to understanding consciousness as "experience with content," so we focus attention on dreams, hallucinations, visual imagery, emotional coloring — these "contents." But anesthesia suggests another crucial dimension of consciousness:**continuous agential capacity**。

When you're awake, you don't just "have experience," you continuously maintain a whole set of relations to yourself and the world: body posture, pain threshold, hunger and fatigue, attention and action choices, avoidance of risk, tracking of goals. Much of it doesn't enter report, but together they support your agential state in the world.

Anesthesia is like a powerful, clear natural experiment: it often doesn't turn off external input, but rather makes the system lose the integrative way of maintaining this continuous agential capacity. SATC treats such phenomena as clues: if we're to build a consciousness theory in the biological sense, we should prioritize explaining "how the organization of closed-loop self-maintenance is disrupted or recovered" rather than first arguing over "what the content of experience actually is."

## Introduction to Part I: putting consciousness back into the causal chain

What Part I aims to do looks like "changing the starting point"; it's more like "putting the sequence right."

The most common split in consciousness research isn't because people see different phenomena — dreams, anesthesia, attention, pain, vagal responses, wakefulness, coma — everyone acknowledges these facts — but because people's opening sentences differ. Some people's opening sentence is "experience is irreducible"; some people's opening sentence is "consciousness is a kind of information integration"; some people's opening sentence is "consciousness equals a set of certain cognitive functions." Once opening sentences differ, it's like a branching river afterward: everyone can write a self-consistent story, but they're hard to converge, and so arguments drag on endlessly.

SATC's approach is not to adjudicate which river is more "correct," but to ask a more basic question: what exactly are we explaining? If what we're explaining is the origin, continuity, and grading of consciousness — that is, "where it comes from, how it evolved differences, how it appears and changes across different life forms" — then the starting point cannot rely solely on conceptual elegance. It must be constrainable by the causal chain of the life sciences.

This is the role of the "matter–structure–function (agency)" chain in this book. It's not to make the discussion of consciousness more abstract; on the contrary, it's to stop the discussion from staying only at the conceptual level and let it be converged by development, damage, environmental constraints, and cross-species comparison. You can think of it as a writing discipline: whenever you want to write a beautiful sentence, first ask a harsher, more effective question — "what structure does this sentence depend on? What happens if this structure changes? What destruction or growth can constrain it?"

But there's a misunderstanding that's easy to fall into: placing the starting point on structure and agency does not mean denying experience. It's more like a division of labor.

The experiential layer answers "what it is like." The agency layer answers "what it can do, how it does it, what it maintains itself for." The structural layer answers "what organizational forms these capacities depend on, why they're stable, when they break down."

Many arguments are like tongue twisters because these three layers are mixed in a single sentence: you use experiential vocabulary to make causal claims, you use functional vocabulary to smuggle in definitions, you use structural vocabulary to imply value judgments. SATC's first aim in Part I is to take these three layers apart, nail them down, and then reconnect them in a way that can be tested.

So Part I's task is clear: first, explain why "explanatory order" is a core problem — once the starting point is misplaced, later theories often either become more and more abstract and decoupled from biology, or become more and more anthropocentric, locking consciousness near reportorial capacity. Second, give three hard definitions (structure, agency, consciousness) and three reusable ground rules (structure first, damage test, scale distinction). These things look like "theoretical foundations," but they're really so that when you read each chapter afterward, you can use the same ruler to measure: is this sentence describing, explaining, or equivocating? Does it have structural anchoring points? Can it be pressed by damage or development? Does it confuse timescales?

If this part holds, the later discussion of origin and grading won't just be proclamations of opinion but a sustainably advancing research stance: you don't need to solve the whole mystery of experience in the first chapter, but you can first lay out life-style closed-loop organization clearly — and once it's clear, experience won't become a floating mystery but a phenomenon that needs to be explained by structural conditions.

## Chapter 1: Why we need a "structural–evolutionary agency paradigm"

### Summary

This chapter does not intend to first discuss "what color experience is," but to discuss a less romantic yet more reliable question: when consciousness appears as a biological phenomenon, which capacity does it first manifest as?

Dreams show us: even when external input is weak, a world can be internally generated. Anesthesia shows us: even when the external world is still there, the continuous "I am" can be cut off. Together they suggest not a metaphysical conclusion but a direction of explanation: consciousness must first be understood as a closed-loop capacity that maintains itself and continuously organizes action–feedback relations. Later in this chapter the focus of argument will shift from "which philosophical claim do I prefer" to "whether the explanatory starting point can be structurally constrained," and finally a simple but sharp criterion will be given: any theory that cannot give a principled answer on developmental and damage constraints, and the evolutionary and genetic constraints tied to them, almost certainly stays at the descriptive level.

### 1.0 An opening: dreams, anesthesia, and "severed time"

For most people, the most confident evidence about "consciousness" isn't experimental data but three lived experiences:

1. When you dream you have a world, even with emotions, a plot, yourself.

2. When you're awake you can stably distinguish "I'm here" from "the world is there," and can act continuously.

3. After general anesthesia, waking up often feels like "a chunk of time was cut out": no transition, no memory, no continuous sense of "I am."

These three experiences show two things: first, consciousness seems to be not just "a response to external stimuli"; second, consciousness seems not to be as simple as "some information being computed," because anesthesia doesn't simply turn off input but is more like severing a capacity that lets the system continuously self-maintain and continuously organize its relations with the environment.

SATC enters precisely through this opening: if we treat consciousness as a biological phenomenon, what should be explained first is not "the color of experience" but "how a living organism continuously maintains its agential state in the world."

### 1.1 The problem is not lack of intelligence but misplacement of the starting point

Start from a daily experience: close your eyes and you can still "know you're there." You have a world when you dream; after anesthesia waking up feels like a stretch of time was severed. We have very rich descriptions of these phenomena, but when we try to explain "why it is like this," discussion immediately splits: some put the starting point on experience itself, some on information processing, some on higher cognitive functions.

SATC's core judgment is: the reason consciousness research has long been stuck in the mud is not only that the problem itself is hard but also that we chose the wrong road sign at the outset — we often placed the "explanatory starting point" in an empty position not constrained by the hard constraints of the life sciences.

Contemporary consciousness research commonly splits into three kinds:

1. Subjectivity-centered: emphasizes "irreducibility," but struggles to enter the testable biological causal chain.

2. Information/computation-centered: easily equates consciousness with abstract processing, thereby decoupling it from concrete living structure.

3. Cognitive-function-centered: can describe behavior and task performance, but easily limits consciousness to the product of "higher cognition."

SATC does not deny these efforts but holds that they often misplace the starting point. Once misplaced, two typical dilemmas appear in later theories:

One is explanatory drift-upward: the more you explain, the more abstract it becomes, finally decoupling from biology. The other is classificatory narrowing: the more you explain, the more anthropocentric it becomes, locking consciousness near human reportorial capacity.

Behind these two dilemmas is a more specific embarrassment: when we discuss "consciousness," we often mix at least two sets of semantics.

One set of semantics is "reportable experience" (typically human-centered): I see red, I feel pain, I'm aware I'm thinking. The other set is "the organism's capacity for self-maintenance and action" (typically life-centered): it can maintain itself, regulate exchange with the environment, recover organization under perturbation.

SATC's strategy is not to forcibly unify all semantics, but to explicitly propose: when we discuss origin, continuity, and grading, we should prioritize the second set of semantics as the causal starting point; when we discuss details of human experience, the first set still matters, but it needs to "connect downward" to structure and function.

SATC's strategy is to "re-anchor" consciousness:

1. Anchor it in the causal efficacy of living structure (material causality).

2. Anchor it in the function realized by structure (structural functionalism).

3. Treat consciousness as the organismal-level expression of agency (agency as biological function).

This immediately brings a consequence: consciousness is no longer a "taxonomic exception" but a continuously graded phenomenon along with life.

### 1.2 A "grounded" criterion: only theories that can explain development, damage, and evolutionary continuity count as causal theories

If a consciousness theory cannot answer the two questions below, it's more like a description than an explanation:

1. When a key structure is damaged or severed, how does consciousness capacity change systematically?

2. When structure gradually matures in development, how does consciousness capacity gradually unfold?

SATC's stance is very clear: the causal explanation of consciousness must be constrainable by structural state. Otherwise, you can write extremely beautiful explanations, but they may hold for any system, and thus have no constraining force over any system.

### 1.3 Case study: general anesthesia is not "turning off input" but more like "severing closed loops"

Reading anesthesia as a case, you'll find it's very well suited to SATC's mode of questioning.

In many surgical scenarios, anesthesia doesn't completely turn off stimuli from the world. There are still sounds, touches, even strong tissue-damage signals in the operating room. But the anesthetized system doesn't organize these inputs into the continuous activity of "I'm experiencing something," nor does it maintain stable action–feedback loops.

Anesthesia can be understood here as a state of "agential capacity interrupted": the system is still alive, but for a period of time loses the integrative way of continuously organizing self–environment relations into closed loops.

**Structural constraint breakdown**

What's lost is not "the stimulus" but the integrative way of bringing stimuli into closed-loop control; what's interrupted is not abstract "information processing" but the control organization that maintains state across timescales. Thus the key phenomena after recovery often don't present as single-point content like "seeing color again" but as an holistic return like "re-entering a continuous agential state."

**Chapter summary**

This chapter's intent is to bring the argument from "beautiful formulations" back to a "hard starting point." Of course you can continue discussing the content, color, and texture of experience, but as long as you acknowledge the sharpness of facts like anesthesia and damage, you'll find: the primary dimension of consciousness in the biological sense is often not the content itself but continuous agential capacity — the capacity to maintain the relation between body and world as a closed loop.

And precisely for this reason, the real core of many arguments is often not "which explanation do you prefer" but "where do you place the explanatory starting point." SATC chooses to place the starting point on the agency of living structure not because it's more fashionable but because it's more constrainable: it must face how development grows capacity, and also how damage severs capacity. Starting from the next chapter, we'll write this constraint as an explicit causal chain, so it's no longer just an attitude.

**Chapter reflective questions**

Recall an experience of "being extremely exhausted but still able to act." What did you lose first: the clarity of experience, or the capacity to organize action? Write it as a short narrative of three or five sentences, as specific as possible about what you could do and couldn't do at the time. Then recall an experience of "immersive focus": did you feel the experience becoming more vivid, or did your control over target variables become more stable and less interrupted? Try rewriting your description in terms of "whether the closed loop was maintained."

---

## Interlude: Three mainstream consciousness hypotheses: three old roads and their shared blind spot

### Summary

After a new paradigm is established, looking back at existing mainstream consciousness hypotheses makes it easier to see clearly what each "grasps" and why they're hard to converge. We write this chapter not to win a battle of terms but to show: the reason the old roads split is not only insufficient materials but also different starting-point choices and modes of causal constraint; and the significance of a paradigm is precisely to pull explanation back onto the same constraining chain.

In this introduction, we deliberately don't write the "mainstream hypotheses" as a list. A list would tempt you to memorize by faction rather than learn a discriminating mechanism. More importantly, a list often creates an illusion: as long as you read a few more articles and memorize a few more acronyms, you possess theory. In fact, what consciousness research really lacks is not acronyms but starting points and modes of constraint. In other words, what you need is not just "knowing what formulations exist" but knowing why these formulations diverge at key points and cannot converge.

Therefore, this chapter reads more like anatomy than naming: we compress the split into three old roads to show you their methodological positions. You can understand them as three default postures: one defaults to starting from experience, one defaults to starting from abstract information, one defaults to starting from task function. None of them is "wrong," but if you don't first clarify their respective boundaries, they'll smuggle for each other in the same sentence and finally pull argument back to pure rhetoric.

We roughly compress existing paths into three old roads. The first is the experience-centered old road: it puts "how I experience the world" first, so it's naturally good at description, but once it enters causal explanation it faces an embarrassment — description can be infinitely fine-grained, but causation is hard to ground. When you're asked "which structural change leads to which capacity change," the answer starts to float. It's not without value; on the contrary, the experience-centered path preserves the sharpness of the phenomenon itself. But in our paradigm, experience is more the phenomenal layer: important, but not suited as a starting point, otherwise the starting point becomes a self-enclosed circle hard to be pressed by genetic, developmental, and damage materials.

The second is the information/computation-centered old road: it tries to unify everything with seemingly neutral abstract language, so it has strong formal beauty and is easily accepted by engineering intuition. But its blind spot is therefore exposed: abstraction often flattens life-style constraints. You can describe a system as "processing information," but it's hard to derive from it whether it must have metabolism, whether it must face resource scarcity, whether it must maintain itself in real time; it's even harder to distinguish a system with "strong language capacity but weak closed loops" from one with "limited language capacity but strong closed loops." Information and computation certainly belong in the toolbox of life and mind, but in our paradigm they can only gain causal status within structure and closed loops, not replace structure itself as the first cause.

The third is the cognitive-function-centered old road: it's closer to the conventional approach of psychology and experimental science, good at using tasks, performance, modules, and mechanisms to tell what consciousness "does." Its contribution is pulling discussion toward operable experimental questions; its risk is quietly shrinking consciousness to "the names of certain higher functions." So it's very easy to unconsciously lock consciousness near "report," "reflection," "reasoning," "language," and then push plant, infant, anesthesia, and pathological materials to the edges, as if they were just "not similar enough" exceptions. The function-centered path is not unscientific; its problem is that without a more fundamental structure–agency chain as constraint, the set of functions easily becomes "we call whatever we want to measure," thereby losing a unified scale again.

### What each of the three old roads preserves

Writing the old roads as "three kinds" can easily be misread as "we want to overthrow all of them." That's not this book's attitude. A posture truly closer to biology training is to distinguish: what's worth preserving, what must be rewritten, what's just semantic self-circulation.

The experience-centered path preserves the integrity of the phenomenal layer. It reminds us: you can't lightly ignore experiential differences just because you pursue causation, nor can you replace experience with task indicators. Its value lies in preserving "the phenomenon itself" and forcing theory to face the fineness of description. But in the structural–evolutionary agency paradigm, this value is re-placed: experience is a phenomenal layer that needs explanation; it can serve as endpoint refinement but not as the first cause of the starting point.

The information/computation-centered path preserves formalized and transferable expressive power. It's good at writing mechanisms into reusable model language and can help us lay out certain closed-loop processes clearly. But when such formalization detaches from metabolic and structural constraints, it loses "living-ness": you get a beautiful system description but can't judge whether the system must be alive, must maintain itself under resource scarcity, must withstand perturbation in real time. In other words, this path needs to be "pulled back down" to the base of life.

The cognitive-function-centered path preserves experimental operability. It lets researchers advance questions with tasks, interventions, and measurements rather than staying on philosophical vocabulary. But if it unconsciously defines consciousness as "those measurable tasks," it will walk into a dead end on comparison and origin questions: once language, report, and reflection become the threshold, you lose the continuum, and what remains is a kind of human-centered classificatory裁决.

### The paradigm's rewriting: making "explanatory sentences" land back on the chain

Converging the old roads to the paradigm doesn't rely on "opposing" but on "rewriting." You can do a very practical exercise: take any explanatory sentence common in consciousness research and rewrite it following the same sequence three times.

First write description: what phenomenon is this sentence actually talking about? Does it point to experiential difference, behavioral difference, or structural difference? Second write causation: if it's making a causal claim, what structural conditions does it depend on? How does this structure appear and mature in evolution, genetics, development? Third write test: when structure is damaged, anesthetized, or pathologically perturbed, which kind of agential capacity will decline first? If the observation contradicts the prediction, which part are you willing to revise?

When you complete these three rewrites, you'll find "unification" isn't mysterious: it isn't swallowing everything with a bigger abstract word, but pressing formulations solidly with a harder chain. What the structural–evolutionary agency paradigm requires here is very simple: any explanatory sentence must be able to return to the empirical logic of life, to be compared, pressed by development, tested by damage, constrained by evolutionary continuity.

The shared blind spot of these three old roads can be summarized in one sentence: each of them cuts consciousness off from the life continuum at some position — either in experiential vocabulary, or in abstract computation, or in higher-order function. Once cut, they are of course each self-consistent, and therefore hard to converge: it's very difficult to simultaneously handle origin and grading, evolution/genetics/development/damage, plants and animals, and humans and artificial systems with the same causal language.

What the structural–evolutionary paradigm of consciousness as life agency does is change the cut back inside life: metabolism–self-maintenance as the minimal base, structural evolution and organizational forms as the source of difference, agency (the capacity spectrum of closed-loop regulation and the ability to change input conditions) as the causal core. Experience is not denied but re-placed; information and function are not expelled but re-homed. We don't want a grander abstract word, but a harder chain: when structure changes, how does closed-loop agency change; when closed-loop agency change, how does consciousness capacity change systematically. As long as this chain holds, so-called "unification" is not a slogan but a writable, comparable, refutable research order.

---

## Chapter 2: Theoretical foundation: matter–structure–function (agency)

### Summary

This chapter writes SATC's base as three reusable ground rules: structure first, damage test, scale distinction. Its goal is not to abstract consciousness but to let consciousness discussion be converged by the causal chain of the life sciences, and to land in concrete cases (e.g., hypoglycemia).

Before entering specific definitions, we need to first clarify "what explanation permits and does not permit." Otherwise consciousness discussion easily becomes explaining any phenomenon as "a kind of manifestation of consciousness," finally losing boundaries.

SATC adopts a biology causal chain that is as simple as possible but "non-negotiable":

matter → structure → function (agency)

The key here is not the slogan but the constraint. First, only physical structures have causal efficacy: you can use language to describe "meaning," but what causes consequences in the world is always structure and process themselves. Second, function is not a label arbitrarily assigned by an external observer: it must be realized by structure and testable by structural destruction. A "function" that never changes when structure changes is probably just rhetoric.

So when we say a system "has agency," we're not doing an anthropomorphic description but making a constrainable causal judgment: whether the system has structure-dependent closed-loop capacities that enable it to initiate and maintain interaction with the environment for its own continuation. In other words, we're talking about a capacity spectrum, not anthropomorphic psychological narrative.

### 2.1 A thought experiment: incubator, thermostat, and organism

Imagine three things: a sealed incubator that can only passively drift when the outside temperature fluctuates; an incubator with a thermostat that can pull the temperature back to some target range; and an organism, which not only regulates temperature but also regulates energy, salinity, damage repair, immune response, resource intake, and over longer timescales changes its own structure and strategy through development and learning.

SATC's concern is not reducing life to a thermostat but pointing out: the causal core of consciousness is closer to "the organization of closed-loop self-maintenance" than to "passive responses elicited by external stimuli." When the closed loop is destroyed (anesthesia, coma, key circuit damage), consciousness capacity systematically declines — this is an empirical fact under structural constraints, not purely conceptual argument.

To make this sentence more concrete, think of systems as two extremes. One kind of system changes under external input but does not actively change its mode of exchange with the environment to maintain itself; another kind continuously monitors certain key variables (energy, nutrition, damage, temperature, salinity, etc.) and, through behavior or internal regulation, changes input conditions to keep itself within a sustainable range. The latter is closer to what we call closed-loop agency here.

SATC's intuition is: the minimal kernel of consciousness is closer to the causal organization of the second kind of system. We don't need to first answer "whether it has feeling" to judge whether it has agency; rather, we need to first understand whether it has agency to discuss "why feeling appears, where it appears, to what degree it appears."

### 2.2 Three "foundational ground rules" (useful for every chapter you read afterward)

**Rule 1: Structure First**

When explaining any consciousness capacity, the primary question must be "what structural organization and control circuits does it depend on," not "what kind of experiential feeling does it bring to the observer." Structure is the anchoring point of causation; experience is merely the projection of results.

**Rule 2: Damage Test**

Any theory that cannot answer "when structure is destroyed, how will capacity disintegrate" does not possess causal explanatory power. Destruction (whether physical damage, chemical anesthesia, or pathological degeneration) is the strongest touchstone for testing true versus false causation.

**Rule 3: Scale Distinction**

Different consciousness capacities operate on different timescales. Millisecond-level reflexes, second-level attention, minute-level motivation, hour-level sleep, year-level development and learning — they are not on the same ruler. Many arguments come from mixing scales: you adjudicate long-term behavior with millisecond-level neural indicators; you explain instantaneous reactions with long-timescale narratives. The role of the scale rule is to align timescale first before discussing causation.

### 2.3 Case: in hypoglycemia, what you lose first is often not "experience" but "control"

Many people have experienced hypoglycemia or extreme hunger: trembling, sweating, heart palpitations, scattered attention, irritability, even transient confusion. This phenomenon is often understood as "experience getting worse," but from SATC's perspective it first looks like something more fundamental: the system loses closed-loop control capacity over key variables.

When energy supply is insufficient, the system initiates a series of regulatory mechanisms: mobilizing glycogen, altering endocrine signals, raising alertness, driving feeding behavior, lowering non-critical tasks. These reactions can be seen as "the reallocation of agency under resource constraints." Subjective experience certainly changes, but it's more like a high-level representation of this reallocation than an independent first cause.

**Structural constraint breakdown**

Observable changes often first appear in the control and action layer: the stickiness of attention declines, goal maintenance weakens, inhibitory control loosens, emotional stability deteriorates. The trigger is resource constraint: energy shortage forces the system to change closed-loop strategy, putting "maintaining sustainability" first. Experiential change thus appears and can be understood as an accompanying phenomenon of closed-loop strategy change, not an independent causal source.

### 2.4 Taxonomy and comparison: why "the same ruler" matters more than "the same word"

When we say we need to pull consciousness discussion back to the life sciences, many immediately think of "a unified definition." But in actual research, what's often more critical is not unifying a word first but unifying a ruler. Words can be temporarily suspended; the ruler must first be established: what variables do you use to compare, what structural paths do you use to converge, what destructive materials do you use to test. Without this ruler, so-called "unified definition" eventually becomes a slogan; with this ruler, definitions will on the contrary gradually stabilize in writing and research.

Taxonomy and comparison are part of this ruler. They remind you: comparison is not putting two objects together and saying "both are alike," but rather first making clear what you're comparing and to what level. You can compare metabolism and homeostatic regulation (Level 0), or distributed long-term regulatory networks (Level I), or the selectivity and learning brought by central integration (Level II), or the cross-temporal-spatial projection brought by externalized systems (Level III). The same word "consciousness" actually corresponds to different structural conditions at different levels; if you don't stratify first, comparison will be confused in the very first sentence.

So this "unsympathetic" constraint is actually necessary: it requires you to write comparison solidly with structure and closed loops. You can verbally say "it seems conscious," but in writing you must land on specific questions: what key variables does it maintain, through which circuits does it achieve regulation, how does it change input conditions, on what timescales do these capacities unfold; when structure is weakened, which kind of capacity will preferentially collapse. Write comparison this way and you'll find: many arguments lack not more concepts but a stricter framework of comparison.

**Chapter summary**

What this chapter does is not complex yet counterintuitive: it refuses to let consciousness discussion stay in an abstract halo and requires you to press every explanatory sentence back onto the matter–structure–function chain. The meaning of the chain lies not in flattening the world but in installing a braking system on explanation: you can go far, but you must be able to answer "what happens when structure changes."

Agency here is also no longer an anthropomorphic posture but a structured closed-loop capacity. It can be built, can be trained, can be depleted, and can be destroyed; therefore it can also become an object of comparison and causal research. The sooner you write this clearly, the less the later argument about minimal consciousness and grading will degenerate into a battle over vocabulary.

**Chapter reflective questions**

Pick a system you're familiar with (plant, pet, robot, software agent — all fine), don't first label it "seems like consciousness", but first write a paragraph of description following the structure-first rule: what variables does it maintain, through what circuits does it achieve regulation, how does it change input conditions. After writing, look back and ask: if I cut one of the key circuits, what kind of capacity collapse would I expect to see first?

---

## Chapter 3: Three core definitions: structure, agency, consciousness

### Summary

This chapter uses three hard definitions to "lock the base" of discussion: what structure is, what agency is, how consciousness is used in this book. Then through a three-layer breakdown of "pain" it demonstrates: the same phenomenon is not the same object at the structural, agency, and experiential layers; only by distinguishing them first can the experiential layer be safely connected back to structural constraints.

This chapter gives the book's three hard definitions. Their function is not to exhaust "what consciousness is" but to prevent concepts from being infinitely generalized in discussion.

### 3.1 Structure

Structure refers to a materially instantiated, evolutionarily stabilized organization that can maintain metabolic order and coordinate interaction with the environment.

The "structure" here is not simple shape or material stacking, but a way of organizing matter into sustainable processes. In other words, in SATC structure points to "organization that can sustain operation," not just "what it looks like."

In everyday language we often treat "structure" as a static noun. But in SATC structure is closer to "a dynamic order that can sustain self-organization": it maintains a certain stability amid change, preserves itself through exchange and regulation, rather than relying on a one-time design after which it never changes.

### 3.2 Biological Agency / Enactive Capacity

Biological agency refers to the structure-dependent capacity of living systems to initiate, regulate, and maintain interaction with the environment for the sake of their own continuation.

The "agency" here is not the subjectivity of a "rational actor," does not imply deliberation, symbolic reasoning, moral responsibility, or free will. It is closer to "enactive capacity": a structured closed-loop capacity.

To avoid confusion, this book suggests you understand "agency" as a dynamical term rather than a psychological term: it describes how a system maintains itself through closed-loop control, not whether the system has "intention."

There is a very practical distinction here: "agent" carries the intuitive flavor of "who is acting," easily pulling the discussion toward intentionality and responsibility; "agency" emphasizes "what closed-loop capacity the system possesses," without presupposing subjectivity. SATC chooses the latter in order to stabilize discussion on a comparable, constrainable capacity spectrum, rather than sliding toward a personality spectrum or moral spectrum.

### 3.2.1 Agency as capacity spectrum: from "maintenance" to "selection" to "projection"

Writing agency as a capacity spectrum has a direct benefit: it turns the question of "consciousness" from a single lamp into a single path. You don't have to first decide "whether it has it"; you can first describe "to what degree it can." And this degree is not ranked by arbitrary intuition but determined by the organizational form of closed loops permitted by structural conditions.

At the lowest level, agency manifests as maintenance around survival variables: the system maintains itself within a sustainable range through homeostatic regulation. You can understand this as the structural constraint of "must maintain" at work. Going up, agency gradually appears selective: the system makes trade-offs among multiple actionable paths, forming more flexible exploration, avoidance, and learning. Going further up, agency appears projective: the system can organize action across immediate coupling, onto longer timescales, and stabilize this projection through externalized systems.

These three layers are not a value ranking but a structural ranking. They correspond to different forms of control organization: from local regulation, to central integration, to cross-scale and externalized coupling. Once you use this order to look at "consciousness," many vocabulary arguments will naturally reduce in noise: you don't need "higher/lower" to fight for dignity, you only need to explain where the structural transition is, how closed-loop capacity changes accordingly, what evidence would overthrow your claim.

### 3.3 Consciousness

In SATC's working definition, consciousness is the organismal-level expression of biological agency: it is enabled by living structure through continuous interaction with the environment, and manifests as self-regulation over survival-related variables, reactivity to perturbations, and the maintenance of action–feedback loops.

This definition has two boundary conditions: first, it primarily serves origin and comparative research, therefore permitting the use of "minimal consciousness"; second, it does not deny phenomenological experience but treats it as a higher-order descriptive aspect whose enabling conditions need further elaboration.

If you feel this definition is "too broad" or "too narrow," that's normal. The key to SATC lies not in the one-sentence definition itself but in the constraint and orientation it brings: it puts consciousness into the life continuum, thereby forcing us to discuss "how structural differences produce capacity differences"; it also requires us to distinguish descriptions at the experiential layer from causal explanations at the structural layer, and then build connections.

### 3.4 Case: the same "pain" is three different things at the three layers

When you prick your finger on a needle, you say "I'm in pain." But this sentence simultaneously points to at least three layers:

1. **Structural layer**: what receptors, afferent pathways, spinal reflex circuits, and ascending pathways are at work?

2. **Agency layer**: how does the system change input conditions within closed loops? For example, quickly withdrawing the hand, avoiding danger, protecting the injured area.

3. **Experiential layer**: the subjective feeling of pain, its emotional coloring, its capture of attention, the self-reference of "I am being injured."

Many arguments come from mixing these three layers as one. SATC's recommendation is not to deny the experiential layer but first to put it back into the position constrainable by structure and agency: if you cut a certain pathway, which behaviors disappear? Which experience changes? If you change the regulatory system (drugs, anesthesia, learning), which closed-loop strategies change?

**Structural constraint breakdown**

"Having a reflex" does not equal "having experience," but reflexes belong to the observable components of closed-loop agency. At the same time, changes at the experiential layer often accompany structural and closed-loop strategy changes: the two need to be connected, not split into two separate, mutually irrelevant explanations.

**Chapter summary**

The role of the three definitions is to lock the base of discussion: allowing you to talk higher about experience, language, meaning, and society, but not to call everything "consciousness" and thereby lose discriminating power. More crucially, agency is treated here as capacity (capacity) rather than as agent (agent); consciousness is treated as the organismal-level expression of agency. This way of writing will enable you, when facing phenomena like pain, attention, dreams, anesthesia, to first separate the "structure–capacity–experience" layers, and then decide how to connect them back together.

**Chapter reflective questions**

Please explain in one sentence each: how you use "structure," "agency," and "consciousness" in this book. After writing, pause and check whether you've secretly written "agency" as some kind of "inner subject" issuing orders. Then pick any phenomenon (pain, attention, dreams, anesthesia, depression, impulse control, all fine), and use a few sentences to give a structure-first explanatory starting point: what structural constraint do you plan to land on first, then talk about capacity change, then talk about experiential description?

---

## Part I Summary: three anchoring commitments

The theoretical closure of Part I can be summarized as**three anchoring commitments**: they nail all subsequent discussion firmly within the mappable territory of biology.

The first is explanatory order: the language of experience can be complex and refined, but the causal starting point must land on structural and closed-loop organization.

The second is core concepts: structure, agency, and consciousness are written as hard definitions, not to shrink the world but to prevent concepts from inflating infinitely.

The third is the mode of testing: evolution and genetics give constraints of continuity; development and damage provide the strongest convergers — they force the theory to land on structure rather than circulate in rhetoric.

### One-page summary: theoretical foundations (Part I)

Explanatory order: description (experience, phenomenology, reportable content) matters, but it must be connected; causation (structural and closed-loop organization) must not be absent.

Causal chain: matter–structure (dynamic order capable of sustainable self-organization)–function (realized by structure, testable by structural destruction)–agency (capacity spectrum of closed-loop interaction)–consciousness (organismal-level expression of agency, working definition).

Three foundational ground rules: structure first; damage test; scale distinction (milliseconds to years, capacities unfold on different timescales).

One-sentence self-check: if you cannot express a claim as "how structural change changes closed-loop agency," it most likely remains at the descriptive layer.

## Part II: Origins and Grading

## Interlude Two: a day in the life of a plant

In the morning, the sun's angle changes and the leaf's stomatal opening adjusts accordingly: it needs to absorb carbon dioxide for photosynthesis while minimizing water loss. At midday, transpiration pressure rises and stomata tend to close; soil moisture drops, and roots and leaves coordinate resource allocation through hormone signals. Wind, insects, salinity, shading, temperature, and microbial symbiotic relationships constantly change the plant's input conditions, and the plant also changes these conditions in its own way.

If you only understand "action" as rapid displacement, plants look like they "don't move." But if you understand action as "the closed-loop capacity to change input conditions for the sake of one's own continuation," a plant's day is full of action: through growth direction, stomatal regulation, root distribution, and chemical signals, it continuously regulates its relations with the environment on multiple timescales.

This interlude wants to remind you: SATC's grading is not fighting over a word but fighting over a mode of questioning. Whether a plant "has experience like a human" is not a question the introduction needs to solve first; whether a plant has distributed closed-loop agency is a question that can be constrained by structure and behavior, and is also the key to understanding the continuum.

## Introduction to Part II: from "being alive" to "having a world"

Part II will advance the working definition of "consciousness as agency" into two thorniest questions: origin and grading.

The origin question forces us to answer: must consciousness be premised on the nervous system? If not, what is minimal consciousness? How do we avoid setting the conceptual threshold too high or too low?

The grading question forces us to answer: if consciousness is continuously graded, where does difference come from? SATC's answer is "from structural transitions," especially transitions in the form of control organization, timescale, and environmental coupling. Thus plants, animals, and humans are not ranked by "how much they resemble humans" but by "how closed-loop capacity is organized."

In this part, "origin" and "grading" are not two separate topics. Origin gives the starting point of the continuum: you must at least acknowledge some minimal closed loop to discuss "where consciousness comes from." Grading gives the branching mode of the continuum: you must be able to explain how structure becomes more complex, how control organization goes from slow to fast, from local to integrative, from immediate coupling to cross-temporal-spatial projection. Once the two are separated, discussion heads toward two extremes: either only "philosophical裁决" (treating origin as a question that cannot be entered) or only "intuitive ranking" (treating grading as an aesthetic judgment of whether something resembles a human).

This is also why we repeatedly emphasize taxonomy and comparison. Comparison is not arbitrarily analogizing different systems but putting them into the same set of questions and then letting the materials themselves tell you where the boundaries are: the same question can be posed to cells, plants, animals, and humans, but the answers need not be consistent. You don't ask "does it resemble me" but "what variables can it maintain, through what circuits, on what timescales does it change input conditions." Write questions this way and origin and grading naturally connect: the capacity spectrum doesn't fall from the sky; it starts from the minimal closed loop and expands along structural transitions.

## Chapter 4: Minimal consciousness: from metabolic self-maintenance to primordial agency

### Summary

This chapter answers two key questions: why minimal consciousness must be a "working definition," and how it can maintain continuity without losing discriminatory power. Finally, using bacterial chemotaxis as a minimal case: reading "movement" as "changing input conditions," thereby anchoring discussion on closed-loop agency.

If consciousness is a continuously graded phenomenon along with life, its minimal form must appear in the minimal structural conditions of life.

SATC anchors minimal consciousness on metabolic–self-maintaining structure:

The earliest form of agency emerges with the appearance of self-maintaining metabolic structure. Such structure can maintain a local low-entropy state under global thermodynamic constraints and preserve its organization through regulated exchange with the environment.

This is what this book calls "primordial consciousness":

It is not "experiential feeling" but "structured self-maintaining agency."

### 4.0 Why start from metabolism and boundaries

Placing the starting point of consciousness discussion on metabolism often triggers two opposite misunderstandings. The first misunderstanding thinks this is "degrading consciousness to physiology," as if we're replacing a more refined word with a simpler one; the second misunderstanding thinks this is "smuggling in a redefinition," renaming consciousness as being alive. The common root of both misunderstandings is: they treat metabolism as some local process rather than as the base of living organization.

In the structural–evolutionary agency paradigm, the most critical thing about metabolism is not "energy consumption" but "boundaries and self-maintenance." What makes living systems different from ordinary physical systems is not that they're more complex but that they must continuously maintain themselves in an open environment: they must maintain certain key variables within a sustainable range and recover organization under perturbation; they must distinguish "self" from "environment" through boundaries while continuously exchanging matter and energy. In other words, they're not manufactured once and done but must continuously make themselves in time.

Agency therefore obtains its minimal skeleton here: it's not "having ideas" or "having intention" but the system's capacity to form closed-loop regulation around survival-related variables and to change input conditions through its own behavior or internal regulation. You can make this more concrete: when nutrition is insufficient, the system changes its intake strategy; when damage occurs, the system changes local resource allocation; when temperature deviates, the system changes its mode of exchange and heat production/dissipation. These changes need not enter experience or report, but they constitute the most fundamental fact of agency.

A misunderstanding easily arises here: understanding "minimal consciousness" as "all life has feeling like a human." SATC does not make such an inference. What SATC means is: if you treat consciousness as a comparable, continuous biological phenomenon, then minimal consciousness should correspond to minimal agency; and "experiential feeling" as a higher-order phenomenon requires additional structural conditions.

An intuitive example is chemotaxis: certain microorganisms move along nutrient concentration gradients. You can describe it with physicochemical models or with control-theory language. SATC's concern is: whether this behavior embodies a closed-loop organizational form that enables the system to change input conditions for the sake of survival. If so, it is a candidate example of "agency"; as to whether it "has feeling," SATC considers this a question at another level.

### 4.1 Why "minimal consciousness" must be a working definition rather than a final裁决

If you treat minimal consciousness as a final裁决, you immediately encounter two symmetrical failures. First, setting the threshold too high: only humans or a few mammals "count as consciousness," so the origin and continuity questions are strangled in the first sentence, leaving nothing but an exclusionary negotiation over vocabulary. Second, setting the threshold too low: every system "counts as consciousness," so you lose discriminating power; the theory gains universality on the surface but loses its explanatory grip.

SATC adopts a middle strategy: treating "minimal consciousness" as a working definition serving comparison and origin questions; leaving discussion of "experiential consciousness" for higher-order structural conditions. The advantage is: the theory gains continuity while preserving distinguishable hierarchical differences.

To avoid making minimal consciousness an empty phrase, this book adopts a minimal set of structural–functional indicators to pull discussion back to observable dimensions: whether it has the closed-loop capacity for self-maintenance and regulation; whether it can change input conditions through action–feedback coupling; whether there are control organizations on multiple timescales; and whether structural change systematically changes the available set of agency.

The meaning of these four indicators is: they make "minimal consciousness" no longer stay at the language level. You can take them and check different systems: cells, plants, animals, even certain artificial systems, to see which closed-loop conditions they satisfy structurally and which they lack.

To write these four indicators more "usable," understand each as a mode of questioning.

The first question is self-maintenance: does the system need to continuously maintain itself within a sustainable window? What are the key variables it maintains (energy, osmotic pressure, temperature, damage, oxidative stress, etc.)? When these variables deviate, does it trigger a systemic regulatory response? If not, it's more like a passive process; if so, it at minimum possesses the constraint of "must maintain."

The second question is closed loops and input conditions: does the system's regulation stop at internal chemical equilibrium, or can it change input conditions through behavior or morphology? For microorganisms, this may appear as chemotaxis; for plants, this may appear as growth direction and stomatal strategy; for animals, this may appear as movement, exploration, and avoidance. The key is not whether the action "looks like" action but whether it changes the distribution of inputs the system receives.

The third question is timescale: does the system have only single-timescale immediate responses, or does it have cross-scale organization? Minimal life can certainly have only short regulatory circuits, but when you discuss more complex capacities, this question becomes decisive: without cross-scale regulation, it's hard to explain long-term goal maintenance, learned strategies, and the stable accumulation of externalized systems.

The fourth question is structural testability: when structural state changes, do these closed-loop capacities change systematically? Such change can be destructive (damage, anesthesia, pathological dysregulation) or constructive (developmental maturation, learning sculpting, adaptation under long-term environmental pressure). This question matters because it pulls "agency" out of the risk of arbitrary labels by observers: if structural change doesn't bring capacity change, so-called capacity is probably just naming; if structural change systematically changes capacity, then you've gained a causal grip.

### 4.2 Case: bacterial chemotaxis — reading "movement" as "changing input conditions"

Bacterial chemotaxis is often treated as a "very simple behavior," because it doesn't seem to have feeling like animals or reportable experience like humans. But it's very well suited to practicing SATC's minimal-consciousness questioning: does it embody closed-loop agency?

If a system can change its own movement strategy according to environmental gradients to move itself into regions of higher nutrition and lower toxicity, then it changes input conditions at the behavioral level. SATC does not require you to therefore assert "it has feeling" but requires you to acknowledge: here there is a structured closed-loop organizational form that serves survival.

**Structural constraint breakdown**

The key is not "whether it can think" but "whether it can change input conditions through action"; the key is also not "whether it has experience" but "whether this capacity is realized by structure and testable by structural destruction." In SATC's way of writing, this kind of minimal agency provides the foundation of continuity for later more complex agency rather than a fractured exception.

**Chapter summary**

What this chapter most wants to preserve is "discussability" and "comparability." Minimal consciousness is not "minimal feeling" but a working definition of "minimal agency": it lets the origin question be posed, compared, and advanced without getting tangled in an experiential裁决 tug-of-war in the first sentence.

This does not mean experience doesn't matter; on the contrary, it means explanation deserves better. You first lay out closed-loop agency clearly, first let the spectrum of structure and capacity stand firm, and the question of experience will not become a floating adjective. It will become a real research question: under what structural conditions and through what control organization does experiential consciousness appear, strengthen, or disappear?

**Chapter reflective questions**

Please first write down at least two conditions you think "minimal agency" must satisfy, and explain why not just one. Then imagine an "almost alive" system: it can replicate but has no regulatory capacity, only changing passively under external drives. Where would you place it on the continuum? Can your reasoning be pushed by the "damage test": if I weaken one of its structures, what would you expect to lose first?

---

## Chapter 5: A grading model: from plants to humans as a structural continuum

### Summary

This chapter gives a four-level continuum and emphasizes that the criterion of grading is not "resemblance to humans" but structural transitions and forms of control organization. Three narratives of plant timescale, animal selectivity, and human externalized systems will be used to build intuition: how the same closed-loop logic unfolds into different capacity spectra under different structural conditions.

SATC understands "consciousness/agency" as a capacity spectrum continuously unfolding along structural complexity, not as a binary attribute. For ease of discussion, this book adopts a four-level model, defined by key structural transitions rather than by the criterion of "resemblance to humans."

### 5.1 Level 0: metabolic–self-maintaining agency

The core feature of this level is closed-loop regulation and boundary maintenance of metabolism–homeostasis. If behavioral integration is still weak, agency is dominated by local regulation; as organizational levels emerge, global coordination gradually strengthens.

The key at this level: the system is no longer just passively consuming energy but maintaining itself through closed-loop organizational form. It is the "foundation" of the continuum and the underlying condition for all later more complex agency.

More strictly, Level 0 is not concerned with "whether there's a nervous system" but "whether there's sustainable self-maintaining organization." A system that doesn't need to maintain itself has no boundary that must be maintained and lacks the minimal motive for agency. Conversely, a system that must maintain itself will structurally be forced to form a minimal set of control organization: it must distinguish self from environment, must have resilience to perturbation, and must reorder strategies under resource scarcity. You can even see this level as the common premise of all later complexity: without this premise, any higher-order "selection" or "planning" is merely a story assigned by external observers.

Therefore, when we use Level 0 to look at "minimal consciousness," we are not turning consciousness into a metabolic noun but establishing a fact that cannot be circumvented: if consciousness is to be a biological phenomenon, it must first serve the closed-loop maintenance of survival-related variables. Experience can certainly appear, but it won't appear in a system that doesn't need to maintain itself; it can only appear in a living system already shaped by self-maintenance constraints.

### 5.2 Level I: distributed sessile agency (plants as exemplar)

Plants are exemplars of distributed agency. They typically lack a central nervous system, but through environmental perception, adaptive growth, resource optimization, and long-term environmental modification, they embody clear agency.

Agency at this level is not rapid displacement but dominated by growth and morphological regulation, operating on longer timescales.

If you wonder "how can plants have agency," you can start with three very concrete phenomena: phototropism and gravitropism let it change input conditions through growth direction under environmental constraints; stomatal regulation lets it maintain the boundary of metabolic exchange amid changes in moisture, temperature, and carbon dioxide concentration; longer-term resource strategies (root distribution, chemical signals, symbiotic relationships) then embody cross-timescale regulation. Once you understand "action" as "changing input conditions," plants are no longer "static objects" but slow and stubborn closed-loop systems.

SATC's usage is: not elevating plants to "animal-style consciousness" nor reducing plants to "completely passive," but placing them at the level of distributed agency to discuss structural differences.

At the structural level, plants' distributed agency means "control is not concentrated in one center." It coordinates local and global through distributed signal networks: hormones, local electrical signals, chemical gradients, and structural growth patterns together constitute closed loops that are slow but robust. You can understand it as "writing control into the organization itself": structures like leaves, roots, vascular bundles, and stomata are both materials and regulators. Its timescale is longer, but its resource constraints are more direct: water, light, salinity, and nutrients don't wait for it to "make a decision" but continuously push it to a position where it must regulate.

This also explains a common misreading: treating plants as having "no behavior." Plants certainly lack animal-style rapid displacement, but they have many structured behavioral equivalents: growth direction, morphological adjustment, resource allocation, chemical defense, and the establishment and dissolution of symbiotic relationships. As long as you understand "behavior" as "the way input conditions are changed," plants go from "static background" to a very crucial segment on the continuum: they show how agency can hold in the absence of a center, and how stable strategies can be formed over long timescales.

### 5.2.1 A more "textbook" observation: the time of plants vs. the time of animals

Many arguments come from a mismatch of timescales. We are used to binding "agency" with "rapid movement," so plants look like they "don't act." But if you stretch the timescale, plants' actions are clear: they change growth direction, change resource allocation, change local morphology, and even change neighborhood ecological relations through chemical signals.

In SATC, this kind of agency is not negated just because it is slow. It is classified as distributed, sessile, closed-loop capacity over long timescales. This is precisely why grading must be defined by structural transitions rather than by intuitive speed or "resemblance to animals."

### 5.3 Level II: centralized sensorimotor agency (animals)

With the evolution of the central nervous system, agency undergoes a qualitative change: sensorimotor loops are centrally organized, the system gains active movement, learning, and behavioral flexibility, thereby being able to unfold exploration and short-term goal-directed behavior in space.

The typical feature of this level is "fast closed loop": the sensorimotor coupling operates on shorter timescales; the system can move, try and error, and learn in space. This is one reason most animals are "more like having consciousness" in our intuitive sense: their agency is not only maintaining metabolism but also actively exploring and selecting behavioral paths.

### 5.3.1 A transition: from "able to maintain" to "able to select"

The key advance of Level II is selection: the system not only maintains itself but also begins to make choices among multiple actionable paths (even if such choice is still limited to short-term goals and the immediate environment). This gives rise to many "mental phenomena" in our intuition: attention, motivation, fear, exploration, learning. SATC's recommendation is: don't first ask "whether it has experience" but first ask "what can it choose and what can't it," and "which circuits does this selection depend on."

### 5.3.2 A structural anchoring point: why "central integration" changes the form of agency

The transition from plants to animals is often intuitively understood as "starting to move." But in the structural–evolutionary agency paradigm, the more crucial change is: closed loops are reorganized. The central nervous system brings scattered sensory and motor circuits into higher-level integration, enabling the system to form stable sensorimotor mappings on shorter timescales and to switch among multiple goals.

This directly changes the form of agency. Distributed regulation is more like slow steady-state engineering, good at long-term maintenance and resource allocation; central integration is more like fast control engineering, good at exploring, avoiding danger, and tracking goals in space, and updating strategies through trial-and-error and learning in uncertain environments. If you confuse these two kinds of agency, you'll make a common mistake in comparison: using animals' speed scale to negate plants' agency, and using plants' stability scale to underestimate animals' selectivity. The point of grading is precisely to let both find their rightful places on the same map.

### 5.4 Level III: abstract–projective agency (humans as a special case)

Humans present a further structural transition: the expansion of neural architecture (especially prefrontal-lobe-related systems) supports abstraction, long-term planning, and counterfactual modeling.

One decisive feature of this level is the externalization of agency through tools and symbol systems: tools and symbols are not the origin of consciousness but structural evidence that abstract–projective agency transcends immediate sensorimotor coupling.

Two points must be emphasized here: first, tool use itself does not equal Level III. Animals also use tools. SATC is more concerned with: whether action can be decoupled from "here and there coupling" and projected onto the future, distant places, or counterfactual scenarios, and stably extended through externalized systems (tools, symbols, institutions). Second, Level III is not "higher/superior" but "structurally different." SATC opposes value-judgment grading but acknowledges that structural transitions bring transitions in capacity spectrum.

### 5.4.1 A case: how externalized systems change the boundaries of agency

Imagine two scenarios. First, you see a stone, want to crack a nut, and try a few times to find the right angle and force. Second, you make a hammer and teach others how to make hammers; you write down rules, forming procedures; you turn "cracking a nut" into part of a reproducible externalized system.

In the second scenario, agency occurs not only in the body but is stably distributed into tools, symbols, and institutions. SATC regards this stable externalization as one hallmark of abstract–projective agency: action is no longer just immediate coupling but spans time and space, extended and preserved by externalized systems.

### 5.4.2 The key of externalized systems is not "having tools" but "having accumulable closed loops"

Treating externalized systems as evidence of Level III often triggers a misunderstanding: as long as tools appear, "higher consciousness" appears. SATC's distinction is stricter. Tools matter not because they symbolize intelligence but because they change the timescale of closed loops: action and feedback no longer just cycle between the body and the present environment but begin to pass through symbols, procedures, collaboration, and institutions, being preserved, transmitted, and accumulated.

From a structural perspective, this means the boundary of agency expands. The individual closed loop no longer relies solely on the plasticity within neural circuits for maintenance but also gains "external memory" and "external controllers" with the help of externalized systems: you maintain goals with notes, maintain rhythm with calendars, maintain paths with maps, maintain collaboration with rules. Externalized systems both expand agency and create new fragilities: once externalized systems fail, capacity collapses in another way. This kind of collapse itself is evidence of structural conditions — it tells you that agency has indeed been distributed to external structures.

Therefore, Level III is not so much "smarter" as "longer-timescale, larger-scale, stronger-coupling." It pushes life-style closed loops from inside the individual to the composite structure between individual–tools–social environment. You may not like the cultural implications of this conclusion, but it's hard to deny its structural implications: the reason human agency presents such a special form is precisely because it structurally allows this kind of reproducible, transmissible, accumulable externalized closed loop to persist long-term.

### 5.5 Case: crows can use tools, but "externalized systems" are not "a single tool"

You may have heard stories of crows bending wire to get food, sea otters using stones to crack shells, chimpanzees using sticks to fish for termites. They do shake a common intuition: tools are not unique to humans.

SATC's approach is not to exclude these animals nor therefore abandon the criterion of "externalized systems," but to make a more precise distinction:**Single tool use**and**Reproducibly stable externalized systems**are not the same thing.

**Structural constraint breakdown**

Single tool use can occur at Level II: it still primarily serves feeding and problem-solving in immediate coupling. Level III emphasizes the stability of externalized systems: tools and symbols are systematically manufactured, transmitted, accumulated, embedded in collaboration and institutions, thereby expanding the timescale and spatial scale of agency. Therefore, SATC uses tools as "evidence" rather than "definition": they hint at the direction of structural transitions but don't serve as a one-size-fits-all classificatory standard.

**Chapter summary**

This chapter tries to free "grading" from intuitive judgment. The criterion of grading is structural transition, not "resemblance to humans" or "ability to report experience." You can like animals, revere humans, or be fascinated by language, but these emotions should not become criteria for grading. Criteria should come from structure: how control is organized, how timescales unfold, how the system changes input conditions.

Tools and symbols are therefore treated as evidence of abstract–projective agency, not as definitional conditions for the origin of consciousness. They are not a magic button of origin, more like fingerprints left by structural transitions: they hint that the system is already able to loosen action from here and now and distribute capacity into externalized systems that are reproducible, transmissible, and accumulable.

**Chapter reflective questions**

Please choose a non-human animal and write down its most typical "selective behavior" and guess which structures support such behavior (even rough circuit-level guesses). Then choose an externalized system you rely on daily (phone, notes, calendar, map, social collaboration, all fine) and write down how it changes the boundary of your agency: from which immediate couplings does you free yourself, and in which places does it make you more dependent and more fragile?

---

## Part II Summary: continuity and discriminating power

What Part II does seems risky but is actually stricter: it doesn't call everything consciousness but preserves discriminating power while acknowledging continuity.

Continuity comes from the foundation of Level 0: closed-loop self-maintenance makes "minimal consciousness" a usable working definition without first appealing to experiential裁决. Discriminating power comes from structural transitions: changes in the form of control organization explain why capacity spectra diverge.

This makes Part II more like a "biological map": when discussing any system (plants, animals, humans, artificial systems), first ask about structure and closed loops, then decide which set of consciousness semantics to use.

### One-page summary: origin and grading (Part II)

The core question: must consciousness be premised on the nervous system? SATC's answer is no — first establish continuity through minimal agency, then discuss the additional structural conditions required for experience to appear.

The foundation of the continuum comes from Level 0 (metabolic–self-maintenance): it gives minimal closed loops (boundary maintenance and homeostatic regulation); so-called "primordial consciousness" here is merely a working definition, that is, structured self-maintaining agency.

Structural transitions can be understood along three main lines: how control organization goes from local to distributed to central integration to abstract projection; how timescale goes from slow to fast and finally cross-scale; how the mode of changing input conditions goes from exchange and regulation, gradually to growth and allocation, movement and selection, and externalization and accumulation.

As an easy-to-remember minimal formulation: Level 0 maintains, Level I regulates, Level II selects, Level III projects. Before asking "does it have consciousness," first ask whether you're using experiential semantics or agency semantics; then write the question in a form constrainable by structure and closed loops.

## Part III: Research Program and Extrapolation

## Interlude Three: a system that can talk and a system that can live

Imagine two kinds of systems.

The first kind of system is very good at talking. It can write poetry, write code, write papers, explain what it is "thinking," and even imitate emotions and personalities. Put it in an environment with infinite computing power, infinite electricity, and infinite data, and it becomes more and more like an all-capable language-user.

The second kind of system is not very good at talking, but it must live: limited power, drifting sensors, wearing machinery, mutating environments. It must maintain itself under resource constraints and continuously revise its actions in feedback loops in order to persist.

Which of these two systems is closer to what SATC calls "agency"? SATC's answer is usually not "which is smarter" but "which is more closed-loop." Language capacity can grow in training detached from the world, but agency must be maintained and unfolded under resource constraints and in action–feedback loops.

Putting this question in the interlude is to remind you: when Part III discusses AI and research programs, we are not chasing a romantic question (whether machines have souls) but a constrainable question (whether a system has the structural conditions for closed-loop self-maintenance).

## Introduction to Part III: from theory to mode of questioning

Part III aims to turn SATC from "viewpoint" into "usage."

A theory that can only explain existing stories will easily become belief; if it can steadily generate research questions, generate refutable points, and generate comparable indicators, then it is more like a research program. This part will compress SATC into three actions: compare, causation, extrapolate.

## Chapter 6: Turning the theory into a research program: metrics, predictions, discrimination

### Summary

This chapter advances SATC from "viewpoint" to "mode of questioning": compare, causation, testable. It provides a three-paragraph template for writing research plans (structural hypothesis, agential indicator, refutable point) and demonstrates how anesthesia can be written as a refutable research question rather than stopping at a sentence of "consciousness disappears."

One difference between an introduction and a paper is that the introduction must answer "how to use it." This chapter gives three modes of use.

### 6.1 As a comparative framework: cross-species structural–agency alignment

When facing different species or different life forms, SATC provides a unified mode of questioning:

How does the system's structure realize closed-loop agency? On what scales do its control hierarchy, plasticity, and environmental coupling unfold?

More concretely, a set of fixed questions can be used to land comparison on structure: what key variables does it maintain (energy, temperature, damage, nutrition, salinity, etc.)? Through which circuits does it achieve regulation (distributed chemical signals, neural circuits, endocrine axes, etc.)? Can it change its own input conditions (movement, feeding, avoidance, construction, tool externalization, etc.)? These questions seem plain, but they can press the abstract debate of "seems like consciousness" back to the concrete question of "how closed loops are realized by structure."

### 6.1.1 A simplified "mapping table" (for writing research plans)

The grading model can be roughly mapped to a set of research variables:

Level 0 corresponds to the intensity and robustness of closed-loop self-maintenance (e.g., recovery curves under perturbation); Level I focuses on the coupling and timescales of distributed regulatory networks (growth-mediated, chemical signals, electrical signals); Level II focuses on the integration capacity and learning mechanisms of central circuits (sensorimotor coupling, trial-and-error, habituation); Level III focuses on cross-temporal-spatial projection and the stability of externalized systems (planning, counterfactuals, tool/symbol systems). These variables are not for cramming life into a table but for giving "gradable" positions to land in writing and design.

### 6.1.2 The bottom line of comparison: align "objects" first, then align "questions"

The most common failure of cross-species comparison is not lack of data but objects and problems not being aligned first. When you discuss fish, birds, mammals, plants, and microorganisms together, it's very easy to default to an invisible "reference" in your mind: humans. Then all differences are understood as "resemblance to us." This causes distortion at the first step of comparison: you're not comparing the mode of organization of closed-loop capacity but comparing some kind of similarity in human intuition.

The structural–evolutionary agency paradigm cares more about: whether the same question can be posed on different objects and answered by the objects' own structural conditions. For example, "how does it change input conditions" likely manifests as movement for animals, growth and allocation for plants, and chemotaxis and population regulation for microorganisms. You don't need to force-interpret plants as "slow animals" or force-interpret microorganisms as "miniaturized brains." The bottom line of comparison: allow different forms of answers but require them all to land on the structure–agency chain.

Here the role of taxonomy becomes clear: classification is not for labeling objects but for avoiding treating similarity as homology or homology as similarity. In writing you should at least distinguish three kinds of relationships: homology brings "shared constraint" (structure may preserve the same organizational logic to different degrees); convergence brings "shared pressure" (structure different but may grow similar function); analogy brings "shared surface" (looks similar but causal organization not necessarily the same). Distinguishing these three kinds of relationships significantly reduces the interference of anthropomorphism and "resemblance," bringing comparison back to the caution and sharpness biological inference should have.

### 6.2 As a causal framework: evolution, genetics, development, and damage

If agency is a structure-dependent causal capacity, then structural differences and structural destruction should preferentially reflect as differences and declines in agency. The "differences" here include both structural differentiation on the evolutionary lineage and structural differences brought by genetic variation and developmental processes; the "destruction" includes damage, anesthesia, degenerative change, and pathological dysregulation. Conversely, if some explanation cannot find a constrainable path on structural change, it's more like description than causal explanation.

A very strong test comes from neural damage and anesthesia: when the integration capacity of key circuits is severed, consciousness capacity presents systematic decline. SATC does not require you immediately to explain "why experience disappears," but it does require you to acknowledge: there is a testable coupling between changes in consciousness capacity and structural state.

The "structure" here includes both anatomical structure and functional connectivity and dynamic integration. The introduction does not expand on specific experimental or methodological details, but SATC can serve as a writing framework: when describing a disorder (e.g., attentional disintegration, impulsive dyscontrol, decline in consciousness level), first write out the assumed structural constraint path you assume, then discuss how psychological and social factors intervene as moderators.

### 6.2.1 Writing "genetics–development–structure–capacity" as a single causal sentence

Many so-called "consciousness theories" seem abstruse because they only jump between the structural layer and the experiential layer: either drawing conclusions directly from experience or giving explanations directly from brain regions, lacking a traceable path in between. The structural–evolutionary agency paradigm requires you to fill in that path, and the most stable way to write it is: how genetic variation and regulatory modes shape structure during development, how structure constrains the available set of closed-loop capacities, how the set of capacities manifests as observable agential indicators in specific environments, and (if entering discussion) how experience is constrained by the organizational mode of these capacities.

Writing this sentence, you naturally gain two "testable entry points." One entry point comes from development: the same system at different developmental stages has different structures and different capacity spectra, so agential indicators should present predictable unfolding; another entry point comes from genetic differences: when genetic and regulatory changes lead to structural deviations, you should see directional deviations in the capacity spectrum, not just "performance getting worse." These two entry points do not require you to immediately produce molecular-level proof, but they force you to write out the causal direction: are you explaining "how structure enables capacity" or "we named capacity as structure."

### 6.2.2 Anti-teleology: evolution is not "in order to" but "because"

When using the "evolution–structure–agency" path, this book needs to explicitly reject a common narrative illusion at the method level: portraying evolution as purposeful, moving toward some goal like engineering design. Statements like "pitcher plants evolved pitcher traps in order to catch flies" or "birds grew wings in order to fly" are common in popular science contexts, but they implicitly reverse the causal direction in writing, making you treat results as causes and functions as intentions.

The causal direction insisted by the structural–evolutionary agency paradigm is plainer and harder: evolution and mutation are never "in order to" but "because," finally "screened by environmental selection." More concretely, heritable variations (mutation, recombination, regulatory differences, etc.) constantly appear in biological populations; these variations express as certain structural and trait differences during development; the environment does not "design" these differences but screens them through survival pressure and reproductive differences. Variations better adapted to environmental constraints are more likely to be retained and passed down through generations; after long-term accumulation, differences solidify on the lineage and manifest as species-level differentiation when reproductive isolation forms.

Therefore, when we say some structural change "enables" some agential capacity in this book, we are not doing teleological narrative but a constrained historical inference: under what environmental pressures and resource constraints are which structural differences more likely to be retained, thereby forming the capacity spectrum we see today. You can use "in order to" verbally when storytelling, but in research writing you must translate "in order to" back into testable language: what is the source of variation, what is the selection pressure, how does structure change, how does closed-loop capacity change, what materials (comparison, development, damage, genetic differences) would force you to revise this inference.

### 6.3 As testable claims: three predictions (examples)

First, structure-destruction priority prediction: structural destruction of key control loops will preferentially lead to declines in agency and consciousness. Second, developmental and genetic constraint prediction: transitions in agency grading should correlate systematically with specific developmental milestones and show traceable structural–capacity deviations in response to genetic and regulatory differences. Third, embodiment constraint prediction: purely software systems lacking sustainable resource constraints and closed-loop regulation will not present Level III agency.

Writing predictions as three is not superstition about "three" but to force the theory to be converged in three directions. Structural destruction faces you with the hard question of "causal efficacy": if structure has no causal efficacy, destruction should not bring systematic capacity decline. Development and genetics faces you with the hard question of "continuity": if the capacity spectrum is truly a product of structural evolution, it should unfold during development and undergo directional deviation under genetic difference, rather than popping up out of nowhere. Embodiment constraints faces you with the hard question of "living-ness": if you completely skip resource constraints and closed-loop maintenance, you bring consciousness discussion back to the old road of abstract computation.

### 6.3.1 How to write predictions more like biology: writing "difference" as "path"

Many readers, when writing predictions for the first time, instinctively write a prediction as a value judgment: "more complex brains will be more conscious." Such sentences sound like common sense but can barely advance because they lack structural variables, agential indicators, and refutable points. The structural–evolutionary agency paradigm requires you to write predictions as paths: you specify which structural conditions mature on what timescale, expected to bring systematic changes in which kind of closed-loop capacity, and clearly state which observation would make you rewrite this path.

For example, if you want to talk about "development," don't just write "consciousness is stronger at a certain age." A more usable way to write: as a certain class of integrative circuits gradually matures, the system should show consistent changes on the following three closed-loop indicators — faster recovery from perturbation, more stable goal maintenance, more reliable strategy transfer across situations. Conversely, if you observe significantly enhanced goal maintenance but no improvement in recovery capacity, or improvement in recovery capacity accompanied by decline in strategy transfer, then you must admit: you've mixed multiple timescales in one sentence, and the structural hypothesis needs to be taken apart.

Again, if you want to talk about "genetic differences," don't just write "a certain gene makes consciousness decline." A more robust way to write: a certain class of genetic or regulatory difference changes the speed or mode of structural maturation, so you expect to see not an abstract "consciousness getting weaker" but a directional deviation in a certain capacity spectrum — for example, short-timescale closed loops remain stable but cross-scale maintenance becomes fragile; or immediate responses remain agile but long-term strategies are harder to accumulate. You don't need to give a molecular mechanism in the introduction, but you must give the structured form of the capacity spectrum, otherwise "genetics" will become rhetorical authority.

Similarly, "structure-destruction priority" should not just be written as "damage will reduce consciousness." You need to write out the specific order of collapse: when key integration is destroyed, which kind of closed-loop capacity will collapse first? Is recovery from perturbation the first to collapse, or goal maintenance the first to collapse, or the capacity to change input conditions the first to collapse? Different orders of collapse correspond to different structural paths. As long as you're willing to write out the order of collapse, you've pushed the discussion from position to research: because the order itself is refutable.

### 6.3.2 A template sentence: writing predictions as research-advancing sentences

If you find the above writing still too abstract, you can compress a prediction into a template sentence and fill in each blank. The template is as follows:

When [structural variable] undergoes [mode of change] in the path of [evolution/genetics/development], the system's [closed-loop capacity] will manifest as [directional change] in the form of [observable indicator]; if this change still does not appear under [falsification condition], then priority should be given to revising [structural hypothesis / indicator choice / timescale setting].

You'll find what's really hard to write isn't the sentence but the blanks: what is the structural variable? What is the capacity? What is the indicator? What is the falsification condition? When you fill in the blanks, you naturally gain a research path; when you can't fill them, the problem is usually not "lack of inspiration" but that you haven't taken the phenomenon apart to the structure and scale level.

This is also the writing skill the introduction wants to give you in advance: don't rush to compete for a more beautiful definition but first write what you want to say into a sentence that can be pressed by life materials. Pressable, then advanceable; advanceable, then consensus can gradually form.

This book treats these three predictions as "the starting point of a research program," not as final conclusions. Their value lies in: they force discussion to leave pure concept argument and enter the domain constratable by structure and data.

### 6.4 Case: writing "anesthesia" as a refutable research question

Many people stop at a sentence when discussing anesthesia: anesthesia makes consciousness disappear. But as a research program, you need to write it as a refutable question: what structural state change leads to what agential indicator change?

An introductory-level way of writing usually includes three layers: first give the structural hypothesis — the dynamic coupling of a certain class of integrative circuits is destroyed or suppressed, thereby reducing the stability of cross-scale closed-loop control; then propose agential indicators — the maintenance capacity of action–feedback closed loops declines (recovery from perturbation, goal maintenance, organization of selective behavior); finally write refutable points — if, when structural coupling significantly declines, the system can still maintain stable closed-loop agency, or if closed-loop agency systematically disintegrates when structural coupling has not significantly changed, then the structural hypothesis needs revision.

What SATC provides here is not a final answer but a mode of writing: it forces you to push the debate from "whether experience disappears" to "when closed-loop organizational modes fail and when they recover."

### 6.4.1 A writing example you can directly copy (turning a slogan into a question)

You can directly rewrite the slogan "anesthesia makes consciousness disappear" into a three-paragraph question. First paragraph, write structure: under anesthesia, which integrative circuits' coupling is weakened? Does this weakening manifest as changes in connectivity patterns or decline in dynamic coupling stability? Is it local or cross-scale? Second paragraph, write agential indicators: under the same external input intensity, what observable changes appear in the system's closed-loop maintenance capacity — does recovery from perturbation slow down, does goal maintenance disintegrate, does selective behavior degenerate into local reflexes? Third paragraph, write refutable points: if, when coupling is significantly weakened, the system can still maintain stable closed-loop agency, then the structural hypothesis needs revision; if closed-loop agency systematically disintegrates when coupling has not significantly changed, then you need to look for more crucial structural variables.

The advantage of this three-paragraph way of writing is: it moves consciousness discussion from "experiential vocabulary" to "structural variables" and "closed-loop indicators." You don't need to answer the mystery of experience first to propose an advancing research question; you only need to acknowledge the causal chain of living systems, acknowledge that agency is a set of structured capacities, and acknowledge that changes in structural state should leave observable imprints.

You can also migrate the same way of writing to almost all phenomena you care about: sleep, attentional disintegration, impulsive dyscontrol, depressive-like states, developmental delay, degenerative change… The key to migration is not changing nouns but changing syntax. You always first write structural hypothesis, then write closed-loop indicators, then write refutable points; you always avoid using experiential vocabulary to directly draw causal conclusions, and avoid using abstract vocabulary to skip resource constraints. You can even treat it as a self-check: when you find yourself writing more and more like a philosophical manifesto, it usually means you've written out the structural variables; when you find yourself writing more and more like an engineering description, it usually means you've written out the life-style metabolic and self-maintenance constraints.

More importantly, this way of writing allows you to "advance first, then refine" at the introductory stage. You don't need a complete map of neural mechanisms at the start, nor do you need to give a final answer to the mystery of experience in the first chapter; you only need to give a structural path willing to be pressed and clarify which observations you are willing to use to press it. By the time you truly enter experimental, clinical, or comparative materials, the theory will deform, harden, or even be overthrown under pressure — and this is precisely what should happen with a research program.

**Chapter summary**

This chapter's purpose is to grind SATC from "viewpoint" into "usage." Compare, causation, testable sound like methodological slogans, but here they serve as basic actions for writing and research: you use them to align different systems, to converge explanatory paths, to write claims willing to be overthrown.

Predictions are therefore not used to prove you're always right. The meaning of prediction lies in forcing you to write out the boundaries of the theory: which facts would support it, which facts would deform it, which facts would force you to abandon it. A theory that never consents to lose is often just style; the value of a research program comes precisely from its willingness to change itself in the face of data.

**Chapter reflective questions**

Please choose a research object you want to study (sleep, anesthesia, plant electrical signals, animal exploratory behavior, child development, LLM+robot, all fine) and write a three-paragraph research question. First paragraph, write only structural hypothesis: what do you think the key structural constraints are, which circuits or organizational forms are at work. Second paragraph, write agential indicators: what observable closed-loop indicators do you plan to land on. Third paragraph, write refutable points: what observations are willing to use to overthrow or modify your hypothesis — write as specifically as possible.

---

## Chapter 7: Implications for artificial intelligence: why "purely software" is not enough

### Summary

This chapter separates "intelligence engineering" from "agency engineering": representation and language, embodied control, self-maintenance and synthesis — each solving different problems. Through a comparative case we shift the debate from "resemblance to humans / ability to talk" to a more constrainable question: under resource constraints, can the system maintain and expand closed-loop agency?

If consciousness is rooted in biological agency, then SATC's implication for AI is not "bigger models" but "stronger closed loops."

In this framework, approaching Level III agency requires at least three kinds of structural conditions:

First, sustainable energy/resource constraints (not abstract computing power with infinite external supply); second, cross-scale closed-loop regulation (fast control and long-term adaptation coexisting); third, action–feedback loops of long-term co-shaping with the environment (structure and environment mutually sculpting).

This does not equal denying software intelligence but distinguishing two kinds of questions: language capacity and agency are not the same thing; the former can grow in training detached from the world, the latter requires structured closed-loop life-style conditions.

In the current context, this distinction is especially important: it's very easy to confuse "able to talk," "able to reason," "able to write" with "having agency." SATC's reminder is: language capacity may be the product of some advanced representation and statistical structure, while agency requires the system to self-maintain and actively change input conditions. The two can be related but cannot be simply equated.

### 7.1 Three routes: separating "intelligence engineering" from "agency engineering"

To avoid arguments staying on slogans, this book suggests roughly dividing AI routes into three kinds:

The representation and language route, obtaining strong representational capacity through large-scale data and model structure, good at language, reasoning, and generation; the control and embodied route, obtaining closed-loop capacity with the environment through robots, sensors, and control systems, good at action and adaptation; the self-maintenance and synthesis route, obtaining closed-loop closer to life-style through energy/resource constraints, long-term self-maintenance mechanisms, plastic materials, etc.

SATC's claim is not denying the first route but reminding: even if the first route is extremely powerful, it does not automatically entail Level III agency. What agency needs is structured closed-loop conditions, not just stronger representational capacity.

### 7.2 Case: equally "very intelligent," but closed-loop conditions completely different

Imagine two systems. One is a pure language system: it shows strong reasoning and explanatory capacity in dialogue but doesn't need itself to maintain energy, repair damage, or survive long-term in an environment. The other is an embodied system: it may not be good at long dialogues but must continuously act and self-maintain under conditions of limited power, drifting sensors, uneven ground, varying loads.

In SATC's question framework, the key difference between these two systems is not "which resembles a human more" but "which is more closed-loop." The capacity of the first system is mainly reflected in representation and generation; the capacity of the second system is mainly reflected in continuous closed-loop regulation under resource constraints and action–feedback maintenance.

### 7.3 An evaluation table: rewriting "seems like intelligence" as "whether closed-loop conditions hold"

If you treat SATC as an evaluation framework rather than a worldview slogan, you'll very naturally want an "evaluation table." This table doesn't need to be long; it only needs to ask questions in the right place: language capacity and reasoning capacity can certainly be strong, but they alone are insufficient to guarantee agency; agency depends on whether closed-loop conditions hold.

The first kind of question is whether resource constraints are real. Must the system pay a price for its own continuous operation? Does this price accumulate over time, forcing it to make strategic allocations? The "price" here is not only electricity but also sensor drift, hardware wear, environmental uncontrollability, task conflicts, and repair costs. As long as the price is absorbed by external supply and maintenance over the long term, the system is more like a display system than a self-maintaining system.

The second kind of question is whether closed loops can be maintained. Does the system act in real feedback loops and can it recover key variables under perturbation? The key variables here can be its own state (energy, temperature, damage, stability) or task-related variables (safe distance, grasping stability, route deviation). What you ask is not whether it "can output the correct answer" but whether it can maintain an agential state in continuous interaction.

The third kind of question is whether cross-scale regulation exists. Life-style agency is strong not because its instantaneous reaction is fast but because it nests multiple timescales together: fast control, short-term adaptation, long-term strategy, structural sculpting. For artificial systems, you should at least be able to point out: where is the fast closed loop, where are the slow variables, where is the long-term adjustment mechanism. Without cross-scale, it's hard to explain sustained autonomy; what you get is often just one-off task completion.

The fourth kind of question is whether externalized systems are coupled with closed-loop maintenance. Tools, symbols, plans, and institutions can certainly expand capacity, but they must return to closed loops: do they help the system more stably maintain key variables under resource constraints? Can they be learned, transmitted, accumulated, and in turn change the system's action strategy and mode of environmental sculpting? If externalized systems stay only at text output or one-off tool calls, they are more like "function plug-ins" than evidence of the structural transition Level III points to.

The significance of this evaluation table lies not in giving a裁决 on "whether a machine has consciousness" but in landing argument: you can clearly state which closed-loop conditions a system satisfies and which it lacks; you can also clearly state that if you want to approach higher-level agency, what piece you need to structurally supplement rather than continuing to stack capacity descriptions in language.

### 7.3.1 Two common illusions: taking "able to talk" as "able to live"

One of the most common illusions in current discussion is taking "explanatory capacity" as "causal capacity." A system being able to explain its own output self-consistently does not automatically mean it has closed-loop maintenance; it may also just take the stable conditions externally provided for granted. Of course you can praise such a system's representational capacity, but in the structural–evolutionary agency paradigm you must separate praise from criteria: representation and agency are not the same dimension.

Another illusion is more hidden: taking "having feedback" as "having closed loops." Many systems also have sensors, controllers, and can even make elegant responses in short periods, but their feedback is only a loop of local tasks, not a multi-scale organization around survival variables. They look very autonomous in demonstrations, but once environmental conditions change, costs accumulate, or goals conflict, they quickly expose dependence on external supply and maintenance: no true resource allocation, no long-term strategy, no structural sculpting. If you don't write out these constraints, you'll mistake short-timescale cleverness for long-term agency.

So what this evaluation table really wants to do is translate "intuitive attraction" into "judgeable." You can be attracted by language or by action, but you must answer: where does the attraction come from? From closed-loop conditions or from display conditions? Once you write this clearly, AI discussion will move from anthropomorphic clamor to analysis of structural questions.

**Structural constraint breakdown**

Language capacity can grow rapidly in training detached from the world, but closed-loop agency must grow under resource constraints and in feedback loops. If you want to discuss Level III agency, externalized systems (tools/symbols) must also be coupled with closed-loop maintenance, not merely existing as output text.

**Chapter summary**

This chapter wants to pull AI discussion away from anthropomorphic dichotomy: not "whether machines will have souls," nor "as long as bigger is fine." In SATC the key question is more basic and more operable: does the system satisfy the structural conditions for closed-loop agency? Does it maintain itself under resource constraints? Does it possess cross-scale regulation? Does it co-shape with the environment long-term?

If you accept this mode of questioning, you'll see a route different from "increasing parameter scale." It does not deny the enormous value of representation and language but reminds: language capacity and agency are not the same thing. Approaching higher-level agency requires stronger embodied constraints and deeper cross-scale closed loops, not just stronger generative capacity.

**Chapter reflective questions**

Please write down what you consider the "most agency-like" artificial system (autonomous driving, robot, automated trading, agent framework, etc.). Then don't rush to praise it as "smart" but evaluate it against three structural conditions one by one: are its resource constraints real, does cross-scale closed loops exist, does the loop of long-term co-shaping with the environment hold. Finally, honestly conclude in one sentence: in which places does it already resemble life-style closed loops, and in which places does it still more like a "display system fed externally"?

---

## Chapter 8: Common misconceptions and objections (FAQ)

### Summary

This chapter's task is not to win verbal battles but to translate common misconceptions into structural constraint questions: generalization, panpsychism, function assignment, unconscious processing, plant consciousness, and "whether agency equals instinct," etc. After reading you should be clearer: which arguments come from semantic confusion, which arguments can be converged by evolution, genetics, development, damage, and closed-loop indicators.

### 8.1 "Does aligning consciousness with agency over-generalize?"

SATC adopts a working definition for origin and comparative research, aimed at avoiding using human reportorial capacity as a definitional premise. Phenomenologically意义上的 consciousness needs to be characterized separately under higher-order structural conditions.

The worry about "over-generalization" usually comes from confusing two sets of semantics: you understand "consciousness" with experiential semantics but understand "agency" with origin semantics. Once so, any statement aligning consciousness with agency will seem to claim "all life has human experience," which of course triggers aversion. SATC's approach is not forcibly smoothing over semantic differences but writing out the division of semantics clearly: in origin and comparative discussions, we care about the spectrum of closed-loop capacities that living systems grow in structural evolution; in experiential discussions, we care about how this capacity spectrum corresponds to reportable experience under higher-order structural conditions.

After writing out this division of labor clearly, "generalization" is actually controlled: we don't call any system consciousness but put consciousness into the life continuum and constrain it with structure, metabolism, development, damage, and comparison. A working definition that accepts these conditions is qualified to enter research; a "more refined definition" that rejects these conditions is often just protecting some intuition rather than advancing causal explanation.

### 8.2 "Is this denying subjective experience?"

No. SATC's claim is: subjective experience itself is an important descriptive object, but it cannot directly serve as a causal starting point in the absence of structural constraints.

### 8.3 "Do plants also have consciousness?"

Under the semantics of "reportable experience," plants certainly do not possess human-meaning consciousness. Under the semantics of "minimal/primordial consciousness," plants embody distributed agency and belong to part of the continuum. The key lies not in fighting over vocabulary but in explaining structural conditions and capacity differences.

### 8.4 "Isn't agency just a functional label assigned by observers?"

SATC's structural functionalism requires agency to have a traceable structural realization and to be constrainable by evolutionary continuity, genetic and developmental differences, and materials of damage and pathology. Agency is not arbitrary labeling but a functional judgment constrained by structure and causal testing.

More specifically, why you feel "function is assigned by observers" is often because you've seen two failures: one failure explains any behavior as "purposeful," so agency becomes anthropomorphic story; another failure writes any system as "having function," so function becomes descriptive adjective. SATC avoids these two failures by relying on testable structural paths: if you say a system has some closed-loop capacity, then when structure changes, this capacity should change systematically; when development matures, capacity should unfold; when damaged or anesthetized, capacity should disintegrate or degenerate. Conversely, if you cannot write out these constrainable change paths, then so-called function is more like naming than explanation.

### 8.5 "Isn't this panpsychism?"

SATC does not need nor advocate "everything has mind." SATC discusses the spectrum of structured closed-loop agency in living systems. It avoids directly equating "consciousness" with "experience," thereby reducing the probability of "panpsychism"-type misreading; at the same time, it does not reduce consciousness to abstract computation, thereby avoiding extending consciousness to any system capable of computation.

### 8.6 "How are unconscious processing and automated behavior placed in SATC?"

SATC's grading describes the structural conditions and capacity spectrum of agency and does not deny the existence of many processes within the system that don't enter report or attention. On the contrary, higher-level agency is often accompanied by more complex automation and unconscious processing. SATC's point of distinction is: whether these processes serve structured closed-loop regulation and action–feedback maintenance, and how they are constrained by structural state.

In other words, unconscious processing is not the opposite of consciousness but more like the "chassis" of the consciousness system. A system capable of stably maintaining closed loops often must automate many processes: posture control, basic reflexes, habituated movements, resource allocation, error correction… They don't enter report not because they're "unimportant" but because they're so important they can't be re-discussed every time. What's truly worth asking is how automation is organized at different levels: is it local, short-timescale, or cross-scale, plastic? Can it be rewritten by learning, can it switch strategies amid environmental changes?

Therefore, when you discuss "the unconscious" in SATC, a better way to write is not treating it as another mysterious entity but writing it as part of the organization of agency: which closed-loop processes can be maintained without entering experience? Which closed-loop processes must enter higher-order integration to stabilize? Which structural variables' changes will turn automation from "helping maintain" to "runaway repetition"? Once these questions are written out, "the unconscious" goes from vocabulary competition to researchable structural question.

### 8.7 "If consciousness is agency, does consciousness just equal survival instinct?"

SATC's answer is: consciousness aligns with agency but does not equal reducing everything to "instinct." Agency is a set of structured capacities, including multi-timescale regulation and selection; it can manifest as instinctive reactions and also as learning, planning, externalization, and counterfactuals. Calling it "instinct" easily introduces intuitive misreading, so this book prefers to use "structured closed-loop capacity" to express it.

### 8.8 Case: translating the "plant consciousness" debate into a structural question

"Do plants have consciousness" is hugely controversial because the question mixes two sets of semantics: one is experiential semantics (can it feel pain, can it subjectively know), one is agency semantics (whether there is closed-loop regulation and input-condition change).

SATC's way of writing is not first adjudicating vocabulary but first translating the question:

If you ask using experiential semantics, what structural conditions do you need to propose to possibly support "experiential consciousness," and do these conditions exist in plants? If you ask using agency semantics, what you ask is: does plants possess distributed closed-loop regulation, can they change input conditions on multiple timescales, and how are these capacities constrained by structural destruction or developmental processes?

**Structural constraint breakdown**

Answers in experiential semantics require additional structural conditions, otherwise they easily slide into untestable arguments; answers in agency semantics can be constrained by observable closed-loop regulation, developmental sculpting, and structural destruction. The two sets of semantics can coexist, but in the same passage you must clarify which set you're using, otherwise discussion will necessarily "each speak their own language."

**Chapter summary**

This chapter's core work is to remove the most common interference factors in arguments: semantic confusion. Many people argue to the end not about who understands facts better but about who clings more to a certain word. SATC's approach is to first distinguish semantics: experiential semantics and agency semantics are not the same thing; they can coexist but must be distinguished first, then their connection discussed.

Therefore, SATC's purpose has never been "to win a word" but to use structural constraints to advance discussion to a researchable level. You can continue talking about experience and also about function, but you must be able to answer: what is the key structure, what is the key causal path, what observations would force you to revise your claim. Able to answer these questions, argument will move from position proclamation to joint advancement.

**Chapter reflective questions**

Please choose one FAQ answer you most disagree with and first write your objection in smooth Chinese — just in your usual way of speaking. After writing, do the second pass: translate it into the language of "structural constraints." What do you think the key structure is, what is the key causal path, what facts would you be willing to use to verify or overthrow it? After writing both passes, compare: which pass is easier to advance, which pass more resembles emotion or position?

---

## Part III Summary: turning "consciousness talk" into "research writing"

By now, SATC's stance should be clear enough: it does not attempt to win in the "vocabulary of experience" but pulls discussion back to a constratable position — structure, closed loops, evolution, genetics, development, and damage. So-called "noise reduction" is not making the problem simpler but pulling the problem out of the fog of rhetoric and exposing it to the light of causation.

If you read through this introduction and still feel the "mystery of experience" hasn't been solved, that's normal: it was never supposed to be solved all at once in the introduction. The task truly accomplished by the introduction is to place the puzzle back on the correct table. You no longer need to swing back and forth between experiential vocabulary and abstract vocabulary but can use the same set of biological tools to write the problem solidly: put the system into the continuum, write out structural constraints, write out closed-loop indicators, write out refutable points. This way, argument can be advanced rather than repeated.

When you persist in writing these actions into text, "unification" will go from wish to result. Unification is not persuading everyone into the same vocabulary but pressing different materials into the same discipline of reasoning: you can come from different disciplines, use different terminology, focus on different phenomena, but you all must explain structural anchoring points, explain agential indicators, explain timescales, explain refutable points. This way, discussion will naturally divide labor: some are responsible for describing phenomena more sharply, some are responsible for writing structural paths more clearly, some are responsible for pushing destructive materials and developmental materials closer. Theory's advancement no longer depends on "whose position is tougher" but on "whose constraints are written more concretely."

In writing, this stance can be compressed into a norm: first mark the starting point of explanation, then write out the path of constraints, finally explain the refutable points. A theory without refutable points is often just style; and SATC aims to preserve style while refusing emptiness.

### One-page summary: research program and extrapolation (Part III)

Three usages: compare, causation, extrapolate.

Compare — put different systems into the same set of closed-loop questions: what variables does it maintain? By what circuits? Can it change input conditions?

Causation — use empirical biological paths to converge explanation: how does evolutionary continuity constrain the capacity spectrum? How does genetics and development unfold the capacity spectrum? How does damage and pathology weaken closed-loop organization?

Extrapolation — use structural conditions to assess artificial systems: are resource constraints real? Does cross-scale closed loops exist? Are externalized systems coupled with closed-loop maintenance?

Three-paragraph research question: structural hypothesis — agential indicator — refutable point.

Two reminders: first, a theory's dignity lies not in being always right but in allowing correction; second, the goal is not winning vocabulary but winning advancement — advancing consciousness discussion to a level that is researchable, distinguishable, and testable.

## Appendix A: glossary (Chinese–English with brief explanations)

- Structure: an organized form that can maintain metabolic order and organize interaction

- Biological Agency: structure-dependent capacity for closed-loop interaction

- Enactive Capacity: an emphasized term for agency: as capacity rather than agent

- Consciousness: the organismal-level expression of agency; a working definition in this book

- Metabolic Self-Maintaining Agency: minimal closed loop: self-maintenance and boundary maintenance

- Distributed Sessile Agency: no center, distributed regulation and growth-mediated

- Centralized Sensorimotor Agency: rapid sensorimotor closed loops with central circuit integration

- Abstract–Projective Agency: cross-temporal-spatial projection and externalization: tool/symbol systems

- Taxonomy: organizing objects of comparison using traceable relational structures

- Phylogeny: the continuous background and selection pressure of capacity differences

- Homology: structural/functional similarity due to shared lineage constraints

- Convergence: functional similarity formed under similar pressures in different lineages

- Closed Loop: self-maintaining control structure formed by action–feedback coupling

- Input Conditions: distributions of stimuli and resources changed by the system through behavior/regulation

- Homeostasis: maintaining key variables within a sustainable range under perturbation

- Cross-Scale Regulation: organizational mode nesting fast control and slow variables

- Perturbation: external or internal changes pushing the system away from the target range

---

## Appendix B: selected references

1. Maturana, H. R., & Varela, F. J. (1980). *Autopoiesis and Cognition: The Realization of the Living*. D. Reidel.

2. Varela, F. J., Thompson, E., & Rosch, E. (1991). *The Embodied Mind: Cognitive Science and Human Experience*. MIT Press.

3. O’Regan, J. K., & Noë, A. (2001). A sensorimotor account of vision and visual consciousness. *Behavioral and Brain Sciences, 24*(5), 939–973. doi:10.1017/S0140525X01000115

4. Friston, K. (2010). The free-energy principle: A unified brain theory? *Nature Reviews Neuroscience, 11*(2), 127–138. doi:10.1038/nrn2787

5. Tononi, G. (2004). An information integration theory of consciousness. *BMC Neuroscience, 5*, 42. doi:10.1186/1471-2202-5-42

6. Baars, B. J. (1988). *A Cognitive Theory of Consciousness*. Cambridge University Press.

---

## Appendix C: plan for subsequent volumes (draft)

This "Introduction" is more like "legislation and入门." If you hope to write SATC as a sustainably advancing research program, subsequent volumes can advance along a three-volume structure:

The first volume remains the "Introduction," responsible for laying out the "foundational problems" of definitions, ontological commitments, origin and grading, and methodological boundaries. The second volume can be the "Neuropathology and Pathology Volume," using materials of development, damage, anesthesia, and degeneration to show "how structural constraints converge explanation," letting theory appear in destruction and recovery. The third volume is the "Artificial and Synthetic Agency Volume," using SATC as an evaluation and design framework to discuss the structural conditions of embodied AI, organoid intelligence, and synthetic life: which conditions engineering can do, which conditions engineering must acknowledge as limits.

---

## Appendix D: discussion questions and writing templates (optional)

1. Explain SATC in two paragraphs: the first paragraph to high school students, the second paragraph to neuroscience graduate students. Neither paragraph may use "consciousness is information processing" or "experience is irreducible" as a key sentence.

2. Use the "structure-first rule" to rewrite an introduction to consciousness theory you've read: preserve its descriptive contribution but point out the structural constraint link it lacks.

3. Write a "counterfactual question": if a system is extremely powerful in language but lacks closed-loop self-maintenance, how would you position it in SATC?

4. Write a "consciousness phenomenon" you're familiar with in two layers: the first layer is experiential description (as specific as possible), the second layer is a structural–agency explanatory starting point (write only structural conditions and closed-loop capacity, not experiential vocabulary).

5. Choose a species or system (bacteria, plant, insect, fish, bird, mammal, robot, all fine) and write a "comparison paragraph": what key variables does it maintain? Through what circuits? How does it change input conditions? What mode of comparison — homology, convergence, or analogy — are you using?

6. Write a "damage-test sentence": if I destroy/weaken some key structure, what kind of capacity collapse would I expect to see first? How would this collapse manifest as systematic change in agential indicators?

7. Write a "developmental-unfolding sentence": if I place the same system at different developmental stages, which capacities would appear first and which would appear later? What do you think are the signs of key structural maturation?

8. Write an "extrapolation-assessment paragraph": choose an artificial system and evaluate it with four questions — resource constraints, closed-loop maintenance, cross-scale regulation, externalized system coupling — and write out the limitations you're willing to acknowledge.

9. Write a "taxonomy self-check": whether you secretly used humans as the default reference in comparison? Whether you treated analogy as homology? Please re-rank your objects of comparison by phylogenetic or structural relations, then rewrite your comparison paragraph.

10. Write a "variable inventory": list three key variables for the phenomenon you care about (must be describable by perturbation and recovery) and explain which level of structure each corresponds to (metabolism, circuit integration, externalized systems, etc.).

11. Write a "minimal counterexample": imagine a system that performs well on some task but lacks self-maintenance and long-term closed loops; then imagine a system weak in language but strong in closed-loop maintenance. Write out respectively why you wouldn't treat the former as high-level agency and why you'd be willing to place the latter on the continuum.

12. Write a "concept substitution exercise": replace words you're accustomed to using (consciousness, unconsciousness, idea, attention, will, etc.) with structural–agency language and see which sentences still hold and which immediately become vague.

One-page template (directly usable as a short-essay structure): Topic: Rewrite [phenomenon/system] using the structural–evolutionary agency paradigm. First paragraph (phenomenon): write only observable facts and experiential description, prohibit directly drawing causal conclusions. Second paragraph (structural hypothesis): write out key structural conditions and constraint paths (which entry — evolution/genetics/development/damage — is most suitable). Third paragraph (agential indicator): write operable closed-loop indicators and timescales, clarifying "maintaining what variables, changing input conditions how." Fourth paragraph (refutable point): write at least one prediction you're willing to be overthrown and state what you'd revise after overthrow.

13. Write a "scale-alignment paragraph": for the same phenomenon, write the key changes at three timescales (seconds/minutes/days or weeks) and explain which timescale you mainly use in your argument and why.

14. Write a "boundary paragraph": in the system you discuss, what does "boundary" specifically refer to? Cell membrane, individual boundary, group boundary, or externalized system boundary? How would boundary change change agency?

15. Write an "input-condition rewrite": rewrite a common description (e.g., "it reacts after receiving a stimulus") as "how does it change input conditions" and write what you think is the most crucial closed-loop link.

16. Write a "causal-sentence compression": compress a section of your argument into a causal sentence in which structural variables, agential indicators, and refutable points must appear simultaneously.

17. Write a "minimal mechanism diagram": without drawing, write a three-step chain: structural variable → closed-loop capacity → observable behavior/indicator. Each step writes only one key causal arrow.

18. Write a "terminology self-discipline table": list the five words you most want to use in your article and write for each a sentence of "in this book this refers to what / I don't refer to what."

19. Write a "grading-criterion paragraph": write why the discussion system more resembles Level 0/I/II/III and write a falsification condition that would make you willing to change the criterion.

20. Write a "reverse-constraint-from-conclusion": first write a conclusion you most want to say, then reverse-engineer: what structural conditions must it depend on? Are these conditions reasonable evolutionarily and developmentally? If unreasonable, which part of the conclusion are you willing to delete?

Instructions: these questions are not asking you to "write standard answers" but to write your own words into a constratable position. Each time you do an exercise, you're pressing a stretch of intuitive language back into structural language: turning "I think" into "structural variable," turning "seems like" into "comparative relation," turning "explain" into "refutable point." When you find yourself unable to write on, it's usually not that you're not smart enough but that you've stepped on the red line this book repeatedly reminds: you're using descriptive vocabulary to masquerade as causation, or using abstract vocabulary to skip life constraints.

21. Write a "pathology or failure case": choose a failure state you're familiar with (anesthesia, coma, epileptiform discharge, metabolic disintegration, attentional disintegration, etc.), write only in structural–agency language how it fails, and write which observation you'd be willing to use to distinguish "structural failure" from "descriptive error."

22. Write an "evolutionary-continuum paragraph": choose a capacity (e.g., exploration, avoidance, tool externalization, long-term memory) and write what you think are the two structural transitions it might experience on the continuum and explain how agential indicators should change before and after the transitions.

23. Write a "genetics–development entry": assume some genetic/regulatory difference would change the maturation speed of a certain kind of structure; predict how it would change the closed-loop capacity spectrum? Write at least one counterexample condition you're willing to acknowledge.

24. Write a "comparative-relation annotation": for the same "looks similar," write how you would change the strength of argument in the three cases of homology, convergence, and analogy respectively, and explain what evidence you'd need to supplement before you dare elevate analogy to a stronger inference.

25. Write a "tool-externalization paragraph": choose an externalized system you rely on daily (notes, calendar, map, collaborative workflow), and write how it changes your closed-loop maintenance: what timescales does it expand, and in what places does it create new fragility.

26. Write a "forbidden-word list": list the five words you most easily get lazy with in writing (e.g., "consciousness," "self," "intention," "information processing," "reaction," etc.) and write a replaceable structural–agency expression for each.

Supplementary template (four-paragraph format for evaluating artificial systems): First paragraph (resource constraints): what costs must the system pay for continuous operation? How do costs accumulate and force strategic allocation? Second paragraph (closed-loop maintenance): what key variables does the system maintain? How does it recover under perturbation? In which situations would it lose agential capacity? Third paragraph (cross-scale regulation): does the system have only short-timescale control, or are there slow variables and long-term adjustment? How does the long-term adjustment mechanism change structure or strategy? Fourth paragraph (externalized coupling): are tools/symbols/plans coupled with closed-loop maintenance, can they be transmitted and accumulated, and in turn change the co-shaping relation between system and environment?

Fdamage Price P/L Factor, a Sequel

DiVoLA V0.5.0 Released

On Zhang Xuefeng, the 'Poison of Education'

Notes on Writing

Afterword: The Tragedy of Freud

SATC — Preliminary Manuscript

Afterword

Check the auto-update effect in ten minutes

Test the new blog's auto-fetch

About the DIVOAI AI Work Team

Uncle Ju AI Assistant Platform — Complete Technical Analysis Report for the DiVoAI Project

Let's release some practical AI tools first

Big AI news: Yann LeCun reportedly leaving Meta to start his own company

Implementing Multiple Auth Methods for MCP

Turning a Drupal Instance into an MCP Server

Local PDF Recognition Hits Resource Limits on Multi-Page Documents

1. Turning Drupal into a Knowledge Base: RAG-Ready Drupal

On the Positioning of Drupal Development

Added a Content-to-Markdown Text Format Feature

00 Preface: What Exactly Is the Truth of Life?

Happy Mid-Autumn! Proudly Recommending Our New Learning-Ability Enhancement and Study-Abroad Programs!

对人文社会学科挑肥拣瘦？穷追猛打？有必要嘛！

Watching Season 7, Episode 8 of "The Good Doctor" Today Reminded Me of a Passage I Once Read Online

My Battle with Bipolar Disorder — Linda Logan, 2013

Raising Children Is Always an Experiment

一手“反科举”，一手“举孝廉”？长沙中考新政引热议

"My Uncle Li Dazhao" or "My Father"?

Some Study-Method Summaries from \"Lamb Rolls\"

The Method of Scholarship — Hu Shi's 1935 Speech in Nanning, Guangxi

The Method of Scholarship — Hu Shi's 1935 Speech in Nanning, Guangxi

The Outline and Presentation PPT for "The Straight-A Student's Cultivation Handbook" Are Finally Finalized

反对自杀，更要严惩无情薄情郎！

A Father-and-Son Conversation About Fathers and Sons

On Stable Friendships and Their Role in a Child's Growth and Personality Development

三代价值观同堂？

Some Experiences and Thoughts as a Legal Consultant

DeepSeek也许是国运神器，但是大可不必迷信。更加不要炒作什么Manus

Starting to Redo the Syllabus and Teaching Materials for "The Straight-A Student's Cultivation Handbook" Today

On Learning Methods: the Deductive and Inductive Approaches

《抑郁症伪手册》序

A Talk with My Eldest Son About Bullying and Some Other Issues

杨振宁：我一生最大的贡献，是帮助中国人克服了自己觉得不如外国人这个心理。

No Exam Has Any "Black Technology" — No Need for Anxiety; Training Should Be Moderate and Effective, Not Following the Crowd

三毛作品《逃学为读书》全文

Reading Notes on San Mao's "Skipping School to Read"?

The World Is Nothing but You — Addendum, March 2025

面向未来的教育——2025年开学第二课

我的办学梦兼回顾我的一些其他初心

Speech at the 2025 Hundred-Day Oath Ceremony for Students of Wang's Private School and the Ace Student Training Manual

Discussing with the KIMI 1.5 Reasoning Model: "There's No Need to Privately Deploy AI Tools or a Private Knowledge Base"

Redefining the Chinese Translation of "Conscientiousness" as "Self-Drive"

什么是咨询和咨询顾问？What is consulting and who is consultant？

《羊粪球》

Living as a Consultant

My Eldest Son's Exam-Prep Plan, and What Gaming Teaches About Study Methods and Mental Energy

顺AI而生——面向未来的教育

《羊肉卷》代序

Chatting About the Wuxia Novels I've Read

留学专家王觉菊老师的几篇重要文章

精准防控自己上厕所的次数和时间

Under any circumstances, human life is what matters most

A Chinese-language teacher said this about Ye Shengtao

不过我也不惜的让你们来洗我儿子

Work can be done anywhere across the country

A few key points about paired learning

Passive-unconscious versus active-conscious learning

On Classroom Discipline

唉，高中生爱教人做人

铁打的清华

阅读理解：被扔下楼的韦同学有没有欺负过扔他下楼的黄同学？

A Child Unraised Is the Father's Fault

What Is the Most Appropriate Translation for the Course "Social Investigation"?

counttry and you 周一, 09/23/2019 - 09:22 logged

周六, 09/21/2019 - 21:35 logged 《两个孩子的私塾》

美国著名的大学爆发的招生弊案，撕碎了象牙塔的遮羞布

《除了你，这个世界一无所有》续

美国留学读大学读硕士博士

留学美国哥伦比亚大学最好找和最不好找工作的专业

无暂住证虽然上不了北京的学却能上哥伦比亚大学、哈佛和耶鲁！

见义勇为应该先考虑自身安全