Deconstructing MBTI, gaining a bird’s-eye view of psychology, gaining a bird’s-eye view of thinking, and gaining a bird’s-eye view of the world—all are products of a single intellectual tool.
Original work: tomzheng (tomz)Network notes compiled and elaborated by: AI based on publicly available notes
This book combines the six main volumes and supplementary volumes of the “Concept Vector Series” into a single edition. While the individual volumes remain independently available, the unified edition has been rearranged according to the order in which readers typically encounter them: starting with the most familiar—human beings (the psychological framework)—entering through thought, then encountering the methods themselves (concept vectors), followed by the world (the philosophical framework), society and life, and finally aligning everything into a comprehensive table (framework synthesis). The supplementary volumes conclude with an appendix featuring a raw-materials repository, self-referential readings, and a one-page reference card for the entire book.
Each section can be read independently, and skipping around won't cause much loss; the only recommendation is to save Section Six (Integrated Framework) for last.
Each section of this book has been compiled, elaborated upon, and expanded by AI based on the author’s publicly available online notes. It is particularly important to note that a significant number of arguments presented in this work do not align with the author’s own views, and some of the additional content added during the compilation process has not been approved by the author. Readers are advised to exercise independent judgment. Any discrepancies in viewpoint or overextension of interpretations shall be the responsibility of the compiler, not the author.
Psychological Framework: Classification of Individuals and Inner Structure
The entire book starts with the most familiar subject: people. How does the human mind work? What are the different kinds of distinctions among people? After all those years of administering the MBTI, what exactly does it really mean?? ((The methodology behind this classification system is covered in Part Three; for now, let’s just get started with this part.)
The MBTI categorizes people into sixteen types using four letters, yet no one has ever clearly explained what those four letters actually mean. This book provides an answer based on a psychological framework:
●The four dimensions—Introversion/Extraversion, Sensing/Intuition, Thinking/Feeling, Judging/Perceiving—correspond one-to-one with MBTI’s E/I, J/P, T/F, and S/N. The four columns formed by the intersection of these four dimensions actually represent a streamlined pipeline: “Information enters the brain → People output externally.”
This book doesn't challenge MBTI; rather, it provides MBTI with a “principles manual.”
Chapters 1 through 3 discuss four dimensions; Chapters 4 and 5 explore the inner circular structure and its operational pathways; Chapter 6 discusses revisions to the MBTI; Chapter 7 compares this theory with other personality theories; Chapter 8 is dedicated exclusively to INTPs; Chapter 9 focuses on applications; and Chapter 10 provides a summary.
Each chapter can be read independently. Readers who have taken the MBTI test are encouraged to read with their own four-letter type in mind.
Everyone has taken the test, yet no one can explain it clearly.
In recent years, MBTI has almost become a social currency. People include “INFP” in their self-introductions, ask during job interviews and chats, “Are you an E or an I?”—even at matchmaking events, you’ll find people discussing the four-letter code.
But as soon as you ask just one more question, the excitement will be exposed:
·Are E and I really about “whether or not one loves socializing,” or “where one’s energy comes from”?
·S and N—why is one called “sensory experience” and the other “intuition”? What is their relationship to the everyday terms “sensation” and “intuition”?
·J and P—officially translated as “Judging” and “Perceiving”—what does “Judging” mean? And why do P-type people dislike making plans?
The more practice tests you take, the more ambiguous the meaning of these four letters remains. MBTI provides categorization but fails to explain the underlying principles. It tells you what an INTP looks like in behavior, yet it doesn’t clarify which specific aspects of the psyche each of the four switches—I, N, T, and P—actually controls.
The psychological framework is here to fill in this instruction manual. Its claim is: The four letters of MBTI are not four random personality labels, but rather four independent switches on a psychological machine.
Take another look at this chart and read your four-letter type again. INTP—Introverted + Intuitive + Rational + Perceiving: a person whose attention is inward, who doesn’t like constraints, who thinks logically, and who enjoys absorbing sensory experiences. Isn’t that much more straightforward than the official explanation?
This correspondence table wasn't pasted in afterward. The author’s notes preserve its origin: the psychological framework came first (derived from concepts such as Freud’s, introversion versus extroversion, and others), and later it was discovered that this framework happens to be isomorphic to MBTI’s letter system—eight terms correspond exactly to eight letters, no more, no less.
There are two possible explanations for how two independently developed systems can align: coincidence, or the fact that they’ve both stumbled upon the same underlying reality. The author is betting on the latter.
But he acknowledged that the referee for this betting game can only be a human: determining whether something is a coincidence or an isomorphism relies on human experience—a specialist in concept vectors, deeply familiar with the theoretical domain, and someone who has undergone a long process of refinement—only such a person can identify truly aligned words. While the Big Five personality traits rely on statistics, aligning concept vectors requires not only statistics but also experience and continuous iterative updates. Alignment isn't a one-time decision; it's a long-term, ongoing process of calibration.
Of course, the correspondence isn't perfect—and it's precisely in these imperfections that the real intellectual challenge lies: The J/P dimension in MBTI is ambiguously defined in the official theory. As argued in Chapter 6 of this book, the “planning” dimension should be removed from MBTI and replaced with “perception/action.” This is a surgical intervention applied to MBTI by the psychological framework itself.
You don’t need to do 93 questions. The author’s minimalist version has only four questions:
Introvert or extrovert? (Introvert is I, extrovert is E)
Are you more inclined to obey or to cherish freedom? (Obeying is S, cherishing freedom is N)
Rational or emotional? (Reasoning is T, emotion is F)
Answer based on your first instinct. Once you’ve finished, you’ll have four letters—and this time, you’ll know what each letter means.
For the company team-building event, HR handed out a set of MBTI questionnaires, and twenty people immediately dove into answering them.
Half an hour later, the results came out—and the room immediately erupted in excitement: “I’m an ENTJ—born leader!” “Wait, I’m an I? But I’m actually pretty talkative!” “Is this really accurate? Last year’s test gave me a different result.”—Everyone had their own four-letter personality type, but upon closer inspection, each person interpreted their letter differently: Some saw E as “socially outgoing,” others viewed J as “neat and tidy,” and still others stared at T for ages, wondering if it stood for “sports.”
This is the current state of MBTI: the tests are being distributed faster than the instruction manuals. Twenty people have received twenty different sets of coordinates, yet few of them even know what the coordinate axes mean—so the letters have become the new constellations, sparking lively conversations but leaving everyone confused when it comes to actually using them.
The approach in this chapter is exactly the opposite: we’ll start by discussing the coordinate axes, not the tests. Only after you’ve figured out what each of the four switches controls will those four letters truly begin to make sense.
Before you continue reading, please bring out the principles from Part Three once more: These categories describe stable biases, not switches; the corresponding characteristics are probabilistic, not definitive. Your four-letter code is a coordinate on a map, not a cage.
In the next chapter, we’ll take apart one switch at a time.
Switch 1: Introverted/Extroverted
This is the least ambiguous switch. The author’s definition consists of only two lines:
Introverted: Likes to be alone.
There’s no need for esoteric explanations like “energy sources.” If you enjoy being alone, that simply means you’re introverted. However, there’s a set of “synonyms” for extroversion and introversion worth remembering—they can help you connect this trait to other things:
In the “Passive/Active” row, the author marked it as “questionable”—this honesty is worth learning from: This corresponds to the draft version, not an ironclad rule.
Switch Two: Perception/Action
This switch is where the psychological framework makes its greatest contribution to MBTI—and it’s worth taking our time to explain.
The perceptive person loves to experience things. Watching a movie, exploring a city, or reading a book—just the act of taking in these experiences is itself a source of enjoyment. The action-oriented person, on the other hand, loves to produce results. If he watches the same movie, he’ll write a film review; if he explores the same city, he’ll complete his check-in checklist.
Perception is input; action is output. The author’s thesaurus widens this switch considerably:
How to read this table: Start from the top and go down, line by line, to see where you stand. Most people’s self-assessments will flip several times along the way—no problem; that’s exactly what it means when we say, “The switch is an axis, not a point.”
This switch corresponds to J/P in MBTI. In MBTI, J is interpreted as “Judging,” and P as “Perceiving”—a rather roundabout way of putting it. A psychological framework offers a much more straightforward explanation: Perceiving (P) means being willing to spend time on experiences; Acting (J) means being willing to spend time on outcomes.
After scrolling through short videos for two hours,
“This switch—‘perception is input, action is output’—can be measured with just a smartphone screen.”
On Saturday night, the two of them each spent two hours scrolling through short videos. The perceptive one put down his phone, feeling completely satisfied—those two hours of lively clips, funny jokes, and adorable pets: just the act of consuming it all was an enjoyment in itself, making the time well spent. The active one, however, put down his phone and felt a hollow emptiness inside—two hours had passed, yet nothing remained; he’d have to get up and create something, even if it meant tidying up the kitchen—only then would this evening feel “not wasted.”
In the same two-hour period, one person feels they’ve gained something, while the other feels they’ve lost—this isn’t a difference in self-discipline; it’s a difference in where their “happiness plug” is plugged in: one is plugged into “input,” the other into “output.”
So don’t take at face value the blanket statement that “scrolling through short videos is a form of moral decay.” For those who are perceptive, it’s like recharging their batteries; for those who are driven to action, it’s actually a drain on their energy. It’s far more helpful to figure out what kind of “socket” you’re plugged into than to blame yourself.
The author also noted a set of tentative correspondences (labeled as preliminary associations, not conclusions):
Happiness keeps people anchored in the present moment (perception), while anxiety pushes people toward action (behavior). If this association holds true, then the “perception/action” switch isn’t merely a preference—it’s a directional drive hardwired into the reward system. What the author means is this: For most of the time, humans are driven by instinct—dopamine (actively seeking out things to do) and anxiety (passively reacting to what comes their way)—and it’s precisely these two directions that shape our behavior.
Two switches intersect to produce four “columns” of the psychological framework. At first glance, they resemble four grids arranged side by side:
But if you carefully examine these four categories, you’ll discover something even more intriguing: they’re not four separate categories at all—they’re actually a single line. Here, “inner” and “outer” carry their most literal meanings: inner refers to the inner world, while outer refers to the external environment.
·Extroverted sensing: Obtaining information from the external world—like a sensor;
·Introverted perception: Information is processed internally—integrating new information into one’s own internal categories.
●Inward action: Engaging in internal planning and decision-making to formulate plans and action goals;
Extraverted action: An action generated through interaction with the external environment—like an actuator.
Read in this sequence: Information enters the brain from the outside (extroverted perception) → is internally processed and digested (introverted perception) → is internally planned into goals and strategies (introverted action) → and finally manifests as actual external actions (extroverted action). From the moment information enters the brain to the moment a person outputs it externally, it forms a complete, unbroken assembly line—with no stations missing.
This assembly line will be ready for use very soon. For those who are extroverted and perceptive, their sensors are highly sensitive—no new development can escape them, making them the most well-informed members of their social circles. As for those who are introverted and action-oriented, they have powerful planning skills; they always keep a detailed blueprint in their minds—but unless you ask, they won’t volunteer any information. The 16-item checklist in Chapter 4 is precisely about fitting these four rows onto this very assembly line.
Switch Three: Rational/Emotional
The definition is equally simple:
Rational: Likes to reason things out.
When rational people encounter disagreements, their first reaction is, “Let’s sort out the logic”; when emotional people encounter disagreements, their first reaction is, “What kind of attitude is that?” Neither approach is superior—one clarifies matters thoroughly, while the other preserves relationships. Both are essential to human society. Synonym list:
Pay attention to the two middle lines. Rationality and emotion are not merely “styles of speech”; within the psychological framework, they point to the deepest structure: rationality represents the force on the side of consciousness, while emotion represents the force on the side of the subconscious (instinct). The inner circle described in Chapter 4 depends entirely on this pair of forces to keep turning.
Switch Four: Control/Random
·Self-control: Likes to restrain oneself.
·Reckless: Doesn't like to control oneself.
The disciplined person sets an alarm, makes a plan, and sticks to rules; the free-spirited person sleeps until they naturally wake up, goes with their mood, and hates being told what to do.
This switch corresponds to the S/N dimension in MBTI. In MBTI, S/N is interpreted as “Sensing/Intuition”—whether one focuses on concrete details or on abstract possibilities. The psychological framework offers a more psychologically grounded interpretation: people who are “controlled” adhere to the rules of reality (hence appearing “real”), while those who are “spontaneous” follow their inner voice (hence appearing “ethereal”). The reason intuition feels “direct” precisely lies in its ability to bypass the constraints imposed by external rules.
This interpretation has an even deeper layer, hidden within the conceptual framework of Part Two. The author’s analysis breaks it down as follows: actual sensation is, in essence, flat thinking, while intuition is, in fact, hierarchical thinking. People with a flat mindset focus on details and fail to see the bigger picture, thus needing guidance from others and appearing more submissive—this is what we call “reality.” By contrast, people with a hierarchical mindset come equipped with their own mental maps; what we refer to as intuition is, at its core, big data that has emerged from the accumulation of hierarchical structures. Such individuals don’t need others to lead them—they lead themselves.
The four switches can also combine in pairs to form the “row” of the psychological framework:
·Reason + Spontaneity = Self (Free Reason): Independent thinking, acting according to one’s own logic—this is consciousness itself;
●Reason + Control = Superego (Reason · Control): Willpower and attention—means by which reason controls instincts;
Sensibility + Control = Non-Self (Sensibility · Control): Sensation, feeling, and emotion—note that this is not an “instinct that’s been managed”; on the contrary, it’s precisely the means by which sensibility manages rationality.
Sensuality + Spontaneity = the Id (free sensuality): desires, habits, beliefs—our instinctive urge to revel and let loose.
Does it look familiar? Freud’s name has appeared. But their positions are different from those Freud arranged—this is the topic of the next chapter.
Four switches may sound abstract, but you can actually figure them out in just one weekend.
On Saturday morning, some people wake up and immediately invite others to play sports (extroverted + action-oriented), while others curl up on the couch binge-watching documentaries (introverted + perceptive). At noon, a friend bursts into tears, saying she’s just broken up with her partner. Some hand her tissues and join in cursing her ex (emotional), while others start helping her analyze the list of problems in the relationship (rational). In the afternoon, they’d originally planned to head to the library—but suddenly everyone decides to go sing karaoke. Some feel “It’s frustrating that the plan has changed” (control-oriented), while others think, “Great! I’ve been dreading reading anyway” (spontaneous).
Perception/Action: Do you prefer input or output?
●Rational/Emotional: Reason or emotion?
Control/Spontaneity: Do you restrain yourself or let yourself go?
Four switches, sixteen possible combinations. In the next chapter, we’ll assemble the last two sets of switches into a rotating ring: the Ego, the Superego, the Non-Self, and the Id.
The engine of psychological activity
The first three chapters have all been about “classification”—that is, what kinds of people there are. This chapter is about “operation”—how a person’s inner world functions.
The psychological framework's response consists of only one sentence:
Human psychological activity is an interaction between instinct and consciousness.
Instinct is the感性 of freedom (the id), while consciousness is the rationality of freedom (the ego). The two argue daily, yet they also collaborate daily. The question is: Since instinct and consciousness are separated by an abyss, what enables them to communicate? They rely on two means:
·Non-I (Sensibility · Control: sensation, emotion, desire, mood, interest)—this is the means by which sensibility manages rationality. Note that it is not an “instinct being managed”; quite the opposite: it is the very hand of instinct reaching out to govern rationality. When emotions take over, even the most meticulously planned schemes can easily be thrown off course.
·The Superego (Reason and Control: Willpower, Attention)—This is the means by which reason controls instinct. It’s through this that consciousness restrains impulses and anchors itself firmly to the chair it’s supposed to be sitting on.
Where is the boundary between means and content? According to the author’s distinction, informational content encompasses both emotion and reason—since both are involved in accumulating external information, they qualify as content. By contrast, the “non-self” and the “transcendent self” lack informational content and remain constant and unchanging; thus, they are classified as means. It’s also important to note that in this framework, “emotion” refers specifically to instincts and the subconscious—according to the author, it’s “a brain operating without any logic.”
Here’s a particularly clever detail: Someone once asked a sharp question: “If this very ‘me’ is brimming with emotions, how can it possibly have anything to do with rationality?” The author’s response was: “Exactly—precisely because this part doesn’t follow logic is precisely why it can help emotion rein in and even control rationality.” Reason cannot control reason—what always stops an eloquent speaker isn’t necessarily a better argument; rather, it’s emotion. That’s precisely why, on the debate stage, appealing to emotions often trumps appealing to logic.
A ring-shaped structure, not a four-story building.
Arrange the four positions, and you’ll find that they connect end to end:
The Id (the instinct itself) is acted upon by the Non-I (sensory means), thereby influencing the Ego (consciousness itself). The Ego, in turn, uses the Superego (rational means) to regulate the Id. And as the regulated Id accumulates energy for its next surge—sensory and rational forces each occupy half of the loop, exerting opposing yet complementary pressures, thus forming a ceaseless circular motion.
Freud's three levels are: the id, the ego, and the superego. The psychological framework has four positions: the id, the ego, the superego, and the non-self, with explicit annotations:
In Freud’s theory, the ego lies between the superego and the id; therefore, it does not correspond to any specific position in the table.
The meaning is this: Freud’s arrangement forms a linear structure (Id—Ego—Superego), whereas the psychological framework is organized as a 2×2 matrix (Rational/Sensory × Spontaneous/Controlled), which, when set in motion, takes on the form of the circular ring shown above. The extra element—the “Non-Self”—does not have an independent position within Freud’s system: it is the toolbox held by the sensory realm—the collection of your sensations, emotions, desires, and moods—those very means by which instincts reach into consciousness and influence decision-making.
Here’s an example: You know full well that you should be writing (your own plan), yet for no apparent reason, you feel restless and can’t seem to sit still (it’s not your conscious self taking action—emotions are managing your rationality). Finally, you turn on the Pomodoro timer and force yourself—with sheer willpower—to stay seated (now it’s your superego stepping in—rationality takes over and starts managing your instincts). It’s a daily struggle; the cycle has turned full circle, and all three “characters”—the id, the ego, and the superego—have taken their places.
What’s placed in each of the four positions?
The four rows represent four characters, and the four columns represent the four stations in the information pipeline. Where the rows and columns intersect, the resulting sixteen cells are the true stars of this table—each cell answers the question: “At this particular station in the pipeline, what exactly does this character handle?” Let’s take a look at them one by one.
My Four-Panel Self (the assets of consciousness itself across the four stations, collectively forming “My Life”):
Ability (Extraverted Sensing): The skills you acquire from the external world—everything you’ve seen and learned comes from outside.
Hobbies (Introverted Perception): Things that keep getting revisited and cherished within—no particular reason, just because they’re enjoyable, and the more you savor them, the richer they become.
Career (Introverted Action): The long-term goals that have grown from your inner planning—what you intend to become.
Relationships (Outward Action): Things formed through interactions with others—relationships can only emerge from reciprocal exchanges; you can’t come up with them just by staying indoors and thinking about them.
These four boxes aren’t a static checklist—Chapter 5 will transform them into a flywheel: Relationships reveal the capabilities you’re missing; capabilities fuel your career; your career sustains your passions; and your passions, in turn, bring people together in meaningful relationships.
Attention (Extraverted Perception): The power of focusing outward. When discussing how to overcome procrastination, the author points out that the very first tool is attention—precisely what a Pomodoro timer is designed to manage.
Perseverance (Introverted Perception): The strength to persist inwardly. There’s a very practical remark in the notes: Behind perseverance lies passion—“a strong desire to do something” leads to action. When you find it hard to keep going, what’s lacking isn’t toughness; it’s simply the desire itself.
Willpower (Inner Action): The strength to endure inwardly. “Used to overcome pain”—the plan is already set; all that’s left is to bear it.
Reactivity (Outward Action): The ability to respond swiftly to external stimuli. “Once a habit is formed, there’s no longer any pain—only inertia.”—The fastest response is one that happens without conscious thought.
When Chapter 9 discusses procrastination, these four weapons will take the stage one by one.
Sensation (Extraverted Perception): The sensory reports directly received from the external world—cold, hot, sour, sweet—vote even before reason has a chance to speak.
Emotions (Introverted Perception): The parts that keep fermenting within—love, hate, closeness, and distance—become ever stronger the longer time passes.
Desire (Inward Action): The impulsive urge to mentally calculate and target what one desires. Desire often disguises itself as a plan, infiltrating the hall of rational deliberation.
Emotion (Outward Action): The part that bursts outward—expressed visibly, directly influencing your interaction with the environment. Even the most rational and well-thought-out plans are often overturned right in front of everyone by it.
Extroverted Sensing: Innate awareness—the part you acquire without learning, pre-installed at birth.
Faith (Introverted Perception): The deepest conviction within—the things you believe without needing to prove them.
Experience (Inward Action): The ruts left behind in your heart by past events—these quietly dictate how you calculate and how you make choices.
Habit (extroverted action): A pattern of behavior executed without much thought—this is the most effortless form of external expression and also the hardest to change.
When a person introduces themselves, what they say almost always falls within the confines of the “ego quadrants”; to truly understand what governs a person, one must look at the “id quadrants”; and whether one side wins or loses in a tug-of-war depends on which—whether the “superego quadrants” or the “non-ego quadrants”—has the stronger tools.
In the author’s notes, there is a rather stark entry: comparing the five techniques for controlling the people from the “Shang Jun Shu” with the “Four-Box Model of the True Self”—
The way to understand it is this: The four dimensions of the "true self"—knowledge, experience, habits, and beliefs—are the deepest, most instinctive parts of a person. Precisely because they are instinctive and require no conscious thought, these four dimensions also happen to be the easiest entry points for external forces to exert influence. Understanding your own inner structure is, in essence, also understanding where your vulnerabilities lie.
Is it a circle or an abstraction? Just watch how it spins throughout the day.
The alarm clock rang in the morning. Ben said, “Just five more minutes—this bed is so cozy!” (desire surging). I replied, “If you don’t get up now, you’ll be late, and you’ll lose money. How embarrassing!” (emotions start taking over your calculations). The Super-Ego chimed in: “Get up! Only self-disciplined people are worthy of talking about their dreams.” (willpower steps in to counteract). The Ego said, “Stop arguing—let me think it over. There’s no meeting this morning, so actually I could afford to be twenty minutes late.” (independent thinking takes charge).
You see, you’re not “just one” person making the decision—you’re four sides of a circle passing the baton to each other. What we call “hesitation” is simply the circle getting stuck; what we call “decisiveness” is just the circle spinning smoothly and swiftly.
But who initiates that “change”? Take, for example, the moment when one finally gets out of bed—the initiator isn’t the ego’s dissatisfaction; rather, it’s rational calculation: rationality figures out that “enduring now will lead to better outcomes,” and thus mobilizes the superego to take action. Weighing pros and cons and anticipating consequences—that’s precisely what rationality has always been good at. Thus, free will does have a place within this framework: managing oneself, a self-referential structure.
What is the relationship between psychological frameworks and cognitive frameworks?
Having read this far, a natural question arises: What is the relationship between the thinking framework discussed in Volume Two and the psychological framework presented in this volume? The author’s perspective is clearly articulated:
●The thinking framework is part of the self—the conscious, rational part. In the second section, the machine centered on consciousness and broadcasting to the memory cluster occupies only the “self” segment on this circular continuum. At its core, thinking is a computational component—a mechanical entity akin to a muscle. If you treat a person like a calculator, asking him merely to input data and obtain results, then he becomes nothing more than a calculator, utterly devoid of any human essence. Thus, discussing the thinking framework alone cannot capture the essence of humanity.
●The Id is an independent structure that operates separately from conscious thought. Desires, habits, beliefs, and perceptions all run automatically without requiring any conscious deliberation—indeed, even when you’re asleep, the Id continues to function. The very act of dreaming provides evidence of this: while you sleep, your subconscious remains active. Dreams are nothing more than the Id’s monologue, freed from the constraints of the ego; it resides right there, in the compartment labeled “Id.”
·The ego and the id interact with each other to form the psychological framework. Only when consciousness—the ego—combines with the instinctive realm—the id—and is further supported by two mutually regulating forces—the superego and the non-self—does a complete psyche take shape.
So, a psychological framework is more macroscopic than a cognitive framework: while a cognitive framework focuses solely on how the computational components are manipulated, a psychological framework takes into account the whole person—the computational components themselves, along with their motivations, their opponents, and their hidden cards.
The amount of information is considerable, but the core message boils down to just one sentence: The mind isn't a pot of porridge, nor is it a stack of floors piled on top of each other. Rather, it’s a circular structure in which rationality and emotion each occupy half of the circle, managing each other through two complementary mechanisms that connect end to end and keep spinning endlessly, day and night.
In the next chapter, equip this ring with finer gears: four rows, each with its own operating track.
The 16 grids in Chapter 4 are static shelves. This chapter brings them to life.
The author recorded the operational route of the “Self” series, with four sentences forming a loop:
Only by building relationships with upstream and downstream partners can you identify the skills you need to improve; only by earning money through your career can you afford to pursue your passions; and only when you’ve developed the necessary skills can you advance your career—using everyone’s hobbies to forge connections.
Draw it:
Relationship → Ability → Career → Hobbies → Relationship (Upstream-Downstream) (Improve) (Earn Money) (Satisfy) (Build)
This is a flywheel of self-growth: Relationships reveal the skills you’re lacking; those skills propel your career; your career, in turn, fuels your passions; and your passions bring people together to form meaningful relationships. These four life themes aren’t a mere list of separate items—they’re part of a cyclical production line. If any one link breaks, the entire chain will slow down—for instance, someone without strong relationships won’t know what skills they’re missing, and someone without a fulfilling career won’t be able to afford their passions.
Now let’s look at the operation between the four lines (the ring discussed in Chapter 4):
·Non-self is a means of rational management through sensibility: when sensations, emotions, desires, and moods surge forth, rational plans get swept along with them.
●The superego is a means of rationally controlling instincts: It mobilizes attention, perseverance, willpower, and responsiveness to bring impulses back within the bounds of social norms.
The author also specifically explained a particularly confusing point: If the square labeled “not-me” is filled with emotions, how does that relate to rationality? His answer was: “Exactly—precisely because this part doesn’t follow logic can it help emotions rein in and control rationality.” After all, reason cannot control reason; what always stops an eloquent speaker isn’t necessarily a better argument—it’s emotion. This explains why, on the debate stage, appealing to emotions often triumphs over appealing to logic.
Another application of the operating line is happiness.
There’s an article on Zhihu discussing “the sources of happiness,” and the author has fitted it into a four-line psychological framework:
Four kinds of happiness, four pathways: satisfaction from bodily senses, fulfillment from external achievements, contentment from inner peace, and harmony in relationships. The author comments: For most of the time, human beings are driven by instinct—this drive is roughly divided into dopamine-driven activity (proactively seeking out things to do) and anxiety-driven reactivity (passively dealing with whatever comes our way). The four sources of happiness precisely reflect the sensations experienced when these four interactive pathways between instinct and consciousness are activated.
A practical inference: People who consistently rely on just one source of happiness over the long term are prone to burnout. Only by rotating among four different sources of joy can we maintain a sustainable inner ecosystem.
Each of the four kinds of happiness has its own dedicated “socket”—and this isn’t a case of theoretical perfectionism; it’s grounded in lived experience.
People who only value external achievements suddenly feel empty once they’re promoted to director—like the plug for achievement has been pulled out, while the other three outlets have never been connected in the first place. Those who rely solely on sensory pleasures find that their happiness comes quickly but fades just as fast, leaving them constantly driven to ramp up the intensity. As for those who seek only tranquility, after hiding away in the mountains for three years, they return to find they’ve lost touch with the world altogether.
Only by cycling through these four aspects can we maintain a sustainable inner ecology: Connect with accomplishment when working, connect with your senses when eating, connect with tranquility when alone, and connect with harmony when spending time with family and friends. If any one of these aspects is left without energy for too long, it will find other ways to “collect its debt”—insomnia, irritability, and emptiness are all reminders of unpaid bills.
The next chapter returns to MBTI: Use this machine to give MBTI a thorough checkup.
The J/P dimension in the MBTI, officially described as "Judging/Perceiving," is popularly referred to as "J types love planning, while P types love flexibility." The author has performed a surgical procedure on this dimension, and here’s how the knife was wielded:
The first step is to distinguish between thinking and psychology. If you treat a person like a calculator—asking them only to input data and obtain results—then they’re merely a calculator, with absolutely no connection to human nature. Thinking is just a computational component, a mechanical entity akin to a muscle; it doesn’t require any aspect of human nature.
Step two: Examine the attribution of “planning.” What is planning? It’s the process arrangement for achieving a goal. Even rocks have their own causes and external factors, ultimately leading to a result—and they, too, have a “process.” Some processes require planning, while others don’t. Thus, whether or not there’s a process, and whether or not planning is needed, are attributes of the matter itself, not psychological attributes. Therefore, the MBTI’s concept of “planning” should be removed, since it has nothing to do with psychology.
Step three: Shift to the genuine psychological dimension. The term “O-plan” can be replaced with “work schedule”—which is transactional. But the real psychological opposition lies between perception (liking sensations and inputs) and action (liking outcomes and outputs).
The effect of this shift: MBTI has transformed from “four vague labels” into “four psychological switches.” J is no longer simply “someone who loves making plans” (many people make plans only because they’re driven by deadlines); instead, J has become “someone who finds joy in achieving results.”
In the MBTI community, the mainstream experience of J/P is actually expressed in another set of terms: “love certainty vs. love openness.” Can these two sets of language be translated into each other? The author’s translation is remarkably concise: People who love openness are those who love thinking but not acting (perception)—they have poor executive function and thus lack good planning skills. On the other hand, people who love certainty are those who love acting but not thinking (action)—if you don’t think at all, of course you’ll feel certain; but if you overthink, you’ll lose that sense of certainty. As for those who “often change their plans”? Well, they’re naturally people with poor execution ability.
The biggest loophole in the term “Love Project = J-shaped” can be spotted right in the office.
Among my colleagues, there are two well-known “planning freaks.” One is an action-oriented type: he makes annual plans, breaks them down monthly, and spends Sunday evenings scheduling the coming week—his reason for planning is that he feels at ease only when he sees tangible results; for him, planning is the scaffolding that leads to actual output. The other, on the other hand, actually hates planning—but with project deadlines looming one after another, he’d be in trouble if he didn’t plan at all. He plans because he’s driven by deadlines; deep down, what he truly longs for is a weekend where he can just pack up and head out on a whim.
The same “Love Plan”—one is J (who finds joy in output), and the other is P (who finds joy in experience but is driven by reality to fill out forms). When measured by “whether or not they love making plans,” the two are similar; yet when measured by the psychological framework’s switch, they turn out to be complete opposites.
That’s why “plans” must be removed from the MBTI: Plans are behaviors, and behaviors can be forced by the environment; switches, on the other hand, represent motivation—motivation is what can’t be hidden. When you measure behavior, what you’re really measuring is one’s situation; when you measure motivation, what you’re measuring is the person themselves.
J/P isn't the only dimension that needs a makeover. S/N also requires a thorough overhaul.
The official explanation goes like this: S stands for “sensing,” which focuses on concrete, detailed, and immediate facts; N stands for “intuition,” which focuses on abstract, possible, and holistic perspectives. This distinction sounds clear in theory but becomes confusing in practice: Among artists who focus on details and engineers who focus on the big picture—whose is S and whose is N? Is “intuition” really a source of information, or is it simply a matter of reaction speed? The official theory has never really clarified this point.
The knife technique of psychological framing is the same as that of J/P: Replace information preferences with psychological switches. The true opposition between S/N isn't "detail versus wholeness," but rather control versus spontaneity.
S (Control): Restrain oneself and abide by external rules;
N (Casual): Go with the flow and listen to your inner voice.
With this shift, the original observation suddenly makes perfect sense. Why do S-type people seem “practical”? Because those who adhere to rules inevitably focus on reality—after all, rules are inherently embedded in reality itself. Why do N-type people seem “intuitive”? Because when one bypasses external rules, all that’s left is the inner voice to listen to. The reason intuition feels “direct” precisely lies in its ability to sidestep the control of external rules.
The author also leaves behind an additional piece of evidence—the observation of the situation faced by S-type individuals:
S is often in a subordinate position: either others are superior to him, or some duty or goal is greater than himself.
Please note that this statement has nothing to do with “details versus the big picture”; rather, it fits perfectly with the dichotomy of “control versus spontaneity.” A person who exercises self-discipline naturally tends to put others, responsibilities, and goals ahead of themselves. An observation about social demeanor happens to hit precisely on the “control” switch—this suggests that at the core of S/N lies self-restraint, not merely an information channel. (Of course, this is a probabilistic observation, not a categorical judgment: it serves as a reminder for S-type individuals to be mindful of the blurring of their own personal boundaries.)
People often say that the MBTI results vary each time you take the test, so it’s unreliable. The author’s response is concise and straightforward: Instead of designing test questions, just ask people directly and let them describe for themselves what type they belong to.
Reason: The questionnaire simplifies the questions to enhance feasibility, while simultaneously introducing greater uncertainty. Rather than letting you guess based on a bunch of vague questions, it’s better to clearly define the types and let you take ownership of them yourself. As a result, the focus shifts from “measurement accuracy” to “whether you understand the definitions of the types”—which is precisely what Chapters 1 through 3 of this book do.
One major critique of the MBTI is, “Last year I was an INFP, but this year I’m an ESFJ.” The author acknowledges this fluidity and provides a list of influencing factors:
The factors influencing MBTI include: innate traits, postnatal experiences, family, work, society, urban environment, education, and entertainment.
The persona one presents to one’s parents, the persona one adopts at work, and the persona one displays around old friends can be seen as three distinct “waves” superimposed on top of each other. This isn’t a sign of failure in testing—it’s precisely the true structure of human beings: the overlapping of innate personality and acquired personality. The sixth section of this book (Framework Synthesis) will devote an entire chapter—titled “Multiple Personalities”—to exploring this issue in greater depth.
The MBTI itself is actually incomplete—beyond the four letters, more letters could be added. In addition to the psychological framework (the four dimensions), we could also incorporate dimensions from the cognitive framework (Part Two) and the philosophical framework (Part Four). The author’s algorithm is as follows: Considering these three frameworks, people can be categorized into 16 × 16 × 16 = 4,096 types, rather than just the original 16 types.
The sixteen-room structure is just the main entrance. Inside the door lie more than four thousand additional rooms.
Jung's Eight Dimensions: Returning to One's Roots and Ancestry
Jung is the ancestor of the MBTI. In "Psychological Types," it is proposed that the essence of personality lies in a system of cognitive functions. All these functions can be categorized along two dimensions: information gathering (Sensation S/Intuition N) and decision-making (Thinking T/Feeling F). When combined with the orientation of introversion (I) or extraversion (E), this yields eight distinct functions: Se, Si, Ne, Ni, Te, Ti, Fe, and Fi.
The psychological framework and its counterpart are direct:
(It is said that Jung once stated that F and T both belong to the “rational” function—that is, the judging function. This differs from the psychological framework’s interpretation, which places T/F along the “consciousness/unconsciousness” axis. This constitutes a point of divergence worth further investigation.)
Another theory called “Jungers” attempts to add a thinking dimension beyond the four psychological dimensions of MBTI, dividing people into 64 types—where the subject corresponds to discreteness, the background corresponds to continuity, the whole corresponds to hierarchy, and analysis corresponds to flatness. This approach aligns perfectly with the book’s expanded framework: “Psychological Framework + Thinking Framework = 16 × 16.”
The Big Five Personality Traits: A Fair Comparison
The Big Five personality model is the most widely recognized personality framework in academia. However, the author’s attitude toward it is both positive and critical:
What’s better about the Big Five than MBTI: It’s simpler and more straightforward. MBTI directly maps onto 16 personality types, which is overly complex and lacks rigor; the Big Five, on the other hand, focuses directly on four or five independent factors, making it much clearer and more precise.
The drawbacks of the fifth year (in the author's view, there are three):
The uniqueness of the combined type has not been recognized. The Big Five describes people in terms of five scores, while MBTI portrays people as a combination—yet this combination carries holistic significance: an INTP is not merely the average of I + N + T + P; rather, it’s a “species” that emerges from the interplay among these four dimensions.
The tone is overly biased and judgmental, failing to recognize that personality itself is neither good nor bad (the term “neurotic” itself carries a negative connotation).
“Neuroticism” is not an independent personality dimension—the author argues that neuroticism is simply a combination of “rationality and self-control”: a person who is both rational and self-restrained is particularly sensitive to loss of control, and thus appears neurotic. It’s not a new dimension; rather, it’s a combination of two older dimensions.
Readers may disagree with this ruling. But please pay attention to the method used in the ruling: checking whether the candidate dimension is already a combination of existing dimensions—this is the standard procedure for “deduplication” of conceptual vectors.
Enneagram: Put it back into the four boxes.
The Enneagram uses nine words—perfect, helpful, achievement-oriented, sensitive, intellectual, loyal, lively, leader, and peaceful—to describe nine personality types. The author maps these onto the four quadrants of a psychological framework:
Eight boxes hold eight types, leaving “Peace” alone in the ninth box. The corresponding interpretation is this: The nine types of the Enneagram are not nine parallel dimensions but rather composite outcomes arising from combinations across several dimensions—eight of them are combinations, while one (Peace) may correspond to a state where all dimensions are balanced or centered. The fact that the nine-box grid cannot be neatly arranged suggests that the very axes used for categorizing the Enneagram deserve scrutiny.
For over a century, personality theories have been all over the place—what’s it like? It’s like having four map apps installed on your phone.
Open the Same City: One app calls that road “Zhongshan Road,” another has layers dotted with snack shops, and yet another insists on using its own color schemes and abbreviations. The interfaces of these four apps differ, their names differ, and their layers differ—but the city itself is the very same city. Before you criticize one of them for “not being drawn correctly,” you’d do well to ask yourself: Is it simply using a different way of labeling?
Jung, MBTI, the Big Five, and the Enneagram—these are the four apps of the city of personality. In the “Psychological Framework” app, the Big Five’s “Neuroticism” is labeled as “Rationality & Control”; in the Enneagram, the “Perfectionist” type is labeled as “Control + Perception.” As for the labeling method, feel free to argue—it’s not our job to join in the debate.
Jung’s eight cognitive functions, the Big Five personality traits, the Enneagram, and MBTI—over the past century, these various personality theories have each offered their own distinct perspectives. Yet a psychological framework has done one crucial thing: it has eliminated redundancy. As a result of this reduction, four key dichotomies keep recurring: Control/Spontaneity, Rationality/Emotionality, Introversion/Extroversion, and Sensing/Intuition. To a large extent, the differences among these theories stem from variations in how they combine and name these dichotomies.
Sixteen personality types—why do I focus specifically on INTPs? Because the author herself is an INTP (as she explicitly acknowledges in many places in her notes), and this type is explored in the greatest detail. Also, INTPs are among the most discussed types online, with the richest available resources. This chapter serves as a demonstration of “dissecting a personality type through a psychological framework”—and the other fifteen types can be analyzed using the same approach.
The four letters of INTP, when broken down, stand for: I—Introverted, N—Love of freedom (spontaneous), T—Rational, P—Perceptive.
Based on these four switches and following the logic of soft causality, the following cognitive characteristics are highly likely: strong awareness but weak memory; strong creativity but weak application; strong divergent thinking but weak convergent thinking; strong hierarchical thinking but weak flat thinking. The combination of divergence and hierarchy constitutes abstract thinking—INTPs are the very embodiment of abstract thinking.
Pushing further down, the following underlying worldview is highly probable: complex, lacking a unified will, immutable, and global. The author refers to this complexity—lacking a unified will—as “jungle thinking”—the world isn’t a play directed by a single director, but rather a jungle where everything grows independently.
The author’s notes contain three sentences that, at first glance, seem like jokes, but upon closer inspection reveal sociological observations:
·“INTPs are spoiled by their families”——selfish, willful, indifferent, and lazy. But on the flip side of these labels lies this: Their upbringing allowed them not to exercise control, not to take action, yet still enabled them to live quite comfortably.
·“Westerners are INTJPs;”
●“Men are INTPs.”
From this, he even came up with a half-joking “MBTI equation”:
INTPs are spoiled + INTPs are consumers + Westerners are INTPs + men are INTPs.
=Easterners are hardworking female workers``
Of course, this equation shouldn’t be taken seriously—it’s an exaggerated way of superimposing four probabilistic observations to help you see clearly where the arrow is pointing: The INTP is a personality type “nurtured by its environment.”
This judgment has a more serious version, for which the author specifically created a separate clause:
·I (Introverted): In an era where socializing is unnecessary, where material abundance is sufficient and we no longer need to trouble each other, solitude becomes the only viable option.
N (casual): Take charge of your own life—don't be a mere tool. Only in an era where everyone doesn't have to blindly obey the collective can freedom truly be an option.
·T (Rationality): Guided by reason, not enslaved by emotions—only when one is secure and affluent enough to no longer have to watch others’ expressions can rationality prevail over sentimentality.
·P (Perception): More enjoyment, less work—only in an era of overproduction can “loving to feel” avoid starvation.
Conclusion: As the economy develops, the INTP personality type is becoming increasingly prevalent. Conversely, the ESFJ personality type—characterized by warmth, responsibility, consideration for others’ feelings, and proactive action—appears to be a “hardship-prone” personality among Chinese (or, more broadly, East Asians). In an era of resource scarcity, this personality profile represents the optimal behavioral strategy.
Thus, a personality type ceases to be merely a psychological fact and becomes an economic one as well: the distribution of personalities is a function of the environment. This explains why intergenerational communication is so difficult—parents’ personalities represent the optimal solution for an era of scarcity, while children’s personalities represent the optimal solution for an era of abundance; these two optimal solutions simply fail to see eye to eye with each other.
In principle, this proposition is testable—for instance, one could examine the correlation between the distribution of types across different countries and their levels of economic development. As for verification, the author simply passed the buck: “This kind of investigation isn’t my forte. I’m only good at daydreaming.”
The author says: “The difference between INTPs and ESFJs is like the difference between the West and the East.”
ESFJ: Extroverted, Sensing (S), Feeling (F), Judging (J)—passionate, rule-abiding, emotionally sensitive, and always eager to take charge. If the INTP is a product of “one can survive alone,” then the ESFJ is a product of “no one can do without anyone else.” Each personality type is a masterpiece of its own era.
The author also noted a rather sharp observation from an online reply (which he calls “the INTP’s mouthpiece”): “INTPs are spoiled at home—selfish, willful, indifferent, and lazy.” — This statement itself is an accusation from an ESFJ perspective. With these two entirely different systems of evaluation, neither side can convince the other.
The entire process of analyzing an INTP can be replicated: break down the letters → infer cognitive traits → deduce their worldview → then place it back within the economic context. A personality type isn't a label—it's a causal tree (a soft causality) that can be traced from a seed all the way to a towering tree.
Four Weapons to Overcome Procrastination
Procrastination is the epidemic of our time. The psychological framework doesn't offer just a single piece of advice—“Be self-disciplined”—but rather four powerful tools—because these correspond to the four squares in the “Superego” row:
On the line for the Superego (reason and control) are four weapons for overcoming procrastination. In reality, they’re about conquering the subconscious mind.
Controlling attention—The Pomodoro Technique is all about controlling attention;
Perseverance—persistently sticking to one thing; the driving force behind this is passion. “Really wanting to do it” leads to action.
Willpower—That fierce determination used to overcome pain;
Reactivity—Once a habit is formed, there’s no more suffering; only inertia remains.
Pay attention: the four weapons correspond to four different kinds of failure—loss of focus, waning enthusiasm, fear of hardship, and failure to develop a habit. Procrastination isn’t a single illness; rather, it’s four separate illnesses sharing the same name. Diagnose which “grid” you’re stuck in, then choose the corresponding weapon—this is how psychological framing works to treat these issues.
All four weapons come from the “super-ego” side. What about the “non-self” side—feelings, emotions, desires, and moods—the very sources of procrastination? Is there a corresponding approach to address these? The author’s answer in just four characters: “Weaken the signals.” His ultimate demonstration: stare at a blank white wall for half an hour, and afterward, you’ll feel brimming with motivation—because anything is more interesting than staring at a blank white wall. Staring at a blank white wall means reducing all signals to zero.
Four procrastination weapons—every day in the gym, everyone is practicing them in full force.
Attention: For those who lock their phones in the locker room, one set of exercises equals one set of actions; for those who leave their phones on the treadmill, one set of exercises brings three messages. The first weapon? Start with physical separation. Endurance: Those who can show up year after year don’t rely on gritting their teeth—they’re driven by an “extraordinary desire”—the desire to fit back into those pants, the desire to come out on top at the class reunion. Passion is the engine that fuels endurance. Willpower: During that final set of squats, your legs are shaking, yet inside you hear a voice saying, “Just two more!” That fierce determination to overcome pain—it’s something the gym provides on demand. Reactivity: The most advanced type are those who head to the gym precisely when it’s time—no hesitation, no self-sabotage; as soon as they change their shoes, they’re out the door. Once the habit is formed, the gym becomes not about pain, but about momentum.
The people applying for the card change every year, and their postures in the four-grid space vary widely: some are distracted (lack of focus), some are apathetic (lack of perseverance), some are fearful (lack of willpower), and some are lazy (slow reaction time).
Before you next time scold yourself for “procrastination,” first ask yourself which box you’re stuck in—four ailments, four remedies. The gym and the desk share the same prescription.
Matching: What kind of people are suited to be together?
The author’s notes include two conjectures about matching, both of which honestly state, “Not necessarily correct”:
●Hypothesis 1: The best match is one where there are differences in various aspects, but the differences aren't too significant. This idea draws inspiration from the notion that “highly sensitive people and moderately sensitive people can get along”—if everyone were exactly alike, it’d be boring; if everyone were completely different, it’d lead to conflict. It’s precisely the “slight difference” that creates just the right amount of tension.
Hypothesis 2: It could also be that certain aspects are complementary while others are consistent—just like the composition of a team’s members.
Another, even bolder statement: “In today’s materially abundant environment, it’s better to find someone who’s similar to you; in harsher environments, you need to seek someone who’s complementary.” (He joked, “That’s why there are so many gay people.”) He then cited a piece of research showing that couples share 80% of their characteristics in common.
The methodological value of these conjectures exceeds their conclusive value: translating the metaphysical question of “compatibility” into a structural question—“Which dimensions should be the same, and which should differ”—is precisely what conceptual vectors are designed to do.
Education: First diagnose, then prescribe.
The psychological framework offers a very direct critique of education:
First, you need to understand the scope of application; the very first step is to screen the individuals. Just like seeing a doctor, making an accurate diagnosis is more important than inventing a miracle cure.
Currently, this standardized, one-size-fits-all approach to education is highly ineffective. There is an urgent need for early diagnosis followed by the provision of multiple educational pathways.
Think about it: From elementary school through middle school and college, we’re subjected to a huge amount of ineffective education. It’s only when many people enter the workforce that they finally encounter effective education—first taking on actual work tasks and then studying specifically around those tasks, without wasting time on anything unnecessary.
The author’s proposal: Evaluating an educational approach could follow the model used for drug approval—first conducting a “diagnosis” (identifying what type of learner this is), then prescribing a “treatment” (determining which method works best for which type of learner), and clearly outlining both the “indications” and “side effects.” Meanwhile, there are many essential pieces of knowledge that school education tends to overlook, and society must step in to fill those gaps.
This approach is now known as “teaching according to individual aptitude,” a concept that has been discussed for two thousand years. The psychological framework fills in the missing piece: What exactly are the dimensions of “individual aptitude” that underlie this approach—extraversion vs. introversion, sensing vs. intuition, rational vs. emotional, and conscientiousness vs. spontaneity—plus the four modes of cognitive framing. Once we have established the coordinates for diagnosis, we can then prescribe the right remedy.
Procrastination needs to be addressed in four distinct stages; compatibility should be assessed across multiple dimensions; and education must begin with diagnosis before prescribing a remedy. The purpose of psychological frameworks isn't to label people, but rather to recognize the unique differences among individuals—this statement now comes equipped with actionable coordinates.
The entire content of the psychological framework can be summarized in two tables.
Four switches:
The Inner Circle:
Psychological activity is the interaction between instinct and consciousness: The id, influenced by the non-self, acts upon the ego; the ego, in turn, is governed by the superego, which manages the id. Rationality and emotion connect end-to-end, forming a circular loop. The ego’s four quadrants also feature another dynamic flywheel—[Relationship → Ability → Career → Hobby].
The three boundaries of this section
Probabilistic: All inferences—from type to feature—are soft causations; whether you believe them or not is up to your own experience to decide.
In the author’s notes, there’s a hand-drawn diagram: the horizontal axis represents introversion/extroversion, and the vertical axis represents control/obedience. This is an unexplored sub-framework—social personality: control versus obedience, introversion versus extroversion—four possible combinations. The diagram contains no further words; it’s like an ungerminated seed, left here to wait for the future.
Now that we’ve established the structure, let’s take the next step deeper: How does this brain actually work? Why is it that even equally intelligent people sometimes fail to understand each other? Please turn to Part Two, “The Thinking Framework: The Four Ways of Thinking.”
Thinking Framework: Four Ways of Thinking
Why do some smart people fail to understand each other?
The previous section categorized people; this section delves even deeper—into thinking itself. What types of human thinking are there? What do they each look like? And why can knowing someone’s thinking type help cure more than half of communication problems?
The information stored in a person’s brain and its organizational structure are different:
·Some people have a spreadsheet in their minds—everything has an attribute column.
Some people have a mental map—a map where everything is neatly categorized, with each thing having a parent and offspring.
Some people have a linear view of the world—everything has a sequence, with causes preceding effects.
Tables, diagrams, linearity, and outlines—these are the four modes of thinking. They arise from the intersection of two dimensions: “discrete vs. continuous” and “hierarchical vs. flat.” When paired with the dimension of “process”—memory/consciousness versus generation/execution—they form a complete typology of thinking. This typology answers the half of the question that MBTI leaves unanswered: It not only reveals what kind of person you are, but also how your mind actually works.
Chapters 1–2 discuss the underlying structure of thinking; Chapters 3–4 focus on two dimensions; Chapters 5–8 each dedicate a chapter to a specific way of thinking; Chapter 9 presents a famous pair of contrasts (people who focus on differences vs. people who focus on similarities); Chapter 10 explores applications (in disciplines, careers, and knowledge management); Chapter 11 examines thinking frameworks and artificial intelligence; and Chapter 12 offers self-assessment and a summary.
Each chapter is short, so you can casually open any chapter at random.
A familiar moment
Have you ever had this experience:
You recognize every single word he says and hear every single sentence he utters—but still, you have no idea what he’s trying to convey. You ask him to “get to the point,” but he insists that every single word is crucial. You tell him to “see the bigger picture,” yet it seems to you like he’s just floating around in the clouds, completely detached from reality.
Both people are smart, both are sincere, and neither harbors any hostility—but still, they just can’t seem to hit it off.
We usually attribute such things to “incompatible personalities” or “misaligned knowledge.” But the author offers another explanation—one that’s much simpler:
Poor communication isn't simply a matter of aligning knowledge points; a major reason is the difference in thinking styles.
A Neglected Category
When it comes to “human categorization,” most people immediately think of personality: introvert or extrovert, impulsive or patient. The MBTI has perfected this art.
However, personality only answers half the question. It tells us “what this person wants and what they care about,” but it doesn’t tell us “how this person’s brain processes information.”
Two people can both have equally gentle personalities, yet their ways of thinking could hardly be more different: One, upon hearing a new idea, first asks, “To which category does it belong?” The other, on the other hand, immediately wonders, “What consequences will it lead to?” The former lives in a world of categorization; the latter lives in a world of cause and effect. When they discuss a problem, it’s like asking an accountant and a novelist to co-author an article—not that one is right and the other wrong, but rather that their brain’s internal file formats are simply incompatible.
The author’s exact words were: “Classifying human thinking types is extremely important, yet currently there are relatively few people paying attention to this issue. Personality types alone are actually insufficient—thinking types are equally crucial.”
This book is precisely here to fill in that missing half.
However, before we proceed with classification, we first need to answer a more fundamental question: Can thought itself be clearly defined?
Modern science finds itself in an awkward position here. It can measure brain waves and perform fMRI scans, yet it remains unable to explain clearly what exactly “a thought” is all about. The author’s judgment is straightforward: precisely because external tools cannot crack this problem, we must build our own instruments—using the approach outlined in Part Three: observe the surface phenomena, infer the underlying mechanisms, and construct a model.
The entire cognitive framework is essentially a “guesswork-based” instruction manual for the brain. It doesn’t involve opening up the skull; instead, it infers the internal structure solely from the external manifestations of thought—how a person reads, speaks, organizes information, and makes decisions.
Are you right or wrong? The test isn't an autopsy report—it's whether it can explain the people around you, and whether it can predict what they’ll excel at and what will cause them pain.
Let’s reveal the answer first, and then we’ll go over the reasoning step by step later.
You can try a little experiment right now: Think back to your most recent note-taking session. Was it organized in columns and grids? Did you draw a tree diagram? Did you keep a chronological log? Or did you first outline the main points and then fill in the details?
The answer will tell you something. See you in Chapter 2.
When you’re thinking about a problem, what’s going on in your mind?
Take it apart and you’ll see that two activities are happening simultaneously.
There’s something called memory—storage. It manages learning, accumulation, and collection. It doesn’t speak; it just stores things.
There’s a thing called consciousness—processing. It governs thinking, analysis, and creativity. It doesn’t store anything; it just keeps running.
This distinction is as trivial as stating the obvious. Yet it immediately provides a useful diagnostic tool:
●Some people have strong memories but weak awareness— they read widely and seem to know everything, yet rarely offer their own insights.
Some people have strong awareness but weak memory— they don’t read much, yet they always manage to come up with their own interpretations and explanations for everything.
The author summarizes these two tendencies in eight characters: “Those who remember love to learn and follow; those who are conscious love to think and have their own views.”
You must have both types of people around you. And you yourself probably lean more toward one of them.
But memory is not “a” machine.
Here, we’d like to correct a common misunderstanding.
As soon as someone says “memory is responsible for storage,” many people immediately picture a warehouse-like machine, a hippocampus, or a hard drive—a single, isolated entity. This mental image is incorrect.
The author’s view is that memory isn’t a single entity—it’s a collection. The brain doesn’t have one unified warehouse; rather, memories are stored all over the place—memory is a multitude of distributed storage units throughout the brain, a collection rather than a single machine. Here, we deliberately avoid making detailed distinctions: we don’t talk about visual memory or motor memory; we simply refer to “memory.” In reality, the brain is likely even more complex than this—memory may not have fixed regions or central hubs. The diagram at the beginning of this chapter labels only “memory” without further specification.
What about consciousness? Consciousness is precisely the central hub that mobilizes this collection of memories.
The author uses a vivid metaphor: Consciousness is the central stage bathed in the spotlight. Whatever is illuminated on that stage is what the entire audience sees. Consciousness “broadcasts” information out into a network of memories scattered throughout the brain—each memory receives the broadcast and then sends back its own echo. From this perspective, consciousness is also a “giant neuron that connects all neurons,” serving as the rational center relative to the instinctive center.
So the accurate statement is: The cognitive framework as a whole is a machine—a machine centered on consciousness, broadcasting outward to a cluster of memories. We can’t say that consciousness is one machine and memory is another; memory is a group, while consciousness is the central hub—only when they’re combined do they form that single machine.
The author later provided additional explanations regarding two details of this machine. First, broadcasting is received by everyone—rather than being spotlighted on just one particular point, every memory receives it simultaneously, each echoing back in its own way. Second, attention itself is situated within this framework: the information that appears in consciousness is precisely what—
Noticed information; attention is the control over “what information appears in consciousness.”
So, “Who’s pushing the information onto the stage?” The echo of the memory group sends up the strongest signal—but consciousness isn’t merely a passive receiver: consciousness can control itself—a kind of self-reference. (The relationship between this self-referential structure and AI will be explored further in a prophetic manner in Chapter 9.)
Whether it's memory or consciousness, each has two states:
Running: Performing daily operations—reviewing old memories, retrieving old ideas;
To produce: to create something new—forming new memories, giving rise to fresh ideas.
Memory and consciousness, generation and operation—each pair intersects to produce four types of cognitive activities:
Let’s take a complete example and see how these four activities come together:
You read an economics book—new information comes in and gets stored in your memory (digesting the knowledge); a few days later, you suddenly recall a model from the book and start using it to think about housing prices around you (reflecting); after pondering it a bit, you come up with your own conclusions and store them in your mind as new memories (again, digesting the knowledge); the next time a friend asks you whether they should buy a house, you blurt out your analysis (putting the knowledge into practice).
Memory and consciousness thus feed off each other: when consciousness acts upon memory, it gives rise to memory; when memory acts upon consciousness, it gives rise to consciousness.
Broadcast: Consciousness sends information to all memory locations → · Echo: All memory locations send back their responses
How New Memories Are Written In: A Calculation Involving ABC and ABD
“Consciousness, when it acts upon memory, gives rise to memory”—is there a more detailed description of this process? In an early manuscript from 2014, the author left behind a rough yet vivid calculation (the full text is reproduced in Appendix Volume V, “Understanding and Cognition of Thinking”):
The brain first receives a piece of information from the outside world, which it records as ABC and stores away. A few days later, it receives another piece of information, recorded as ABD. At this point, the brain compares ABD against ABC: the common part is AB, while the differing parts are C and D. Once the comparison is complete, the new information is written into memory in the form of “the similarities and differences between the new and old information”—this is the most fundamental action of “digesting knowledge”: identifying what’s the same and filtering out what’s different.
A few days later, a string ABCDE arrived. The brain split it into two segments: the cooked segment ABC and the raw segment DE—first dividing the information into segments, then assigning them to their respective categories: ABC went to the alphabet branch, while 123 went to the number branch. Each segment was placed in its proper category, and then reorganized according to their original relationships.
The calculations are done using letters, yet what’s being talked about are things that happen every single day: When you remember a stranger, it’s because of “who he resembles—and who he doesn’t”; when you learn a new dish, it’s because of “how it differs from the dishes you already know.” New memories never just get dumped into the brain all at once—rather, they’re built by comparison, distinction, and categorization.
Being good at studying doesn't mean you know how to think.
It may seem like just a matter of noun classification, but it can immediately clarify a common misunderstanding.
Conversely, “This person is brimming with ideas—how come their exam results aren’t good?”—his consciousness is strong, but his memory is weak. Exams test the functioning of memory, not the generation of consciousness itself.
Traditional education almost exclusively measures the memory side. Yet whether a person is better suited for research or execution, for creativity or management, lies more deeply on the consciousness side.
Two types of people you can see right from your office.
Think about the two types of colleagues in your office.
One type is the “living dictionary”—they can recite the company’s historical data backward and forward, knowing everything there is to know about it. But as soon as you ask, “What should we do about this?” they just stare at you, waiting for you to make the decision—strong memory, weak initiative.
The other type is just the opposite: ideas keep popping up one after another, and every meeting brings new insights. But ask him to report last quarter’s figures, and he’ll scratch his head and say, “Roughly speaking.” — Strong in awareness, weak in memory.
By now, we’ve covered the two underlying dimensions of this book (memory/consciousness, operation/creation). These two dimensions intersect to form a total of four categories—there’s not much more to it than that.
The real highlight is in the next chapter: In this machine, information is stored in entirely different forms. Those four distinct shapes are what truly mark the dividing line between one person and another.
The first dimension: Discrete and Continuous
Close your eyes and think about, “How do you remember a city?”
Some people remember a collection of places: my home, the office, that noodle restaurant, a friend’s apartment... Each place is a distinct point, and these points don’t have to be connected to one another. This is discreteness—the view of the world as a pile of loose sand, from which we extract both its commonalities and its individual characteristics.
Another type of person remembers just a few routes: starting from home, walking along the river, crossing the bridge, turning right, and arriving at the office. Each location lies precisely along a single line. This is continuity—succession in time, contiguity in space, and causal relationships among them—in short, a seamless thread connecting everything together.
The author provides a brief yet significant correspondence: discrete corresponds to abstract, while continuous corresponds to concrete.
Why? Because continuous information is the information itself—just record how things actually happen, without any processing. On the other hand, discrete information requires processing: you have to cut the connected world apart, pick out individual, independent points, and then compare these points to identify their similarities and differences—this is precisely the “abstraction” process discussed in Part Three.
The discrete person: divergent, accumulates knowledge, has a science-oriented temperament. The continuous person: holistic, leads to action, has a humanities-oriented temperament.
Pay attention to the point about “continuous causation of action”—we’ll come back to this later: People who think in a continuous manner have an advantage when it comes to taking action—because they think continuously, they don’t need to overthink; they simply follow the flow, allowing them to act swiftly.
The second dimension: hierarchy and flatness
Let’s think about another question: “Does the knowledge in your mind have a hierarchical structure?”
People at the hierarchical level: Yes. Knowledge is like a tree—with a trunk and branches and leaves. When discussing a topic, first lay out the main points and then expand on the details, working your way down from the roots. The author’s criterion is very practical: commonalities give rise to hierarchy—those who see commonalities among things can’t help but extract these commonalities and treat them as parent categories, thus giving birth to hierarchies.
The flat person: No, and there’s no need for it. Knowledge is like a flat plain—every single piece of knowledge is roughly equally important; the details themselves are treasures. As for personality, it’s flat—someone who sees personality in everything regards each detail as worthy of being kept separate, reluctant to surrender it all to a general summary.
The author also offers a highly practical observation: “People who think in terms of hierarchy are at least not directionally challenged. After all, hierarchy implies a global map—seeing the big picture from broad to specific, from far away to close up. A person who’s directionally challenged doesn’t lack memory; rather, the mental map in their brain lacks layering—every intersection is crammed onto the same level, so naturally they can’t find their way.”
There’s also a compromise reminder: Continuity has hierarchies too—for example, large regions and smaller subregions on a map, or major and minor steps in an action. These two dimensions are independent and can be freely combined.
Is that “continuous hierarchy” (such as an outline for writing a book) considered abstract? The author later made a distinction: hierarchy isn’t necessarily abstract. Abstractness requires discreteness—necessitating the comparison of multiple entities, whereas in a continuous structure, there aren’t “multiple entities” to compare. A continuous hierarchy is merely an organization of continuous information: small.
The narrative and its chapters constitute continuous information; the novel’s outline is not abstract. By contrast, the content of an academic book itself is discrete and abstract, and only its outline can be considered an abstraction.
Physicist Dyson divided scientists into two types: birds and frogs.
From a high vantage point, birds focus on the broad landscape and don’t need many details; a frog, lying flat on the ground, fixes its gaze only on what’s right in front of it and loves to delve into minute details.
In the words of this author: Birds represent hierarchy, while frogs represent flatness. Science needs both birds and frogs—but the trouble is that birds and frogs look down on each other. Once we understand that these are two distinct dimensions rather than a matter of superiority or inferiority, the bird-vs.-frog rivalry can finally come to an end.
These two dimensions may sound abstract, but just open your phone’s photo album— they’re right there.
Let’s first look at the first dimension: How do you organize your photos? Some people rely on time: “That one from Qingdao last November”—they scroll backward along the timeline, and the photos are arranged sequentially along it; this is continuity. Others rely on themes: “In the ‘Food’ album”—the photos are broken down into individual, discrete items and grouped together based on shared characteristics; this is discreteness.
Now let’s look at the second dimension: Does your photo album have a hierarchical structure? For those with a hierarchy, their albums are nested within albums: Under “Travel,” there’s a subcategory for “Domestic” and another for “International”; and under “International,” there’s yet another subcategory—organized by year—a tree-like structure. For those who think in a flat, linear way, all five thousand photos lie neatly stacked together on a single camera roll; when you want to find something, you just search directly for “cat”—a simple, flat expanse where retrieval relies on keywords rather than structure.
Two types of people find it uncomfortable just to look at each other: those with hierarchical thinking feel restless, as if staring at a pot of porridge-like film rolls; those with flat thinking feel drained, as if gazing at layers upon layers of folders. In fact, there’s no right or wrong when it comes to albums—dimensions merely reflect positions. The four ways of organizing introduced starting from Chapter 4 are precisely the four approaches that emerge from the intersection of these two dimensions.
Now, let's cross the two dimensions:
Tables, charts, linear structures, and outlines—these four forms of information storage represent four distinct ways of thinking.
The next four chapters each focus on one type. When reading, I recommend bringing along two people: yourself and someone you just can’t seem to connect with no matter what.
Coordinates: Discrete + Flat
Imagine an HR professional’s computer screen: a staff table, with one row per employee, and each column dedicated to a specific piece of information—name, age, department, and performance.
For table thinkers, the entire world is just one big table. All things and phenomena are represented as “rows,” and each row carries several “columns” (attributes). The way they come to understand a new entity is by asking: What attributes does it have? And what value does each attribute take?
The author’s summary is just one sentence: “There’s a spreadsheet in my head.”
What does this kind of thinking look like?
Daily characteristics of table thinkers:
The key difference lies in their instinct. When they see two people, they unconsciously align them—this one tall, that one short; this one hurried, that one steady. The ability of painters, musicians, and wine tasters to “sense the slightest nuances in every brushstroke and every note”—that’s precisely the top-end configuration of this level.
Their approach to accumulating knowledge is horizontal: they don’t seek to integrate it into a system, but rather aim for an abundant stock of information.
The author also points out the corresponding forms of knowledge management: tables, attributes, and tags. People who tend to tag both people and things are often characterized by table-based thinking.
Skill and Cost
Specializing in: discerning details. The author’s corresponding aptitude is “Art”—those who can distinguish details have an advantage in any field that requires fine perception: art appreciation, quality inspection, debugging, and user experience design.
Cost: You see the rows but not the trees. A hundred rows of data lie there, yet the table itself can’t tell you which row is important or how the rows relate to one another. Thus, those who think in terms of tables need to be on guard: inventory keeps growing, yet the underlying structure remains stubbornly undeveloped.
In Holland’s theory of vocational interests, “Art” is assigned to this category by the author for the same reason: the core competency of art lies in sensing differences.
If you want to see the grassroots headquarters of table-based matchmaking, take a stroll through the dating corner in the park.
One umbrella, one A4 sheet of paper—no prose on the page, just a column heading: “Male, born in ‘89, 178 cm tall, holds a master’s degree from a ‘985’ university, works in the public sector, owns a home with no mortgage.” A single person is reduced to a single line of attribute values; hundreds of umbrellas lined up side by side form a giant outdoor chart. Parents who come to browse scan their eyes horizontally—first they zero in on the age column, then on the property ownership column, and only after that do they bother to take a closer look at the remaining rows.
Is this table cold and ruthless? It certainly is. Yet it lays bare both the instinct behind tabular thinking and its price: The instinct is that aligning hundreds of people quickly—and efficiently—relies on consistent columns; the price, however, is that we see only rows, not individuals—the living, breathing person behind each row’s attribute values simply doesn’t fit within the confines of a table.
That’s why matchmaking corners always manage to bring people together yet are consistently disliked by young people. They’re a handy sieve—but not a heart capable of truly feeling. This very same sentiment could be applied, word for word, to table-based thinking.
If you create comparison charts when shopping, if you prefer Excel to Word, and if you remember people’s names by focusing on their characteristics rather than on stories—
Welcome to the world of tables.
Coordinates: Discrete + Hierarchical
If a table thinker’s mind is like a flat, two-dimensional table, then a graph thinker’s mind is like an upside-down tree.
Is the world just a pile of scattered grains? Fine—let me start sorting them into categories. Group together those that share common features, and separate those with distinct differences on either side. Above these groups, there are even larger groups, and we keep layering them upward until we finally arrive at a single root. The author’s summary: “They can abstractly categorize.”
Biological classification (kingdom, phylum, class, order, family, genus, species) is a monument to diagrammatic thinking. Library classification systems, corporate organizational charts, and computer folders—all are diagrams.
What does this kind of thinking look like?
They strive for unity, prioritize the bigger picture, and love to organize. When they see chaos, they can’t help but feel the urge to put things in order.
The author describes how these individuals move along the tree of knowledge: they pursue toward the trunk—constantly asking, “What does this belong to?” and “What principle lies behind this?” They relish the thrill of seeing a single principle applied in multiple contexts, thereby building a comprehensive mental framework.
“Imagine” this machine—the earliest drawing of it.
The conscious activity of graph-based thinking is “understanding and imagining.” Imagine exactly how this machine operates. In an earlier draft from 2014, the author included a more detailed drawing (the full text is reproduced in Volume Five of the supplementary materials):
When we hear someone say, “Character A hit Character B,” we can vividly picture the scene in our minds, and the action of “hitting” itself is also conjured up in our minds. This is imagination.
The logic behind imagination is this: Imagination doesn't arise out of thin air—it’s feedback stored in a discrete manner. First, your mind accumulates enough well-organized, categorized old information—such as what A looks like, what B looks like, and what the action of “hitting” entails. When a new sentence comes along, the central processing unit retrieves the relevant old pieces, compares them, and puts them together, thereby creating a scene that has never been seen before. Thus, the essay asserts: Discrete storage is the foundation of imagination—the richer the inventory and the more neatly organized the categories, the more abundant the material available for imagination.
This also explains why people who think in terms of graphs (discrete + hierarchical) excel at “conducting experiments in their minds”: their mental inventory is organized like a tree, making retrieval the fastest and integration the smoothest.
Skill and Cost
Specializes in: abstraction and simplification. The author’s corresponding discipline is “Science”—believing that although the world is complex, it is governed by simple underlying principles; first identify these fundamental laws, then use them to analyze and make predictions. Particularly well-suited for physics: adept at conducting experiments in one’s mind.
Cost: In order to fit into the system, one can’t help but try to force oneself into its mold. — A new phenomenon that doesn’t quite fit into any existing framework; the first reaction of a pattern thinker is to doubt the phenomenon itself, rather than to question the tree. The more perfect the tree, the stronger the resistance to anything that doesn’t conform.
The corresponding form of knowledge management: attribute tag inheritance—subclasses automatically inherit attributes from their parent classes. Once you master a parent class, you instantly gain an entire set of child classes—this is the most rewarding experience for those who think in terms of graphs and networks.
Why do Chinese people clearly distinguish between “uncle” and “maternal uncle”?
Graphical thinking isn't the exclusive domain of bookworms—China's kinship terminology itself is a nationwide graph-based system.
In English, the term “uncle” blurs together uncles, paternal uncles, maternal uncles, and even in-laws like father-in-law and uncle-in-law. But Chinese doesn’t do that—first, it distinguishes between paternal and maternal relatives, carving out a broad category (paternal/maternal cousins); then, based on whether the relative is older or younger than one’s father, it further divides into subcategories (paternal/maternal uncles); and finally, depending on whether the relative is from the spouse’s family or one’s own family, it branches out again into yet another layer (father-in-law/uncle-in-law). Each relative has an exact coordinate on this tree of kinship terms; get even one position wrong, and you’ll be considered impolite.
The practicality of this tree became evident during the Lunar New Year: As the child was led around to pay New Year’s calls, he mumbled under his breath, “Uncle Da, Uncle San, Uncle Er, Auntie Da”—in reality, he was simply traversing a tree. And the tree rewarded him handsomely: Once he learned that “uncle” refers to one’s father’s older brother, the entire generation’s honorifics were automatically passed down, eliminating the need for him to memorize each one individually.
A single tree has managed to keep track of the relatives of over a billion people. The exhilarating feeling that comes from being a graph-based thinker—“mastering the parent class and effortlessly inheriting from it”—is something Chinese people have experienced since childhood.
Table-based thinking and graph-based thinking both belong to the category of “discrete” and both deal with a collection of points. The key difference lies in this: one is content simply to collect the points, while the other can't help but assign them into layers.
So these two types of people actually get along quite well—they’re both interested in discrete information, work in the same field, and can exchange insights with each other—but they also tend to look down on one another: the graph people think the table is “unsystematic,” while the table people think the graph is “too focused on the big picture and ignores the details.”
Don’t jump to conclusions just yet. Chapter 9 will carefully weigh this pair of rivals on the scales.
Linear thinking (continuous + flat): a river
For linear thinkers, information in the brain is connected from beginning to end—like a novel, like a river, like a movie.
The author’s exact words: “The information in my mind flows from beginning to end just like writing a novel.”
The key word for this way of thinking is sequence: first what happens, then what happens next, and finally, what will happen in the end. They understand a thing by telling it as a story; and when they predict what will happen, they simply extend the story backward—“What comes next?”
The conscious activity of linear thinking is analysis and reasoning: causal chains are their native language. Because A, therefore B; because B, therefore C. The author’s assessment of the individual’s abilities is: highly skilled in mathematics, as they can anticipate consequences—playing chess, conducting deductions, and playing dominoes are all games that rely on linear thinking.
Most readers would be taken aback: Isn't mathematics the most abstract of all disciplines? How could it possibly fit into the category of “continuous and flat”? Later, the author provided an additional explanation: Mathematics has nothing to do with abstraction—mathematics is detached from reality and thus cannot even be described as abstract. What mathematics demands most is a long chain of reasoning—from axioms to theorems—a seamless, step-by-step progression. And precisely those who think in a continuous, flat manner excel at exactly this kind of reasoning.
It corresponds to the “humanities” temperament: directly drawing on historical experience, focusing on “how to apply” knowledge, and paying close attention to the origins and context of matters. The ability to tell stories, the capacity for empathy, and the imaginative power to immerse oneself fully—all are gifts from the river.
The cost is also obvious: it’s difficult to reroute rivers. People who think linearly are least adept at “stepping outside the script”—when things aren’t arranged in chronological order (such as in a large table or a hierarchical taxonomy), they tend to lose patience easily.
Outline Thinking (Continuous + Hierarchical): One Book
Outline thinking is an upgraded version of linear thinking: the river is still the same river, but now it’s marked with navigational aids, segmented sections, and mileage markers. As the author put it: “Articles and books, too, have tables of contents.”
This kind of thinking processes a continuous stream of information—but rather than letting it flow freely, it organizes it into layers: This task has three stages; that project has five steps, and each step is further divided into smaller substeps. Maps have large regions and smaller neighborhoods; actions have major steps and minor steps—continuous yet hierarchical.
The conscious activity of outline thinking involves generalization and decomposition: breaking down a large, continuous task into discrete chunks (decomposition) and merging a collection of small chunks into a single overarching theme (generalization). The author’s assessment of the individual’s abilities is: “Good at completing large tasks step by step because they excel at decomposition.”
Project management, large-scale engineering, writing books, and governing a country—whenever something is “large and continuous,” it all falls within the realm of outline thinking. In Holland’s vocational interest model, the “enterprising” type corresponds precisely to this category.
Cost: The outline is the skeleton; the flesh and blood must be filled in by other forms of thinking. Outline-oriented thinkers tend to mistake “having planned it out” for “having already done it,” and they also tend to lack flexibility when faced with unexpected events that aren’t covered in the outline.
Quoted verbatim from Volume Five of the Supplementary Collection—The metaphors “a river” and “a book” only emerged many years later; yet the seeds had been sown long ago.
Each cell is independent and comparable in size and orientation.
The “Memory Production” column is worth reading on its own: The four modes of thinking each offer a distinct entry point for understanding the world—comparing differences, comparing similarities, comparing adjacency, and comparing subordination. Your very first act of perceiving the world fundamentally determines what kind of thinker you are.
The same accident, two different accounts.
Two cars scraped against each other at the intersection, and the traffic police arrived. An elderly bystander spoke first, offering a linear account: “I saw it all clearly—first, the white car pulled out of the residential complex without signaling; then, the black car happened to glance down at its phone for a moment. When it looked up again, bam!...” From start to finish, not a single second was missed—a complete movie unfolding before our eyes.
The traffic police’s report follows an outline-based approach: “1. Basic Information: time, location, and parties involved; 2. Sequence of Events; 3. Cause Analysis; 4. Liability Determination.” The same accident is broken down into four sections, each with its own heading. The sections are linked logically rather than chronologically.
There’s no superiority between the two accounts. The old man’s version is rich in detail and immersive, yet it would take half an hour to tell it all; the traffic cop’s version can be handed off to the next person in just thirty seconds—but you won’t feel that one little “shh” sound. One takes charge of the experience itself, while the other handles the handoff—river takes charge of flowing, and books take charge of being turned.
Now that we’ve covered the four ways of thinking, we can go back and explain the question we started with: Why is it so exhausting to talk to him?
Most likely, you’re outline-oriented, while he’s linear. You want to “start with the structure,” whereas he wants to “begin from the very beginning.”
Imagine you’re watching the news together on TV. You can’t help but say, “Just tell me the bottom line—has this thing actually worked out yet?” But he replies, “Don’t rush me! Let me start from the beginning—first, last month…” Neither of you is at fault; it’s just that the way information is presented differs: one of you wants a tree, the other wants a line.
Without a format converter, it’s like talking to chickens and ducks. In the next chapter, we’ll provide a comprehensive exploration of the conflict between these two types of people (as well as tables and graphs).
Let’s start with the name: The author refers to this theory about “comparison” as the “Comparative Theory.” In his later self-assessment of his system, he considers it the part he’s most confident in—and the one he’d most strongly recommend—“even more important than the philosophical framework itself.”
Quick sketches of two types of people
This chapter comes from one of the longest essays in the original notes, titled “Comparing People with Different Traits.” It depicts a pair of mirror images in a world of “displacement”:
People who compare differences—such as painters and musicians, who accumulate vast amounts of material and can perceive subtle nuances in their work; or domain experts, who master all the information within a particular field and can retrieve every relevant piece of information on any given topic. In everyday life, when we meet people, we tend to compare things like the size of their eyes, their height, their build, and so forth, thereby describing them in various aspects.
People who focus on similarities: When faced with a hundred things, they see three categories; they enjoy generalizing, abstracting, labeling, categorizing, and removing duplicates; they relish the thrill of mathematics—where “a few axioms can lead to millions of results”—and use abstraction to compress the world.
The author uses a beautiful metaphor: For the great tree of knowledge, those who focus on differences tend to delve into the leaves, while those who focus on similarities tend to delve into the trunk. Those who head toward the leaves pay attention to applications—after mastering specialized knowledge, they aim to generate practical benefits. Meanwhile, those who head toward the trunk focus on principles—understanding the underlying reasons and building a comprehensive framework in their minds.
(In the words of this section: This roughly corresponds to a comparison between tabular thinking and graph-based thinking. However, the original article’s intent is broader, referring generally to two different orientations within the larger tendency of “discreteness.”)
A Tree of “Soft Causality”
The main body of the original article is a massive comparative chart: starting from the way of thinking, it systematically reveals the differences between these two types of people in terms of cognition, values, and interpersonal relationships. The introduction includes an important statement that you must read first:
The causal relationships here are all probabilistic and not deterministic. This is easy to understand: there are many possible causes that can lead to a particular outcome, so the final result is less predictable. We can call this “soft causality.”
With this agreement in mind, we’ve selected several of the most informative comparative chains (each chain is read as “with a relatively high probability of tendency”):
The way to understand the world
People who compare similarities tend to: abstract, simplify, identify patterns, and pinpoint causes. The scientific mindset is this: Although the world is complex, there are simple underlying patterns and many deterministic factors at play. Thus, we analyze the causes and then use these fundamental principles to make predictions and analyses.
One in, one out—there’s no logic to it.
Why does this difference exist? The author explains that it stems from differences in the subjects of memory. When comparing distinct points, the subjects of memory are different instances rather than simplified, generalized outcomes—thus rejecting simplification and tending instead toward complexity, mystification, and total acceptance.
The way we approach the world
People who focus on differences: They enjoy acquiring various skills, with the ultimate goal of mastering them. The purpose of mastering these skills is to put them into practice—because they can see improvement with each exercise. They place great value on networking and spend a lot of time nurturing relationships.
Principles and Results
People who emphasize similarities: They hope the world is like a soccer field—clear rules and fair play; they stick to principles and bottom lines; they don’t focus on outcomes but on the process itself; and they treat everyone equally.
People who focus on differences: They are tolerant and able to compromise; they focus on outcomes, results, and calculating benefits. They treat different people differently—because they constantly compare differences, they naturally assume that people are inherently different, and treating different people differently isn't a problem at all.
Aesthetics and Text
People who focus on differences: They value aesthetics, pay attention to details, are meticulous about word choice, and place great importance on sensation—yet aesthetics is an area that requires extensive accumulation, precisely what “collecting” excels at.
Information intake method
The final point is particularly insightful: Those who focus on differences can only grasp information by drawing on rich, sensory inputs—such as bodily experiences and on-the-spot observations. In contrast, those who focus on similarities can derive vast amounts of meaning from a small amount of abstract information, excelling at deduction and recognizing universal connections—for instance, when they see a row of statistical figures, they can already conjure up an entire world in their minds.
No praise or criticism.
The notes in the original article are written with utmost seriousness and must be paraphrased:
·Listing so many differences doesn't mean that the two are diametrically opposed. On the contrary, the two often interact extensively—because they both focus on discrete information and operate in overlapping domains, enabling them to exchange insights with each other. However, their underlying approaches are different, making it impossible for them to communicate effectively about their respective approaches.
This article presents neither praise nor criticism, nor does it claim right or wrong. Both perspectives are understandable—each person can simply stick to their own viewpoint.
Here’s a gentle reminder: People who constantly compare others “wear many masks”—not out of hypocrisy, but because in their eyes, everyone is different, and every situation calls for a different approach. It’s only natural for them to say one thing to one person and another to another.
Two Ways to Explore a Museum
The museum is an open-air laboratory for observing these two types of people.
As soon as he entered the exhibition hall, he headed straight for the first exhibit and began examining them one by one: This piece had a lustrous glaze; that one boasted delicate crackle patterns; and the third featured a rare inscription on its base—each exhibit was unique in his eyes, and he read every single word on the display labels. Three hours later, he still hadn’t left the first exhibition room.
Another visitor walked in and immediately looked for the exhibition hall map: This museum is divided into five halls, each dedicated to a different dynasty. He started by getting an overview of the overall structure, then zeroed in on the highlights. His pace was brisk—pausing only ten seconds in front of each exhibit—but his mind had already pieced together a clear mental map: What exactly is celadon? What’s white porcelain? And what, precisely, is blue-and-white porcelain? The origins and development of these major categories were crystal clear in his head.
As they were closing up, the two of them met at the door and both felt the other had come all this way for nothing. One said, “You’re walking so fast—did you even figure out what’s going on?” The other replied, “You did figure it out all right—but you didn’t even get close enough to take a good look at the signature on that museum’s prized treasure.”
Actually, a museum can accommodate two different ways of exploring it: those who follow the leaves take away the texture; those who head toward the trunk take away the map. What’s truly worrisome is spending a lifetime exploring only one way.
It’s a real-world application of the entire “thinking framework”: starting from a binary comparison (comparing differences vs. comparing similarities), we use soft causality to derive dozens of observable differences, and then use these differences to explain the countless instances of “mutual misunderstanding” in everyday life.
The next time you feel like saying, “Why is this person acting this way?” you could rephrase it as another question: Is he someone who focuses more on differences, or someone who focuses more on similarities?
Four Ways to Approach a Problem
When something is right in front of you, the first question asked by four different types of thinking varies. The author’s comparison chart has only one line, yet the more you ponder it, the more accurate it becomes.
What is it, why is it, what will happen, and what should we do—these four questions almost encompass all the problems humanity faces in confronting the world.
The reason for subject bias in schools lies right here: a particular subject is essentially a training ground for a specific type of question. Those who excel at asking “why” tend to fall in love with science; those who excel at asking “what if” tend to gravitate toward the humanities; those who excel at asking “what is” tend to be drawn to the arts; and those who excel at asking “what should we do” often find school itself rather uninteresting—because school doesn’t teach them how to figure out what to do.
The author’s corresponding remarks (as a draft, for reference only and not definitive):
Here’s an additional sentence to explain why “discrete hierarchies excel in physics”: Physics requires first constructing a mental model (conducting experiments in one’s mind) and then comparing it with reality. As for the more counterintuitive correspondence—that “continuous, flat systems excel in mathematics”—Chapter 6 already provides a complete proof; there’s no need to repeat it here.
The author himself left room for interpretation: “If you’re not good at these, it might be more memory-oriented or more thinking-oriented. This is just a preliminary reflection.”
The Dewey Decimal Classification system was also brought into the picture. Influenced by Bacon, Dewey divided human knowledge into three major categories: memory (history), imagination (literature and the arts), and reason (philosophy and science). The author correlates tabular thinking with art (imagination), networked thinking with science (reason), and linear thinking with memory (history)—thus finding the intellectual origins of this tripartite division of human knowledge within different types of thinking.
An interesting application is knowledge management. People with four different thinking styles naturally adopt distinct approaches to managing knowledge:
Think about your note-taking software: Some people use nothing but tags and tables (Notion databases); others rely entirely on nested folders; still others keep everything as long-form journals; and yet others stick strictly to hierarchical outlines.
There’s no such thing as a “superior” approach—instead, we should align tools with our thinking, rather than forcing our thinking to conform to tools. If someone who naturally thinks in outlines insists on using the card-box method, they’ll only end up feeling frustrated. The debates over knowledge-management methodologies often boil down to different types of thinkers simply promoting their own preferred approaches to one another.
When asking for directions in an unfamiliar place, the way you respond—based on four different thought patterns—can reveal your personality type in just ten seconds.
For those who think in tables, the landmark is: “Turn right after you see the red mailbox, then spot the bank with the blue sign—on your left.” — Just follow the sequence of attributes, and you’ll find it easily. For those who think in graphs, they first give you the structure: “This area has a grid-patterned road system; streets running east-west are called ‘streets,’ and roads running north-south are called ‘roads.’ You’re headed to the intersection of Third Street and Second Road.” — First grasp the big picture, and the specific route will naturally emerge on its own. Those who think linearly can’t wait to lead you step by step: “First, walk straight ahead, cross one traffic light, then keep going. When you see an overpass, don’t go up—it’s better to pass underneath. Then…”—a complete, detailed path laid out from beginning to end. For those who think in outlines, they break it down for you into segments: “[Divide it into three parts. In the first segment, exit this alleyway; in the second segment, head onto the main road and keep walking straight; in the third segment, enter the residential complex across the street. Once you reach the end of each segment, come back and ask me about the next one.]”
All four fingerings are possible. But the one you prefer to hear reveals what kind of thinking you have; the way a person habitually fingers also honestly reflects their thought patterns. Asking for directions is the cheapest—and most accessible—test of one’s thinking style.
There are also two small finishing dimensions—scattered notes from the original notebook—that fit perfectly here:
·Creation and Execution: The “Bo” type—creators—enjoy fresh innovation and new insights; the “Zhuan” type—executors—prefer mature solutions and classic works.
There’s no inherent superiority or inferiority between breadth and depth. It just makes life a lot less painful when you’re looking for a job or building a team—knowing clearly who’s responsible for expanding the frontier and who’s responsible for holding the ground.
What is consciousness?
Consciousness is one of the greatest unsolved mysteries in modern science. Yet the author’s view is remarkably straightforward:
I believe that, if we don’t hold onto mystical ideas, consciousness isn’t hard to understand. Consciousness is simply a broadcasting mechanism (the Global Workspace Theory)—it’s just that memory, instinct, and sensation can reach consciousness.
Let’s break down this metaphor: The brain is like a grand theater, with different brain regions serving as the backstage crew—memory, instinct, sensation, and emotion. Consciousness, meanwhile, is the central stage bathed in the spotlight: whichever piece of information is brought onto the stage can be instantly received by every department in the entire theater. So-called “being aware” is simply “being broadcast.”
The author positions himself within the existing academic landscape: There are several schools of thought on the theory of consciousness (such as the Global Workspace Theory and Integrated Information Theory). He explicitly states, “My approach belongs to information-processing cognitive psychology and general psychology”—consciousness is the central stage bathed in the spotlight, a gigantic neuron that connects individual neurons, and—compared to the instinctual center—a rational center. He even said something even more pointed: After examining various schools of thought, “I feel that research on consciousness is still in the prehistoric era.”
AI is inadvertently validating thought frameworks.
Even more interesting is the second half. From recent advances in AI, the author observes that the structure of cognitive frameworks is being rediscovered by the engineering community:
·The MoE (Mixture of Experts) architecture of large models, much like his proposed “mind as a star-shaped structure”—where a central router distributes information to various expert modules. Bitcoin also exhibits this kind of structure: multiple entities compete for broadcasting rights.
360AI’s CoE (multi-model collaboration) approach involves invoking multiple large models to perform three steps: problem-solving, supplementation, and summarization. Following this logic, he reasoned: If we were to further develop this process—where a question is sent to tens of thousands of large models, several results are returned, the best one is evaluated, and then the evaluation result is sent back to those same tens of thousands of models, creating an endless loop—this would become the consciousness model he had envisioned: a continuous cycle of broadcasting and converging information.
●1AI has already begun to engage in causal inference—in the language of cognitive frameworks, this means AI is starting to “compare adjacent relationships.”
He also noted directions such as “continuous learning” and “nested learning”: Nested learning involves numerous submodules, which aligns with the broadcasting principle of cognitive frameworks; as well as designs that separate memory from the model and distinguish between deep and shallow memories—this separation of memory and consciousness is precisely the structure discussed in Chapter 2.
Put these together: In the engineering world, in order to build smarter machines, they are painstakingly reconstructing the structure of “memory/consciousness and broadcasting/collecting” piece by piece. It’s not because engineers have read this theory; rather, it’s because the very shape of the problem determines the shape of the answer.
The four ways of thinking = four types of AI
The author also draws an even more striking analogy: the several technological approaches to artificial intelligence happen to correspond precisely to four different modes of thinking:
·i Masks details and lets AI fill in the missing details (the principles behind large language models and image generation)—this is a table: completing an entire row from fragmented attributes;
AI Based on Knowledge Graphs and Logic—This is the Genealogy: Reasoning on a Classification Tree;
Obstruct the results and let the AI predict what will happen next (next-word prediction)—this is linear: following the chain of cause and effect forward;
Autonomous driving and maze navigation, gradually building a comprehensive map—this is the outline: extracting a hierarchical structure from continuous experience.
However, the author later added a safeguard to this comparison chart: Large language models themselves are not equivalent to any one of the four types of thinking—they represent a mechanism that is even lower-level than “comparison-based thinking.” Rather, as the scale of model parameters increases, abilities akin to these four types of thinking tend to emerge. Thus, the correct way to interpret this chart is as follows: The four task categories serve as question formats for AI—each of the four question types corresponds to one of the four types of thinking—but it does not mean that a particular model “is” a specific type of thinking.
The evolutionary history of machine intelligence is retracing the map of human thought patterns.
“Four types of AI”—sounds cutting-edge, but actually, you’ve got the whole package right in your pocket.
When you type, the input method keeps guessing your next word—following along with what you’ve said so far. That’s linear AI. The photo album automatically groups photos of the same person into a single album and even asks, “Who is this?”—completing a line of attributes from fragmented clues. That’s tabular AI. Your phone map plots a route for you that says, “40 minutes total—first by subway, then on foot”—breaking down a continuous journey into two separate segments. That’s outline-based AI. And some Q&A apps are backed by knowledge graphs: when you ask, “Who was Li Bai’s father-in-law?”, it traces the relationship chain layer by layer—reasoning through a classification tree. That’s graph-based AI.
Four distinct technological approaches—each company pursuing its own path—none of them seems to be validating a single overarching theoretical framework. Yet, when viewed together, these approaches turn out to be nothing less than engineered versions of four different modes of thinking. The very shape of the problem itself determines the shape of the solution: the machine has taken a long, circuitous route, only to end up back at the old mental map of the human brain.
The author also notes a key insight, which I’ll place here as the concluding remark of this chapter: Consciousness is a self-referential structure—thinking must involve some form of self-monitoring. Currently, AI is managed through external feedback loops; the next logical step, however, is to use AI itself to replace these loops, enabling the system to manage itself autonomously. Using AI to manage AI—this capability, within the framework of thinking, is referred to as “control of thought,” standing alongside breadth, depth, and information capacity of thought.
By that day, even the last square in my mental framework was filled in.
Four sets of self-test questions (from the original notes)
Below are four sets of minimalist self-assessment questions designed by the author. Instructions: For each set, choose one of the two options—don’t think about “What I should do,” but rather focus on “How I actually tend to behave more often.”
A. Consciousness or Memory? 1. I enjoy the pleasure of pondering and solving problems more. 2. I prefer expanding my knowledge base and becoming an expert in my field.
B. Creation or Execution?? 1. I prefer novel creations and fresh insights. 2. I prefer mature solutions to problems and classic books.
C. Continuous or Discrete? 1. I prefer stories to fragmented pieces of knowledge. 2. I enjoy mastering knowledge without worrying about the details of how it all fits together.
D. Hierarchical or Flat? 1. I prefer to grasp the knowledge of a discipline as a whole and am not interested in overly detailed information. 2. I enjoy practical, actionable details but am not very interested in the big-picture puzzle of the discipline itself.
After you’ve made your selection, cross-reference the results of Questions C and D: C1D1 represents outline thinking, C1D2 represents linear thinking, C2D1 represents graph-based thinking, and C2D2 represents table-based thinking. As for Questions A and B, they’ll reveal which side—consciousness versus memory, or generation versus execution—is stronger in you.
This self-assessment is merely an entry point—the author’s principle (see Chapter 10 of Part III): Self-judgment is more reliable than questionnaires; questionnaires merely provide a mirror.
Four measuring tools. The capacity of working memory—each of the four types of thinking has its own way of measuring: Linear thinking assesses how far one can reason ahead; diagrammatic thinking checks how many things one can recall at a glance; tabular thinking determines how many features one can keep track of simultaneously; and outline thinking gauges how finely a task can be broken down. Whichever tool is shortest reveals your weakest thinking style—this approach is faster than taking self-assessment quizzes.
At this point, all the components of the thought framework can be summarized in a single table:
Add two additional dimensions: memory/consciousness (inventory or processing), and generation/operation (creating something new or using something old).
The line “Memory Production” is worth reading on its own: The four modes of thinking each offer a distinct entry point for understanding the world—comparing differences, comparing similarities, comparing adjacency, and comparing subordination. Your very first act of engaging with the world fundamentally determines what kind of thinker you are.
This table is 4×4, with sixteen cells representing sixteen intersections. The rows represent four types of activities—“Memory/Consciousness × Production/Use”—while the columns represent four types of thinking. The intersections themselves are the true stars of this table; they answer the question: “When engaging in this particular activity, what exactly does this type of thinking entail?” Let’s examine them one by one.
Memory Production—Four Ways of Encoding New Information into Memory:
Comparing Differences (Tabular Thinking): Compare new information with existing inventory item by item to identify discrepancies. A painter can sense even the slightest differences in each brushstroke, and domain experts can retrieve all relevant information on every topic. In everyday life, we compare people—eyes big or small, height, weight, and build—describing them from every possible angle—this is precisely comparing differences. They delve deeper and deeper into “details,” gradually recording the world with ever greater precision.
Comparing Similarities (Graphical Thinking): Categorize new information under an existing class. Generalization, abstraction, labeling, and deduplication—all done in one smooth flow. They pursue the direction of the “trunk,” relishing the thrill of “applying one principle to multiple contexts,” and gradually simplifying their understanding of the world.
Comparing Adjacent Relationships (Continuous Flatness): Connect new information to its preceding and following events—what event it follows, and what event it will lead to next. A note from the author in 2023 can serve as testimony for this concept: “Now, artificial intelligence has already begun to engage in causal inference—that is, it’s starting to ‘compare adjacent relationships.’”
Comparing hierarchical relationships (continuous levels): Place the new element within its superior and subordinate categories—determine to whom it belongs and who, in turn, belongs to it. First establish the position, then discuss other aspects.
Consciousness Production—Four Ways of Generating New Ideas Through Thinking:
Collect and Apply (Table-Based Thinking): Gather numerous examples of a particular type of thing, then simply use them as they come. In your notes, there’s a comparison: When you “understand” the similarities among people, it’s to digest and reflect; when you “master” the differences among them, it’s to put that knowledge into practice. The more you see, the more naturally you’ll start using it—collecting is precisely the creative approach behind table-based thinking.
Understanding and Imagining (Graphical Thinking): Use abstracted “classes” to infer things you’ve never seen before. Once you understand a class, you can imagine members of that class that haven’t yet appeared.
Analysis and Reasoning (Continuous Flatness): Follow the chain of cause and effect forward. The notes state that people with continuous flatness excel in mathematics “because they can anticipate consequences”—once the first link in the chain is established, subsequent steps unfold on their own.
Generalization and Decomposition (Hierarchical Structure): Generalize upward into an overarching outline, and decompose downward into specific steps. It’s precisely this approach that enables even a major undertaking to be accomplished methodically and step by step.
Memory application—the way information is stored in the mind. This section focuses on Chapters 3–6; here, we’ll just mention the key points:
Table: There’s a table in your mind, with rows and columns for easy comparison.
Ontology: A categorized tree in which subclasses automatically inherit the attributes of their parent classes.
Linear: A river flowing from beginning to end, with each segment seamlessly connected to the next.
Outline: A book with a table of contents—first build the framework, then fill in the details.
Consciousness in Action—The Four Directions of Thought That Guide Our Daily Lives:
Focus (Table Thinking): Pin your attention on a single cell and delve deeply into it.
Divergent Thinking (Graphical Thinking): Making associative leaps between different categories. There’s a reminder in my notes worth copying here: “The side effect of focus is that it conflicts with divergent thinking”—concentration and divergence represent a battle between two opposing modes of thought.
Linear (continuous flat): As you follow the chain all the way through, what you fear most is being interrupted.
Decomposition (Outline Thinking): When encountering a task, break it down into several parts and tackle them one by one.
Finally, a note from the author: He once asked himself whether the line “consciousness utilization” should be revised to read “breadth of thought, depth of thought, information capacity of thought, and control of thought.” Although this revision wasn’t adopted in the final draft, the very question mark itself reveals that the “consciousness utilization” line remains an area he’s still refining to this day.
It answers: Why communication is exhausting, why people tend to specialize in certain subjects, why the debate over tools is boring, and why different AI approaches are like the four types of personalities.
It didn’t answer the question: “Can your way of thinking be changed?” Later, the author added an answer: “Thinking can indeed be trained—but compared to psychological types, thinking types are less easily changed and more stable.” Simply recognizing your own cognitive bias is already the beginning of change: a person with outline-oriented thinking, knowing that the other person thinks linearly, will no longer say, “Could you please get straight to the point?” Instead, they’ll say, “Just start from the beginning—I’ll take care of summarizing it myself.”
He also clarified a point that’s often misunderstood: A person certainly masters all four types of thinking, though they’re naturally more adept at one particular type. In different fields, everyone tends to rely on the type of thinking they’re best at—but no single field ever relies exclusively on just one type of thinking. Go is the best example: Some players have a keen sense of the big picture, others excel at shaping good positions, and still others are brilliant at calculation—yet no single player can master only one of these skills. Even those who excel in the big picture can’t be poor at calculation. Of course, the relative importance of the four types of thinking varies across different fields. As for more precise measurements, the author admits he hasn’t yet figured it out clearly and can only rely on self-reflection based on the descriptions provided in the book.
The framework of thinking clarifies “how the brain works.” After reading both “Humans” and “Thinking,” one question has become impossible to suppress: Where on earth did these categories come from?
The answer lies in the very foundation of the entire book—the method itself. Please turn to Part Three, “Concept Vectors: The Method of Concept Definition.”
Concept Vectors: A Method for Defining Concepts
A set of thinking tools that arrange concepts as clearly as elements.
In the first two sections (People, Thinking), we’ve been using a single tool all along—though we’ve never explicitly named it. In this section, we’re finally bringing it center stage: this is the foundation of the entire book, and it explores “how classification is created.”
We use concepts every day—introversion versus extroversion, rational versus emotional, simple versus complex, whole versus part... But where do these concepts actually come from? Why are some concepts clear, while others remain vague? And why is it that two intelligent people can argue for ages only to realize at the end that they’re actually talking about the same thing—or perhaps not even talking about the same thing at all?
This book introduces a method called “concept vectors” (formerly known as “binary genes”). Its proposition is both bold and concise: any concept can be expressed as a combination of binary dimensions, just as any substance can be expressed as a combination of several elements. The MBTI uses four letters to describe sixteen personality types, and the Bagua uses three lines to depict eight trigrams—both are examples of concept vectors; it’s simply that previous scholars failed to explicitly articulate this method.
This book will systematically answer the following questions: What exactly are concept vectors? Where do they come from? What rules do they follow? What can they be used for? And what are their limitations? After reading this book, you’ll develop a pair of “concept-deconstruction” eyes: whenever you encounter any pair of antonyms, any four-quadrant diagram, or any classification system, you’ll be able to recognize the underlying common structure behind them all.
Chapters 1 and 2 explain “why it’s needed,” while Chapter 5 explains “what it is.” Readers who are eager can start by reading Chapter 5 first.
Chapters 3 and 4 cover the methodological prerequisites (abstraction, modeling, analogy) and serve as the scaffolding for understanding the entire book.
Chapters 6 through 10 cover rules and applications; Chapter 11 explores the origins of the ideas; and Chapter 12 provides a summary.
A common debate
The family is discussing whether their child should enroll in piano lessons. The dad says: “First, the child isn’t interested; second, two classes a week are too exhausting; third, even if they do learn, it might not be very useful.” The mom says: “First, all the kids at my colleague’s house are taking piano lessons; second, music helps cultivate one’s character; third, if we don’t start now, they’ll regret it later.”
Both individuals could each come up with ten more points. As you sit there listening, the more you listen, the more confused you become: Should you really be paying attention to numbers one, two, and three, or to numbers four, five, and six? Is this debate really a clash between “one, two, three” and “four, five, six”?
The author of this set of ideas says: No.
His point is this: Whether it’s one, two, three or four, five, six—behind it all lie a, b, and c. The reason A and B ended up arguing is that both of them were attacking each other using the finished products assembled from these components, without taking the products apart to check whether the internal parts came from the same batch.
If you break down “one, two, three” and “four, five, six” into the levels of A, B, and C, the situation changes: Dad’s “lack of interest” and Mom’s “cultivation of temperament” both stem from the same underlying factor—what exactly is the purpose of learning the piano? Dad’s “too much hassle” and Mom’s “you’ll regret it later” reflect yet another underlying factor—the question of whether the time and effort invested are truly worth it. The real disagreements boil down to just one or two key factors, which can be discussed objectively and pragmatically. As for everything else, what the two of them are really talking about is essentially the same thing.
This is the starting point of this set of ideas: concepts are composed, and compositions can be disassembled.
Question 7: The birth certificate of this set of ideas.
In the author’s notes, there remains a chain of inquiries that he himself calls “the origin.” Seven questions, step by step, led to the development of the entire methodology.
Question 1: Is human thinking really indescribable?
The current state of psychology seems to support the notion of “not being able to explain clearly.” Psychology encompasses dozens, even hundreds, of different subfields, each with its own distinct perspective. Take personality alone: there’s MBTI, the Big Five, Enneagram, Jung’s eight cognitive functions, and various theories of temperament—all of which have complex relationships that no one can fully articulate.
But the author’s judgment is precisely the opposite: it’s not that difficult. The problem lies in the approach—we’re lacking a handy tool.
Second question: If you can’t explain it clearly, could you use the method of “guessing”—“guessing”—to infer the underlying principle based on the surface phenomena?
That’s exactly how Traditional Chinese Medicine works. Since ancient people didn’t have the means for anatomical dissection, they observed external manifestations, speculated on the internal mechanisms of the body, and then tested their hypotheses through clinical efficacy.
When confronted with fundamental particles that temporarily cannot be further divided, physicists can only rely on speculation and conjecture—first developing a workable model and then seeking more evidence to support it.
Since both Traditional Chinese Medicine and physics can work effectively even when dealing with objects that remain “opaque,” why can’t thinking and consciousness do the same?
Third question: With so many appearances, which ones are fundamental?
A person’s outward appearances are endless: the words he uses when speaking, his gait, his spending habits, the way he loses his temper... If each of these appearances were studied individually, even a hundred years wouldn’t be enough to exhaust the research.
There must be a way to identify, from the vast sea of appearances, which are “parts” and which are “finished products.”
Fourth question: How do we determine the mutual relationship between nouns?
Let’s first look at the simplest case. Antonyms represent a relationship of opposition; synonyms represent a relationship of similarity; and what dialectics studies is the relationship of unity in opposition between two terms.
The relationship between these two words already involves so many nuances—so then...
Fifth question: Can a more complex relationship among four words describe things more precisely?
Four words—each pair is related to the others, and the total number of relationships suddenly multiplies. If these four words are chosen wisely, they can serve as four pillars, supporting an entire conceptual space.
Sixth question: Where did these four-word phrases come from?
The answer has emerged: words grouped in sets of four generally consist of combined concepts formed by pairing two “concept vectors.”
A concept vector consists of a pair of opposite terms—such as big and small, inside and outside, dynamic and static. Each concept vector has two possible values; when combined, they yield exactly four possibilities. With three concept vectors, the number of combinations rises to eight; with four, it climbs to sixteen—and this process can continue indefinitely.
Question 7: What are the benefits of doing this?
The benefits are immediate and obvious: Once you’ve clarified the combinatorial relationships among concepts, the number of terms you need to study is drastically reduced. What used to be hundreds of concepts that you had to memorize one by one now requires only a few dozen “genes” to understand—everything else is just permutations and combinations.
How big is this matter?
At the end of this chain of follow-up questions lies a rather ambitious goal:
Concept vectors actually address the fundamental question of “how concepts are defined and how they arise.” Conceptual abstraction is the very foundation of philosophy, the core of everyone’s thinking, the bedrock of culture, and the root of everything related to abstraction—be it language, programming, information, or cognition itself.
This may sound grand. But first, let’s keep in mind what it looks like in everyday life.
Before the periodic table was introduced, chemists were confronted with dozens of seemingly unrelated substances; discovering a new element felt like winning the lottery. But after Mendeleev arranged them into a single table, even the empty spaces in the table took on meaning—it predicted elements that had yet to be discovered.
The conceptual world of humanity today is still in a state “prior to the advent of the periodic table”: amid an ocean of knowledge, vast amounts of content merely reiterate the same ideas in countless different ways; meanwhile, the truly important conceptual combinations remain overlooked and unmentioned by anyone.
The concept vector aims to be a periodic table of concepts. It may not achieve complete success, but its logic is simple enough—so simple that everyone can learn it in ten minutes and spend a lifetime testing it.
A component from a Sichuan dish.
The word “parts” may sound like something from a factory, but actually, you can find them right in the kitchen.
Open the Sichuan restaurant’s menu, and you’ll find dozens of dishes: shredded pork in fish-fragrant sauce, Kung Pao chicken, Mapo tofu, twice-cooked pork... To outsiders, the array of dishes is dazzling—each one seems like a whole discipline unto itself. But chefs don’t see it that way. What chefs focus on is the row of seasoning bowls arranged beside the stove: scallions, ginger, garlic, broad bean paste, sugar, vinegar, Sichuan peppercorns, and chili peppers. Break down those dozens of dishes, and you’ll find that they all boil down to different combinations of these same seasonings—the fish-fragrant flavor comes from a specific ratio of sweet-and-sour broad bean paste; Kung Pao relies on a blend of numbingly spicy chili paste with sweet-and-sour flavors; and Mapo tofu is made by combining broad bean paste, Sichuan peppercorns, and minced meat.
So when a chef learns a craft, it’s not about memorizing a hundred recipes—it’s about becoming intimately familiar with each individual component. Once you’ve mastered the components, cooking becomes a matter of arranging and combining them; if you just rely on memorizing recipes, you’ll find yourself unable to cook anything once you switch ingredients.
Concepts are just like dishes. Most of the grand principles we argue over, study, and memorize are “ready-made dishes”—the whole idea behind this line of thinking is to take you right up to that row of seasoning bowls by the stove.
Map of this book
This booklet will unfold along three main lines:
The Problematic Line (Chapters 2–4): Why is the scientific method insufficient? And when science falls short, what other methods can we turn to?
Methodology (Chapters 5–8): The formal definition of conceptual vectors, the fundamental relationships among concepts, dimension axes and grayscale, and the naming and encoding of concepts.
The Value Line (Chapters 9–12): What it can and cannot do, how others might challenge it, and who its intellectual forebears are.
Before we begin, we need to accept an agreement made by the author. He says: It’s impossible for every single word to be 100 percent precise—there simply aren’t enough words in the world to achieve that level of accuracy. Readers must themselves “grasp” what each word is pointing toward.
This isn’t about shutting people out—it’s about letting go in advance: You don’t need to pore over every single word of this book; instead, just follow along with the examples. The examples are all drawn from everyday life—once the meaning clicks, the right words will naturally come to you.
The agreement is now in effect. Let’s now turn to the first real question: With science already so advanced, why does this book even come into play?
First, let's pay tribute to science.
Let me start with a fair point: the reason we’re picking apart the science in this chapter is precisely because we respect science.
Science is the most successful tool of knowledge that humanity has ever devised. The temperature at which water boils, the load-bearing capacity of bridges, the dosage of medications—in all these fields, scientific answers are almost never wrong.
The author is not opposing science itself, but rather a mindset he calls “Scientism”—the belief that only science is right, and everything else is wrong; the belief that wherever science cannot reach, there’s nothing worth thinking about, or even nothing that exists at all.
The problem is that the areas where science falls short are far greater than we realize.
The Six Limitations of Science
In the author’s notes, there is a list of the limitations of science.
Focus on the big picture of science; don't dwell on details.
Science focuses on significant, large-scale phenomena observed in large samples and does not devote much attention to excessively small-scale phenomena.
Take, for example, the pulse diagnosis. In Traditional Chinese Medicine, the pulse can be categorized into dozens of types—such as floating, sinking, slow, and rapid—and each type corresponds to a wealth of clinical experience accumulated over time. However, these subtle differences, which rely heavily on the individual practitioner’s tactile sensitivity, are difficult to bring into the scientific realm—not because they don’t exist, but because they’re simply too “small,” so tiny that it’s impossible to design large-scale experiments to verify them.
II. Look at the cause scientifically, not at the preceding factors.
Scientific inquiry asks, “What causes what?” and demands a clear causal chain.
But in reality, many things are influenced not by “causes” but by “antecedents”—a series of historical conditions, environmental preconditions, and serendipitous encounters. Although these antecedents don’t establish a strict causal relationship, they nonetheless have a tangible impact on the outcome. The choice a person makes today about which city to develop their career in is often not determined by a single “cause,” but rather emerges from decades of accumulated antecedents.
Third, science emphasizes certainty and is not adept at dealing with complex issues influenced by multiple factors.
This is the most critical one.
The ideal scenario for the scientific method is to control all other variables, manipulate only one variable at a time, and observe the results. However, when a single event is influenced simultaneously by dozens of factors—and these factors also feed back on each other—using the method of controlling variables becomes ineffective.
Psychology, sociology, and economics all deal with precisely this kind of object.
Four, science relies heavily on deduction, but deduction is not the exclusive domain of science.
Conversely, we should also be fair: even non-scientific fields can employ deductive tools—and when they do, the results become more rigorous, certain, and easier to reach consensus on. The concept of vector spaces, which will be discussed later in this book, is precisely an example of a non-scientific approach that makes extensive use of deduction.
The most typical example is history. History cannot be replayed as a controlled experiment.
All research that relies on one-time events—certain aspects of history, archaeology, geology, and each individual’s unique life—are outside the comfort zone of science.
Six, science places greater emphasis on “low-end” tools such as physical chemistry rather than macroeconomic regulation.
“Low-end” isn’t a derogatory term—it simply refers to a level of study: physics and chemistry investigate the fundamental layers of the material world.
But many real-world problems are high-level regulatory issues—managing a city, maintaining a healthy body, or nurturing a relationship. The role of underlying principles in addressing these high-level issues is akin to the role quantum mechanics plays in cooking: in principle relevant, but in practice utterly unhelpful.
Repairing computers and mediating disputes
Where are the boundaries of science? Just compare two things at home, and the answer will become clear.
One task is fixing computers. When a computer won’t turn on, those who know what they’re doing follow a standard procedure: First, try replacing the power cable; if that doesn’t work, try replacing the memory module; and if that still doesn’t help, disassemble the graphics card—working on only one variable at a time, pinpointing the exact point where the problem lies once you’ve made progress. This is the perfect setting for scientific inquiry: variables are controllable, experiments are reproducible, and the causal chain is crystal clear.
The other thing is mediating a quarrel. A young couple is arguing, and they’ve asked you to settle the dispute. After listening carefully, you realize that the root of the conflict lies in a single remark made last week—but that remark somehow ties back to last month’s accounting records, which in turn involve the mother-in-law, who then brings up the seating arrangement at the wedding three years ago... Dozens of factors are all intertwined, each one feeding into the next. And yet, you can’t possibly run even a single “single-variable” experiment—after all, you can’t just temporarily remove the mother-in-law and see whether the argument still continues.
Fixing computers is something science can handle; mediating disputes, however, is beyond science’s reach. Yet in life, mediating disputes far outnumber fixing computers. Thus, the “Six Limitations” aren’t just a list for armchair theorists—they’re an everyday reality in every household.
Why Psychology and Economics Just Can’t “Get Together”
The author makes a sharp observation: Disciplines such as psychology and economics always try to align themselves with science, but in reality, they just can’t quite make it.
Why? Because the research objects of these disciplines share several common characteristics: there are too many uncontrollable factors, and any change in conditions can alter the results. Surveys, statistical analyses, and experiments—all of them are merely “pretending to be science.”
Survey statistics certainly have their value, but they cannot be falsely labeled as science—they can only be called summaries of phenomena, and they should be content with being just that. If we dig deeper one layer further, asking why these fields cannot become sciences, the author offers two reasons.
First, individual differences are too great, and both individuals and their environments are changing with the times.
Physics studies electrons—every electron in the world is identical. Psychology, on the other hand, studies human beings; no two people in the world are exactly alike. Not only are individuals different from one another, even the same person isn’t the same from year to year. And this generation is even more different from the previous one.
Second, once complexity surpasses the critical threshold, the most important phenomenon is “mutual influence and positive and negative feedback.”
It’s not just that induction is no longer sufficient—it’s that the nature of the objects themselves has changed. The human body is one of the systems with relatively high stability, yet even so, half of its phenomena already represent outcomes beyond the critical threshold of complexity.
Thus, the author argues that psychology and sociology—just like history—are disciplines that “grasp phenomena and then interpret those phenomena.” What they lack is an underlying model. This intellectual framework—particularly the cognitive frameworks, psychological frameworks, and philosophical frameworks that will be discussed later—is precisely an attempt to provide such an underlying model for these fields.
Science, Philosophy, and Theology: A Division of Labor Chart
If science isn't sufficient, then who's responsible for the areas where it falls short? The author provides a remarkably concise chart illustrating the division of labor:
What science can’t solve must be tackled with other approaches—namely, philosophy. And if philosophy can’t convince you, you’re left with no choice but to pick one belief to embrace—that’s theology. (Just kidding: Behind theology comes classification—because classification explains why you believe this rather than that.)
This diagram deserves a few extra words of explanation. It doesn’t贬低 theology; rather, it points out a fact: the problems humanity faces are naturally divided into three layers.
·The first level, which is verifiable, should be left to science—let experiments and observations decide.
The second level involves propositions that cannot be verified but can be debated. This is the domain of philosophy, where logic and speculative reasoning are used to compare the strengths and weaknesses of various claims. The abstract modeling discussed in this book falls precisely within this second level.
●The third level is one where neither verification nor resolution is possible. Each person can only choose one to believe in. There’s nothing shameful about this—after all, the most important decisions in life (which values to embrace and what kind of life to lead) are made at this very level.
The error of the “Scientology” organization lies in declaring that the second and third levels do not exist, or in pretending that they can be subsumed into the first level.
In reality, however, the issues of the second and third levels not only exist but often have a greater bearing on human happiness.
One morning, a three-level question.
The three-tier division of labor may sound abstract, but you’ll actually encounter it all in just an ordinary morning.
In the morning, when I opened the fridge, I found that the milk had expired by one day—should I drink it or not? That’s the first-level question: it’s verifiable. Let’s check what the “best-before date” actually means, and give it a sniff to see if it’s gone bad. The answer is objectively available; no matter who checks, the result will be the same.
This morning, a friend sent you a message, inviting you to join them in a small business venture—would you go for it? This is the second level of inquiry: it can’t be verified, but it can be debated. No instrument can tell you outright, “You should go,” yet you can lay out the facts, calculate the risks, and think through potential fallback options—step by step, carefully weighing the pros and cons. The abstract modeling discussed in this chapter takes place precisely at this level.
At lunchtime, an elderly family member suddenly asked, “What’s it all about—what do we strive for in this life?” The table fell silent for a couple of seconds. This was the third level of questioning: it couldn’t be verified, nor could anyone offer a definitive answer. In the end, each person could only choose one answer to believe in—and those who picked “seeking excitement” and those who chose “seeking peace of mind” ended up living entirely different kinds of lives.
What’s so troubling about “Scientology” is its insistence on denying the existence of second- and third-level issues. But you can’t fool people for an entire morning: each of these three levels has its own unique solution—no one can ever replace another.
A specific debate: Can Bayesianism define science?
In recent years, a popular view has emerged: The essence of science is not falsifiability, but Bayesianism—science is simply the process of continually updating one’s degree of belief in light of evidence. Some science communicators have even declared that “falsifiability is already obsolete.”
The author explicitly opposes it.
His view is: To determine whether something is scientific, we should look at its falsifiability. It is precisely abstract models like conceptual vectors that make use of Bayes' theorem.
Why? Because Bayesian degrees of belief differ from person to person—they’re subjective and, in many cases, unverifiable. The very reason science is called “science” is precisely because it strives for intersubjective consistency—anyone performing the same experiment should arrive at the same result. If we were to include “subjective belief updates” as part of science, the boundaries of science would dissolve, and anything could be lumped into it.
This debate is discussed here precisely because it foreshadows the position taken in this book:
The scientific belongs to science—falsifiable, reproducible, governed by deterministic laws or the law of large numbers;
●Each has its own specific scope of application; neither can subsume the other.
The boundaries of science are not flaws in science itself, but rather the conditions under which this tool—science—can be applied effectively.
Acknowledging that science has boundaries isn't about opening the door to ignorance; rather, it’s about making room for other legitimate modes of thinking—abstraction, modeling, analogy, and deduction.
In the next chapter, we’ll take a close look at exactly what these “non-scientific” ways of thinking are all about.
You’ve been using it since you were a child—only no one ever told you.
The word “abstraction” sounds like jargon from the philosophy department. But in fact, it’s quite the opposite—it’s something you use every single day, yet you’re so accustomed to it that you don’t even realize it.
Language is abstraction. When you say the word “chair,” you’re not referring to the wooden chair in your home or the swivel chair in your office—but rather to a whole category of things that can be sat on.
Grouping together things that share certain common features—that is abstraction. Conversely, dividing a group into two subgroups based on a particular distinguishing feature: this distinguishing feature itself is also an instance of abstraction. The entire process of dividing a set into two subsets according to both shared features and distinguishing features constitutes the full scope of abstraction.
There is a famous case in personality psychology.
The method behind the Big Five personality theory is essentially the “lexical hypothesis”: Researchers collect all words in human language that describe personality traits, perform statistical analyses to identify which words consistently appear together, and ultimately derive the five major personality factors from these patterns.
In other words, the Big Five personality traits were “dug out” from the abstract sediment of language. Our ancestors condensed millions of observations into vocabulary, and researchers then unpacked these vocabularies to derive classifications.
Abstract—compress—then abstract again—this is the assembly line of humanity’s conceptual factory.
A single word, “eaten,” is a condensed package of an entire meal.
When Chinese people meet, they greet each other with: “Have you eaten?” “I have.”
The two words “ate” encapsulate an entire meal—what was eaten, where it was eaten, who ate it with, and whether it was delicious—all packed into one concise statement, without a single detail left out. No one thinks this way of speaking is lazy, because both parties understand: in this context, that level of compression is exactly what’s needed.
If the other person asks, “What did you eat?” the compressed file starts to unpack: “Noodles.” “What kind of noodles?” “Beef noodles—medium-thin strands, with an egg.” Back and forth we go, gradually unpacking from the abstract all the way down to specific details. How deep the conversation goes depends entirely on how interested the other person is.
Everyone is familiar with this set of moves—though no one calls it “abstract.” The very ability of language to adapt and adjust its level of compression according to the context is what keeps it functioning: we use big words when writing reports, and concrete examples when discussing details. The frameworks discussed in this book work in much the same way—first compressing vast, complex entities like thought, psychology, and society down into a tiny package called “four”; then, when we need more detail, we unpack it layer by layer right before your eyes.
Why abstraction is necessary: because the world is layered.
Abstraction is not a luxury for scholars—it is a necessity forced upon us by the structure of the world.
The world is layered. When the quantity reaches a certain scale, there’s a certain probability that higher-level structures will emerge: molecules aggregate to form macromolecular organic compounds; macromolecules aggregate to give rise to single-celled organisms; cells assemble into organs; organs combine to form humans; and humans come together to form societies.
Each level has its own set of new characteristics, which differ significantly from the underlying characteristics. The atoms that make up objects obey quantum laws; the objects themselves follow Newtonian mechanics; the Earth adheres to geological laws; and interstellar space obeys astronomical laws.
You can’t use quantum mechanics to directly predict the weather any more than you can use Newtonian mechanics to explain love.
The existence of hierarchy brings about two consequences.
First, because there are multiple levels, abstraction is necessary. Each level requires its own set of concepts. Attempting to study the nation using the concepts developed for studying electrons is bound to fail.
Second, the higher we go up the hierarchy, the harder it becomes to maintain scientific certainty. As we move along the chain—from physics and chemistry to biology, psychology, and sociology—the complexity steadily increases, and factors such as randomness, probability, sensitivity to initial conditions, and irreproducibility become increasingly prominent. For objects that are too complex, we can only offer coarse-grained descriptions—qualitative rather than quantitative.
This is precisely the moment when abstraction makes its appearance: When an object becomes so complex that it can no longer be measured accurately, humans resort to a “simplification of thought” in order to grasp it.
Abstraction is modeling, and modeling inevitably involves error.
The cost of abstraction must first be laid out on the table.
Abstraction is about simplifying complex things—just like drawing a picture: with just a few strokes, you outline the general shape and identify the most important elements of the whole. If you fail to do this successfully, you’ll end up drawing only a foot or a head.
More precisely, abstraction is mathematical modeling: using the changes in a few variables to simulate reality. Since it’s a model, there will inevitably be errors; it can only “fit”—that is, approximate—rather than be exactly equal.
The quality of a mathematical model is judged by how small the error is in its ability to fit reality; however, this kind of fitting and the associated fitting errors do not come with the same kind of inevitability or the guarantees provided by the law of large numbers that are characteristic of science.
From this, two inferences can be drawn.
Inference 1: Abstraction is a creation of the human brain—it does not exist objectively; rather, it is more akin to mathematics than to science.
The information in the brain is always the result of abstraction—it is never the objective reality itself. The notion of “objectivity” is actually a stance rooted in reductionism—the assumption that there exists a world that can be endlessly broken down and ultimately fully determined.
Inference 2: Abstract models cannot test “truth or falsity”; they can only test “goodness or badness.”
Once abstraction is performed, it will inevitably differ from the real entity. Therefore, when evaluating whether a model is valid, the criterion for assessment is not falsification—but rather: Is it easy to use? Is it effective? Are there better models available?
This effectiveness is probabilistic. Whether a traditional Chinese medicine model holds up depends on its therapeutic efficacy; whether concept vector modeling is valid depends on the extent to which it helps people clarify their thinking about a problem.
There’s also a quantitative intuition regarding the quality of abstract models: the longer the gene length (equivalent to having more parameters in a multivariate equation), the higher the fitting accuracy. A four-dimensional model is more refined than a two-dimensional one; an eight-dimensional model is even more refined than a four-dimensional one—but at the cost of increased complexity and greater effort required to work with it as the number of dimensions grows.
Good modeling involves striking a balance between “model complexity” and “real-world fit”—the comparison is made based on the degree of fit achieved at the same level of complexity.
Instances and Abstractions: Two Complementary Paths
When faced with a problem, people have two paths to choose from.
One approach is the case-based method: a car—what’s its payload capacity, horsepower, and wheelbase? Mastering these details allows you to arrive at a precise solution. The advantage of the case-based method is its precision; its disadvantage, however, is that cases are regionally specific and temporally limited—change the location or the era, and you’ll have to start all over again.
The other approach is the abstract one: a car is simply “wheels plus a sofa.” Starting from this rough model and gradually adding details can also lead to relatively precise solutions. Abstraction possesses universality and permanence.
When studying a specific instance becomes impossible—because the object is too complex, too distant, or non-replicable—abstraction becomes the only viable path. It allows us to derive numerous results from a relatively small amount of information.
Complex entities have another characteristic: each individual is unique, and no two “people” have exactly the same configuration. Abstracting over the commonalities among multiple individuals is precisely what this kind of object is designed for.
Currently, one approach in artificial intelligence involves using the learning outcomes of Model A to benefit Model B—this is known as transfer learning. Abstract thinking has long been able to employ this method. The ancient Chinese concept of “harmony between heaven and humanity” is precisely an example of applying insights gained from nature to the human body—this represents humanity’s earliest instance of cross-domain transfer learning.
Zang Xiang: The Study That Isn't Disassembled
In Traditional Chinese Medicine, there’s a concept called “Zang Xiang,” which means: The functions of the internal organs (Zang) manifest as external signs (Xiang).
The key point of Zang-Xiang thinking is: without dissection, we abstractly summarize from external observations and design (or speculate on) the internal mechanisms.
The author argues that the Zang-Xiang thinking is a methodology that has been seriously underestimated. He offers a compelling defense: Just as physics, faced with currently indivisible fundamental particles and fundamental forces, must rely on conjectures and hypotheses to reach consensus, so too does the quest for a unified theory require even more conjectures and hypotheses.
In terms of methodology, Traditional Chinese Medicine and quantum mechanics are actually isomorphic: through repeated trials and explorations, one gradually develops a workable formula. When the formula proves effective, it means the initial guess was correct. The relationship between medicine and symptoms has been refined over thousands of years through repeated practical experience.
The concept of "zang-xiang" and abstraction inevitably lead to a holistic perspective—because you can't dissect an object into its components for study; instead, you can only observe it as a whole.
Western medicine only focuses on the physical (surgical), chemical, and cellular (pharmaceutical) levels; yet many things simply cannot be disassembled.
·Humans currently do not yet fully understand all the links in the chain of feedback loops;
Therefore, “inferring internal structure from external manifestations” is not a backward-looking expedient but rather a legitimate approach when dealing with complex objects.
The thinking frameworks and psychological frameworks that will be discussed later in this book are precisely the internal structures of the brain “guessed” using this very same method—exactly the same methodology employed by Traditional Chinese Medicine to infer the structure of the human body.
Abstraction relies on consensus to take effect.
Finally, there’s a sociological fact that’s often overlooked: abstraction cannot be empirically verified; for it to exert any effect, it must rely on universal acceptance.
Language is the best example of this. Language is an abstraction; humans cannot do without it. Yet the emergence of language wasn't simply because it made logical sense and turned out to be useful—rather, it was adopted and recognized by everyone precisely because it proved practical and effective in real-world use.
Mathematics is also abstract; in mathematics, deduction plays a particularly large role, and when deduction dominates, it’s easier to gain recognition. Philosophy, too, is abstract—but in philosophy, induction plays a larger role, making it harder to reach consensus.
Thus, the fate of an abstract model is akin to the fate of a language: it doesn't hinge on whether it's "true," but rather on whether enough people find it useful and are willing to use it.
The Deming Cycle (PDCA) is an abstract model, and so is MBTI. Once these models are widely recognized and accepted, they start to function—meaning that everyone essentially builds a program in their own minds. These universally acknowledged abstractions serve as a third-party code library for all of humanity.
Concept vectors are a set of techniques and theories concerning “how to build programs in the brain.”
The shredded pork in fish-flavored sauce doesn't contain any fish.
“Abstraction only works with consensus”—and the Sichuan cuisine menu is the best proof of that.
The dish "Yu Xiang Rou Si" doesn't actually contain any fish—at least nine out of ten first-time diners will pause for a moment upon ordering it. But Sichuan locals aren't taken aback: in their shared understanding, "Yu Xiang" no longer refers to fish at all. Instead, it denotes a specific flavor profile—a taste created by blending pickled peppers, ginger, garlic, sugar, and vinegar according to a particular ratio. "Yu Xiang" is an abstraction; its meaning is sustained by collective agreement, not by the literal definition of "fish" as found in dictionaries.
There are countless dish names like this: “Spicy Sliced Beef and Offal” doesn’t actually contain any married couple, “Ants Climbing Trees” has no ants at all, and “Wife’s Cake” doesn’t include a wife. To outsiders, taking these names literally would lead to hilarious misunderstandings, but locals, guided by shared understanding, use them with perfect ease. That’s exactly how language works: the meaning of a word isn’t embedded in the word itself, but rather resides in the mutual agreement and conventions held by those who use it.
So the fate of an abstract model is akin to the fate of a language—“Yu Xiang” can survive not because it’s “true,” but because enough people find it useful and are willing to use it. The same principle applies to whether this conceptual vector menu can be passed down through the ages.
Abstraction is classification; classification is abstraction. Starting from similarities and differences, it groups the world into categories.
The world must resort to abstraction precisely because of its hierarchical structure. Abstraction, in turn, inevitably involves error due to simplification. This error means that the only valid criterion for evaluating abstraction can be “usability,” rather than “falsifiability.” And ultimately, “usability” is determined by the consensus of its users.
Having understood this property of abstraction, we can confidently move on to the next chapter: In the toolbox of abstraction, there are three most commonly used tools—guessing, modeling, and analogy.
Guessing is unavoidable.
As people live in this world, they are constantly making guesses—guessing tomorrow’s weather, guessing the other person’s intentions, and guessing the consequences of a particular choice.
People speculate about many things that cannot be confirmed—not out of weakness, but as a fundamental survival instinct.
The real question isn't whether to guess—or not—but rather, “Which path should we take to make the guess?”
A living textbook in this regard is stock price forecasting. Open any stock forum, and you’ll find all sorts of theories fiercely competing with one another—some rely on chart patterns, others on fundamental analysis, some follow market rumors, and still others go with their gut feelings. For the same stock, different prediction approaches each have their followers and their own share of successes and failures.
Studying “how to guess” is just as important as studying “what to guess.”
The author’s notes outline three speculative pathways.
Path One: Learn from History
The first approach is to predict the future based on history: what happened yesterday will likely happen again tomorrow in a similar way; methods that worked last year will probably still work this year.
That’s exactly what neural networks do: You give them a large number of historical samples, they statistically identify patterns from these samples, and then use those patterns to make predictions about new samples. Empiricists take this approach, and so does machine learning.
The strength of this approach lies in its stability and reliability—as long as the environment remains unchanged. But its weakness also stems from this very point: once the environment changes, past experiences become a burden.
Pathway 2: Analogy
The second approach is analogy: using general principles to predict specific cases.
The logic of analogical thinking is this: If A and B are similar in certain respects, then the principles that hold true for A may also hold true for B.
For example, plants absorb energy and consume energy; animals also absorb energy and consume energy; cars likewise require energy input and consume energy—thus, the “input-transformation-output” model applies equally to all three of them. Research findings based on this model can therefore be applied simultaneously to all three.
Traditional Chinese Medicine is a master of analogical thinking. TCM holds that nature operates in cyclical patterns, and the human body likewise exhibits cyclical rhythms. Thus, it hypothesized that the human body possesses structures similar to those found in nature—leading to the concept of the Five Elements.
Using nature’s “white box” to explain the human body’s “black box”: If a natural water flow gets blocked, what might happen? Similarly, if the body’s qi and blood get blocked, something similar could occur.
Analogy and classification are twin brothers: things with the same “genes” can be analogized. The rules governing one thing can be applied to all concepts and entities that share the same genes.
It is precisely because of analogy that the study of the most universal and abstract concepts carries the greatest significance—once you’ve thoroughly understood the pair of concepts “motion and stillness,” you’ll simultaneously have thoroughly understood countless myriad phenomena that encompass both motion and stillness. That’s why this book repeatedly asks, later on, about the reasons behind the most universal concepts.
Even art may be grounded in the principle of analogy: Why do we associate various emotions with a piece of music? Perhaps precisely because the rise and fall of sound—its pitch, tempo, and intensity—share structural similarities with the ebb and flow of emotions. Such similar structures evoke similar experiences.
Path Three: Design
The third approach is the most distinctive and best embodies the methodological hallmark of this set of ideas: to observe the manifestation of a given phenomenon and, without disassembling it, design your own entity that can produce exactly the same manifestation.
To give you an analogy: Suppose you have an opaque machine in front of you. You enter a few numbers, and it spits out a result. You can’t see its internal structure, but you can build your own machine that produces exactly the same input-output behavior as the original one.
Once it’s built, you’ll have reason to speculate: the internal structure of the original machine might be quite similar to the one I’ve built.
This is precisely the engineering version of “Zang-Xiang”—and it’s also how the major frameworks discussed later in this book came into being: The thinking framework and the psychological framework emerged in exactly this way—by observing the external manifestations of human thought and psychology, and then designing an internal structure that can produce the same kinds of outward expressions.
The principle behind this line of thinking corresponds to the philosophical framework of "teleology": inferring structure from function—guessing "what kind of structure it needs" based on "what purpose it aims to achieve."
Design-based speculation and anatomical study each have their own roles: For things that can be disassembled, taking them apart is the most straightforward approach; for things that can’t be disassembled, are too expensive to take apart, or wouldn’t make sense even if taken apart (because they’re so complex that you’d still not understand them after disassembly), we rely on design to make educated guesses.
Modeling: Fitting high complexity with low complexity.
Putting the three paths together, they all point to the same thing: modeling.
Modeling is about using low-complexity models to approximate high-complexity reality. Reality is always more complex than any model—after all, a map is always simpler than the territory it represents; otherwise, the map would be useless.
Good modeling makes efforts in two directions:
Reduce the complexity of the model. The simpler the model, the easier it is to use. Concept vectors significantly lower complexity: a single framework involves only four genes. Moreover, if several frameworks map onto one another (as is precisely the case among the philosophical, cognitive, and psychological frameworks discussed later in this book), then you don't even need four genes—remembering just two or three will suffice.
Enhance the fit to reality. The degree of fit depends on repeated deliberation—carefully selecting the most critical dimensions.
So, when comparing two models, the question is: At the same model complexity, which one has a higher degree of fit to reality?
This standard is different from science. Scientific theories aim for truth; models aim for goodness.
The author repeatedly emphasizes: Concept vectors are not truths—they are merely models. Just like programming, they’re a tool for understanding the world, an algorithm for making sense of it. As for the modeling itself, “no matter how you build it, it’s all valid.”—I build my model this way because I’m confident it will help me think and yield better results. If you find it inconvenient or ineffective, feel free to create your own model and let’s compare which one works better for everyone.
In “General Systems Theory,” there’s a passage that puts it exceptionally well—a passage the author specifically excerpted in his notes: “If a model, though unable to solve a problem or capable of solving it only in a rough manner, repeatedly and independently emerges across diverse problems in various fields through its mathematical patterns or relational structures, then such a model will never become obsolete—it seems to have already revealed some deep-seated, universal law of nature and society.”
The concept vector aspires precisely to be such a “general model”: its four-quadrant structure repeatedly emerges across various fields—including personality, cognition, philosophy, society, economics, and medicine.
The subway route map is a deliberately misdrawn map.
“Modeling inevitably involves errors”—the most famous example is displayed in every subway station.
The actual subway lines twist and turn underground: here they loop around an old building, there they swerve to avoid a stretch of riverbed. Yet on the map in your hand, all lines are either perfectly horizontal and vertical or slanted at forty-five degrees—and station distances are drawn with painstaking precision. Not a single detail on the map matches the real geography. The cartographers deliberately got it wrong: they knew that what passengers really wanted wasn't geographic accuracy, but rather, “Where am I? Which line should I transfer to? And how many stops do I need to ride?” If you can answer these few questions clearly and concisely, then the map is a good map; but if it’s drawn with the same precision as a satellite image, it’ll actually become impossible to read.
This is a textbook example of “fitting high complexity with low complexity”: the error isn’t an accident—it’s something the designer and illustrator deliberately discarded, tossing away 90% of the geographic information in exchange for making the map instantly comprehensible to everyone in just three seconds.
Next time you hear the criticism that “this model isn’t accurate,” first take a moment to think about subway maps: Inaccuracy can sometimes be a failure of modeling, but other times it’s a badge of honor for the model. The key question is this: Has the information it discarded delayed or hindered what you’re trying to accomplish?
The Legitimacy of Modeling: From TRIZ to Traditional Chinese Medicine
Someone might ask: Is this kind of “as long as it works” modeling considered esoteric or mystical?
The author’s answer is to look at precedents. TRIZ, the innovation theory widely recognized in the engineering community, is itself a form of modeling: Traditional Chinese Medicine is a form of modeling; the Deming Cycle is a form of modeling; and the MECE principle (Mutually Exclusive, Collectively Exhaustive) from the Pyramid Principle is also a form of modeling.
None of these models has ever been subjected to a “falsification” test; the only way they’ve been tested is by seeing whether they work in practice.
Whether a model is valid or not, the ultimate criterion for testing it is certainly not falsification—after abstraction, a model is bound to differ from reality. It’s far too easy to “falsify” a model; all you need to do is find some minor detail that doesn’t quite match. The real test lies in this: Is it practical and useful? Are there even better models out there? That’s the real question.
It is effective. This effectiveness is probabilistic.
Whether traditional Chinese medicine modeling is valid depends on its therapeutic efficacy; whether concept vector modeling is valid depends on how much it helps you clarify confusion.
So what exactly is the difference between modeling and metaphysics?
The key difference lies in interpretability. Esoteric theories often make numerous assertions without ever clearly explaining how those assertions came about. To avoid falling into esoteric thinking, we should strive to explain thoroughly the entire process by which these assertions are formed.
Concept vectors also make numerous assertions, but each assertion can clearly explain its origin—what dimensions it draws upon, which steps were involved in the combination, and what correspondence principles were applied. By laying out the reasoning process openly for anyone to examine, we draw a clear line between modeling and mysticism.
Complexity: Why the Scientific Method Fails Here
The chapter began by stating, “Science isn’t enough.” Now we can offer a more theoretical explanation for this.
The reason was already discussed in Chapter 3: When a system reaches a certain scale, new emergent properties begin to appear at higher levels. As we move up the biological-social continuum, randomness and probabilistic factors become increasingly prevalent. Objects that are too complex can only be described at a coarse-grained level—qualitative rather than quantitative.
The author also offers a critique of complexity science itself: Complexity science has taken the wrong path because it is still trying to solve problems using scientific tools. On the contrary, artificial intelligence represents one of the right directions, since AI is “unrelated” to traditional science—it doesn’t rely on analytical solutions; instead, it relies on fitting and modeling. Even genetic algorithms, which originated in complexity science, have ultimately been categorized as belonging to the field of artificial intelligence.
Whether this judgment holds true is open to debate, but it highlights a trend of our times: in the face of complexity, “fitting” is increasingly replacing “analytical” approaches as the mainstream methodology. And abstract modeling is precisely the brain’s version of fitting.
Why didn’t he speak today?
You don’t need to learn the three guessing paths specifically—every day, they’re running right here in the office.
Today, my colleague Xiao Wang walked in without saying hello and immediately sat down, staring at the screen. You’ve already started guessing what’s going on in his mind.
The first guess draws on historical patterns: He does this every Monday—probably because he didn’t get enough rest over the weekend. This approach is grounded in history; it’s worked in the past, and most likely will work again in the future.
The second guess involves looking for similarities: After his previous proposal was rejected, he wore exactly the same expression—this is an analogy; similar situations point to similar causes.
The third approach to guessing is the most challenging: You put yourself in his shoes, immersing yourself completely in his situation—“If I’m weighed down by mortgage payments and my child has a fever, I wouldn’t even feel like saying a word when I walk through the door.” This is a form of design: It doesn’t pry into his heart directly, but instead creates within your own heart a version of him capable of exhibiting the same kind of behavior.
None of the three guessing methods is guaranteed to be correct—after all, he might just not have gotten enough sleep. But pay attention to the phenomenon itself: When confronted with an opaque black box that’s impossible to fully understand, people never simply stare blankly into space; rather, three different approaches spring into action simultaneously. What this chapter does is simply give names to the very same set of actions you’ve already been employing all along.
When confronted with an opaque object, humans have three paths of inference: drawing lessons from history, applying analogical transfer, and designing replication.
Three paths converge in modeling—using low-complexity models to approximate high-complexity reality, replacing “truthfulness” with “usability” as the criterion for evaluation, and clearly distinguishing “explaining the origins and context” from pseudoscience.
With this methodological toolkit in hand, we can finally introduce the protagonist of this book.
A one-sentence definition
After four chapters of buildup, the protagonist can finally make their debut.
Concept vectors: A method of defining a concept using several “binary dimensions.”
A concept is equivalent to a vector; each dimension of the vector, like an axis on a number line, has two poles—for example, black and white—this is precisely where the term “binary” originates. Multiple dimensions combine to form a vector, which the author refers to as a “gene.”
The formal name for this method today is “concept vectors,” a term inspired by the “word vectors” used in the field of artificial intelligence—but please note that concept vectors have nothing to do with artificial intelligence. Artificial-intelligence word vectors are automatically generated by algorithms from text data, whereas concept vectors are manually constructed word vectors, where humans define the values of each dimension through reasoning rather than statistical analysis. In its early days, this method had a more vivid nickname: “binary genes”—with dimensions serving as the “genes” of concepts. This earlier name has since been retired and is now used only as a historical term.
Let’s start with the simplest example. The concept of “urgent and important matters” has two dimensions: “urgency” and “importance.” Each dimension has two possible values, resulting in four quadrants: urgent and important, urgent but not important, not urgent but important, and neither urgent nor important—this is the well-known Four Quadrants Rule in the field of time management.
one
You may have been using it for years without realizing that its underlying structure is simply the combination of two concept vectors.
The most famous example: MBTI
Using a method similar to concept vectors to represent concepts, the most obvious example is the MBTI personality classification.
For example, the personality type INTP is composed of four letters: I, N, T, and P. The four binary pairs—I/E, N/S, T/F, and P/J—represent two polarities each. With four binary pairs, each having two possible values, there are a total of 2⁴ = 16 combinations, which perfectly correspond to the sixteen personality types.
Please note several subtle yet ingenious features of the MBTI notation system—each of which will be inherited by the conceptual vector.
First, there’s no need to assign individual names. If each of the sixteen personality types were given a name—like the Enneagram’s “The Perfectionist” or “The Helper”—no one would ever remember them. Instead, we can represent the sixteen concepts using combinations of four letters; just by looking at the letters, you’ll immediately know their meaning.
Second, the similarities are immediately obvious. INTPs and INTJs differ by just one dimension, and INFPs and ENFPs also differ by only one dimension—whose personalities are similar and where they differ is written all over their faces.
Third, it allows for identifying and filling gaps in knowledge. There are exactly sixteen distinct types—no more, no less—and there’s absolutely no risk of “missing a certain personality type that no one has ever studied.” Every single combination is bound to have someone assigned to describe it.
The MBTI is, in essence, a conceptual vector system with a “gene length of 4.” And this is precisely where its weakness lies: Why these four dimensions and not others? Are four dimensions sufficient? We’ll delve deeper into these questions when we discuss the relationship between MBTI and psychological frameworks (Part One).
An older instance: Bagua
If the MBTI is a modern instance of conceptual vectors, then the Eight Trigrams are its two-thousand-year-old counterpart.
The Eight Trigrams are composed of three lines, each of which can take only two values—yin or yang. This is a conceptual vector system with a "gene length of 3"; since 2³ = 8, we obtain the Eight Trigrams. If we double the gene length to 6, then 2⁶ = 64, yielding the Sixty-Four Hexagrams.
The author repeatedly emphasizes: Concept vectors represent a development of dualism, yet they are no longer confined to duality—they now take into account quaternary, octal, and even higher-dimensional structures.
People often talk a lot about duality—yin and yang, dialectics, and the interplay of opposing forces—but they rarely mention “genes”—yet genes are equally important. Indeed, genes represent an endless journey: the very path by which Taiji gives rise to the Two Principles, the Two Principles give rise to the Four Symbols, and the Four Symbols give rise to the Eight Trigrams.
“I believe genes represent a breakthrough from duality”—and the key to this breakthrough lies in the fact that dimensions can keep on extending indefinitely.
More isomorphic instances
Once you grasp the structure “concept = a combination of several binary dimensions,” you’ll find it’s everywhere:
●The periodic table: With n electron orbitals, each containing several electron slots, dozens of elements are formed. These elements then combine to create an even greater number of molecules. The chemical properties of an element are determined by its “coordinates” on the table—these coordinates are essentially its genetic code.
●Composite primary key of a database: In a table, a row is uniquely identified by the combined values of multiple columns; each column represents a genetic locus.
·1 Semantic web triples, Prolog, and Lisp: expressing concepts using structured fields.
In programming, classes and interfaces, as well as attributes and methods: object-oriented programming requires that each function be orthogonal and non-redundant with respect to other functions. An information node carries several attributes, and these attributes constitute the gene chain.
Four-quadrant diagrams: Various fields feature four-quadrant diagrams, with each quadrant representing a combination of two conceptual vectors.
These examples span mathematics, biology, computer science, linguistics, and writing; the author has summarized their commonality in a single formula: multiple fields jointly define an ID.
Concept vectors, the periodic table of elements, composite primary keys, and word vectors are all instances of this formula.
The three details of the definition
Detail 1: Each dimension is first defined qualitatively, and coefficients can be added afterward.
In general, dimensions are used only for qualitative descriptions—for example, black/white. If you want precision, you can add coefficients—for instance, grayscale. Concept vectors are, first and foremost, a qualitative tool.
Detail 2: Dimensions should be as orthogonal as possible to each other.
This requirement is borrowed from programming: Instead of coding a single large class, break it down into many small functions and achieve the same functionality by combining these small functions. Only when the dimensions do not overlap can the resulting conceptual space be used efficiently without waste.
Detail 3: The gene length can be arbitrarily long.
A concept can be continuously refined by describing it with a longer vector, where each dimension of the vector represents a specific attribute. The transition from the Eight Trigrams to the Sixty-Four Trigrams involves extending the vectors; Chapters Seven and Ten will specifically address the new challenges arising from this extension—such as sparse matrices, fine-grained genes, and novel genes.
Why it has “strict beauty”?
The author makes a remark: Conceptual vectors, like mathematics itself, possess a rigorous kind of beauty. The Pythagorean school in ancient Greece held that mathematics is perfect, while reality is imperfect.
This beauty is reflected in three aspects:
●Comprehensive. Given a specified gene length, all possible combinations are exhaustively covered—no omissions exist.
Concise. A large number of concepts are generated from a small combination of genes, significantly reducing the amount of material that needs to be memorized.
●Operable... Comparing the similarity between two concepts boils down to comparing how many dimensions the two vectors align—this is precisely the principle behind recommendation algorithms that suggest similar content to you.
Of course, the perfect ones are tools, not applications. Reality is always rougher than any model.
The author is fully aware of this: modeling inevitably involves errors, and concept vectors are no exception. Their value does not lie in perfectly mirroring reality, but rather in bringing order to the chaotic world of concepts—just as the value of a map does not reside in its exact size compared to the territory, but in its usability.
E-commerce websites have long since turned it into a filter.
The term "concept vector" may sound new, but actually, you’re using it every time you shop online.
Open the e-commerce app and search for “smartphone.” Immediately, a column of filtering options appears on the side of the page: brand, price range, memory capacity, body color, screen size... Each filter item represents a dimension, and beneath each dimension are listed the available values to choose from. If you check “256GB,” “white,” and “under 5,000,” you’re essentially assigning values to three dimensions of the conceptual vector. What’s displayed on the page then is precisely all the instances that match the combination of these dimensions you’ve selected.
The website has already experienced the benefits of this structure: You don’t have to give each phone model a separate name (Benefit 1); you can immediately see at a glance how the two models differ across several dimensions (Benefit 2); and if a particular combination doesn’t have any products listed, it’s most likely an unoccupied niche that no one has yet filled (Benefit 3).
E-commerce engineers haven't read this book, yet in order to manage millions of products, they independently arrived at the same underlying structure. That’s exactly how good structures are—different people, in different places, repeatedly discover them on their own. We’ll devote an entire chapter—Chapter 11—to discussing this point in greater detail.
Concept vector = a concept defined by several binary dimensions.
It’s not a new invention—MBTI, gossip, the periodic table, and database composite keys are all examples of it. It’s the first time this universal approach has been explicitly articulated. Once articulated, it has transformed from a technique confined to specific domains into a tool that can be actively applied to virtually any concept whatsoever.
In the next chapter, we’ll look at several basic relationships between concepts: mutual exclusivity, equivalence, and combination.
First, put up a warning sign.
Please take a look at this sentence: “China maintains friendly relations with Trump.”
This sentence poses no problem in everyday language—it’s perfectly understandable to everyone. However, from a conceptual-structural perspective, it’s a grammatically incorrect sentence: “China” is a concept referring to a country, while “Trump” is a concept referring to a person. The two concepts are not on the same level and cannot be placed side by side.
It’s perfectly fine to speak this way in everyday language. However, the human knowledge system is riddled with discussions that aren’t truly parallel—many debates drag on endlessly not because the two sides hold different views, but because what they’re actually talking about isn’t even on the same conceptual level to begin with.
Therefore, before examining the relationships among various factors, we must first ensure that they are presented side by side and on the same level—this is the foundation of rigorous argumentation. If the foundation is unstable, anything built on top will inevitably be skewed.
The primary purpose of concept vectors is to ensure the rigor of the objects you are discussing.
Mutual exclusion: the two poles within the same dimension.
The first basic relationship is mutual exclusivity. A mutually exclusive relationship is one of antonymy: big and small are mutually exclusive, inside and outside are mutually exclusive, and motion and stillness are mutually exclusive.
The essence of mutual exclusivity is that two concepts occupy opposite ends of the same dimension. On a given dimensional axis, the closer you are to one end, the farther you are from the other. Black and white are mutually exclusive because they represent the two ends of the grayscale axis.
Note: Mutual exclusion does not equate to “incompatibility.”
For any individual, duality is inherently present—each person simultaneously possesses both introverted and extroverted aspects. At the same time, duality is inherently biased; and with bias comes an infinite variety of types.
Mutual exclusion refers to the structure of dimensions, not the reality of individuals. This point will be further elaborated upon in the next chapter when we discuss “Dimensional Axes and Grayscale.”
Allelic: Neither includes the other.
The second basic relationship is that of equivalence. An equivalence relationship means that several concepts are on an equal footing, none of which includes any other:
All seven concepts are the same size; none is larger or smaller than the others.
The concepts all belong to the same field;
Typical example: In Jung’s eight cognitive functions, S/N and T/F are equivalent, and E/I are equivalent.
An important theorem is hidden within the concept of allelic relationships: If there are more than two alleles, they can be decomposed into multi-locus genes, each with a length greater than 1.
For example, S, N, T, and F are four equivalent concepts—four in number, which exceeds the scope that a single pair of antonyms can encompass—so MBTI breaks them down into multi-gene factors with a gene length of 2: S/N constitutes one dimension, while T/F forms another. Four concepts = two dimensions.
Binary.
The reverse is also true: When you encounter a set of concepts whose quantities happen to be powers of 2 (2, 4, 8, 16, 64...), you should be on alert—these are very likely the combined products of some conceptual vector system, and it’s worth looking into the underlying dimensions behind them.
The Eight Trigrams are 8, MBTI is 16, and the Sixty-Four Hexagrams are 64—none of these are coincidences.
Combination: From Gene to Concept
The third type of relationship is combination, which serves as the driving force behind the entire system:
Two binary combinations yield four concepts (gene length 2);
Three binary combinations yield eight concepts (gene length 3);
Once the combinational relationships among concepts are clarified, the number of nouns that need to be studied is greatly reduced—this marks the end of the “dependent origination” chain discussed in Chapter 1.
Now we can get a bit more technical: The number of conceptual combinations in the knowledge domain has been reduced from an exponential scale—where each combination had to be studied individually—to a linear scale, where only the dimensions themselves need to be examined. Instead of studying the eight hexagrams eight times, it suffices to study just three lines; instead of studying the 16 personality types 16 times, it’s enough to examine four dimensions.
Between the concepts of combination and dimension, a parent-child relationship is established: “Urgent and Important Matters” is the child class, while “Urgent” and “Important” are the two parent dimensions it inherits.
In object-oriented programming, if two classes share identical attributes and methods, you can abstract them into a parent class. The significance of concept vectors lies in the fact that, once you have these vectors, you can perform abstraction at different levels: the number of dimensions in which two concepts align determines how closely they are related and whether they can be grouped into the same parent class.
A对照 table: The division of labor between this book and logic
Logic also studies conceptual relationships, but the division of labor is different.
The conceptual relationships discussed in logic refer solely to the relations of identity and difference among concept extensions: identity, genus, species, intersection, and mutual exclusion—five types in total.
In addition to extensional relationships, the relationships among concepts also include causal-logical relationships, spatial relationships, temporal relationships, and intensional relationships. The author cautions: Once you’ve created a concept map, you must carefully clarify exactly what kind of relationship exists between the connected concepts—mistaking an extensional parallel relationship for a causal-logical one is the root cause of countless confused arguments.
The core relationship addressed in this book is “mutual exclusion within the same dimension, and combination across dimensions.” It overlaps with the extensional relationship (equivalent concepts are completely distinct and at the same level in terms of extension), yet their objectives differ: logic focuses on the validity of reasoning, whereas conceptual vectors focus on the definition and generation of concepts.
An operational mnemonic: Group the labels into one column.
Mutual exclusion and equivalence may sound abstract, but in practice they involve a very specific action:
Grouping labels into a single column of a table is a process of identifying mutually exclusive relationships among concepts; at the same time, the concepts in this column are equivalent in status.
Imagine you’ve collected 100 labels describing “human personality”: cheerful, steady, introverted, talkative, cautious, reclusive…
You group them into several columns of the table. Labels grouped into the same column should be mutually exclusive or nearly mutually exclusive—a person cannot simultaneously lean toward “outgoing” and “reserved,” at least not in the same context. Different columns, meanwhile, correspond to different dimensions.
After you’ve organized everything, you’ll obtain a matrix: the columns represent dimensions, and the rows represent the values for each dimension. This matrix is what we call a concept vector system.
Conversely, labels that can’t be neatly categorized also have value: they suggest either that there’s still an undiscovered dimension, or that the label itself is a “composite concept” and should be broken down into its constituent parts.
Waiting-discipline in a reimbursement form
Allelic, mutually exclusive, hierarchical—financial department’s expense reports, I’ve got it all figured out.
The reimbursement form has several columns labeled: Transportation, Accommodation, Dining, and Office Supplies. These columns represent separate dimensions—none of them overlaps or contains the others; each one covers its own distinct area. Just fill in your high-speed rail ticket under “Transportation” and your hotel invoice under “Accommodation,” and everything will be fine.
But if you accidentally categorize a high-speed rail ticket under “Office Supplies,” the expense report will immediately be returned by the finance department—this is an example of filling in the wrong dimension. There’s also a more subtle kind of error: writing “Beijing” in the account field. Beijing is a location, not an account—it’s simply not on the same hierarchical level as “Transportation.” This is precisely the kind of reimbursement form we mentioned at the beginning of this chapter: “China maintains friendly relations with Trump.”
Don’t underestimate this piece of paper. With hundreds of employees across the company and thousands of transactions processed each month, it’s these few neatly organized columns that keep everything running smoothly. “Filing concepts into the right columns”—while it may sound like an academic exercise—actually represents a fundamental discipline that every expense report adheres to on a daily basis.
There are three types of relationships among concepts that are relevant to this book: mutual exclusivity between poles at the same dimension, equivalence among concepts at the same level, and a generative relationship arising from the combination of dimensions.
When the number of alleles exceeds two, it can always be decomposed into multiple genes; when the number of alleles is precisely a power of 2, there’s often a hidden dimension lurking behind it.
Having mastered these three relationships, in the next chapter we’ll address a common misconception: Is “binary” too rigid a way of dividing the world? — No, because conceptual vectors actually go beyond just two dimensions.
The Most Common Misconceptions About "Duality"
Upon hearing the term “conceptual vectors,” many people’s first reaction is one of caution: The world is so complex, and humans are so nuanced—how could we possibly fit it all into a binary “black or white” framework? Isn’t that just oversimplification and labeling?
This criticism misses the mark. After all, concept vectors aren't actually “binary”—they’re rather a single-dimensional axis.
A number line spans any value between -10 and +10. When we say someone is “slightly static,” it doesn’t mean they’re 100% static and 0% dynamic—rather, it means their position on this number line could be around +6—that is, 70% static and 30% dynamic.
The binary represents merely the two endpoints of an axis, used to define direction; when truly describing a body, it’s the position along the axis that matters.
Here, the author specifically quotes the Dao De Jing: “The Dao gives birth to One; One gives birth to Two; Two gives birth to Three; Three gives birth to all things.” He then offers his own interpretation:
·Two—meaning duality, the two poles of the axis;
●Three—this is the mixed state where the middle 50% is dynamic and the other 50% is static;
The alignment system in the tabletop game "Dungeons & Dragons" provides a ready-made example: it features two axes—“Lawful–Neutral–Chaotic” and “Good–Neutral–Evil”—each of which specifically includes a “Neutral” position. Neutral isn’t simply “indefinable”; rather, it represents the rightful midpoint on each axis.
So the accurate statement is: Concept vectors describe dimensions, not switches. It’s just that when describing things qualitatively, saying “more inward” or “more outward” is more convenient than saying “coordinates along the inner-outer axis +3.7”—that’s why we commonly use polar terms in everyday language.
So why not use three poles or five poles? The author’s answer is straightforward: a ternary system doesn’t actually add an extra axis—it’s merely the zero point on the number line; the same goes for five-star ratings—no matter how many stars there are, they all still lie on the same number line. If you want greater precision, the way to achieve it isn’t by adding more poles, but by increasing the precision of the numbers themselves.
Someone asked: “Both elements are indispensable—how can we classify them based on that?”
This is a question that skeptics love to ask: Everyone has both introverted and extroverted traits; a person might be introverted at home but extroverted at dinner parties—so what’s the point of categorizing people in the first place?
The author’s response is divided into four layers, each one more detailed than the previous one.
Level 1: The classification is based on preferences, not on whether something is present or absent.
Introversion and extroversion—every person possesses both simultaneously. The categorization reflects preferences: whichever situation occupies a greater proportion of your time, whichever tends to dominate, that’s where you’ll be placed.
Second layer: Deviation from the average—even a deviation of just 1%—is meaningful as long as it’s stable.
The key isn't the magnitude; it's stability. A person who consistently remains at 51% introverted and 49% extroverted is one type of person, whereas someone who fluctuates dramatically between 20% and 80% is entirely different.
Third level: Add numbers if you find it too rough.
If some people don’t agree with the simple binary classification, we can add numbers to make it more precise—for example, appending four digits to a four-letter code representing a specific type, where each digit indicates the degree of bias along a particular dimension. After all, concept vectors inherently view number lines and dimensions as more precise; letters are merely shorthand representations.
However, the author himself prefers to stick to the two extremes, for a very practical reason: it’s already difficult enough to reach consensus on just two values. The more precise the numbers become, the more arguments arise—leading to endless debates over the exact magnitude of any given number. Binary division is meant to simplify things and conserve brainpower. Only computers, with their abundant computational power, can afford to work with so many digits.
Fourth level: Acknowledge probabilistic nature.
Some people argue that “a certain MBTI type does not necessarily imply the presence of certain personality traits”—and indeed, this is not a necessary connection; it’s merely a matter of probability. Some types carry a higher probability, while others have a lower one; in fact, for some types, the probability is less than 50%. In such cases, this claim should be considered incorrect.
The attitude toward these probabilistic assertions is as follows: Based on one’s life experience, believe in those points that feel meaningful and disregard those that don’t; accept points that can be explained by causal reasoning, while ignoring those that remain inexplicable; and even completely overlook specific characteristics of certain types, focusing solely on the fact that such categorizations exist in the first place.
The same attitude applies to questionnaires. There’s no inherent connection between the questionnaire and its results. The questionnaire simplifies the issues to make them feasible, while also introducing greater uncertainty and less rigor. Those who know their own minds can simply assign numbers to themselves along the dimension axes; for those who aren’t sure, it’s perfectly fine to refer to the questionnaire for guidance.
Gene Lengthening: Subdivision and New Genes
Another misconception about concept vectors is: “With just four dimensions, that’s too few.”
In fact, the length of a gene is open-ended and can keep getting longer. There are two distinctly different scenarios for lengthening, which must be treated separately.
Scenario 1: Subdivision —— A more detailed gene.
Within the scope of the older framework, continue to further subdivide the original dimensions. For example, personality could be expanded beyond the four dimensions of MBTI to include a fifth dimension; the sixteen types could be further subdivided into thirty-two or even sixty-four types. The sixty-four hexagrams are simply an extended chain of the eight trigrams.
Scenario 2: New Gene—Something New Has Emerged.
Whenever a new technology or new phenomenon emerges, it invariably leads to an exhaustive exploration and combination with existing categories. For example, consider the evolution of information media: from stories passed down orally, to theatrical performances, to radio broadcasts, to television and movies, to video games—each new generation of technology gives rise to entirely new categorical dimensions that simply didn't exist in earlier eras.
Biological evolution works the same way: first nerves appear, then the central nervous system, then the brain, then the prefrontal cortex, and finally language—each new gene that emerges divides organisms into two distinct categories.
The method for distinguishing between the two situations is simple: just look at whether the instance is old or new. A refined breakdown of an older framework is a refinement based on an established, well-established instance; only when a new instance emerges will new genes appear.
That’s why studying old genes isn’t pointless: after changes occur, the old genes don’t disappear—rather, new genes are simply extensions built upon the foundation of the old ones.
As genes grow longer, the world becomes a sparse matrix.
Lengthening comes at a cost. If genes are relatively long, not every possible gene combination will have an actual instance in reality.
Mathematically speaking, it’s quite clear: with a gene length of 4, there are 16 possible combinations; in reality, we can find numerous living individuals who embody each of these 16 personality types. However, when the gene length increases to 12, the number of possible combinations skyrockets to 4,096—and many of these combinations may never have existed at all. Those combinations that lack any real-world examples might have existed in the past or the future, or they may simply have very little chance of occurring under current spatiotemporal conditions.
Biological genes are sparse matrices: not every possible base combination is explored. The authors are very candid about this:
If I only look for a handful of key genes—just the few that matter—then every possible outcome from my exhaustive search will generally exist. The classification can keep going deeper and deeper; all I’d need to do is lengthen the gene chain. It’s just that my brain isn’t up to it anymore—I simply don’t have the energy left to keep extending the chain.
Here’s an honest trade-off: short gene systems (three or four dimensions) have few combinations, abundant instances, and are easy to remember and use—but they’re coarse. Long gene systems, on the other hand, are highly detailed, but their combinatorial explosion leads to sparse instances that even the human brain can’t fully grasp.
Artificial intelligence doesn't have this limitation—today's AI models represent a "car" as a vector with hundreds of dimensions. Thus, the author predicts a division of labor between humans and AI: by adding more dimensions and shifting from binary representations to multi-valued ones, AI will be able to encode all knowledge; whereas the concept vectors used by the human brain itself can remain in the single-digit dimension range.
Why is metaphysics prone to making mistakes on this issue?
Metaphysics often employs combinations of vague concepts linked by extremely long chains, making it impossible to specify either the dimensions or the specific values involved. The resulting combination space is so vast that it cannot be empirically tested. Concept vectors, by contrast, take the opposite approach: they aim for few, well-defined dimensions, with mutually exclusive and clearly delineated poles for each dimension, and their combinations must be countable.
A conceptual system—if you can’t even count how many combinations it has, then it’s probably already slipping into the realm of metaphysics.
A principle that electric fans have long understood
“The idea that ‘binary isn’t a switch, it’s a number line’—my home’s electric fan has understood that long ago.”
The buttons on old-fashioned electric fans were marked with three settings: 1, 2, and 3. The two ends of the axis represented “full wind” and “no wind,” yet no one ever used the fan exclusively at either end—during midsummer, they’d set it to level 3; in early spring, they’d choose level 1; and most of the time, they’d stop somewhere in between. Later, air-conditioning fans with stepless speed control came along, and the button was simply replaced by a sliding lever: from minimum to maximum, you could stop it wherever you liked.
The same goes for people. When we say someone is “introverted,” it doesn’t mean they’re 100% introverted—rather, it means they’ve consistently stayed at a particular point along this axis, much like how your home fan is usually set to speed level two. Occasionally, when guests arrive, they’ll temporarily switch it up to speed level four and liven things up a bit; but as soon as the guests leave, the needle snaps right back down to speed level two.
The classification is based on “which quadrant you consistently land in.” So don’t argue with people about whether “you’re introverted or extroverted”—this question is like asking a fan, “Do you actually have wind or not?” The answer will always lie on the axis itself, never at either endpoint.
Binary isn't a switch—it's a number axis. Classification divides stable preferences rather than simply whether something is present or absent. Genes can be lengthened, and this lengthening falls into two categories: “subdivision” and “new genes.” Lengthening leads to combinatorial explosion, which is why reality becomes a sparse matrix—human brains are well-suited to short genes, while AI can make good use of long genes.
Having understood the elasticity of dimensions, the next practical question is: How should we name these dimensions—these concepts?
Why naming is a big issue
Once a concept is created, it must be put to use; and to be put to use, it needs a name.
And names are the biggest source of misunderstanding.
When you say “introverted,” Zhang San’s understanding of introversion may not be the same as yours; when you say “sensitive,” Li Si might picture tears, while Wang Wu might think of intuition... Polysemy—having one word with multiple meanings—and homonymy—having one meaning expressed by multiple words—are the everyday realities of human knowledge’s inherent ambiguity.
The author sighs: Names really are hard to come by—better just use coding instead.
Four naming schemes, from worst to best
Option 1: Everyday vocabulary.
The advantage is that it’s easy to understand; the disadvantage is that it’s riddled with ambiguity. Everyday vocabulary is like an old wrench—everywhere it slips and slides.
Option 2: Proper nouns.
If binary names are prone to misunderstanding, use proper nouns—just like the Bagua’s “Qian” and “Kun”—first creating a term that has no everyday meaning at all, and then providing an explanation and definition for it.
The advantage is that the terminology is clear and unambiguous; the disadvantage is that it comes with a high learning cost and can easily slip into “jargon.” The author imposes strict self-discipline on term creation: opposing the excessive invention of new terms, striving instead to use language that everyone can understand—rather than resorting to jargon or clever wordplay.
Option 3: Define using a set of instances.
Rather than attempting to define a concept in a single sentence, we provide a collection of examples—examples that share the common feature of embodying the concept’s definition. This is the most precise method of definition, but it tends to be verbose.
Option Four: Coding.
The most precise approach is to avoid naming altogether and rely solely on the encoding of conceptual vectors—just like with MBTI. The four letters INTP have no everyday meaning and cannot be misinterpreted; anyone familiar with the rules will recognize the structure at a glance when they see the code.
The author also offered a semi-joking yet very serious mnemonic device, borrowing the notation from the programming language Eve:
<#Psychological framework a:1 b:1 c:0 d:0>
The first half indicates that the concept belongs to a psychological framework, while the second half specifies the values along four dimensions. A concept = the system to which it belongs + the dimension vector. This is the notation for “defining an ID jointly across multiple fields”—machine-readable and不易 causing human misunderstanding.
The reality behind names: One thing can have multiple aliases.
“Actually, the thing is just that thing—only it can go by different aliases and be described in various ways.” Though this statement may seem simple, it’s a powerful remedy for the “battle over nouns.”
A significant portion of the debates in the history of human knowledge has been about terminology: the same thing was called “qi” by the ancients and “energy” by moderns; the same thing was referred to as “ren” by the Confucians and “compassion” by Buddhists.
From a coding perspective, nouns are merely pointers—only the memory object that the pointer refers to is the actual entity. When two people argue about nouns, rather than getting into a debate over the nouns themselves, each could simply state the set of instances their respective pointers are pointing to and see whether these sets represent the same object. If they’re the same, then it’s just a matter of translation; no need to argue further. If they’re not the same, then there’s still time to keep arguing.
In this book, the same method is referred to by different names in various places: concept vector, concept vector, and “Divide into Four” (an earlier name). This is not confusion—it’s a deliberate demonstration: the name is merely the outer shell, while the vector structure represents the essence. Readers should practice early on “seeing the structure through the nouns.”
Named Meta-Rule: Parent Node Definition Method
How should that new concept be defined? The author introduces his own definition method, which can be called a “structured definition”:
Define using the parent node: For example, “My non-parent node is ‘Emotional Control’”—the parent-node definition method is essentially the same as the gene-definition method; specify in which dimensions it takes which values.
Defined by child nodes: “My non-own child nodes are sensations, desires, emotions, and feelings—list the instances under each of these.”
Defined by the relationships among model elements: “The ‘non-self’ is the pathway through which the ‘self’ acts upon itself.” – State its position and function within the overall structure.
None of the three definitions relies on isolated noun definitions; instead, they anchor the concepts within a structural framework.
The meaning of a word is defined by the context in which it appears, rather than by borrowing someone else’s definition from elsewhere. The reason this particular word was chosen is simply that it comes closest to conveying the intended meaning.
This is also the right approach to reading this book—and indeed, to reading any conceptual work: Don’t dwell on individual terms; instead, focus on the structure.
Why does the hospital call-number machine announce the numbers?
As soon as the announcement sounded in the outpatient lobby: “Patient A137, please go to Internal Medicine Clinic 3 for your appointment.”
Why not just call out names? Because there might be three people named Zhang Wei sitting in the waiting area. Names are everyday words—collisions are common. But a number is a one-time code, assigned uniquely to each patient, ensuring there’s absolutely no room for misunderstanding. Take a closer look at this number’s structure: the letter prefix indicates the department, the digits represent the sequential order, and the number at the end specifies the examination room—this single code encapsulates all the necessary dimensions: “which department, which patient, and which room to go to.”
This is Scheme Four discussed in this chapter: the most precise form of naming is coding. The queuing system doesn't rely on everyday descriptions like “Zhang Wei—the one wearing the gray coat”—nor does it bother crafting specialized proper names. Instead, it simply assigns each person an INTP-style code.
Of course, hospitals do use names—when a doctor softly calls out “Auntie Zhang” at the bedside, it’s not “A137.” Coding is all about avoiding errors; names are about keeping things human and personal. Ultimately, when deciding between these four naming approaches, it comes down to that old adage: Consider the context and ask whether it works well in practice.
Naming is the area of conceptual systems where problems most easily arise.
From everyday vocabulary to proper nouns, then to collections of examples, and finally to coding, the level of precision increases step by step. The best form of naming is coding, because coding directly reveals the structure. As for a reliable way to define a concept, it’s better to anchor it within its parent nodes, child nodes, and structural relationships, rather than getting caught up in mere nouns.
Up to this point, we’ve finished discussing the rule-based part of the method. In the next chapter, we’ll move on to the value-based part: What exactly can this method be used for?
The four real-world problems it aims to solve
Before discussing its applications, let’s first take a look at the problems it addresses. The author lists four:
Many fields lack a complete picture. It’s like trying to piece together a map from several scattered sheets of paper—each expert holds one sheet and believes that their own piece represents the whole world.
Scientology. By rejecting other thinking tools, it leads to the abandonment of many questions that science cannot address.
The concept is described only in words, without a strict definition. It remains at the fuzzy stage of natural language.
The concept of non-parallelism is discussed in a confused manner—such as the hierarchical confusion exemplified by statements like “China maintains friendly relations with Trump.”
For these four issues, conceptual vectors have six signature strengths.
Skill 1: Identify and Fill Gaps
Concept vectors are both a map of knowledge and a periodic table of knowledge elements.
The power of the periodic table lies not in the elements already known, but in the empty spaces: When arranging the elements, Mendeleev noticed several vacant slots and predicted the existence of elements yet to be discovered, along with their properties—predictions that were later confirmed one by one.
The same holds true for conceptual vectors: When you organize concepts from a particular field into a matrix of “several dimensions × corresponding values,” the gaps themselves will speak volumes—certain dimension combinations that should exist in reality yet have never been studied before represent knowledge gaps waiting to be filled.
The book includes a beautiful example at the end. It is often said that humans possess both “naturalness” and “sociality”—a point already covered in textbooks. After completing the concept vectors, the author points out: focusing solely on these two aspects is incomplete—humans also possess “interest-basedness” and “idealism.” These four dimensions correspond precisely to the four pillars of the philosophical framework: neither more nor less.
Whether this completion is correct is open to debate among readers; however, the very fact that “missing items can be identified” demonstrates the tool’s power.
Skill 2: Deduplication
De-duplication is about realizing, “You express your thoughts, I express mine—actually, we’re both saying the same thing.”
The author makes a bold assertion: After removing duplicates, one will discover that the vast body of human knowledge is actually not as extensive as it seems—much of the content simply reiterates the same ideas in tens of millions of different ways.
In every culture, every discipline, and every era, people are using their own dialects to convey a few identical truths. Once we eliminate redundancy, we break down barriers and cease to be isolated islands. Just like languages across nations and programming languages of all kinds—when we identify commonalities and focus our research on them rather than each group talking past one another about the same thing, that’s simply a waste of effort.
So what’s left after removing the duplicates? The author’s answer is this book series itself: the comprehensive framework of Volume Six is precisely the streamlined checklist he laid out. The four quadrants of the philosophical framework cast shadows across numerous disciplines, compressing the knowledge of each discipline into a single, concise structure.
The method for deduplication is to compare conceptual genes: If two concepts have identical values across all dimensions, they are simply two different names for the same concept (a synonym relationship). If they match only in some dimensions, their degree of similarity becomes quantitatively apparent.
The author even suggests that instead of verification, we should focus on deduplication. It’s difficult to verify a claim, but when we discover that “this very claim was already made by ancient thinkers—and has been echoed in other disciplines as well”—we can achieve a consensus that transcends time and space. For example, by applying “group theory” to de-duplicate the Five Elements in Traditional Chinese Medicine, TCM suddenly becomes much easier to understand. And that, precisely, is how consensus is reached.
The current state of psychology serves as a cautionary tale: psychology boasts dozens, even hundreds, of different subfields, each with its own distinct perspective and conclusions—conclusions that are merely preliminary summaries of experience, designed to obscure rather than reveal the underlying truth.
Between appearance and essence, what’s needed is precisely a process of deduplication—two dogs have similar genes, one black and the other white; there’s no need to study them separately.
Skill 3: Classification
Classification is not about labeling; rather, it’s about the right to treat things differently.
The author puts it very bluntly: The idea that a species has uniform structure and uniform behavior, and therefore should be treated with uniform methods—that’s simply foolish.
Treating all children with a uniform, one-size-fits-all approach in education yields meager results. When it comes to medical treatment, diagnosis is crucial—and diagnosis itself is a form of categorization. If the diagnosis is wrong, no matter how expensive the medicine may be, it will be ineffective. When constructing an argument, one must first clarify concepts, and clarifying concepts is inseparable from eliminating redundancy through categorization; only by ensuring that concepts are orthogonal can we guarantee their clarity and precision.
The byproducts of categorization are also highly valuable: Categorization reveals mutually exclusive and equivalent relationships among concepts; it helps preserve individuality—allowing us to recognize those who are different from ourselves and to find our own kind; and it enables us to grasp the principle of moderation—if we know we’re at +6 on the axis, we’ll also be aware that there are people at -6 on the same axis.
Of course, classification has its limitations. As the author candidly acknowledges: focusing on commonalities can easily lead to overlooking individual differences, and concentrating on the big picture may cause one to neglect crucial details. The number of biological genes runs into the hundreds of millions, whereas classification can only encompass a limited number—in the single digits. The gap is simply too vast, which is why classification can serve only as an auxiliary tool for thinking. For truly nuanced considerations, we must—just as in the maritime legal system—rely on case-by-case analysis and ad hoc deliberations.
Classification satisfies humanity’s desire for a comprehensive overview—having a rough map is better than groping around in the dark.
Skill Four: Completing and Avoiding Extremes
The structure of conceptual vectors compels one to “look at the other side:”
Those who speak only of the “individual” are reminded that there is also the “whole”;
·Those who focus solely on science are reminded that there are also domains of philosophy and theology to consider.
Many problems arise because we’ve gone too far down one path. The antidote isn’t at the end of the road—it’s on the forgotten other side.
The author uses the metaphor of a T-square: both specialization and versatility are indispensable. Choose one path to delve deeply into, but don't become overly specialized—ensure you remain adaptable, so that you can adjust yourself when the environment changes.
In other words: When judging anything, don't rely on just one single dimension—don't use a single ruler to measure everything; it's a pity if you do that.
Skill Five: Analogical Quantification
Two things “kind of resemble each other”—but just how much do they really resemble each other?
Concept vectors can transform vague intuitive feelings into computable structures: objects with shared genes can be analogized, and the more dimensions they share, the deeper the degree of analogy.
The recommendation algorithm is a daily-life version of this principle: information nodes carry several attributes (gene chains), and similarities are determined by comparing these gene chains. The system then recommends similar information to you.
Skill Six: Dealing with Complexity
Faced with complex problems that reductionism cannot solve, the author’s approach is to view them from an indivisible holistic perspective and tackle them through abstraction—this is what we call the concept vector.
Here’s a point that’s easily misunderstood and needs to be clarified: Concept vectors also involve “decomposition,” but this is not reductionism. In reductionism, decomposition means taking something apart—breaking the whole into parts that can exist independently of the whole. Today’s science is largely reductionist in its approach, and its primary method is precisely this kind of decomposition. However, the decomposition involved in concept vectors is quite the opposite: Each binary division represents a partitioning of the whole. The “inside/outside” distinction isn’t about slicing a person into two separate pieces; rather, it’s about dividing the entire person along a single dimension into two polarities. When these two poles are combined, they still constitute the original whole. Thus, the stance of concept vectors is holism: They can be decomposed, yet they don’t fall apart.
The author argues that this is an important feature of conceptual vectors—and also their significant contribution. Note that this is not the same thing as the holism emphasized in Part Four (the philosophical framework): there, the focus is on viewing the complex aspects of one’s worldview in a unified, holistic manner; here, the emphasis is on a method—one that breaks things down yet keeps them from falling apart.
As for the term “appearance theory” he used in his early years, this explanation now gives it a precise place: Some wholes cannot be disassembled for the time being; we can only stand outside and describe their manifestations—this is an inevitable, downgraded version, not a stance itself.
It is based on the philosophical viewpoint of “indivisibility”: duality is indivisible and mutually interdependent—without unity there is no dispersion, and without stillness there is no motion.
This is a top-down, holistic way of thinking—exactly the same methodological approach as traditional Chinese medicine.
A reminder: Don't use a sword to kill a chicken.
The author specifically emphasizes: Working with concept vectors is a very challenging task.
It’s difficult to identify the most important concept in any given field, and it’s even harder to find concepts with exactly equal gene lengths. It might take years just to develop a conceptual vector for a particular field.
Concept vectors are high-level, generalized statements that are very useful for identifying the most critical aspects and analyzing positive and negative feedback. However, when dealing with specific problems, it’s best not to rely on concept vectors—because the analytical costs become too high.
The ox-knife is used for butchering oxen, not for peeling apples.
The four-character mnemonic before heading out is a checklist for spotting gaps.
Many people silently recite a mantra before going out: “Reach out and ask for money.”
Reach—ID card, hand—mobile phone, take—key, money—wallet. These four words form a homophone that neatly covers all four essential items. The reason this mnemonic works so well isn’t just because of its rhyme; it’s precisely because it serves as a “4-length gene checklist”: identity credentials, communication device, means of transportation, and payment tool—each of these four dimensions handles one specific task, neither too many nor too few. People who often forget things when they go out usually miss one item from these four dimensions; as soon as you recite the mnemonic, the missing piece will reveal itself.
In recent years, someone has added a fifth item to the mnemonic: power banks. Under the category of communication devices, a new and critical factor—“battery level”—has emerged. This is precisely what Chapter 7 refers to as the “new gene”: when new things appear, the old checklist naturally grows longer in response.
Identifying gaps, filling in the blanks, and extending the content—these three signature skills of this chapter are all encapsulated in a simple folk rhyme. After all, a good tool never distinguishes between the court and the streets.
Concept vectors address four challenges—lack of a comprehensive overview, scientific methodology issues, conceptual ambiguity, and hierarchical confusion—by relying on six key capabilities: identifying gaps, eliminating duplicates, categorizing, completing missing information, quantifying analogies, and handling complexity.
It’s a high-level tool that doesn’t address specific issues. In the next chapter, we’ll examine the objections—what kinds of challenges this approach might face, and how the author responds to them.
The value of a set of methods is, in part, reflected in the skepticism they encounter. The author’s notes specifically include entries labeled “Doubt” and “Q&A”—he anticipated that people would doubt his ideas and took those doubts seriously. All the following answers are drawn directly from the original notes; the editor has only made them more coherent.
Question 1: “I don’t believe in personality typing.”
Question: Many people simply don't believe that personalities can be categorized.
Response: Many people who don’t believe in personality types actually just have different ways of thinking—some are naturally good at reading maps, while others are naturally good at memorizing street names. It’s inherently difficult to expect someone who’s good at memorizing street names to trust maps.
For example, if someone pays close attention to details, they’ll tend to resist “broad categorizations”; others, meanwhile, aren’t interested in either categories or individual cases—they’re solely concerned with cause and effect.
Not believing in categorization itself is a type of thinking—precisely the kind of phenomenon that categorization can explain.
Question 2: “The type description is not necessarily valid.”
Question: A certain MBTI type does not necessarily possess those personality traits.
Response: Indeed, it’s not necessarily the case—it’s merely a matter of probability. Some probabilities are high, others are low, and if some probabilities fall below 50%, then that statement is incorrect and can be deleted.
The method of use is as follows: Based on life experience, trust the points that feel meaningful and disregard those that don’t; accept what can be explained by causal reasoning, and ignore what cannot be explained; even focus solely on the categories themselves, without paying attention to specific details.
This isn’t called being lax—it’s called the correct way to use probabilistic knowledge.
Follow-up question: Probabilistic things are meaningless—focus on deterministic laws instead!
To respond again: While the law of necessity is indeed crucial, how many such laws actually exist in the world? Can you be 100% certain that a particular politician will develop dementia? Have you broken him down and measured him atom by atom? To focus solely on the law of necessity while ignoring the law of large numbers—the statistical law governing probabilities—is to limit one’s thinking too narrowly.
Question 3: “The questionnaire isn’t accurate.”
Question: There is no inherent connection between the questionnaire and the results; moreover, many people get different results each time they take the test.
Response: I agree with half of that. Surveys do indeed simplify questions in exchange for feasibility, while also introducing greater uncertainty.
So the author simply skipped designing questionnaire-style test questions—instead, they just asked people directly to identify their own type. In this way, the question boils down to: “Do you understand the definition of the type?” Then, detailed explanations of what each type means are provided to help respondents assess themselves. This approach is better than introducing a bunch of test questions that would bring about greater uncertainty.
(Later, the author did conduct a small number of type-testing questions along with the “framework questionnaire”; see the supplementary notes. However, these were intended to elicit self-observation rather than serve as scientific measurements.)
Question 4: “Why is your categorization correct?”
Question: What if the conceptual vector framework proposed by a certain individual fails to achieve universal consensus?
Response: Everyone proposes their own classification, and the classifications are compared against each other to determine the winner—the market is open.
In addition, the concept vector also suggests that each binary pair can be broken down separately—for example, by isolating the “closeness” dimension alone for categorization and examining whether the resulting conceptual divisions under this category are appropriate, thereby achieving partial consensus.
If a comprehensive consensus cannot be reached, settle for a partial consensus instead. If no one is right or wrong, just see which option works better.
Question 5: “Is this esoteric?”
Response: Esotericism often makes numerous assertions without ever clearly explaining how these assertions came about. To avoid esotericism, one should strive to clarify the entire process behind the formation of any assertion.
Concept vectors also make numerous assertions, but they can clearly explain how these assertions come about—how the dimensions are chosen, how concepts are combined, and whether the corresponding relationships are definitional or probabilistic—all of it laid out explicitly.
Moreover, the interpretability of metaphysics now faces new demands: even artificial intelligence is exploring the issue of interpretability—so metaphysics should even more so focus on studying interpretability.
Question 6: “Is the correspondence between these concepts reliable?”
Question: You say that Confucianism corresponds to this, and blood type corresponds to that—how credible is such a correspondence?
Response: Here, it’s essential to distinguish between two terms: definition and correspondence.
The definition is 100% certain (the structural relationships within the concept vector); the correspondence, on the other hand, is a lateral connection across different domains and is probabilistic—in fact, the author frankly admits that in the worst-case scenario, its accuracy might be as low as 55%. (He later clarified: The 55% was just an example—meaning “higher than random.” If random chance were 25%, then 26% or 27% would already no longer be considered random.)
Things that are computable and deducible have necessary connections; whereas connections like “a certain MBTI type exhibits certain behaviors” belong to the universal interconnectedness and mutual influence among things. One event is affected by numerous factors, each with varying weights, and its causal chain is complex and unreliable. Among the factors ultimately determining the outcome, those that are more personal and temporally and spatially closer tend to exert greater influence.
Therefore, the correct way to use it is as a hint—not as a proof. It suggests to you that “these two fields might be isomorphic,” and thus worth investigating—but it doesn’t guarantee the conclusion.
The author’s exact words were: “The correspondence itself is not rigorous… You’ll have to judge this correspondence relationship for yourself—correspond however works best for you. It’ll need repeated revisions. Right now, this is just a draft, and everyone is free to hold their own different views.”
If there’s a 100% correspondence, then it’s no longer a separate dimension.
Question 7: “Categorization can lead to stereotyping.”
Response: Quite the opposite.
The author himself has emphasized most often the limitations of classification (see Chapter 9): classifications are limited to single-digit categories, while individuals are infinite; classification is merely an auxiliary tool for thinking—fine-grained judgment must be made on a case-by-case basis.
Moreover, humans are a superposition of “multiple personalities” (discussed in detail in Part Six): In addition to the innate personality, there are also multi-layered personalities shaped by influences from parents, profession, education, and socio-cultural factors—much like the superposition of waves.
Categorization helps you see the structure clearly—it doesn’t confine you to rigid boxes.
Probability Class in the Comment Section
“The best practice for correctly using probabilistic knowledge is right here—in the comment sections of online shopping reviews.”
A product has a rating of 4.9 out of 5, based on 10,000 reviews. If you keep scrolling down, you’ll still find negative reviews like, “It broke after just three days.” Does that mean you shouldn’t buy this item? Not at all—dozens of negative reviews out of 10,000 people reflect probability, not a verdict. On the flip side, even in a product with a rating of 3.8, some buyers take the time to write earnestly, “This is worth every penny!” Probability never guarantees any single outcome; it merely tells you the shape of the distribution.
Experts in the comment section all have their own method of reading reviews: They pay attention to which issues receive the most negative feedback—whether it’s about product quality or damage during shipping—and they check whether positive reviews are just generic templates or actually include specific details. They also note how long after the initial review the follow-up comments were posted. This very approach—the one described in this chapter—is precisely about trusting points that can be explained by cause-and-effect relationships while dismissing those that don’t make sense. As for any single review, they neither fully believe nor completely disbelieve it.
After acknowledging that “the world is probabilistic,” one actually becomes more at peace: a score of 4.9 doesn’t need to be mythologized, and a score of 3.8 doesn’t have to be outright dismissed. This holds true for comment sections, for personality classifications, and for most of the knowledge in the world.
The main criticisms of concept vectors—namely, disbelief in categorization, the non-necessity of types, inaccurate questionnaires, difficulty reaching consensus, suspected esotericism, loose correspondences, and the tendency to take death too seriously—all were anticipated by the author, who provided responses to each of them in advance.
The overarching key to responding is encapsulated in a single sentence: Conceptual vectors are about modeling, not truth; the criterion for evaluation is whether they’re useful, not whether they’re true. Once we embrace this meta-rule, all doubts will find their proper place. In the next chapter, we’ll look back at the intellectual forebears of this approach.
If a set of methods cannot explain where they come from, they inevitably risk sounding mystical or esoteric.
The author’s notes include a special “Sources” entry that explicitly acknowledges every intellectual ancestor of the concept vectors. This very list itself serves as a demonstration of “deduplication”: you’ll see that across various fields—both ancient and modern, Eastern and Western—many disciplines have independently arrived at the same underlying structure.
Ancestral List
I. Existing four-quadrant diagrams in various fields.
The four-quadrant time management model, the Boston Matrix, various four-category personality classifications... The four-quadrant—comprising two binary dimensions—is the most common manifestation of conceptual vectors.
II. MBTI: A concept represented by a combination of four letters.
This is a mature template for “coding naming” and “multiple genes.”
III. The Eight Trigrams and the Sixty-Four Hexagrams.
The yin and yang lines form the eight trigrams of three lines each and the 64 hexagrams of six lines each—more than two thousand years ago, ancient Chinese were already using a binary system with gene lengths of 3 and 6.
IV. Word Vectors in Natural Language Processing.
Map a word onto a vector in several dimensions. The key difference is: word vectors are automatically determined based on statistical analysis of the text, whereas concept vectors manually—and through deliberate reasoning—determine the “distance” between words.
The author jokingly refers to himself as having “also worked on some projects that use human effort to mimic artificial intelligence”—for instance, concept vectors are人工 word vectors; his other project, “Speaking Like Programming,” is an artificially structured language; and his personal Weibo account is an artificially labeled subscription—all of which serve as对照 experiments demonstrating how “the human brain can do what AI does.”
V. Dialectics and the Dualism of Zoroastrianism.
The unity of opposites is an ancient global idea. Some even claim that Plato’s classical dualism is identical to the concept of vectoriality—so if that’s the case, why bother revisiting it? This question will be answered at the end of this chapter.
VI. Programming: Classes, Interfaces, Properties, and Methods.
Object-oriented programming requires that each function be orthogonal and non-redundant with respect to other functions; object databases feature the concept of attributes. From a computer science perspective, the author summarizes the concept vector as follows: rather than encoding a single large class, we break it down into many small functions, and achieve the same functionality by combining these small functions together.
In object-oriented programming, each person abstracts a different class model, leading to model complexity. Programming addresses this issue through “interfaces”—a single class model can implement multiple interfaces. The same principle applies to academic modeling: everyone builds models that differ from one another; in fact, these models should also be decomposed into multiple interfaces.
The author refers to this kind of decomposition as “genetic dissection of concepts”—just like how, although there are many biological species, once we conduct genetic analysis, the origins and relationships among them—and their degree of similarity—become crystal clear.
Hierarchical Discipline in Writing.
The writing requirement stipulates that each heading at a given level carries roughly equal weight—this is precisely how the principle of equivalence is reflected in writing.
Table.
A table defines a row by the values in multiple columns—each column represents a gene locus.
The computer tradition of knowledge representation.
The Periodic Table of Elements.
The ultimate model for identifying and filling gaps.
The author says, “There are no other origins.”
This candid statement is crucial: Concept vectors weren't invented out of thin air—they are instead a unifying method distilled from isomorphic practices that already existed across ten different fields.
The name “gene” itself has an origin. In 2017, the author noted in his blog: Inspired by the sixteen personality types—each type is underpinned by four letters—he dubbed this structure “the gene behind the concept.” For example, the concept of “hobby” is underpinned by four genes: introversion, passivity, rationality, and stubbornness. Concepts can be broken down; genes can be combined—and thus, the name was established.
The relationship with mathematics
The author speculates that set theory and category theory in mathematics may be related to conceptual vectors.
A concept is a set; dimensions are the coordinate axes of a Cartesian product. Category theory studies mappings that preserve structure—and the “correspondence” performed by concept vectors is precisely an effort to preserve structure.
The author did not elaborate on this direction, leaving it as an open question.
The relationship with the “big picture” tradition
Professor Wu Jinshan from Beijing Normal University advocates “understanding-based learning,” the core of which lies in the hierarchical structure of knowledge and the broader disciplinary context: Knowledge is divided into four levels—factual and procedural knowledge, disciplinary conceptual knowledge, knowledge of the broader disciplinary context, and general human thinking skills.
The general process of mathematical modeling is as follows: focus on real-world objects, identify the problem, collect and understand data, translate the real-world problem into a mathematical problem (focusing on the primary contradiction while ignoring numerous details), solve and verify the solution, and then systematize and generalize the findings.
Author’s comment: Mathematical modeling originates from reality; conceptual vectors also arise from reality and move toward abstraction. Conceptual vectors themselves represent a “big picture”—precisely the kind of knowledge at the fourth level: general human thinking.
Why do ancient ideas get lost?
Since dualism, the four quadrants, and the Eight Trigrams have existed since ancient times, why is it necessary to reintroduce the concept of vector today?
The author’s diagnosis is this: Ideas have existed since ancient times, yet their application has been woefully inadequate. Since modern times, all attention has been focused on science, leaving philosophy to shrink; as a result, ancient ideas have lost their transmission, leaving modern people without effective tools for thinking, and the use of good thinking tools is far from widespread.
The limitations of thinking have turned many questions that could have been answered into insoluble ones—science has imprisoned human thought.
This passage can be regarded as the book’s motivating statement: It’s not about inventing new ideas, but rather about making ancient yet effective intellectual tools available once again—and restating them in a modern, rigorous, and discussable manner.
Yogurt, soy sauce, and kimchi—none of them know each other.
The same structure keeps popping up in different places—kitchens are the liveliest of these spots.
On the grasslands, herders make yogurt; in the Central Plains, brewers produce soy sauce; in households across the peninsula, kimchi is pickled; and in Europe’s mountainous regions, cheeses are aged. Though separated by tens of thousands of miles and centuries of history—so far apart that no one has ever known the others—they’ve each independently discovered the same fundamental principle: letting microorganisms take the first bite, transforming perishable foods into delicious, long-lasting treasures.
No one has ever convened a “Fermentation Conference,” nor has any patent office ever issued a certificate for it. The structure is objective—anyone who takes food seriously will inevitably run into it.
The same holds true for conceptual structures. The four quadrants, the Eight Trigrams, MBTI, and the periodic table—all emerged in different eras and across diverse fields, yet none directly descended from the others. This isn’t a case of one borrowing from another; rather, it’s akin to fermentation: the structure is simply there, and the more you tinker with it, the more naturally you’ll come to grasp it. Reading this list of ancestors feels like perusing a roll of honor—but in reality, it’s more like a “collision record.” And the more collisions you experience, the stronger the evidence that this path is truly the real deal.
The spectrum of conceptual vectors spans across time, cultures, and disciplines: the four quadrants, MBTI, the Eight Trigrams, word vectors, dialectics, object-oriented programming, writing techniques, tables, the Semantic Web, and the periodic table of elements.
What it aims to do is to eliminate duplicates and achieve unification of these isomorphic practices. Its mission for our time, in an era when scientific attention is overly concentrated, is to reinvigorate the ancient thinking approach of “abstract modeling.”
The next chapter is a summary of the entire book: a map of knowledge.
Condense the entire book into three paragraphs.
The first paragraph, the problem: The conceptual world of humanity is chaotic. Concepts are described using natural language without rigorous definitions; concepts that should be kept separate are mixed together in discussion; different disciplines and various cultures repeatedly reinvent the same truths yet remain unaware of each other’s discoveries; and the success of science has given rise to “scientism,” leading to the abandonment of thought in vast domains of issues that lie beyond science’s reach.
The second section, on methodology: The concept vector approach posits that concepts can be represented as combinations of several binary dimensions—a concept is equivalent to a vector. These dimensions are mutually exclusive yet equally valid; their combinations give rise to concepts. Moreover, dimensions are not limited to just two poles—they form continuous number lines. Gene lengths can be extended, with such extensions further subdivided into new genes. Concepts are best named using codes rather than nouns. This method is not scientific but rather an abstract modeling approach: its criterion for evaluation is “how well it works,” not “whether it’s true.” It clearly articulates the origins of each of its claims, thereby distinguishing itself from metaphysics.
The third paragraph: Value. With concept vectors, knowledge can be used—just like the periodic table—to identify and fill gaps, eliminate redundancies akin to gene sequencing, and provide a comprehensive overview comparable to a map. Categorization grants people the right to treat things differently; multi-dimensional evaluation replaces single-dimensional assessment, enabling individuals to avoid extremes and maintain moderation.
The nature of this method
You can use the author’s own words to characterize this method.
Induction → Deduction: Conceptual vectors are minimalist, free of unnecessary, groundless assertions; more deduction, less induction—this leads to a minimalist effect. Once concepts are clearly organized, we can avoid ad hoc empirical generalizations and instead adopt unified, deductive explanations.
●From Randomness to Unity: Conceptual vectors have revealed the underlying unity of human thought and language. Human thinking is an abstraction of the objective world; the process of abstraction involves comparing two conceptual vectors and identifying their commonalities. Language consists of symbols that correspond to concepts; an event is simply a concept endowed with parameters.
Theories also come in soft and hard varieties: mathematics and sciences that have withstood falsification are not easily changed, whereas economics is more prone to change—its underlying principles are relatively stable, but its popular applications tend to evolve rapidly. Concept vectors belong to the “principles that generate concepts”—the layer that is least likely to change.
There’s another point worth remembering: conceptual vectors should be categorized as belonging to a domain closely related to logic. Like mathematics, they exist objectively and are independent of the human brain—just as logic is a tool of thought, so too are conceptual vectors. They don’t study the world itself; rather, they focus on “concepts” themselves. And concepts represent the very first step in abstraction—a crucial tool of thought. The nature of a tool determines its character: it isn’t tasked with discovering new facts; instead, it ensures that your existing concepts are properly organized and aligned. These tools of thought are the original meta-framework—the very foundation upon which the subsequent sections are built, all of which are their products.
From Method to System: A Map of the Entire Book
The concept vector itself is merely a tool. The value of a tool becomes evident only through its use. Around this tool, the entire book has built these frameworks (arranged according to the book’s reading order):
Psychological Framework (Part One): Reinterpret the MBTI using four “genes”—“Introversion/Extraversion, Sensing/Intuition, Thinking/Feeling, Judging/Perceiving”—and rearrange Freud’s concepts of “Ego/Super-Ego/Id/Unconscious” to answer the question: “What types of people are there, and how does the human psyche function?”
Thinking Framework (Part II): Using genes such as “Discrete/Continuous × Hierarchical/Flat,” we analyze the four modes of thinking and answer the question: “What are the different types of human thinking?”
●Philosophical Framework (Part IV): Using the two dimensions of “Simple/Complex × Unified/Fragmented,” we define four worldviews—reductionism, teleology, holism, and collectivism—and answer the question: “What are the different types of forms that the world can take?”
●Social Framework and Life Framework (Part V): By applying the philosophical framework to society and life, we derive four attributes of human beings—naturalness, sociality, instrumentality, and idealism—and thereby answer the question: “What is humanity’s place in the world?”
Next step: Get the concept up and running.
Concept vectors are not the endpoint—they are the starting point.
Breaking down a concept into dimensions and placing it on a coordinate system still yields only a static table. The author’s next step is to connect these concepts with one another, enabling them together to form an organization that can actually function and operate.
This step has actually already begun. Looking back at the book’s framework, none of its categories are rigid or static: In the psychological framework, there’s a circuit of operation—where the id acts upon the ego through the non-ego, and the ego, in turn, is governed by the superego; the beginning and end connect to form a circular loop. Outside the ego’s four quadrants lies another flywheel—a spiral that links “relationships → abilities → career → hobbies.” In the cognitive framework, there’s a cycle of generation and operation: new information is stored in memory, and that memory then takes center stage. As for the social framework, each of its four lines represents a distinct operational pathway: in economics, it’s production → consumption → circulation → distribution; in politics, it’s the genesis of power → its execution → the establishment of order → and its guarantee. Ultimately, all of the author’s frameworks are interconnected and fully functional.
Vectors define the position of concepts; yet, this position is merely a standby. What this approach truly aims for is the interlocking and collaborative operation of concepts themselves.
Some unfinished words
Finally, the unfinished matters are also on the table:
●The mathematical representation of the 7-concept vector is currently just a single vector—“I don’t know if there’s any possibility of refuting it”—the author welcomes any information that could refute it or build upon it further.
●The to-do list also includes: alignment between columns and rows in the顺势 and psychological frameworks, as well as the correspondence between thinking frameworks and psychological frameworks.
Once a tool is set up, the most important thing is to use it to get the job done.
Where do you start? The author’s answer isn’t “Find dimensions for a given concept”—rather, it’s the reverse: Find a set of words you’re familiar with, and then create an element periodic table for them. First, decide on the number of dimensions—two, three, or even four—and draw the quadrants. Then, place the words into the table one by one. If a word can’t fit into any quadrant, that signals you’re missing a dimension; if several words end up crowded into the same cell, that suggests your current dimensions could be further subdivided. The moment you finish drawing the table, you’ll see—for the first time—the shape of your own knowledge.
Please turn to Section Four, “Philosophical Framework: The Four Forms of the World,” and see how the concept vector makes its grand debut for the first time.
Philosophical Framework: The Four Forms of the World
Use two dimensions to clearly explain the origins and development of the world.
The previous section discussed methods; this section represents the first large-scale application of those methods: using two binary dimensions to classify the “form” of the entire world.
The world we live in, put simply, is composed of various forms of matter. So, what are the different types of these forms?
The answer provided in this book is: By using two dimensions—“simple/complex” and “unified/decentralized”—we can categorize all forms into four quadrants: the reductionist world, the teleological world, the holistic world, and the collectivist world. These four perspectives are not four new terms; rather, they are four different lenses through which we can view the world. Whether we’re looking at an atom, a stool, a human body, a forest, a company, or a nation, each can be placed precisely within its proper framework.
Furthermore, this book also presents a “matrix” of philosophical frameworks: constancy/change, individual/collective. From the intersection of these two dimensions emerge sixteen fundamental categories—matter, interaction, space, time, internal causes, external causes, goals, realization, existence, development, cooperation, competition, independence, openness, structure, and process—and establishes preliminary correspondences with the Nine Palaces of the Luo Shu, the Eight Trigrams, and Eastern and Western cultures.
You don’t need any prior knowledge to read this book: after you’ve finished it, the very first thing you’ll ask yourself when you look at anything is: Which “theory” does it belong to?
Each chapter is self-contained. Readers who just want to get a taste of the story can grasp the main plot by reading Chapters 1, 2, and 4 only.
The boldest opening line
The purpose of a philosophical framework is to provide a holistic understanding of the world.
This sentence sounds like the opening line of all philosophy. But the very next sentence reveals one’s character:
The world we live in, put simply, is composed of various forms of matter. To put it concisely: the world is form itself.
Please note the ambition of this definition. It does not say that the world is material, nor does it say that the world is ideal; rather, it says: The world is form.
Matter is the base material; form is the pattern. The very same atoms, arranged in one way, become a stone—a certain form; arranged another way, they become a flower—a different form; arranged yet another way, they become a human being—a third form; and arranged in still other ways, they give rise to a company, a nation, or a civilization—each representing yet another distinct form.
To study the world, at its root, is to ask: What are the different kinds of forms? How do forms arise, how are they sustained, and how do they transform?
Here, the author quotes a Buddhist saying: “In Buddhism, all things are merely forms, mere illusions.” This statement is often interpreted as a pessimistic nihilism, but within a philosophical framework, it is understood as a structural assertion: since all things are forms, there must exist a theory that clearly explains the origins and nature of these forms.
The philosophical framework is used to clarify the origins and development of the form.
From “What is the world?” to “What kinds of worlds are there?”
Traditional philosophy loves to ask, “What is the essence of the world?” and attempts to encapsulate the entire world in a single word: matter, spirit, the Dao, qi, existence...
This mode of inquiry has a common flaw: the answer is singular, yet the world is plural. The result of trying to encompass the world within a single word is that all the richness of the world is flattened and reduced to uniformity.
The philosophical framework has shifted its approach: instead of asking “What is the world?”, it now asks, “What are the different kinds of forms?”
Switching from singular to plural questions is the first step in the conceptual vector approach: Only by categorizing first can we talk about structure; only by recognizing that the world comprises multiple types can we discuss the distinctions and connections among these types.
This is precisely the ambition behind the “Categorization and Deduplication” discussed in Part Three: Categorization isn’t about slicing up the world; rather, it’s about drawing up a “parts list” of the world. Once this parts list is drawn up, everything can be described as a combination of these parts—just as, once we know the chemical elements, any substance can be expressed in terms of its molecular formula.
What classification should we use? Two dimensions.
What are the different types of forms? The author uses the tool of concept vectors to answer this question. Two dimensions:
·The second dimension: Unity/Dispersion (Non-unity)
“Unity” means that each part is indispensable and collectively serves a single purpose. For instance, the gears of a watch are all essential and form a unified whole; if even one gear were missing, the watch would stop working. On the other hand, a pile of sand is dispersed—whether it has one grain more or one less makes no real difference. Note that this is an inherent property of the thing itself, not a matter of how we observe it. Whether the gears serve the same purpose is determined by the watch itself, not by the observer.
Two dimensions, each taking its two extremes, combine to produce four forms:
These are the pillars of the philosophical framework: reductionism, teleology, holism, and collectivism—collectively known as the “Four Theories.” In the next four chapters, we’ll step into each of these four worlds one by one.
one
Dissipation → Unification
In addition to being arranged in four quadrants, the diagram can also be read vertically. Place an object before you: If you disassemble it downward, you’ll see its components—reductionism lies below; if you ask upward about its purpose, you’ll find teleology at the top; and if you look at it horizontally, you’ll perceive it as a whole (holism), while its surrounding context reveals a sense of community (collectivism)—with holism and collectivism occupying the middle ground. When you scan from bottom to top, you’ll find precisely the complete perspective of “components—self—purpose.”
Before that, there’s one more thing to mention:
These four dimensions—simple vs. complex, unified vs. fragmented, and the dynamic vs. static and internal vs. external that I’ll discuss later—are the most fundamental, the most universal, and the most abstract. In a sense, they’re arranged according to their universality. This is my personal viewpoint; there’s no standard by which we can measure or judge them. Just like there are many cultures, each with its own perspective. Why do these cultures exist? Because without abstraction, humans wouldn’t be able to think at all. When people share common abstractions, they reach a consensus—and that’s precisely what constitutes culture.
The author doesn't claim that these four dimensions are the sole truth of the universe; rather, he asserts: “These are the four pillars I’ve chosen—the most universal ones. Let’s set them up first and see how big a roof they can support. If they can hold up the roof, great—let’s keep them. If they can’t, we’ll just have to find new ones.”
This is the modeler’s honesty—and also the modeler’s confidence.
The Philosophical Framework Within a Box of Building Blocks
“The material is the base, and the form is the pattern”—this single sentence can be fully illustrated with just one box of children’s building blocks.
The same set of building blocks—this morning they’re assembled into a house, this afternoon they’re taken apart and reconfigured into a car, and by evening they’re scattered all over the floor in a jumble. The blocks themselves haven’t changed; the material hasn’t changed either. What’s changed is only their arrangement—their form keeps shifting again and again. “The world is form” refers precisely to this: what’s worth studying isn’t the pile of blocks themselves, but rather “how many different configurations you can create—and what kind of character each configuration has.”
One hundred yuan is indispensable—remove the one supporting the weight, and the whole structure will collapse—this is the world of holism. But the pile lying on the ground—whether it has one more or one less piece doesn’t matter—that’s the world of collectivism.
A box of building blocks, four different shapes. What the child is playing with are the building blocks themselves; what the philosophical framework focuses on is the variety of patterns. The next three chapters will examine these four patterns one by one, in detail.
The world is form; the question of the types of form serves as the gateway to philosophy. Using the two dimensions of “simple/complex × unified/dispersed,” forms are categorized into four types: reductionism, teleology, holism, and collectivism.
In the next chapter, let’s first clarify “reductionism”—the worldview we’re most familiar with.
What is reductionism?
Reductionism is not a new term. It refers to the approach of continually breaking things down:
For example, if a material is composed of molecules, and we thoroughly understand the molecules themselves, then we’ve fundamentally grasped the nature of that material. Similarly, if molecules are composed of atoms, and we gain a clear understanding of the atoms and their interactions with one another, then we’ve fundamentally mastered the nature of the molecules. In a reductionist world, everything is simply made up of basic building blocks.
Within the framework of philosophical coordinates, reductionism = simplicity + fragmentation.
“Dispersion” implies a loose relationship among parts: viewed separately, each part remains itself; put back together, the whole is simply the sum of its parts. 2 + 2 = 4—this is the arithmetic of a reductionist world.
Reductionism is humanity’s most brilliant path of cognition. Physics is its standard-bearer: from objects to molecules, from molecules to atoms, and from atoms to fundamental particles—each step down in the hierarchy of decomposition brings a deeper level of understanding. The remarkable achievements of modern medicine—such as surgical procedures, chemical drugs, and cell-level targeted therapies—are also largely the fruits of reductionism.
The Four Categories of Reductionism
The philosophical framework assigns four sub-concepts to reductionism, which are the result of further dividing “static/dynamic × internal/external”:
Reading: In a reductionist world, when we take things apart, the static core is matter, the static extension is space, the dynamic core is interaction (force, mutual action), and the dynamic extension is time.
Matter, space, interaction, and time—these four terms happen to be precisely the four fundamental stages of physics. Physics is nothing less than the complete unfolding of the reductionist world: matter exists in space, changes over time, and is interconnected through interactions.
This four-part framework is not a wordplay. It reminds us that any reductionist question can be broken down into four key inquiries: What substances does it consist of? How is it distributed in space? What interactions exist among its various components? And how does it evolve over time?
Once the four questions have been answered, a reductionist object is essentially clarified.
The operational pathway of reductionism
The four categories of reductionism are not static; there are operational pathways among them:
Through time, matter disperses into space; through interaction, space coalesces into matter.
In everyday terms: Everything spreads out across space over time (think of a drop of ink falling into clear water), and then, through interactions, comes together again in space to form matter (think of a nebula collapsing to become a star).
Dispersion and aggregation are the breath of the reductionist world.
The boundaries of reductionism
The power of reductionism lies in decomposition; the limits of reductionism also lie in decomposition. The author repeatedly points out its limitations in numerous notes.
First, some things shouldn't be taken apart.
If you want to know a politician’s next move, you can’t arrive at the answer through dissection—humanity has yet to fully explain thought based solely on anatomical analysis.
Second, once it’s taken apart, you can’t put it back together.
Reductionism does not pay sufficient attention to the interrelationships among the various parts of a whole. The author provides an excellent example: Although humans are now able to collect and analyze vast amounts of data on every individual and every enterprise, they still cannot predict macroeconomic trends—partly because complex systems are inherently too random, and partly because humans have yet to fully grasp all the links in the chain of interconnected feedback loops.
A chain reaction cannot be studied solely from a microscopic perspective; it also requires a macroscopic, holistic mindset.
Third, there is a qualitative change at the hierarchical transition point.
When the quantity reaches a certain scale, higher-level structures emerge. These higher-level structures possess their own novel properties that differ significantly from those at the lower levels. If you break down the brain into neurons, you’ll find that the laws governing neurons don’t contain the word “consciousness”; similarly, if you break down society into individuals, you’ll discover that the laws governing individuals don’t contain the word “revolution.”
The relationship with “objectivity”
Finally, a philosophical footnote that can help us see the nature of reductionism: “Objectivity” is a reductionist perspective.
To say that something is “objective” means that it exists independently of the observer and can be measured without ambiguity—this holds true essentially only in a “simple, dispersed world”: the mass of an atom does not vary from person to person.
Once we enter the realm of complex unity or complex fragmentation, the observer’s abstraction, modeling, and stance come into play, and “objectivity” begins to lose its edge. All information in the brain is the result of abstraction—it is never objective reality itself.
This doesn't mean the world is fake; rather, it means that the higher we go up the hierarchy, the more “pure objectivity” becomes an approximation.
Take apart the alarm clock and the two extra parts.
Many people did one thing when they were kids: they took apart the alarm clock at home.
The process of disassembly is a triumphal parade of reductionism. As the back cover opens, the mainspring, gears, hands, and alarm mechanism are revealed—each layer meticulously broken down, every single part laid out on the table, clearly recognizable. At that moment, you’ll truly realize: the alarm clock has no secrets—it’s simply these parts. When you take it apart completely, you’ve mastered it entirely.
Then came the second half: reassembling everything in reverse. As I kept putting the pieces back together, two extra parts suddenly appeared on the table. Strangely enough, the alarm clock still worked—though from then on, it ran five minutes slow every day. Those two extra parts were the memento that reductionism left behind for every disassembler: when you took it apart, you recognized each individual part; but when you tried to put it back together, you found yourself utterly clueless about how those parts fit together.
Even a simple thing like an alarm clock is like this. Now imagine a person, an economy, or even a whole stretch of history— the more you take apart, the more pieces you can separate, yet the more parts you’ll find that can’t be put back together. Those boundaries discussed in this chapter? Anyone who’s ever taken an alarm clock apart has already felt them with their own hands, long ago.
Reductionism = a simple, fragmented form; the approach is decomposition. The four categories are matter, action, space, and time, and the process involves a cyclical interplay of dispersion and aggregation.
It took over the physical world and made science possible; yet, when faced with complex, unified entities, three additional theories are needed to take over. The next chapter introduces a new term coined by the author: teleology.
A newly coined term
Teleology is a neologism coined within the philosophical framework. Some people refer to it as “functionalism”—and indeed, functionalism and teleology are equivalent; they are talking about the same thing.
The example given by functionalism is: Animal legs are for walking, and the digestive system is for digestion. Another example from teleology is: A stool is for sitting on.
Pay attention to the common feature of these three examples: Their explanations all take the form of “What is it for?” rather than “What is it made of?”
When asked what a stool is, the reductionist answer is: “Wood, nails, glue—combined in a certain structure.” The teleological answer is: “It’s for sitting on.” Both answers are correct, but they address different questions.
In the philosophical framework, teleology is characterized by: simplicity + unity—simple in structure, yet with each part working together to serve a single purpose.
The Four Categories of Teleology
Reading: The achievement of a goal involves two aspects: on the static side, there are internal factors (one’s own conditions, abilities, and enthusiasm) and external factors (the environment, opportunities, and others); on the dynamic side, there is the process of setting and attaining the goal.
Internal factors, external factors, goals, and realization—this constitutes a complete “cycle of purpose.”
For any task that “has a purpose,” you can ask these four questions: What is its goal? What is the path to achieving it? Are the internal factors sufficient? Are the external factors favorable?
Exams, starting a business, seeking medical treatment, and dating—these all have the same underlying structure.
A Contemporary Echo: The Coordinates of the Law of Success
The author interprets Barabási’s “The Laws of Success” through the lens of teleology—a great example of how to apply the “Four-Theory Coordinate System.”
·First Law: Capability performance drives success, but when capability performance cannot be measured, social networks drive success.
●Third Law: Initial Success × Social Fitness = Future Success;
Law Four: Successful teams possess both diversity and balance, and they often make it easier for a super leader to stand out.
Law Five: Success can happen at any time and at any age—just as long as you persistently stick with a good idea.
The author’s correspondence is as follows: The First Law corresponds to internal factors (performance capability); the Third Law corresponds to external factors (both initial success and social adaptability are environmental factors); the Fourth Law corresponds to goals (finding goals requires broad interests and diverse teams); and the Fifth Law corresponds to achievement (perseverance).
As for the second law—“capability performance is bounded, but success is limitless”—the author bluntly states that it’s “not a law at all”¹—it’s a commentary on the first two laws rather than an independent mechanism.
The value of this interpretation lies not in right or wrong, but in its demonstration: Take any theory at random, and first ask which of the four quadrants each of its propositions falls into—and which quadrant is missing.
The theory of gaps is incomplete: only by filling in the gaps can the theory become three-dimensional.
Teleology and the “Ideal Nature” of Human Beings
Teleology offers yet another, even more significant outlet: it serves as the philosophical coordinate for understanding “why humans act.”
After people have secured their basic needs, fulfilled their social obligations, and satisfied their physical requirements, they still pursue careers and strive for success. To avoid boredom, they seek out all sorts of activities—some of which may be related to survival and others may not.
This attribute of acting purely “for the sake of a goal” is what the author calls human idealism, which corresponds to teleology.
Huawei’s strategic planning methodology, “Five Looks and Three Determinations,” is an idealized enterprise-level approach. The “Five Looks” refer to examining trends, the market, competition, oneself, and opportunities (gaining a clear understanding of both internal and external factors). The “Three Determinations” involve setting future goals, defining strategies, and identifying strategic control points (establishing objectives and outlining the path to achieve them).
From personal ideals to corporate strategy, the skeleton of teleology remains unchanged.
The Eternal Struggle Between Teleology and Reductionism
There has been a long-standing debate in the history of science and philosophy: Is the world “driven by causes” (reductionist causality), or does it “evolve toward purposes” (teleology)? Although Aristotle’s theory of the four causes originally included a “final cause,” modern science banished it.
The philosophical framework’s attitude is: This isn’t a matter of who’s right and who’s wrong—it’s a matter of coordinates.
A stool is made for sitting—this statement holds completely true at its own level and does not require reduction to the science of wood to be valid. A reductionist explanation connects downward (by asking about the structure of lower-level components), while a teleological explanation connects upward (by asking about the function of higher-level systems).
A comprehensive understanding must be able to break down effectively from the bottom up and also connect smoothly from the top down.
Why Don’t the Flowers Bloom? Four Questions from the Balcony
Anyone who’s ever grown flowers has asked this frustrating question: “I water it every day—why won’t it bloom?”
The four-step framework of teleological reasoning happens to be precisely the four steps needed to answer this question. The first step asks about the goal: Are you sure this is a flowering variety? Someone might be eagerly waiting for a pothos plant to bloom—but if the goal itself is wrong, everything that follows will be off track. The second step asks about implementation: Is the path correct? Watering is an action, but some people end up watering their plants three times a day—meaning the method of implementation is flawed; the harder they try, the worse it gets. The third step asks about internal factors: Is the seedling itself capable? If the seed isn't the right variety or the root system hasn't been properly nurtured, then the plant's own conditions simply aren't up to par. The fourth step asks about external factors: Is the environment conducive? A north-facing balcony lacks sufficient light, and a heated room is too dry—meaning the external conditions aren't favorable.
Goals, implementation, internal factors, and external factors—90 percent of posts in flower-enthusiast groups over the course of a year can be categorized under these four headings.
The scope of this four-part inquiry clearly extends beyond just balconies. Whether it’s taking an exam, starting a business, seeking medical treatment, or dating—each scenario shares the same underlying structure, as we’ve already discussed. The beauty of growing flowers lies in their rapid feedback: if a particular “box” isn’t working properly, after just a few months, the flower will “bloom” or “fail to bloom,” giving you a clear score.
Teleology—the simple, unified form—has the core sentence structure: “It is used for…….” Its four categories are internal cause, external cause, goal, and realization. It explains function, explains pursuit, and explains strategy; it serves as the philosophical foundation of human “idealism.” The next chapter will focus on the theory we’re most familiar with—and also the one most easily misapplied: holism.
Holism: Complex and Unified
Alongside reductionism, holism is often mentioned. In the philosophical framework, holism is defined as complexity plus unity: many parts, a complex structure—but each part is indispensable and collectively serves a single purpose.
The most typical example is the human body: the entire organism is designed to serve the purposes of survival and reproduction, with the heart, liver, spleen, lungs, and kidneys each performing their specific functions—no more, no less.
A company is also holistic: market, R&D, production, and finance—without any one of these departments, the whole system simply cannot function smoothly. In a holistic world, members are not “parts” but rather “organs”—parts can be interchanged or added/removed, whereas each organ has a unique and irreplaceable function.
The four categories of holism: independence, openness, structure, and process.
A whole must first be independent (with boundaries and capable of self-determination), yet also open (exchanging matter, energy, and information with its environment—any whole that is not open is merely a dead object). It has an internal structure (a static arrangement of parts) and processes (dynamic collaboration among parts).
Independence, openness, structure, and processes—these four questions can be used to examine any “whole”: Is this organization independent? Is it open? Is its structure clear? Are its processes smooth?
Regarding the holistic operational framework, the author left only one sentence—but it’s profoundly meaningful:
Openness, when complemented by the environment, can support an independent arms race.
Independence is not the product of isolation; rather, it is the result of openness. The more a person, a company, or a country can draw resources from their environment, the better they can sustain their own independence.
This motto can be posted on the desk of anyone who aspires to “independence and self-reliance.”
How does holism study a whole? The author offers two rather clumsy approaches. First, don’t take it apart: Treat the whole as a black box and observe it only from the outside—Newton didn’t need to know about atoms or molecules; by observing the system from the outside, he was able to formulate his three laws of motion. This is precisely the stance of holism. Second, find equivalents: Since the whole is indivisible, start by identifying two completely equivalent concepts, then describe their relationship—“production” and “consumption” are equivalent, whereas “production workshop” and “consumer behavior” are not. If the concepts you identify turn out to be unequal, it means you haven’t yet pinpointed the right place to make the cut.
Group Theory: Complex and Diverse
The theory of groups is complex and inconsistent. The theory of groups is also a neologism coined within the philosophical framework.
Group theory may be related to concepts such as systems theory and universal interconnectedness, but the author speculates that systems theory might be a broader term encompassing both holism and group theory.
An example of systems theory is the ecological environment: carnivores, herbivores, plants, and bacteria together form an ecosystem. The organization of these organisms is contingent—unlike human organs, which are neither more nor less than necessary. On a grassland, whether there’s one more or one fewer flock of sheep, the ecosystem remains an ecosystem nonetheless.
The four categories of group theory: existence, development, cooperation, and competition.
Members of a group first and foremost exist as individuals (none of them live for the sake of some unified purpose shared by others); each seeks their own development; and they sometimes cooperate and sometimes compete with one another.
Existence, development, cooperation, and competition—this is the grammar of the jungle, and also the grammar of the market.
Please note the distinction between holism and collectivism: a whole has a shared purpose, whereas a group does not. The human body is a whole—everything serves the purpose of survival and reproduction—while the market (in the general sense) is a group: there is no single “overarching purpose” guiding all buyers and sellers. Each individual acts solely in pursuit of their own self-interest, and order emerges spontaneously from countless local interactions.
There are two challenging cases along this dividing line. The author’s ruling is brief: the concept of “keystone species” in ecology—species whose absence would cause the ecosystem to collapse—is not contradictory to population-level theory; whereas the variable microbial communities in the human body are better explained by holism.
The relationship among the four arguments
At this point, the four arguments are complete. Arrange them in a row, and you’ll see a clear lineage:
In describing any hierarchical structure, there is a division of labor: reductionism serves to connect lower-level structures (what this entity breaks down into when dissected), while teleology serves to connect higher-level structures (what role this entity plays within the larger whole). Holism looks inward for collaboration, whereas systems thinking looks outward for interaction.
In his blog, the author also mentioned a classification of approaches: reductionism involves dimensionality reduction, teleology involves dimensionality increase, and holism and collectivism occupy an intermediate dimension—breaking down toward the lower end represents reduction, while ascending toward the higher end signifies elevation.
A set of correspondences casually scribbled down by the author in his notes can serve as quick sketches for the Four Arguments:
·As for tools: In the past, tools were all reductionist—every state could be calculated in advance. AI is the first tool created by humans using a teleological approach—it emerges from training, and we cannot predict its future states.
●To elaborate: A market economy is holonomic, while enterprise management is systemic.
In programming: functions are reductionist, structured programming is teleological, modules are holistic, and object-oriented programming is sociocentric.
These four statements are worth reading several times. They mean that the Four Discourses are not merely philosophical word games—they are four genuinely existing “modes of organization” that remain equally valid in today’s most modern fields, such as tools, economics, and programming.
The soccer team and the vegetable market
The dividing line between holism and collectivism can be observed twice just by taking a weekend outing.
First, take a look at the team on the field: eleven players—forward, midfielder, defender, and goalkeeper—each with their own specific role, all of which are indispensable. If the goalkeeper were to run up and play as a forward, the whole team would fall into complete chaos. Everyone serves the same overarching goal: to win the game. This is a living embodiment of holism: the members are “organs,” not “parts.” Because they share a common purpose, they need a coach, a tactical plan, and coordinated playing patterns.
After the event ended, I turned into the nearby vegetable market, and the atmosphere suddenly shifted. The vegetable sellers, meat sellers, and fish sellers—each one sources their own goods, sets their own prices, and bears their own profits and losses—none of them works toward a “general purpose” for the market, nor does the market itself have any overarching goal. The stallholders each exist and develop independently; sometimes they cooperate (exchanging small change or collecting packages for each other), and sometimes they compete (if you sell something for two yuan, I’ll sell it for 1.8 yuan). Whether there’s one more stall or one fewer, the market remains a market nonetheless.
That same morning, separated by just a wall: on one side of the wall lies the logic of “nothing can be missing”; on the other side, the logic of “one more or one less doesn’t matter.” To understand this thin dividing line is to grasp the two orders that govern our complex world.
Holism focuses on “how organs collaborate” (independence/openness/structure/process), while systems theory focuses on “how members interact” (existence/development/cooperation/competition).
With the four arguments in place, the world’s forms have been given a basic classification. The next chapter will address a natural question: What happens when we apply this framework to human beings themselves?
The First “Reflection” on the Philosophical Framework
The philosophical framework deals with the forms of the world. Since humans are also a form within the world, they should likewise possess the characteristics of reductionism, teleology, holism, and collectivism.
When the four arguments are “reflected” back onto the person themselves, we obtain the four attributes of human beings:
·Humans possess a natural quality—characteristics and instincts common to animals. This is reductionism: if you take humans apart, you’ll find cells, organs, and instincts, all governed by biological laws.
·Humans are social beings—each person assumes social roles such as parent, child, superior, and subordinate, and is embedded in various social structures, with corresponding benefits and contributions. This reflects a holistic perspective: humans are organs within the “larger whole” that is society.
Humans are driven by self-interest—within the jungle of human society, they forage for sustenance, which means earning a living. There will be competition as well as cooperation. This is the theory of groups: the interactions among individuals in markets and societies.
Humans are endowed with idealism—after having secured basic sustenance, fulfilled their social obligations, and satisfied their physical needs, people still pursue careers and strive for success. To avoid boredom, they seek out all sorts of activities. These pursuits may be related to natural, social, or utilitarian factors, or they may be entirely unrelated. This is teleology: humans can act for the sake of purely intrinsic goals.
The author particularly emphasizes: the natural and social aspects of human beings are already covered in textbooks; the aspects of self-interest and idealism, however, were introduced by him personally.
This once again validates the concept vector’s ability to complete missing information—focusing solely on naturalness and sociality leaves gaps in the picture.
Why is “fourness” stronger than “twoness”?
“The unity of human beings as both natural and social beings” is a classic statement found in textbooks. It’s not wrong, but it’s insufficient.
A young man who works overtime late into the night just to earn more money—his behavior cannot be explained by “sociality” (he’s not sacrificing for the collective), nor by “naturalness” (he’s not driven by survival instincts)—rather, it’s driven by self-interest, as he forages in the jungle of the market. The pursuit of self-interest is an attribute at the level of group dynamics.
A middle-aged man who has already achieved financial freedom yet has begun delving into philosophy, planting trees in the desert, and writing novels that no one will ever read—his actions are driven neither by self-interest nor by a sense of duty; rather, they are fueled by his ideals. Ideality is an attribute at the teleological level.
Without the element of self-interest, all economic behavior would have nowhere to stand; without the element of idealism, all of humanity’s great and mad endeavors “for no particular reason” would have nowhere to find their place.
Only when the four quadrants are fully filled out will there be no blank areas on the map of human behavior.
Revisiting “The Meaning of Life” Through the Lens of the Four Qualities
The author offers a deconstruction of “the meaning of life,” employing precisely the four characteristics:
This passage is worth reading slowly.
The contemporary “success” ideology essentially expands the “utility-based” quadrant to become the sole defining feature of life. However, the Four-Quadrant Framework reminds us that naturalness (physical health and instinctive fulfillment), sociality (family roles and social connections), and idealism (the pursuit of goals beyond mere utility) are all legitimate and integral components of a fulfilling life.
Marcuse spoke of the “one-dimensional man”; the author speaks of the “person who knows nothing but making money”—the diagnosis is the same. Yet the Four-Quadrant Framework provides a quantitative prescription: four quadrants, none of which can be left empty.
The Four-Quadrant Framework can serve as a self-assessment checklist. When analyzing your own situation or that of anyone else, ask four questions:
Naturalness: Is your body doing okay? Are your instinctive needs (eating, sleeping, exercising) being met?
Sociability: Have you fulfilled your family and social roles? Is your sense of belonging still strong and stable?
Profitability: Is foraging in the jungle going smoothly? Are the relationships between competition and cooperation healthy?
Idealism: Is there any activity you engage in purely for the sake of avoiding boredom and getting things done—without any expectation of real-world benefits?
The four empty squares in the puzzle are precisely where problems in life tend to arise.
There are several items that readers often ask which category they should be placed in. Here’s the author’s judgment: Emotion is natural; so is aesthetics—indeed, he defines aesthetics as “human beings’ big-data intuition”: When we find a landscape beautiful, at its core, it’s our genes voting—our instinct telling us that this environment is healthy and superior. Religion, on the other hand, is about idealism—it deals with goals that transcend mere self-interest.
It is said that management, too, has its own corresponding formula: For the grassroots level, employ the Legalist approach—emphasizing discipline and heavy rewards and punishments—focusing on people’s natural inclinations; for the middle management level, adopt the Confucian approach—promoting self-awareness and placing great importance on ethics—focusing on social norms; for senior management, apply the Daoist approach—advocating cooperation and guiding according to the natural flow of things—focusing on interests; and for all employees, cultivate a corporate culture—focusing on ideals. If an organization’s management methods are monotonous, it usually means that they’re putting all their effort into just one particular aspect.
This also foreshadows the content of the fifth section, “The Framework of Society and Life”—where the Four Characteristics will unfold into a comprehensive social framework and a life framework.
The Four Characteristics of Mobile Screen Time Reports
Every Sunday, many people receive a “Screen Time Report” on their phones. Few realize that this report is actually a four-dimensional ratio chart.
Open it up and take a look—where has all your time gone? Fitness apps, sleep trackers, food-delivery apps—these are the natural side of your phone usage. Family groups, WeChat Moments, those endless “Read and reply” messages—all these reflect your social side. Office software, stock market updates, part-time job recruitment groups—these are driven by practical interests. And then there’s something no one else knows why you’re so obsessed with: drone-modeling forums, Kunqu opera albums, a column you’ve been writing for three months that no one’s ever read—this is the idealistic side of your phone usage.
The “not a single one can be left empty” principle of the Four-Aspect Framework has been transformed into a straightforward bar chart in this report. Some people have a pillar representing their interests standing tall and unyielding, while the other three pillars are so short they’re virtually invisible; others, meanwhile, have an idealistic pillar that stretches high and proud, yet their pillar of practical interests is suspiciously short. The proportions may vary from person to person, but whenever one pillar completely drops to zero, that particular area of life begins to wither and dry up.
Next time the report pops up, don’t rush to scroll it away—it’s not just traffic stats; it’s your weekly report on all four aspects of your life. If any “column” is drying up, it’ll know before anyone else does.
When reflection is returned to the human being, it yields four attributes: naturalness, sociality, utilitarianism, and idealism. This framework is more comprehensive than the dualistic theory of “naturalness + sociality,” because it can accommodate both profit-seeking behavior and dream-chasing behavior.
The four characteristics also serve as the seeds for the “Social Framework” and “Life Framework” discussed later in this book. In the next chapter, we’ll return to the philosophical framework itself and explore its other set of dimensions: motion and stillness, and inner and outer.
The "column" of the philosophical framework
The previous chapters discussed the “actions” of the philosophical framework: the four perspectives arising from the intersection of simplicity/complexity and unity/dispersion. Now we’ll turn to its “categories”: constancy (stillness) / change (motion), and the individual (internal) / the whole (external).
The rows and columns intersect once more, yielding the complete table of the philosophical framework—the sixteen basic categories:
This table is worth pausing to look at for a minute. We’ve already covered each of its rows (Chapters 2–4); now let’s take a closer look at each column.
Constant·Individual Changing·Individual Constant·Entirety Changing·Entirety
Teleology Internal Cause
Group Theory Existence
Holism Independence
·Invariable Entity: Matter, Internal Cause, Existence, Independence—these are the names of a thing when it “calms down and returns to itself”: its substance, its internal cause, its existence, its independence.
Changing individual: action, realization, development, openness—when a thing “starts moving yet remains the subject of itself,” it is called: it exerts action, it carries out realization, it undergoes development, it strives for openness.
·Invariable Whole: Space, External Factors, Cooperation, Structure—Take a look around at the static environment: space, external factors, collaborative relationships, and structure.
Rows are scaffolding; intersections are the building itself. Each of these sixteen categories represents the focal point of “one angle × one state”—viewed from the perspective of this row, it focuses on the state represented by this column, and ultimately lands on the concept where attention settles. These are precisely the protagonists that the philosophical framework aims to introduce—and they’re worth examining one by one.
Reductionism—breaking things down into their component parts:
Substance (Immutable·Individual): The part itself. When an object settles into stillness and we look only at itself, it becomes “material”—the raw substance left after everything else has been disassembled.
Action (Change·Individual): The movement of a part. When an object starts moving while the subject itself remains unchanged, this manifests as action—force, reaction, or propulsion.
Space (Invariance·Wholeness): The static arrangement of parts. When observing the environment and the environment remains stationary, what is perceived is space—a specific location and extension.
Time (Change · Whole): The dynamic unfolding of the parts. When you observe the environment—and the environment itself is in motion—what you see is time: sequence and process.
This line features a runtime description in the author’s notes, using all four words: “Through time, matter disperses into space; through interaction, space coalesces into matter.” Diffusion relies on time, while aggregation depends on interaction—within these four focal points, the reductionist world breathes and lives.
Teleological One Line (Staring at the Goal):
Internal Factors (Immutable·Individual): The roots of goals within oneself—motivation, innate talents, and intrinsic drive. You’ll encounter these again later in this chapter when we discuss “Life Choices”: Individuals with strong internal factors harness their agency to continuously pursue and achieve their goals.
Realization (Change·Individual): The action of turning goals into reality—execution and implementation. This is the focal point of teleology’s “action inward.”
External Factors (Constant·Collective): Static conditions outside the target—whether the environment provides this opportunity, or whether this wind is favorable.
Goal (Change · Whole): The unfolding of purpose ahead—It guides you through change.
In his commentary on "Barabási’s Law of Success," the author employed precisely these four categories: Law 1 corresponds to internal factors, Law 3 to external factors, Law 4 to goals (finding goals requires broad interests), and Law 5 to realization. The fact that an externally introduced theory of success could so neatly fit into these four categories represents an unexpected validation of this intersectional framework.
Group Theory—A Line (Viewed from a Group Perspective):
Existence (invariant individual): In a group, it “has its number”—membership and presence. For an individual within a group, the first question is whether it’s there or not.
Development (Change · Individual): The growth of an individual within a group. After determining whether it exists or not, the second question is whether it’s growing or not.
Cooperation (unchanged · collective): A stable collaborative relationship among members—a mutually supportive structure that emerges when the group is at rest.
Competition (Change · Overall): The dynamic ebb and flow among members—a ranking competition that becomes apparent as the group operates.
Holistic approach (viewed as a whole):
Independence (Unchanging·Individual): Self-sufficiency of the whole—being able to function on its own, without relying on external assistance.
Openness (Change·Individual): The overall exchange with the external world—matter, energy, and information flowing in and out. In the author’s notes, this pair is described as cause and effect: “Openness draws on environmental inputs to sustain an independent arms race.” Openness represents the individual undergoing change, while independence represents the unchanging individual—using openness to nurture independence is the driving force behind the holistic approach.
Structure (Immutable·Entirety): The static arrangement of each part within the whole—the skeleton. The author has another corresponding note: assigning corresponding structures, corresponding spaces, and corresponding political structures is precisely the aspect of the whole where “each occupies its rightful place.”
Process (Change·Whole): The dynamic flow of each internal component within the whole—circulation of blood. Same note: Circulation corresponds to process, corresponds to time, and corresponds to economy—the process is precisely the “ever-flowing” aspect within the whole. (These two sets of correspondences will reappear when we discuss society in Part Five.)
The very same “individual is in flux”—from the perspective of reductionism, it’s called “function”; from the perspective of teleology, it’s called “realization”; from the perspective of group theory, it’s called “development”; and from the perspective of holism, it’s called “openness.” These four terms represent the distinct facets of the same phenomenon, each viewed from a different angle. The right way to approach the sixteen-grid isn’t to memorize the terms—it’s to reflect: Why, when we shift our perspective, must the very same state be given a different name?
The usage of the sixteen-grid system
The Sixteen-Grid is a “concept checklist.” Take any given subject and place it into this table—each of the sixteen boxes represents one question:
Is it composed of matter? How do its various parts interact? What space does it occupy? What kind of time does it experience? What internal factors does it have? What external factors influence it? What are its goals? How does it achieve them? Does it exist (and if so, what kind of member is it within the group)? Is it evolving? With whom does it collaborate? With whom does it compete? Is it independent? Is it open? What is its structure? What are its processes?
Once the sixteen questions are asked, the philosophical aspect of a given thing is fully revealed.
This is the operationalized version of “the world is form”: form is no longer esoteric—it’s simply coordinates within a sixteen-grid framework.
The reader might ask: Is the sixteen-grid system exhaustive? Could there be a fifth dimension—say, “real/virtual”—expanding the table to thirty-two grids? The author’s answer comes in two parts: You can add as many dimensions as you like—four dimensions are merely his preference, suited to his intellectual level. But moving from one dimension to two represents a major breakthrough—from a single axis to an entire table, from a scattered list to searchable coordinates—this step marks a qualitative leap. Any further additions beyond that are merely quantitative changes.
Motion and Stillness, Inside and Outside, and Life Choices
The dimensions of a column not only describe the world but also describe human beings. The author points out: The columns within the philosophical framework pertain to the major directions of human choice.
People who hold a changing perspective (hereafter referred to as “change-oriented people”) believe that without change, one wouldn’t feel truly alive. In contrast, “status quo-oriented people” prefer to maintain the status quo and stay comfortably within their comfort zones, rather than actively striving to break out of them.
People driven by internal motivations will proactively strive to achieve a goal—yet the risk is that they’ll “easily let others dangle a carrot right in front of their eyes,” spending their entire lives chasing goals set by others. As for those driven by external factors, they should go with the flow and “refrain from doing what’s impossible” (jokingly referring to this as “giving up”). They…
Will question the necessity of the goal: Who set this goal? Why is it worth pursuing?
These four orientations are neither superior nor inferior—change brings vitality as well as turmoil, while constancy brings stability but also stagnation; internal factors can lead to great achievements yet are easily controlled, whereas external factors, though enabling clarity and freedom, may also cause missed opportunities.
The attitude of conceptual vectors is always this: First, recognize the existence of the axis; then, understand your own position on the axis; and finally, comprehend the person at the other end of the axis.
Four names for overtime work
At ten o'clock in the evening, there was still one person typing away in the office. The same action—each of the four lines in the dictionary had its own name.
Reductionism sees the action: fingers tapping, electric currents flowing back and forth, data being rewritten on the hard drive—a single individual exerting force in the process of change. Teleology sees the realization: the proposal due at month’s end is growing longer page by page—a single individual moving steadily toward the goal. Holism sees the development:
Qualifications, accomplishments, and industry reputation are quietly accumulating—a member is growing within the group. The holistic perspective reveals openness: email cc’s, synchronized progress updates, and mutual alignment with the team—an organ is exchanging information with the whole.
There’s only one action—typing on the keyboard—but four different names. This isn’t a wordplay: the name you choose determines what question you’ll ask next. If you call it “Function,” you’ll ask about efficiency; if you call it “Implementation,” you’ll ask whether the goal is the right one; if you call it “Development,” you’ll ask whether this accumulation makes sense economically; and if you call it “Openness,” you’ll ask whether it’s still in sync with the bigger picture.
At its core, the art of the sixteen squares is precisely this “name-changing” skill—only when you place the same fact into the right square will the correct question naturally emerge on its own.
By combining the four categories of "action" and "process" with "motion," "stillness," "internal," and "external," we arrive at sixteen fundamental categories: matter, action, space, time, internal causes, realization, external causes, goals, existence, development, cooperation, competition, independence, openness, structure, and process.
They serve both as a parts list for the world and as coordinates for the four orientations of life. — In the next chapter, we’ll equip this static table with an engine: a running circuit—and explore its correspondence with the Luo Shu and the Eight Trigrams.
Static classification isn't enough.
Up to this point, the philosophical framework has provided only “categories”: the Four Discourses and the Sixteen Grids. But the world is alive; the relationships among concepts are not merely parallel—they also involve processes—how one concept gives rise to another, how one state transitions into another.
The author’s notes include a series of “runs,” each recorded with a brief maxim summarizing the route:
●The operation of reductionism: Through time, matter disperses into space; through interaction, space coalesces into matter. (The cycle of dispersion and aggregation is discussed in Chapter 2.)
·The operation of holism: Openness, when supplemented by the environment, can sustain an independent arms race. (See Chapter 4)
●Self-operation (this is a psychological framework—take a quick look ahead): Only by managing upstream and downstream relationships can you identify the skills you need to improve; only by earning money through your career can you afford to pursue your passions; only with the right skills can you advance your career; and it’s through everyone’s shared passions that you build meaningful relationships.
The operation of the economy: One can only consume after having produced. (In earlier notes, there was also a second half to this statement—“Only what can circulate can be distributed”—but the author later withdrew it—once an organization receives funds, it can distribute them internally within the organization. See Chapter 3 of Part Five for details.)
The meaning of “operational chain” can be summarized in one sentence: Production must come first before consumption—there is a sequence, dependency, and cyclical relationship among these categories.
A static table tells you what the world is made of; a running route tells you how the world lives.
Luoshu: Arrange the four quadrants into a 9-grid.
The author found that the four-quadrant diagram can also be arranged in a more sophisticated manner—the Luoshu style:
The horizontal axis runs from simplicity to complexity, while the vertical axis goes from dispersion to unity. The four corners represent the four quadrants (the Four Theories); the two ends of the number axes—simplicity, complexity, unity, and dispersion—are also crucial concepts, no less important than the four quadrants, so they are placed in the central cells along the four sides. Right in the middle, we put “Form”—the entire diagram is all about form.
The author’s awareness of this diagram is: “I’ve realized that not only are the four quadrants important, but the two ends of the number axes are also crucial concepts—no less significant than the four quadrants themselves. By including them all within the overall diagram, we achieve completeness.”
This sense of “completeness” is worth noting: The Lo Shu Nine Palaces is China’s oldest magic square, which ancient people used to symbolize the balance of heaven and earth. By embedding the Four Symbols into the Nine Palaces, we are essentially saying that the Four Categories are not isolated entities but rather a unified whole—each influenced by the four cardinal directions and revolving around a central axis.
Hidden within this central space is an idea that the author holds particularly dear: The four-part structure often implies a fifth position—a position that can serve as either the midpoint of a number line or the overall essence of a given phenomenon. Each of the four directions governs its own segment, yet the midpoint neither belongs to any one segment nor is exclusive to any particular segment; rather, it represents the common destination shared by all segments. Within the philosophical framework, “form” embodies precisely this midpoint: It is not a fifth worldview in itself, but rather the overarching totality of the philosophical framework. The Four Theories did not arise out of thin air—they stem from form itself and, through four distinct approaches, give rise to further forms: Teleology transforms form into a goal-oriented structure; reductionism breaks down form into simpler components; holism integrates form into a unified whole; and systems theory reduces form back into complexity. Although each of these four approaches yields its own unique results, when they are brought together and converged upon at the center, what emerges once again is none other than “form.”
This “four plus one” arrangement doesn’t appear only here. We’ll encounter it again in the fifth section, when we discuss the economy: production, consumption, circulation, and distribution are the four stages of the economic cycle, while inventory constitutes the fifth stage—the midpoint of the cycle and also the location of the overall economic aggregate. Let’s just get a preliminary feel for it now; we’ll come back to this correspondence in full detail in Chapter 3 of the fifth section.
The intellectual significance of this correspondence lies in its translation of the Eight Trigrams from “divinatory symbols” into a “categorical system.”
Heaven is the totality in change (time); Earth is the totality in stillness (space); Water is the substance of individual entities in stillness (matter); Fire is the force of individual entities in motion (action); Mountains are the reliable external conditions (external causes); Marshes are the internally conducive conditions that can be retained (internal causes); Wind is the direction that carries things far away (goal); Thunder is the action triggered by vibration and resonance (realization).
The only criterion provided by the author is this: The eight trigrams are divided into two groups, each containing four trigrams. The “Heaven” group is more common—thus, the four most frequently used categories (time, space, matter, and interaction) correspond to the “Heaven” group.
Readers can judge for themselves whether they agree with this correspondence—after all, the author himself has said that the correspondence is probabilistic and preliminary; whatever works best is how it should be mapped. Yet this example offers a new approach to reading ancient texts: treat the hexagrams as categories rather than as prophecies.
Correspondence with the Dungeons & Dragons alignment
A more surprising correspondence comes from tabletop gaming. In “Dungeons & Dragons,” character alignments are divided along two axes: “Lawful—Neutral—Chaotic” and “Good—Neutral—Evil.” The author’s corresponding:
And it was accompanied by two brief comments: “A person who is not in harmony with themselves is an evil person,” and “Complexity brings chaos and grayscale.”
— The Lawful seek a world that is simple and predictable; the Chaotic tolerate—or even relish—complexity. At its core, goodness seeks unity (community, empathy), while evil leads to fragmentation (separating oneself from the community and using others as mere means to an end).
Translating the moral dimension through the lens of worldview—this represents a “cross-disciplinary de-duplication” of the philosophical framework.
The operating route inside the dumplings
The operating procedure sounds mysterious, but you’ll understand it once you’ve made a batch of dumplings yourself.
If the dough isn’t properly mixed, you can’t roll out the wrappers; if the wrappers aren’t rolled out well, you can’t wrap the filling; and if you haven’t finished wrapping the dumplings, even when the water boils, you’re just left waiting. Each step is stuck right behind the neck of the previous one—this is the kitchen version of “you can only consume what you’ve produced”: the order isn’t dictated by anyone—it emerges naturally from the inherent dependencies of the tasks themselves.
Now let’s look at the division of labor. If one person handles everything from start to finish—performing each step one after another—it’s like a single worker going through the entire process alone. But when a family works as a production line—grandma kneading the dough, mom rolling out the wrappers, and the child arranging the filled dumplings—the efficiency immediately improves, because there’s handover between different stages. And it’s precisely at these handover points that problems are most likely to arise: if the wrapper is rolled too slowly, the person wrapping the dumplings has no choice but to stop and wait for more dough—this is exactly the same principle as in the economic cycle, where “even the best production is useless if circulation is blocked.”
The last batch of dumplings was brought to the table, and the whole family shared them. The distribution was decided by how quickly each person picked up their chopsticks—and also by the deference shown across generations. That single meal of dumplings on New Year’s Eve perfectly embodied the concepts of sequence, dependence, cyclical patterns, and fair allocation. The flow of operations never exists solely in books—it begins its work wherever there are processes to be carried out.
The category operates as follows: matter and space come together and disperse; independence relies on open sustenance; production precedes consumption.
The four quadrants are embedded within the Luo Shu Nine Palaces, with the “form” at the center; the sixteen grids correspond to the Eight Trigrams, allowing us to interpret the trigram patterns as categories; the factions correspond to the Four Theories, enabling us to read morality as a worldview.
With static classification combined with dynamic operation, the philosophical framework comes to life. The next chapter will explore the echoes and tests of the Four Discourses in various contemporary fields.
A way to examine a philosophy
Whether a philosophical framework is good or not isn't judged by how autonomous it is internally, but rather by how many places it can illuminate once it goes out into the world.
This chapter gathers the echoes of the Four Discourses across several widely separated fields. Most of these echoes originate from the author’s personal notes; the editor has organized them into a “Pilgrimage of the Four Discourses.”
In programming
In programming: functions are reductionist, object-oriented programming is holist, structured programming is teleological, and modules are systemic.
Explain one by one.
Functions are reductionist: a function does only one thing, with deterministic inputs and deterministic outputs. Functions are combined through simple calls—this is the approach of continually breaking down a program into its smallest units.
Structuring is teleological: sequence, branching, and looping—every control flow serves a clear purpose—the program is “for what it’s intended to do.”
·Object-orientedness is a theory of collectives: a collection of objects each exists and evolves independently, cooperating and competing with one another through messages. There is no central controller; instead, the system’s behavior emerges from these interactions.
The four programming paradigms represent four different worldviews. When a programmer chooses a paradigm, they’re actually selecting which “theory” to use to view the world.
In the history of tools
Some people say: In the past, all human tools were invented using a reductionist approach—every state of these tools could be calculated in advance. AI, however, is the first tool that humanity has trained using a teleological approach—a tool whose states we simply cannot predict.
The author adds: As for tools, all four approaches are available—though artificial intelligence is a black box trained through data, rather than a product of logic itself.
The significance of this observation lies in the fact that it places the contemporary anxiety about AI being “unexplainable” within a philosophical framework.
In the past, machines were reductionist—gears, circuits, and code; each component could be disassembled and scrutinized, so in principle, a machine’s behavior could be fully predicted. AI isn’t “designed”; it’s “trained”: we specify only the goal (the loss function), while its internal structure emerges from the data itself. We’re more like nurturing than assembling.
A teleological tool has a feature that a reductionist tool lacks: it can lead us down paths we never imagined and produce behaviors we didn't anticipate. The supplementary volume will also return to the theme of "Humans and AI."
In the economy and society
The market economy is a theory of groups: countless buyers and sellers each pursue their own interests, with no central commander—prices emerge from competition.
●Enterprise management is holistic: Each department performs its own duties while serving the company’s common goal.
·Stock trading methods can also be divided into four categories: Technical analysis looks at charts, believing that history tends to repeat itself; fundamental analysis focuses on companies, holding the view that market trends are unpredictable and can only be evaluated based on valuation; macro analysis examines cycles and policies; and liquidity analysis scrutinizes the博弈 (game) within the market—this old saying about an industry has itself evolved into a four-box table.
The author has repeatedly emphasized the nature of these correspondences: they are suggestive, not conclusive; at best, they have a success rate of only 55%; and they should be applied in whatever way proves most useful. Their value lies in sparking associations, rather than providing definitive answers.
In the world of success studies
By identifying which grid it falls into, you’ll know which theory it belongs to; by pinpointing which grid is missing, you’ll understand where its blind spots lie. Thus, the Four Theories serve as a “theoretical medical checkup.”
A one-sentence positioning
In the author’s notes, a quote by Moriwa is cited: “Philosophy is the speculative component of psychology.”
This sentence can be read in reverse: Psychology is the empirical part of philosophy. And what the philosophical framework aims to do is to give structure, coordinates, and discussable components to the speculative part as well—so that speculation is no longer merely a wandering of inspiration, but rather an engineering endeavor that can be mapped out with blueprints.
The Four Discussions in the Renovation Team
Even people who don’t program can encounter the “Four Theories”—everyone who’s ever installed or repaired a device has run into them, they just haven’t given them that name yet.
Reductionism in the materials list: how many meters of plumbing pipe, how many square meters of electrical wire, how many buckets of paint, how many boxes of bricks—each item is quantifiable, countable, and interchangeable. Breaking down an entire house into its constituent materials is the first step in renovation.
In the pre-construction discussion, teleology comes into play: “What’s this house intended for? Is it for elderly residents, for children attending school, or for rental purposes? Once the intended use is clearly defined, thousands of subsequent decisions will fall into place. Any renovation that proceeds without first clarifying the intended use will ultimately end up requiring costly rework.”
In the collaborative work of different trades—plumbing, electrical work, carpentry, masonry, and painting—each link is interdependent. If the plumbing and electrical systems haven't been inspected and approved yet, laying tiles prematurely is like planting a time bomb for later problems. Each trade serves the same overarching goal—completing this apartment renovation—and none can be omitted; this reflects the logic of “organs.”
The theory of crowds in the building materials market: Dozens of stalls each offer their own prices and compete independently, with no central authority guiding the process. Prices emerge through a back-and-forth exchange among buyers and sellers. Every single price negotiation you engage in at the market is a daily example of the grammar of crowd dynamics.
When a house is furnished, the Four Discourses take turns being featured. Thus, the Four Discourses represent the “general model”—they’re not limited to philosophy books: wherever there’s form, they resonate there as well.
The Four Paradigms have left an enduring impact across programming paradigms, the history of tools, socio-economics, and the study of success—demonstrating that they represent a “general model”—a structure that repeatedly emerges across diverse fields.
The next chapter will summarize this section and answer the question: What is the position of the philosophical framework throughout the entire book?
Condense this section into a single table.
The philosophical framework is the level of “worldview.”
The four frameworks of the entire book have clearly defined divisions of labor:
·The philosophical framework governs the worldview: What are the different types of forms of the world?
●The Framework of Thinking Governs Epistemology: What Kinds of Thinking Modes Do Humans Use to Grasp the World (Part Two):
●The Psychological Framework Governs Human Nature: The Inner Structure of Humans and Their Classification (Volume One):
The Social and Life Framework Governing Practice: A Person’s Place in Society and Throughout Life (Volume Five).
In the author’s notes, there is a sentence that can serve as an overarching principle for the relationship among the four elements:
The mind and thought are machines, while external information is the object of their processing; philosophy, too, governs external information.
The philosophical framework is thus the “highest and most abstract” level within the entire system.
Precisely because it is the most abstract and highest-level framework, it actually has the fewest direct practical applications—you wouldn't use the Fourfold Model to decide what to have for dinner. But when you need to make a judgment between “disassemble or integrate,” “purpose or composition,” or “whole or group,” the coordinates of the Fourfold Model will give you a sense of direction.
Can four arguments fit into four arguments?
A person who takes things seriously would ask: If the world is merely form, and the Four Discourses analyze all forms—then are the “Four Discourses” themselves a form? If so, where exactly should “reductionism” itself fit into this framework? Can a system really contain itself within its own conceptual framework?
The author’s reply put an end to this self-referential snake: The Four Discourses are not forms themselves—they are tools—thinking tools akin to logic or mathematics. Logic studies thought itself and is not itself included in the classification of objects of thought; similarly, the Four Discourses study forms themselves and are not themselves included in the classification of forms.
He also doesn’t shy away from the few self-referential elements hidden within the system: the conceptual vectors give rise to the Four Discourses and also to a cognitive framework; and within that cognitive framework lies abstraction—precisely the kind of abstraction for which the conceptual vectors themselves serve as a tool. The holistic method of “hidden imagery” (designing the structure inside the black box from outside the black box) is precisely the approach used when the conceptual vectors define concepts. The tools grow out of the system itself, yet they in turn return to participate in the very construction of the system—such entanglements, the author argues, belong to the same family as self-reference phenomena in mathematics (he openly admits his lack of familiarity with mathematics; the mathematical family of self-reference—including Russell’s Paradox, Cantor’s Diagonal Argument, Gödel’s Incompleteness Theorems, and the Halting Problem—can be found in an accessible compendium in Volume Six of this work).
The notion that “the four arguments are tools” was later further elaborated by the author, who laid out the entire genealogy of this conceptual framework in one go: The primordial root is the tool of thought—the thinking tool. Conceptual vectors should be categorized alongside logic—many philosophical ideas eventually evolved into mathematics, and logic serves as a prime example. As for those aspects of philosophy that are less rigorous and more closely tied to concrete human experience, they remained within philosophy itself. Mathematics encompasses many forms: geometry is a form, and so is logic. And these various frameworks are products of the thinking tools—mathematical modeling. He has broadened and extended the term “mathematical modeling”: it’s no longer just about modeling a specific system, but rather about modeling the very form of the entire world.
Thus, to accurately describe the position of the philosophical framework, we need to say it twice: as a tool, it stands alongside logic and mathematics; as content, however, it is far removed from them—while mathematics and logic deal with the forms of thought, the philosophical framework deals with the forms of the real world. It’s distant from thought yet close to the world.
Open questions left by this section
In keeping with the author’s honest style, the unfinished aspects of this compilation are also listed as they are:
The choices among the four dimensions—“simple/complex, unified/decentralized, dynamic/static, internal/external”—are subjective. The author states, “This is my personal perspective; there’s no standard way to measure or judge it.” Is it possible to find more objective criteria for selecting these dimensions? (See the book series “The Inquiry Checklist.”)
In the original drafts of teleology and reductionism, there was a version—before and after—that showed inconsistency in the attribution of “simple/dispersed.” (In the introductory text, reductionism had been described as simple and unified.) This book adheres to the framework table: reductionism = simple and dispersed; teleology = simple and unified.
The correspondences among the Four Discourses and social formations, the Four Discourses and the Eight Trigrams, and the Four Discourses and factions are all draft versions. Their validity remains to be seen; readers and authors are invited to continue refining them.
The philosophical framework categorizes the world. Next, let’s zoom our lens back from the world to the human being himself: What emerges when a group of people comes together? Throughout one’s lifetime, what dimensions are available for us to choose from? Please turn to Part Five, “The Framework of Society and Life: Humanity’s Place in the World.”
The Social and Existential Framework: Humanity’s Place in the World
From one person, to a group of people, and then to a lifetime.
The first four sections discussed humans, thinking, methods, and worldviews; this section answers two very practical questions: What is society (what kinds of things emerge when a group of people come together?), and how should one live one’s life (what dimensions does a person have to choose from throughout their lifetime?).
Humans possess four fundamental attributes: naturalness, sociality, interest-driven nature, and idealism. When these four attributes are scaled up to the societal level, they give rise to society’s four major components: human reproduction, politics, economy, and culture. Each of these components has its own distinct operational framework: in the economy, it’s production → consumption → circulation → distribution; in politics, it’s the generation of power → its implementation → maintaining order → ensuring stability. This book also conducts a cross-disciplinary experiment: applying this very framework to revisit the Five Elements theory of Traditional Chinese Medicine.
The framework of life is a shadow of the philosophical framework: The four attributes of human beings, paired with the two axes—“relying on oneself vs. relying on the environment” and “craving change vs. craving stability”—form a map of life choices. At the end of this book, we address a question that no one can avoid: the meaning of life—and propose a worldview that “explores gene functions comprehensively.”
Chapters 1–5 discuss the social framework (covering four dimensions: politics, economy, culture, and the reproduction of human beings); Chapter 6 explores cross-disciplinary comparisons between the Five Elements in Traditional Chinese Medicine; Chapters 7–9 focus on the life framework and the meaning of life; and Chapter 10 provides a summary. Each chapter can be read independently.
Review: Humans have four natures.
As discussed in the fourth section: this “form” of human beings has four attributes.
·Naturalness: Possesses animal characteristics and instincts—a one-to-one reductionist approach;
·Sociality: Involves roles such as parents and children, superiors and subordinates, and belonging to various social structures—corresponding to holism;
·Idealism: After achieving basic needs, people pursue career and seek to avoid “boredom”—a teleological approach.
Now, let’s replace one individual with a group of people. An intriguing question arises: What has emerged at the societal level from each of the four attributes of human beings?
Social Framework: A Four-Row Table
The author’s answer is a table:
person
Just by reading the first line, you’ll understand how to interpret this table: When humanity’s natural attributes are magnified to the social level, what emerges is “the reproduction of human beings themselves”—birth gives rise to new individuals, work is humanity’s material activity, migration is humanity’s movement through time, and the family is the organizational unit of human beings in space.
For the remaining three sections, we’ll each devote a chapter to them: Politics (Chapter 2), Economy (Chapter 3), and Culture (Chapter 4).
This table also consists of sixteen intersections arranged in a 4×4 grid: the rows represent four social dimensions, while the columns correspond to “action, substance, time, and space”—does this ring a bell? Indeed, this is precisely the row from the sixteen-grid reductionist framework presented in Part Four of the philosophical framework. Each of the social dimensions unfolds along four stages—“action, material, time, and space.” The sixteen squares themselves are the true protagonists of this table; let’s examine them one by one.
Humans (from naturalness to human reproduction):
Fertility (Function): The movement of this line—new people are brought into existence. It’s the starting point of all societal cycles: without it, the three lines below would be mere empty talk.
Work (matter): The material form of human beings—labor. The physical embodiment through which humans act upon the world—the “material” of this human line.
Migration (Time): The Flowing Momentum of Human Beings in Time—Generation after generation, place after place, humans unfold themselves along the timeline.
Family (Space): The organizational unit of human beings within a space—the smallest spatial unit where reproduction takes root. Why is the family considered a “space”—a single grid? Because it is the smallest grid in which humans are situated and placed.
Politics (sociality → circulation of power):
Generation (Effect): The dynamics of power—how power is produced: violence, authorization, tradition, and ballots are all pathways through which power is generated.
Execution (Material): The material form of power—edicts, institutions, and administrative actions—transforms power into visible and tangible actions.
Order (Time): The endurance of power through time—rules are upheld day after day. Order is not merely a piece of paper; it is the daily proof of power’s presence.
Guarantee (Space): The Power’s Hidden Trump Card in Space—The violent institutions stand behind the order, rarely making an appearance in everyday life; when they do step forward, it’s to determine the final outcome. (Expanded in Chapter 2.)
Culture (Idealism → Circulation of Thought):
Creation (function): The act of thought—new meanings are produced: writing books, formulating theories, making movies.
Activities (material): The material form of ideas—rituals, courses, festivals, publications—where meaning becomes something one can actively participate in.
Transmission (Time): The flow of ideas through time—from person to person, from generation to generation—along with the diffusion and preservation of meaning.
Organization (Space): The Spatial Arrangement of Ideas—Schools, Churches, Associations, and Publishing Houses. Once meaning has a home, it becomes difficult to die. (Expanded in Chapter 4.)
Economy (profitability → circulation of money):
Production (function): The act of value—things are created.
Consumption (material): The material destination of value—once an object is used up and a need is satisfied, the cycle is complete. No matter how much is produced, if no one consumes it, the cycle breaks down.
Circulation (Time): The flow of value over time—the process of transportation, trading, and turnover, in which money and goods are constantly exchanged.
Allocation (of space): The placement of value in space—how much each person receives, and the distribution of wealth among individuals and social classes.
Remember that note from the fourth section: Circulation corresponds to processes and to time; allocation corresponds to structures and to space. Here, the “time” stop on the economic line happens to be circulation, and the “space” stop happens to be allocation. The correspondence between these frameworks isn’t merely rhetorical—it’s structural. (Expanded upon in Chapter 3.)
What is this table saying?
It’s saying: Society isn’t one thing—it’s four things.
The reason many debates never seem to end is that the two sides are speaking about different “societies”: One person laments that “society has deteriorated,” referring to culture; the other counters that “society is clearly progressing,” referring to the economy. The two are not talking about the same society at all.
With the four lines separated, the benefits immediately become apparent:
A strong economy does not necessarily mean a thriving culture, and a weak economy does not necessarily mean political collapse.
A person’s position across four different dimensions can be entirely distinct: politically, a commoner; economically, a member of the middle class; culturally, a creator; and within the family, the family’s breadwinner.
The four lines also present a dynamic issue: What happens if one of them breaks? The author’s answer: There would be a chain reaction, but there’s no fixed sequence—how the chain propagates and who goes first remain unclear. Among existing theories, the notion that “productivity determines production relations” can be considered a speculation about one possible transmission pathway.
A test: Can you find the fifth line?
Following the convention of conceptual vectors, after setting up the framework, we should test the blank space: Does society have a fifth constituent element?
You might try looking at: Military? The author categorizes the military under the political axis (as a guarantor of power). Technology? It’s placed under the cultural axis (as the creation and dissemination of ideas). Religion? The author later revised this view: State-sponsored religion is part of politics—after all, the church once had armies and taxes; that wasn’t about ideas, but about power. Only when religious faith sheds its authority does it properly belong to the cultural axis. Law? It’s the enforcement and maintenance of order within the political axis.
The effort to find the fifth line has all fallen back within the bounds of the four lines—this doesn't prove that the four lines are complete, but it at least shows that this table has a pretty big appetite.
A Spring Festival dinner table—with all four dishes fully present.
If you still find the four lines abstract, think about the New Year’s Eve dinner table.
The array of dishes on the table represents the year’s economic harvest; who sits at the head of the table and who takes the first bite reflects the political order; the auspicious words exchanged during toasts and the spring couplets hung after the meal embody our culture. And the reason we gather around this table year after year? It’s because families must carry on from one generation to the next—this is the reproduction of human life itself.
At a single dinner table, four different threads are all in motion. Society isn't that mysterious—it's just a big, amplified New Year's Eve dinner.
The next chapter begins with the toughest of the four lines: politics. What is the essence of politics? The framework’s answer consists of just one word: power.
Political line
The social framework amplifies “sociality” into politics. What politics is all about is:
To put it in plain language: Politics is the study of power, and power itself involves four key aspects: how it arises (who holds power and where it comes from), how it’s exercised (how policies are implemented), how order is established (how power stabilizes and becomes the norm), and how its authority is ensured (with violence serving as the ultimate guarantor).
The use of the four stages
These four stages serve as a measuring tool that can assess any kind of power phenomenon:
To see whether a policy can be effectively implemented, look at its execution.
To see whether a society is stable or unstable, look at its order.
Looking at the ultimate bottom line of order means looking at the guarantors—the military and the police, the monopolists of violence.
The reason many political discussions end up being completely at cross-purposes is that the four key components are mixed together in the conversation: One person questions how power is generated (Is the election legitimate?), while another responds by citing performance metrics (The economy has grown!). It’s a case of talking at cross-purposes, because they’re simply not addressing the same issue.
Readers might ask: Where do familiar concepts like democracy and authoritarianism fit into this framework? The author’s answer: Both fall squarely within the realm of “the genesis of power.” They address the question of “where power originates,” rather than how power is exercised or sustained. Thus, no matter how heated the debate between democracy and authoritarianism may be, it still represents only one-quarter of the political landscape.
Country: Larger grid
The author’s notes also include an unfinished draft on the composition of the nation:
Politics, diplomacy, military affairs, economics—and also society and culture? The exercise of power, ownership, order, and guarantees.
This marks the beginning of imposing the social framework onto the entity of “the nation”: the skeleton of a nation consists of politics (internal power), diplomacy (external politics), the military (the external manifestation of violence), the economy (the circulation of interests), and culture (the circulation of ideas).
It’s still a draft (the author himself has put a question mark next to it), but the direction is clear: a nation is an integrated entity of four social dimensions. To analyze a nation is to examine each of its four dimensions individually—this is precisely how the term “nation” is used in Chapter 6, “Correspondences,” under the entry “Nation” (diplomacy = action, military = material, economy = time, politics = space).
Politics in a Residential Community
“Power” may sound like a matter of national importance, but actually, you can find the whole package right here in your neighborhood.
The election of the owners’ committee is a confirmation of ultimate authority—ownership power belongs to the homeowners, not the property management company. Once the owners’ committee is elected, it issues notices and sets rules; this is the stage of implementation. When illegal constructions are demolished or parking spaces are occupied, it’s the owners’ committee that steps in to enforce order. And underpinning it all—and ensuring that these measures actually work—is the simple yet powerful principle: “If you don’t cooperate, we’ll take legal action.” That’s the guarantee.
If a residential community can’t be managed effectively, it’s 80% likely that one of these four links has broken: either the residents are lax (all rules are ignored), or the owners’ committee talks a good game but fails to take action (resulting in paralysis of implementation), or no one obeys the rules (leading to chaos and disorder), or the property management company is too powerful and untouchable (ensuring that things remain skewed).
Now that we’ve finished the political thread—political power is a cycle—let’s move on to the next chapter, which will explore the cycle of money—the economic thread: production, consumption, circulation, and distribution.
Economic line
When the pursuit of profit is scaled up to encompass society, it becomes the economy. The economy is defined by four well-known terms:
The economy is a cycle of interests: first, goods are produced, and only then can they be consumed; circulation moves goods around, and distribution determines ownership.
The author’s recorded route—only half a sentence is an iron rule:
You can only consume once you’ve produced.
Why does production come before consumption? Consumption without prior production is living beyond one’s means—borrowing from the future and spending it today.
In earlier notes, this statement had a second half—“Only what can circulate can be distributed.” Later, the author retracted this assertion: Once an organization receives funds, it can immediately distribute them within the organization itself, without having to wait for the funds to complete their full circulation. Circulation and distribution are not an iron law in which one must precede the other; rather, they are two adjacent stations in a cyclical process: one governs how goods move, while the other determines who gets the profits.
Four stages, four diseases
This framework can immediately be used to diagnose economic problems:
Production illness: unable to produce (technology, resources, and manpower fall short);
Circulating diseases: Factors are stuck (logistics, cash flow, and information flow are blocked);
●Allocated diseases: Unequal distribution (wealth gap, structural imbalances).
Most of the current debates on economic policy can be summarized as follows: stimulating consumption is the remedy for the second stage; smoothing logistics is the remedy for the third stage; transfer payments are the remedy for the fourth stage; and industrial upgrading is the remedy for the first stage. Saying “the economy is in trouble” is vague; only by pinpointing exactly which link is malfunctioning can we talk about prescribing the right medicine.
A detailed account of an investment
The author also noted a more detailed deduction regarding “Where do the produced goods go?”:
In fact, the resources obtained after production have three destinations: investment, consumption, and accumulation.
Investment is about buying the future; consumption is about sustaining the present; and accumulation is about saving for the future. An economy’s character is reflected in the proportions of these three spending categories: an economy that emphasizes investment is betting on tomorrow; one that emphasizes consumption is living for today; and one that emphasizes accumulation is preparing for the day after tomorrow.
(This clue will reappear in Chapter 5—it just happens to explain the correspondences between several organs in the Five Elements of Traditional Chinese Medicine.)
Fifth element: Inventory
Production, consumption, circulation, and distribution are the four stages of the economic cycle. Yet there’s another position beyond these four stages—often overlooked: inventory.
The products produced flow into inventory, and consumption draws from that inventory. What circulates and moves is the inventory itself; distribution determines who gets access to it. The inventory isn't located at any single point, yet it’s everywhere—acting as the midpoint of this circular loop and also representing its total volume.
Here’s a more general principle worth remembering on its own: Whenever there’s a quartet structure, there’s often a fifth element—a point that can be the midpoint of a number line or the overall entity itself. The four stations represent the process, while the fifth represents the whole that has accumulated throughout the process.
This pattern is not unique to this particular context. In the Luoshu arrangement described in Volume Four, the four “arguments” occupy the four corners, while the four “extremes” are positioned along the four sides; at the very center lies the “form”—the form represents the overarching philosophical framework from which the four pathways originate and give rise to further forms. In the economic structure, inventory itself embodies the “form” of the economic cycle: the four stations revolve around it and also revolve for its sake. When we match the five internal organs with this table in Chapter 5, this central position will appear once again.
Your balcony is economics itself.
The three directions—investment, consumption, and accumulation—you don’t need to look at the national ledger; just take a look at your own balcony and you’ll understand.
The scallions you grow on your balcony—cutting some to eat is consumption; leaving a bunch to keep growing in the ground is investment; and drying them up and storing them in jars for winter use is accumulation.
If you chop it all down to eat, you’ll have nothing to eat next week; if you keep it all for investment, you’ll go hungry today; if you dry it all and store it away, the green onions will never be able to reproduce themselves. Just like a family and a country, the real challenge isn’t making money—it’s getting the right balance among three “pots.”
The economy is a cycle of interests; politics is a cycle of power. The next chapter will explore the third dimension: culture—a cycle of ideas—and the oldest dimension of all: the reproduction of human beings themselves.
Cultural Line
When idealism is magnified to the level of society, it becomes culture. The realm of culture:
Culture is a cycle of ideas: First, someone creates—writing a book, proposing a theory, or directing a film; then comes the activity—ideas are put into practice and rehearsed; next, they spread—from one person to a group of people, from one generation to the next; finally, they become embedded in organizations—schools, churches, associations, publishing houses—the “houses” of ideas.
These four stages can be applied to evaluate any cultural undertaking: If a doctrine fails to spread, it dies; if it spreads but lacks original creation, it becomes mere repetition; and if it has an organization but no actual activities, it becomes an empty shell.
The human reproduction line
The oldest line, originating from naturalness:
The primary task of society is neither the economy nor politics—it is, rather, the reproduction of human beings themselves: childbirth gives rise to new individuals; work is humanity’s material activity; migration is the movement of people through time—from rural areas to cities, from one’s homeland to a foreign land; and the family is the spatial organizational unit of human beings.
This trend is now becoming a global topic: declining birth rates, urban migration, and shifting family structures—all four aspects are undergoing dramatic changes. The framework itself can’t provide you with the answers, but it can show you that these issues are all interconnected—move one piece, and the whole system shifts. For instance, the birth rate has never been just a matter of “having children” in isolation; rather, it’s intricately linked to work (which squeezes time and energy), migration (which pulls people away from familiar social networks), and family dynamics (as family structures become more fluid).
Final assembly of four lines
With the four lines combined, the social framework is complete:
People, power, ideas, and money—the four major cycles of society. In political textbooks, society is divided into four key areas of development: economy, politics, culture, and society—remarkably mirroring the four rows of the societal framework itself (“social development” roughly corresponds to the “people” row in the framework). This is no coincidence: good categorizations tend to emerge repeatedly across different contexts.
Four sticks threaded onto a candied hawthorn skewer.
People, power, ideas, and money—the four major cycles of society. It may sound like the titles of four different courses, but in reality, they’re all spinning on the same street every single day.
A new noodle shop has just opened downstairs from your apartment building: The owner works tirelessly from dawn till dusk—making noodles (economy), handling permits, paying taxes, and complying with regulations (politics), posting videos of noodle-making on short-video platforms (culture), and then heading home in the evening to put away the shop, pick up the kids, and spend time with them (people). One candied hawthorn skewer holds four hawthorns; every day, one person weaves back and forth along these four threads.
In the next chapter, we’ll conduct a fun experiment: Using this framework to “de-duplicate” a two-thousand-year-old body of knowledge—Traditional Chinese Medicine’s Five Elements.
Why touch Traditional Chinese Medicine?
Traditional Chinese medicine is the most divisive topic: some hail it as a national treasure, while others dismiss it as superstition. The author’s stance represents a third way—
First, human thinking cannot be entirely based on science—there are many areas that science simply cannot cover. Second, we should strive to explain Traditional Chinese Medicine (TCM) using both science and philosophy; we must avoid blind adherence. Only if you can personally find a logical explanation within yourself, or if there are successful case studies to support it, should you consider accepting TCM.
The approach he takes is to treat the Five Elements of Traditional Chinese Medicine as a model and “de-duplicate” it using his own framework—to see exactly what the Five Elements are trying to convey.
The Correspondence Between the Five Elements and the Economy
Author's comparison table (revised according to the latest discussion):
Reading: Think of your body as an economic system. The liver governs growth and development—this is “production,” where qi and blood are generated; the lungs govern respiration—this is “consumption”—each inhale and exhale represents a ceaseless expenditure; the heart governs the circulation of blood and qi—this is “circulation,” transporting qi and blood throughout the entire body; the kidneys govern the storage of essence—this is “distribution and reserve”; essence is the very last resource that’s carefully conserved and stored away. Meanwhile, the spleen is known as the “official of the granary”—it’s the “inventory” where ingested food is first transformed into usable reserves.
The corresponding operational flow also holds true: consumption is possible only after production has taken place (“distribution follows circulation”—the latter half of this statement has been withdrawn by the author in Chapter 3)—and inventory serves as the transit hub for all of this: if the spleen fails to transform and transport, all four organs will be left without nourishment. As for the “mutual overcoming” relationship, the author explains that, within two distinct dimensions, the liver and the heart are both complementary and opposite, while the spleen and the kidney are both opposite and complementary—thus, mutual overcoming is simply another way of describing “the mutual restraint between opposing ends on the same dimensional axis.”
The “five” in the Five Elements: It’s not one more—it’s four plus one.
A careful reader will notice: the Five Elements are five, while the economic cycle is four. What should we do about that extra one?
In earlier approaches, the merger was proposed: Among the three destinations of resources (investment, consumption, and accumulation), the lungs were categorized as consumption. Thus, the lungs could be merged into the heart—“There are several different proposals on exactly how to carry out this merger.” While this approach works in principle, it always feels somewhat forced and ill-fitting.
Later, a more natural answer emerged: The “five” in the Five Elements represents four plus one—the “one” being the center. Production, consumption, circulation, and distribution are the four stages of a cyclical process; the spleen corresponds to the fifth element—the element that is neither located at any particular stage nor confined to any single station, yet it permeates everything. Production flows into it, consumption draws from it, circulation moves it around, and distribution determines its ultimate destination.
The principle of this “Fifth Element” is universal (as discussed in Chapter 3—let’s revisit it here): it can be the midpoint of a number line or the overall essence of a given entity. In the Luoshu arrangement described in Part Four, the “Form” placed right at the center represents the overarching philosophical framework—the four pathways (the Four Doctrines) originate from it and give rise to further forms. Similarly, in the human body, the spleen occupies the central position—a single hub that governs the overall circulation of qi and blood, with the four internal organs revolving around it. The Five Elements are not five parallel organs; rather, they constitute “four stations plus one central hub.”
Pay attention to the significance of this methodological approach: He didn’t pretend that “four” and “five” fit together perfectly. Instead, he first acknowledged the existence of the merging proposal and then sought a structural solution that felt more natural. This is precisely what the third section refers to—the correspondence is an initial draft, probabilistic in nature, and whatever works best is how it should be matched. The “five” in the Five Elements itself may well be the result of historical accretion (in the era when there were five celestial stars, “five” was a trendy number). Reading it as “four plus one” reveals its structural skeleton rather than merely the literal five words themselves.
Some readers still disagree: Why not just let the framework accommodate one less grid, instead of insisting on “five accommodating four”? The author’s response is very practical: Because four is easy to use—so we’ll stick with four. To those who’ve mastered the tool of “four,” the five can be broken down into four. But no tool has inherent superiority or inferiority—any tool can be used. Just pick the one that works best for you.
The significance of deduplication
The significance of this practice lies not in Traditional Chinese Medicine itself, but in demonstrating a reading method:
By applying group theory (social framework) to reinterpret the Five Elements in Traditional Chinese Medicine, TCM becomes much easier to understand, and consensus is reached.
An ancient discipline ceases to be mysterious once each of its concepts can be translated into clearly defined concepts within another system. Once we grasp that “the liver = production, the spleen = storage, the heart = circulation, and the kidneys = distribution,” the Five Elements’ cycle of generation and control transforms from a mere incantation into an economic flowchart—both for the human body and for the nation.
The Joy of Translation
The joy of “removing duplicates” is a bit like discovering that dialects from different regions are actually just different ways of saying the same thing.
In Beijing, tomatoes are called “xīhóngshì”; in some places, they’re known as “fānqié,” while in others they’re referred to as “yángshìzi.” Some even call them “chòushì”—but after all the debate, they’re all talking about the same thing. Once you’ve figured out which one it is, the recipes will work seamlessly, allowing chefs from Beijing and Yunnan to prepare the very same dish.
The concepts found in ancient texts are themselves dialects. After translating “liver” as “production,” “spleen” as “inventory,” and “heart” as “circulation,” medical records from two thousand years ago suddenly become capable of engaging in dialogue with today’s economics. Removing duplicates isn’t about negating the ancients—it’s about standing in their shoes.
Apply for a passport.
Having finished discussing the social framework, we also conducted a cross-disciplinary experiment. In the second half of the book, we shift direction—from society back to the individual and the personal life framework.
What is a life framework?
The author’s definition of “life framework” is just one sentence:
The framework of life is the shadow of the philosophical framework.
The philosophical framework describes the form of the world (four arguments, sixteen categories); the life framework, on the other hand, reduces this same set of dimensions back to “a single person’s lifetime.” It has very few components—just three:
The four attributes of human beings: naturalness, sociality, idealism, and interest (old friends that repeatedly appear in Part Four and Part Five);
What can you build with three parts?
Let’s take a look at the second point: Should we rely on ourselves or on our environment? This is the fundamental fork in life strategy: Some people believe “My destiny is in my own hands,” while others believe “Circumstances are stronger than individuals.” Note that this isn’t just a matter of motivational clichés—it’s a genuine divergence in how we perceive the structure of the world. Those driven by internal factors constantly set goals for themselves (though the risk is being perpetually led around by a carrot dangling right in front of their eyes); those driven by external factors go with the flow (though the risk is simply “not doing what can’t be done,” sliding into apathy and giving up altogether).
Now let’s look at the third point: Do we yearn for change, or do we yearn for stability? Those who crave change say, “If nothing ever changes, it feels as if we’re not really alive.” Meanwhile, those who prefer stability argue, “It’s nice to keep things as they are and stay comfortably within our comfort zone.” This idea has also appeared in the philosophical framework of “the dichotomy”—the distinction between motion and stillness permeates both the world and our lives.
Finally, let’s look at the first item. The Four Qualities are the four accounts of life’s content:
Naturalness Account: The grounding of the body, health, and instincts;
Social Accounts: The Structure of Family, Roles, and Belonging;
●Profit-oriented accounts: outcomes of revenue, competition, and collaboration;
●Idealistic account: Not driven by the pursuit of practical interests.
Four Characteristics and Capability Map
The author left a small table, mapping the four characteristics onto the four major areas of self-management:
This may seem puzzling at first glance: after all, time management seems most pragmatic. The author explains that the “time management” he defines is about “where you choose to spend your time.” When less time is spent on the other three aspects, the remaining time naturally becomes devoted to idealism. Thus, its essence isn’t a set of techniques—it’s rather a question of “What do you intend to do with your entire life?” Someone without clear time goals doesn’t need time management; they simply need to be told what to do. Meanwhile, financial management pertains to practical interests, interpersonal relationships to social dynamics, and health to natural well-being—each finds its rightful place.
The four major categories of self-improvement available on the market—fitness, social skills, productivity, and financial management—are precisely the four cultivation paths corresponding to the four fundamental aspects of human nature. According to the “Life Framework” theory, you should pursue all four paths, but you can adjust the order based on your own personality traits.
Why do we understand the道理 but still can’t live well?
Self-improvement books on the market can be neatly categorized into four main types according to this framework: fitness (naturalness), social skills (sociality), productivity (idealism), and financial management (practicality).
Many people “understand the道理 but still can’t live well.” From the perspective of life’s framework, the problem often isn’t a lack of effort—it’s rather that they’re lopsided: They’ve read stacks of fitness books but haven’t touched a single personal-finance book; or they’ve installed eight different productivity tools yet haven’t contacted a friend in a year. These four aspects are connected in parallel—if one line is broken for too long, even if the other three remain bright, there’ll still be a dark spot in the room.
The framework is in place. Can it answer that big question—what is the meaning of life? Next chapter.
First, downgrade the issue.
“The question ‘What is the meaning of life?’ has overwhelmed many people. The author’s first reaction was to downplay it: Actually, the meaning of life isn’t that important. What matters most is how we live our lives.”
Let’s look at another layer: The “meaning of life” is always defined by “others”—it has meaning for oneself, for one’s family, and for society. Meaning isn’t an intrinsic attribute of life itself; rather, it’s an evaluation from the perspective of others.
Deconstructing “Meaning” Through the Four Qualities
Any phenomenon can be analyzed from four perspectives: holism, systems thinking, reductionism, and teleology—though these four perspectives are not rigidly fixed; they’re simply the ones I’ve currently thought of. Humans, too, cannot escape these four dimensions. For humans, these dimensions are sociality, naturalness, idealism, and self-interest.
Thus, the big term “the meaning of life” was broken down into four smaller terms:
·From the perspective of naturalness: to live, to live healthily;
From a social perspective: being needed as a family member, friend, or member of an organization;
●From the perspective of self-interest: In the jungle, securing resources and status (success is merely a small part within this framework);
●From an idealistic perspective: Accomplishing something you’ve always wanted to accomplish.
The author particularly emphasizes a point about contemporary society: success is merely one small aspect within the group’s pursuit of self-interest. Focusing solely on this one aspect reveals a severely narrow mindset—indeed, it could be said that we’ve been brainwashed by society. As for the Indian notion—that “life is about experiencing”—this comes from a psychological (perceptual) perspective, and represents yet another distinct viewpoint altogether.
Another outlook on life: traversal
If “the four-character breakdown” is about defense (breaking down grand terms), then the author also offers an offensive approach—a positive outlook on life called “traversal”:
As a manifestation of genes, humans possess the capacity for survival. To fully explore the functions of one’s genes is to live a fulfilling life: dopamine, endorphins, bodily functions, brain functions... For instance, someone like Wang Sicong, who merely indulges in pleasure without exploring these aspects, has yet to fully tap into many of his genetic potentials—and thus hasn’t lived a truly complete life.
As one among all things, humans can experience the functions inherent in all things: participating in competition, being an integral part of the whole, achieving a common goal, and maintaining good physical health.
How to read it: To live is not merely about “reaping the rewards”—it’s about fully activating every single function of the machine that you are. Think when you should think, compete when you should compete, love when you should love, and create when you should create. Those who simply lie back and enjoy life are like devices that use only 5% of their full potential.
Readers might ask: Can the functions of genes really be “traversed” completely? Could it be that this list is, in fact, infinite? The author’s answer is surprisingly simple: Just look at lions and tigers—every day they engage in only a few basic activities. They do what they’re meant to do, and in doing so, they’ve already traversed their entire genetic repertoire. After all, traversing genes isn’t inherently a complicated task.
The opposite of traversal is specialization: while pets’ function is to please others, some people’s life function has been reduced to mere enjoyment. But the author immediately adds a caveat: specialization is equally important—“In other words, everyone should play their own role well.” Traversal brings richness; specialization leads to excellence—both paths are valid; it just depends on which one you choose.
Feel free to read.
As for meaning, that’s about all a framework can do: break down big terms into smaller ones and change single-choice questions into multiple-choice ones. The rest is up to you—your sorting criteria: four accounts, traversal or specialized approaches, relying on yourself or on the environment, seeking change or seeking constancy.
When you put your answers together, they form the framework of your life.
Social framework—Fruit—The four attributes of a person are amplified into the four components of society:
The Framework of Life—A Shadow of the Philosophical Framework:
Four dimensions: naturalness, sociality, utilitarianism, and idealism;
The Four Areas of Cultivation: Health Management, Interpersonal Relationships, Time Management, and Financial Management.
The Methodological Reflections of This Volume
This volume has done something rarely seen elsewhere: it has placed the two largest concepts—“society” and “life”—both within a four-panel framework. Of course, these concepts cannot possibly be fully explored within just four panels—but the purpose of such a framework has never been to “say it all”; rather, it’s about:
Leak-proof: Talking about society no longer forgetting the cultural dimension, and talking about life no longer being solely driven by self-interest.
Alignment: Treat political ailments and economic ailments separately; manage life’s four accounts separately.
Translation: Traditional Chinese medicine’s Five Elements can be translated as economic cycles, and a nation can be broken down into four lines.
By now, we’ve covered Methods (Part III), the World (Part IV), Thinking (Part II), the Human Heart (Part I), and Society and Life (Part V). All that’s left is one final grand undertaking: aligning them all side by side to reveal their astonishing correspondences—blood types, temperaments, vocal tones, academic disciplines, civilizations, climates, modes of production... all neatly arranged in one comprehensive table.
Please turn to Volume Six, “Dividing One into Four: Framework Synthesis and Correspondence with All Things.”
Divided into Four: Framework Integration and Correspondence with All Things
When all categories are aligned into a single table.
The first five sections respectively cover humans, thinking, methods, worldview, society, and life. In this section, we’re taking a bold step: aligning all of these concepts and then comparing them against existing classifications from around the world (such as MBTI, blood types, temperaments, vocal tones, academic disciplines, civilizations, climates, modes of production, and more) to create one comprehensive table.
If the cognitive framework, psychological framework, and philosophical framework each stand independently, is the correspondence among them merely a coincidence? If MBTI’s four types, the four temperaments, the four blood types, the arts vs. sciences, Confucianism, Taoism, Legalism, and Buddhism, as well as hunting-gathering, agriculture, and pastoralism—all exhibit a similar four-part structure, what does this suggest?
This book first establishes a fundamental rule: “correspondence” is not “definition.” Correspondence is probabilistic, with a maximum success rate of only 55% at worst—it serves as a heuristic rather than a proof. Under this rule, the book presents the author’s comprehensive “Divided into Four” master table, offering a line-by-line interpretation and tracing out the famous civilizational chain: from East Asia’s humid and cold climate, through the rise of agricultural civilization, to holistic social organization, to linear thinking, and finally to the humanities-oriented temperament.
This book also addresses three extended issues: the interpretability of metaphysics (the translation chain between星座 and MBTI—a psychological framework-concept vector, along with a corresponding mapping later retracted by the author himself); repositioning psychology’s seven major schools of thought within a three-framework model; and “multiple personalities”—the notion that human types are not singular but rather multi-layered waves superimposed upon one another. The book concludes with four authentic framework questionnaire samples and a list of honest boundaries.
Chapters 1–2 present the overarching principles and rules; Chapters 3–4 introduce the three-framework alignment and the grand overview of “dividing one into four”; Chapter 5 explores the chain of civilizations (from climate to the humanities); Chapters 6–7 delve into two applications (the interpretability of metaphysics and the seven major schools of psychology); Chapters 8–9 examine multiple personalities and empirical questionnaire studies; and Chapter 10 summarizes the ambition and limitations of this framework.
Reading this book requires a bit of patience—and a touch of playfulness. It’s a large-scale “find the isomorphism” game.
The origin of the framework
In the author’s notes, there is a brief self-description of the “framework synthesis” worth quoting in its entirety—these few sentences embody the spirit worth recording in full:
The three-framework origin lies in the fact that modern science cannot clearly explain how the brain and mind work.
With the tool of concept vectors, we’re eager to tackle even more problems. Moreover, thinking and psychology are not easily separable.
The result-oriented philosophical framework can once again return to the human being itself and acquire human attributes. It’s truly remarkable—this allows for a more comprehensive description of what it means to be human.
20230609: The three-frame system has been changed to a four-frame system.
A few sentences can succinctly summarize the entire system’s developmental history: First, to tackle the challenge of “the inexplicability of thought,” we created a framework for thinking. Then, we realized that thought and psychology are inseparable, so we added a psychological framework. Next, we discovered that the same set of dimensions could be used to describe the world itself, leading us to construct a philosophical framework. As this philosophical framework was reflected back onto human beings, it gave rise to the four fundamental attributes of humanity. Finally, when we extended this framework to encompass society and life itself, we arrived at a fourth and fifth framework.
The system’s growth isn’t planned—it emerges like a snowball rolling downhill, as soon as a tool becomes handy, you’ll want to use it to solve more and more problems.
Division of labor among the four frameworks
The author also offers a meta-level metaphor: the psyche and thought are like machines, external information is the material these machines process, and philosophy serves as the overarching framework that directs and organizes this external information. To put it in more technical terms: neurons represent the micro-level, thought occupies the meso-level, the psyche embodies the macro-level, and culture is the data loaded onto the system after birth.
A handy screwdriver
Why does the system snowball? Anyone who has a handy screwdriver at home understands.
The day you first brought it home, all you wanted to do was tighten that one loose screw on the chair. But the screwdriver was so easy to use—comfortable in your hand and with a blade that gripped firmly—so you looked around: The door handle was loose, so you tightened it; the outlet cover was wobbly, so you tightened it; even the kid’s toy car and the kitchen utensils...
Shelf after shelf... By the end of the week, you’ve tightened every single screw in the entire house. It’s not that there are suddenly more screws around the house—rather, once your tools feel natural and effortless in your hands, you suddenly “notice” them everywhere.
The author’s tool is the concept vector. At first, he’d only intended to use it to tackle that one particular “hard-to-articulate” screw; but once he’d unscrewed it, he realized that psychology, philosophy, society, and life itself—all were riddled with screws of the same type. That’s how the four frameworks came about—not planned out in advance, but rather like a handy knife that he kept using as he worked his way through them.
Why align?
After the four frameworks were established, a natural impulse arose: Could they be aligned with each other?
If we could align them—say, the “discrete” nature of the cognitive framework precisely corresponds to the “perception” of the psychological framework, and the “invariance” of the philosophical framework—then it would imply that these four frameworks might not be four separate theories at all, but rather four manifestations of the same underlying structure unfolding across four distinct levels. A single structure manifesting itself four times is far more credible—and far more concise—than four separate structures each standing on their own.
That’s exactly what this section is about. But before we proceed with alignment, we first need to establish clear rules—because “correspondence” is the operation most easily prone to abuse. The next chapter will focus on laying down these rules.
Rule corresponding to Chapter 2: Inspiration, not proof.
Corresponding = Definition
In the entire system, what’s most likely to be misinterpreted are those elegant correspondence tables. The author has specifically established a rule:
Corresponding and synonymous words are different. Synonymous words have essentially the same meaning; in contrast, corresponding words often come from different domains but may share some logical causal relationships or conceptual vector similarities. This correspondence is probabilistic—only about 55% of such correspondences are likely to hold true.
I think it’s important to distinguish between definition and correspondence. As for definition, it’s 100%; as for correspondence, it involves a certain degree of probability.
Things that are computable and deducible have necessary connections; whereas connections like “a certain MBTI type exhibits certain behaviors” belong to the universal interconnectedness and mutual influence among phenomena—where one event is affected by numerous factors, each with varying degrees of weight, and the chain of causation is complex and unreliable.
What does 55% really mean?
Let me first clarify the number itself. The author later admitted: 55% was merely an example. The whole point is that it represents “above random”—if “random” means 50% in a coin toss, then 55% is already not random; if “random” means 25% in a four-choice scenario, then 26% or 27% would also no longer be random. This number makes no claim to precision—it only promises direction: better than pure guesswork, but weaker than an ironclad rule.
Under this clarification, “the worst-case scenario is only 55%” means:
Correspondence cannot be used as a basis for argumentation. “The Confucian holistic approach cannot serve as the basis for any conclusion.”
You can discover this through correspondence. It suggests that the two domains might be isomorphic and worth investigating—once confirmed, the correspondence is elevated to a formal definition; if not confirmed, the correspondence is downgraded to a metaphor.
The correspondence is a living draft. The author has demonstrated this kind of honesty: in the entry “Why Confucianism Corresponds to Individual Change,” he himself lists several conflicting reasons (e.g., Confucianism is outwardly active, thus implying individual change; yet Confucianism also emphasizes politics, which seems to suggest an unchanging whole...). Finally, he concludes, “You’ll have to judge this correspondence for yourself—use whatever works best, and feel free to revise it repeatedly. For now, this is just a rough draft.”
There’s also a deeper methodological awareness: if there’s a 100% correspondence, then it’s no longer a single gene. Perfect alignment implies redundancy—two frameworks would become one and the same thing. The value of correspondence lies precisely in the fact that it represents “similarity between different things.”
Is there a “hard” component to this system? The author’s answer is unequivocal: the positions are hard, while the words are soft. The sixteen positions simply exist—this is a hard deduction; but which word fits which position exactly requires repeated reflection and constant refinement—this is the soft correspondence. To put it in the words of the third section: the structure is definitional, whereas naming is probabilistic.
A contrasting example: Monism and Dualism
The author provided a correspondence table for “monism/dualism,” which can serve as an example of the “correct usage” of these terms:
The way to read this table is not to conclude that “all leftists are in control” (which would be an abuse of logic), but rather to recognize that the psychological structure known as “monism” can manifest itself—in various domains such as prediction, decision-making, politics, and worldview—as optimism, a lack of risk awareness, a demand for equality, conformity, and a sense of purpose... There’s a probabilistic undercurrent linking these traits. It’s precisely from this perspective that the author critiques monists: monists fail to see that everything has both advantages and disadvantages and tend to believe that only they themselves are right. Those who are capable of employing categorical thinking should, by contrast, be anti-monist.
You can only believe half of what the introducer says.
“This rule—correspondence is inspiration, not proof—is something marriage agencies understand best.”
When a matchmaker introduces two people, their opening remarks always involve analogies: “This young lady has such a lovely personality—just like your cousin.” “That young man is so down-to-earth—he’s exactly like your dad was when he was young.” Can you really take these words at face value? You can trust them—at least halfway. What they’re really saying is this: There might be some familiar pattern or structure in this stranger that’s worth exploring further—and that’s precisely the clue that might prompt you to give this person a chance.
But if you take the analogy as a conclusion—“Since she’s like my cousin, they’ll definitely get along”—you’re likely to be disappointed. The resemblance is only 55%; the remaining 45%? You’ll have to sit down and eat three meals together to find out. Even the matchmaker knows this clearly: the purpose of the analogy is to give you a gentle nudge into the right direction—not to do the actual living for you.
This is the proper way to approach “correspondence”: It’s a stepping stone, not a guarantee of stability. All those elegant correspondence charts throughout the book should be read with this mindset—treat them as introductions, not as marriage certificates.
Telescopes and Microscopes
Now that the rules are set, let me wrap up with a metaphor: The right approach is like a telescope—not a microscope; it’s the starting point for exploration, not the endpoint for proof.
A telescope lets you spot “what looks like a mountain in the distance”—but to find out what’s actually on that mountain, you’ll have to go there and take those few steps yourself. It’s precisely those few steps you take that constitute real research. Treating the mountain image seen through the telescope as if it were the mountain itself is exactly how constellation columns are written; but setting off to investigate once you’ve spotted that mountain image—that’s the proper way to use this framework.
With this rule in mind, the next chapter will officially begin aligning the four frameworks.
Align large tables
This is the heartland of the entire system: the dimensions of the thinking framework, psychological framework, and philosophical framework, aligned into an eight-row table.
Reading: Each line represents a “cross-framework isomorphism.” For example, the first line states: “Consciousness in thought, rationality in psychology, and simplicity in philosophy—are all manifestations of the same structural tendency at three different levels.”
Line-by-line interpretation (with the author’s explanation)
The author left several explanations in the notes, which we’ve embedded in each line:
Consciousness = rationality = simplicity. Consciousness and rationality both pursue clarity, singularity, and definiteness—simplicity is their worldview’s flavor.
Memory = Sensibility = Complexity. “Memory doesn’t need logic.” Memory is a warehouse that accepts everything whole; sensibility is emotion that defies logic; complexity is a world that refuses to be simplified—each of these three rejects the approach of “stripping away the complex to arrive at simplicity.”
Generation = spontaneity = dispersion. "Spontaneity gives rise to creation." New things always emerge from places that don't adhere to rules; dispersion means there is no unified will—and precisely because there is no unified will, room is made for the unexpected.
Operation = Control = Unity. Operation means making good use of what already exists; it requires discipline and focus. Control and unity are closely interconnected, sharing the same vital essence.
Discrete = Perception = Invariance. Discrete thinking views the world as a collection of points, where each point is static; perception is the straightforward reception of the world as it is.
Continuity = Action = Change. “People with a continuous mindset have an advantage in action: because of continuity, they don’t need to overthink, allowing them to act swiftly.” Continuity is flow; flow is change; and change is action.
Hierarchy = Introversion = The Whole. “Introversion cares about the whole but not about individual differences—just as with hierarchy, it involves viewing issues from a holistic perspective.” Hierarchy implies taking a bird’s-eye view; taking a bird’s-eye view means embracing the whole.
Flatness = Extraversion = The Individual. “In the philosophical framework, the interaction among individuals corresponds to extraversion in the psychological framework.” To a flat person, there are no superiors or subordinates—there’s only one level playing field of individuals; for an extraverted person, every interaction involves a specific, concrete individual.
What is this table claiming?
It claims that how a person thinks (cognition), how they feel (psychology), and what they believe (philosophy) are all underpinned by the same deep-seated configuration.
What is the basis for this claim? The author’s answer is causal, not rhetorical: just like the geographical determinism we’ll discuss later, these alignments are underpinned by chains of causality. For example, “People with flat thinking tend to be obedient”—those who can’t see the bigger picture are even more in need of guidance—a psychological causal link that welds “flatness” and “obedience” together. Thus, this alignment isn’t merely a wordplay involving “the same author naturally self-governing within the same system”; every single line asserts the existence of a causal pathway that is both debatable and testable.
A person with “strong consciousness” is likely to be rational and tends to prefer a simple worldview; a person with “strong memory,” on the other hand, is likely to be emotional and tends to believe that the world is complex. Pay attention to the wording—“likely,” soft causality, and the 55% warning all apply.
If this framework generally holds true, it leads to a practical conclusion: By observing a person’s performance in any one of the three dimensions, we can make rough predictions about their other two dimensions. For instance, by looking at how someone organizes their notes (thinking), we can guess their spending habits (psychology); by observing whether they believe in fate (philosophy), we can infer their habitual approach to planning (psychology). This kind of “cross-dimensional prediction” is precisely what makes the entire system both most intriguing—and also the part that requires the greatest care in its application.
A live demonstration of “cross-layer prediction”
The most fun way to use this table is to conduct experiments on acquaintances.
I have a friend whose note-taking app is filled with mind maps—he never makes to-do lists—typical of someone who thinks in visual patterns. When I asked him how he predicts things, he said, “Most likely, I’m intuitive; most likely, I feel the world is too complex to predict with certainty.” Sure enough, that turned out to be true: his go-to phrase when making decisions is, “I’m just going with my gut,” and his closing remark whenever we discuss global affairs is always, “It’s not that simple.”
Of course, there are times when predictions turn out to be wrong—that’s precisely what the 55% figure means. If you guessed correctly, it suggests the hidden clue might actually exist; if you guessed incorrectly, it serves as a reminder that there could be other variables at play with that person. The game is fun to play—just don’t take the conclusions too seriously.
The eight dimensions have been aligned. But the author didn’t stop there—he took the alignment even further, extending it to MBTI, tone of voice, blood type, temperament, academic disciplines, and civilizations. In the next chapter, take a look at that even larger table.
The most famous table in the system
On April 17, 2020, the author completed the alignment work and produced the largest correspondence table in this system—“Divided into Four:”
How do I read this table?
Read vertically: All the rows in the same column have a “corresponding relationship” with each other (the kind associated with 55%). For example, in Column C—continuous flatness, linear thinking, extroverted action, EJ, sensory-control, ST, non-self, single-syllable, sociability, complex unity, holism, changing individuality, Confucianism, humanities, East Asia, damp and cold climate, agriculture, melancholic temperament, Type A—these are all placed in the same column, suggesting that they share structural similarities.
Read horizontally: The four boxes in the same row represent the four quadrants of a certain classification system. These quadrant systems come from completely different fields: psychology (MBTI), linguistics (tones), physiology (blood types), geography (climate), anthropology (modes of production), and philosophy (the Hundred Schools of Thought).
What this table claims is: These four divisions across different domains all refer to the same “four.” The fourfold structure of the world repeatedly emerges at various scales—this is precisely the “general model” definition cited in Part Three: When a particular structure independently emerges multiple times across different domains, it reveals some deep-seated, universal law.
A few points worth mentioning
The MBTI types—J/EP/EJ/IP and SF/NF/ST/NT—are two quartiles of the 16 MBTI types (one divided according to E/I × J/P, the other according to S/N × T/F). The 16 types = four quartiles × four quartiles, neatly fitting into four columns.
Temperament types: Choleric, Sanguine, Melancholic, Phlegmatic—this is the ancient Greek theory of the four humors, a “division into four” dating back two thousand years.
The blood-type personality theory—asserting that blood types O, AB, A, and B correspond to specific personality traits—has received virtually no support in mainstream research. The author’s confidence stems from two sources: First, his methodological stance—that everything is interconnected, and “mainstream research is overly dogmatic about science”; second, a rhetorical question: Can mainstream research truly demonstrate that there’s absolutely no correlation between blood type and personality? Of course, he himself acknowledges that this correspondence is unreliable and may require constant revision and adjustment.
Tone Row: The first, second, third, and fourth tones of surnames—this is the most surprising row. Later, the author provided an explanation: Tones represent emotions; surnames with different tones are borne by people with different emotions; and people with different emotions are of different types. As for the source of evidence? He claimed it was his own “big-data observations,” and even offered concrete examples: Surnames with the first tone—such as Zhang and Zhou—those he observed tended to have minds that were predominantly continuous and flat (linear), precisely falling into Column C of this table. He also stated that, to date, he has not encountered any real-life counterexamples to this pattern. These two rows remain the softest in the entire book; when you read them, be sure to keep the rules from Chapter 2 handy.
The correct approach to handling this table
Let me reiterate the rule from Chapter 2: This table is an exploration map, not a judgmental verdict. Its proper use is to generate hypotheses—“Type A and the melancholic temperament both correspond to Column C; does the description of Type A blood really carry a hint of the melancholic temperament? Let’s take a look.” Its improper use, on the other hand, is to draw conclusions—“Therefore, blood type determines personality.”
The author’s own attitude toward it can be seen in the entry “Why Confucianism Responds to Individual Change”: The author lists both supporting and opposing arguments, acknowledges the draft nature of the work, and welcomes repeated revisions.
What does this table read like?
The first time you look at this chart, you’ll probably be reminded of your experience browsing a flea market: On one stall, there are pocket watches, abacuses, ration coupons, and old telephones—each from a different era and serving a different purpose. Yet when they’re displayed together, they somehow evoke an indescribable sense of “cohesion.”
The difference is that the sense of completeness in a flea-market find relies on luck, whereas the sense of completeness in this table relies on coordinates: each row answers the question, “In which quadrant—simple vs. complex, unified vs. fragmented—do you find yourself?” Thus, this isn’t a collection; it’s an index.
On the table, there’s a single row containing an entire chain of reasoning: Confucianism, humanities, East Asia, damp and cold climate, agriculture, and holism—why do they all appear in the same column? The next chapter will unfold this chain; it serves as the “model project” for the entire table.
In the master table from the previous chapter, there’s one row that stands out particularly:
Confucianism, humanities, East Asia, damp and cold, agriculture, holism—these are all listed in the same column.
This wasn't something the author just jotted down casually. In his notes, he specifically wrote out an entire chain of reasoning:
Confucianism is an East Asian culture; East Asia has a humid and cold climate; the humid and cold climate gave rise to an agricultural civilization; the agricultural civilization led to a holistic form of social organization; and perfectly suited to this holistic organizational structure is a continuous, flat (linear) mode of thinking—precisely the kind of thinking characteristic of the humanities.
Six stages—each one progressively tracing the thread from weather all the way to the intrinsic character of the discipline. This chapter dissects this chain, examining it link by link. It serves as the “model project” for the entire table: if this particular chain holds water, then aligning the other columns in the table will no longer be a matter of guesswork—it will become a hypothesis that can be questioned and tested.
5.1 First Ring: A Damp and Cold Climate
The starting point of the chain is climate. The author further divides climate into four types: hot and dry, hot and humid, humid and cold, and dry and cold—combining two binary dimensions to create a fourfold classification.
East Asia (eastern China, Japan, and the Korean Peninsula) has a humid and cold climate: abundant precipitation, distinct four seasons, cold winters, hot summers, and high humidity.
Why can climate serve as a starting point? Because in the early stages of civilization, climate was the one thing humans had the least ability to resist. You could move house or change your political system, but you couldn't stop the rains from falling in the Yangtze River basin. Climate is a given constraint—everything that follows is built upon it.
5.2 Second Ring: From Damp and Cold to Agriculture
A humid and cold climate is ideal for farming. Abundant water, frosty periods, and distinct four seasons—this is precisely the perfect setting for rice and wheat cultivation.
Compared with the other three zones, it becomes quite clear: dry and hot regions are ideal for hunting, while humid and hot regions are suited for gathering (in tropical rainforests, fruits are available year-round, making farming less necessary). Dry and cold regions are perfect for pastoralism (grasslands don't grow crops—only grass grows there, and the grass feeds livestock). Four climates, four modes of production—another case of division into four.
East Asia has been “locked” into the category of agriculture.
5.3 The Third Circle: From Agriculture to Holism
This link is a crucial leap in the entire chain: How does the mode of production influence one’s worldview?
Agriculture has several characteristics.
First, agriculture is a collaborative endeavor. Building irrigation canals, constructing dams, and rushing to harvest and replant—all these tasks cannot be accomplished by individual households alone. This is especially true when it comes to flood control—take the Yellow River and the Huai River, for example: upstream and downstream regions must coordinate closely. If one village builds a dam, the resulting floodwaters could inundate neighboring villages. No one can remain unaffected on their own.
Second, agriculture is a long-cycle endeavor. You sow seeds in spring and harvest in autumn; during the intervening months, you can’t just up and leave, nor can you switch to another line of work. People are tethered to the land, bound to the agricultural calendar, and locked into the rhythm of “the whole year as one unified game.”
Third, agricultural output depends on the state of the entire system. Whether a harvest is good or bad doesn't just depend on how diligently you tend your own plot of land—it also depends on the weather, water availability, pest infestations, and the conditions of neighboring fields. Local optimization is useless; only when the overall system performs well can we truly say it’s good.
Living under such a mode of production for thousands of years, a society will naturally evolve an organizational structure that prioritizes the “whole”: the family-state isomorphic structure, collectivism as the guiding principle, and putting the bigger picture first. This is holism—a complex unity that focuses on the individual’s position and changes within the larger whole.
Just compare the two—and you’ll see the difference: Hunting emphasizes individual bravery and timing (a reductionist temperament: simple, fragmented, focused on a single target); pastoralism involves moving with the water and grasses, with highly mobile individuals (a holist temperament: complex, dispersed, where individuals face an unchanging environment); and gathering leans even more toward a teleological approach.
5.4 The Fourth Circle: From Holism to Linear Thinking
Worldviews filter ways of thinking. This is a principle repeatedly employed throughout this book—Part Four discusses the philosophical framework, while Part Two explores the cognitive framework; both are consistently listed together in the alignment table found in Part One and this part.
How does a holistic society think? To fit the “complexly unified” world into its mind, the most suitable approach is a continuous, flat one: The world unfolds continuously (rather than in discrete, grid-like steps) and exists as a seamlessly interconnected whole on a single plane (rather than as a hierarchical pyramid). This continuous, flat perspective is precisely linear thinking.
The hallmark of linear thinking is to trace a narrative from beginning to end, explaining the full context and relationships, and placing great emphasis on “insight”—gaining an overall understanding through continuous storytelling rather than dissecting a single definition in isolation.
5.5 The Fifth Circle: From Linear Thinking to the Humanities
If we move further down the spectrum of thinking styles, all the way to education and disciplinary ethos, we arrive at the humanities.
What is the mindset of the humanities? Reading long texts, grasping the overall structure, understanding the context, sensing implied meanings, and comprehending specific details within the broader narrative—this is precisely what we call continuous, flat, linear thinking. In the master chart, the author has also arranged “learn, comprehend, think, and apply” into four separate boxes; the humanities column corresponds exactly to “learn.”
Thus, the chain is complete: damp and cold → agriculture → holism → linear thinking → humanities-oriented temperament → Confucianism.
Why is Confucianism included in this category? Because Confucianism has been the operating system of China’s two-thousand-year-old agricultural civilization: it emphasizes the Five Relationships (the position of individuals within a web of social connections), the concept of family and state (where the individual is embedded within the larger whole), harmony (with the whole taking precedence over the part), and moral education (through continuous verbal and practical example rather than discrete rules and regulations). It represents the complete unfolding of holism in ethics and politics.
5.6 How to Use This Chain: Three Reminders
This necklace is beautiful, but when you wear it, you must put on three “safety helmets”—otherwise, it could turn from inspiration into misguidance.
First, it’s a probabilistic chain, not a deterministic one. Each link represents “more likely” rather than “certain.” Even in wet and cold regions, there are pastoral communities; in agricultural societies, reductionist thinkers emerge as well; and among humanities students, many tend to think in hierarchical terms. The chain describes a shift in statistical centrality—just like saying, “Southerners love to eat rice”—which is broadly true, yet often inaccurate when applied to any particular household or individual case.
Second, it’s a historical snapshot—not an eternal label. The chain of events explains the formation of “civilizational genes,” yet these genes tend to lag behind reality. Today’s China has long since become industrialized and information-driven; its modes of production have changed, but the inertia of its cultural psychology remains. It’s like how a person’s childhood can influence their entire life, yet it doesn’t determine every single step they take.
Third, it is an interpretive tool, not a value hierarchy. To say that “East Asia tends toward holism and linear thinking” does not at all imply that European-style reductionism and tabular thinking are either superior or inferior. These four approaches represent four sets of adaptive strategies that humanity has developed in response to four different existential conditions—each with its own strengths and domains of expertise. Whether it’s the fine-grained analysis of division of labor (reductionism) or the systemic grasp of interconnectedness (holism), civilization cannot advance far without either one of them.
Fourth, can counterexamples be digested? East Asia has also produced scholarship akin to textual criticism, and Europe boasts a rich tradition of holism—yet the author’s verdict is this: textual criticism is not reductionism; rather, it reflects an emphasis on history. Reductionism is defined as the downward analysis into physics and mathematics, whereas textual criticism does not break things down to that level. As for the notion that “a single country can accommodate all schools of thought,” he considers this perfectly normal: absolute 0% and absolute 100% are exceedingly rare; the continuum merely describes mainstream tendencies.
Fifth, how large is “large” enough? Even under similar conditions of dampness and cold, the social temperaments of Japan and China differ significantly. Similarly, despite both being agricultural societies, China exhibits substantial north-south differences—so at what scale does the chain begin to break down? The author offers geographical isolation as a criterion: the effective scale of the chain is determined by the degree of isolation among geographic units. As for China’s north-south differences, he provides further analysis in “Superposition of Waves” (Chapter 8). What makes China unique, however, is its vast territory, long-term political unity, and deep-rooted Legalist influence—not a purely feudal system.
The most fun thing about this chain is that it gives “regional snark” a chance to level up.
Next time you hear someone say, “Chinese people are just like this and that,” or “Westerners are just like this and that,” you can break down their statement and ask: Which link in the chain are they referring to? Is it climate? Mode of production? Social organization? Way of thinking? Disciplinary ethos? What’s the probability of each link being valid? And does this entire chain still hold true today?
Most people, when they label something, flatten an entire chain into a single assertion and toss it out. But right now, you’ve got six distinct steps in your hands—you can examine them one by one.
The label is a compressed file, and the chain is the decompression tool. This chapter doesn't teach you a new label—it teaches you a decompression action instead.
If the civilization chain from the previous chapter holds true, it leads to an ambitious inference: those things regarded as “esoteric” may not be entirely without merit—perhaps the explanatory chain simply got broken. In this chapter, we’ll conduct an experiment using constellations.
6.1 A Bold Hypothesis: Constellations and Climate
The author's exact words are:
The zodiac is not determined by the stars’ positions, but rather by the temperature and humidity of the climate at the time of birth, as well as the seasonal influences on food and living conditions.
The meaning of this sentence is: If there really is some statistical correlation between星座 and personality, then what’s likely at play isn’t the stars themselves, but rather the season of birth.
The reasoning goes like this: Babies born in different seasons experience vastly different climates during their first few months of life. Babies born in winter spend their first three months in cold weather, wearing heavy clothing, spending little time outdoors, and having mothers whose diets consist mainly of stored foods. By contrast, babies born in summer experience the opposite conditions. Yet it is precisely during these first few months of life that the nervous system undergoes its most rapid period of development.
Thus, a short chain of explanation was formed:
Season of birth → Early climate and food, living conditions → Early developmental environment → Statistical shift in temperament.
Note that this remains a hypothesis—and a weak-effect hypothesis at that. At most, it can explain only a slight statistical tendency; it cannot account for any individual case. However, its value lies in the fact that it transforms the untestable claim—"stars influence personality"—into a potentially testable one: "seasons influence early development." The latter at least puts us on the threshold of scientific inquiry.
The author’s attitude toward this hypothesis is even more cautious than the words he has written. He admits frankly: “I haven’t done much research on constellations, and there’s no authoritative theory about them to begin with. This is just a guess at the underlying cause—though it might be right, or it might be wrong.” He goes even further: In his early years, he tried matching the Four Symbols with the MBTI (see 6.2), but later realized that the Four Symbols didn’t correspond to the seasons—at this point, he completely rejected that part of his analysis and has since withdrawn it. As for any unexplained residuals, rather than sweeping them under the rug, he simply cuts them out altogether.
6.2 A corresponding entry that was withdrawn by the author himself
This section originally documented an attempt to match the Western astrological “Four Symbols” with the four-letter categories of MBTI.
Western astrology divides the twelve zodiac signs into four elemental groups based on the "four elements": the Fire signs (Aries, Leo, Sagittarius) are passionate; the Earth signs (Taurus, Virgo, Capricorn) are practical; the Air signs (Gemini, Libra, Aquarius) are curious; and the Water signs (Cancer, Scorpio, Pisces) are emotional. Each of these four elemental groups governs three zodiac signs, which again suggests a "four-part structure." As a result, some people—including early authors—have attempted to correlate this fourfold division with the four-letter categories of the MBTI personality typology.
The author has now explicitly retracted this correspondence. The reason is straightforward: the Four Symbols don't align with the seasons. If the meaning of zodiac signs truly lay in the birth season (as suggested by the June 1 hypothesis), then the three-season categorization—“creative, fixed, mutable”—would at least still make some sense. But the Four Symbols span across seasons instead—thus, the chain breaks here. Once it’s broken, there’s no repairing it, no trying to salvage it; the author simply withdraws it.
This withdrawal remains in the book—not as a flaw, but as an on-the-spot implementation of the rules laid out in Chapter 2: The corresponding entry is a living draft. Once it’s found to be invalid, it’s downgraded or revoked. Being able to locate in the book the “corresponding entry withdrawn by the author” demonstrates even more vividly that this method’s teeth are firmly rooted in its own flesh than merely finding the “entry the author was proud of.” The explanatory chain approach in the next section remains unchanged—this particular case precisely illustrates what to do when the chain breaks.
6.3 The Chain of Interpretation: Why Can Occultism Be “Translated” Layer by Layer?
What this chapter really wants to convey is not constellations, but rather a method: the chain of interpretation.
The author mentally set up a string of translators like this:
Constellation ← MBTI—Psychological Framework ← Concept Vector
This means: The personality traits described by astrological signs can be restated using MBTI typology language; MBTI types can be explained in terms of the four dimensions of the psychological framework—extraversion vs. introversion, sensing vs. intuition, thinking vs. feeling, and judging vs. perceiving—which together constitute the structure of these types; and the four dimensions of the psychological framework themselves are products of combinations of conceptual vectors.
Each level builds upon the one below, becoming increasingly rigorous with each step. Constellations are empirical descriptions of phenomena drawn from folk experience; MBTI is a semi-standardized questionnaire-based classification system; psychological frameworks are dimensional structural models; and conceptual vectors represent a method based on generative models. The further to the right you go, the stronger the explanatory power becomes, and the fewer the assertions made.
The value of this approach lies in the fact that it doesn't readily dismiss any form of folk wisdom; rather, it asks: “What more fundamental concept might this be an approximation of?” Not all esoteric practices are scams—many of them represent summaries of phenomena from the pre-scientific era. Ancient people observed certain statistical patterns but lacked the tools—no variables, no models—to fully understand them, so they ended up attributing these patterns to the stars. The stars may have been wrong, but the underlying patterns themselves might have contained a grain of truth.
Of course, conversely, this approach by no means endorses the occult. Whatever can be translated, we keep; whatever can’t be translated, we discard. As for those specific daily horoscopes and compatibility scores found in astrology—since they don’t find any meaningful foothold in the explanatory chain—we’ll just leave them in the entertainment section.
The stance of this chapter can be summarized in one sentence: Regarding metaphysics, we neither bow down nor burn it at the stake—we instead subject it to interrogation.
The interrogation method is to ask repeatedly: What phenomena have you observed? Is it possible to explain this phenomenon using more fundamental variables? How far can we go in explaining it? And what’s the size of the remaining residuals?
In fact, the history of science has always been engaged in precisely this kind of inquiry. Alchemy was interrogated into chemistry; astrology was interrogated into astronomy; homeopathy was interrogated into the placebo effect. While interrogation doesn't guarantee that every accused person is innocent, it does ensure that no one who knows the truth will be wrongly executed.
The zodiac may have brought only a tiny bit of truth—the tiniest hint of the season of one’s birth. As for the Four Symbols chart that has been circulating for two thousand years, the author’s decision was to withdraw it. The conclusion of the interrogation could also be: “Not established—this is precisely the difference between an interrogation and a ritualistic homage.”
6.4 By the way, the I Ching
Beyond astrology, the author offers a non-superstitious interpretation of the I Ching: Yin and Yang are simply opposites. From Yin and Yang to the Eight Trigrams and then to the 64 Hexagrams, we see a process in which simple concepts are continually combined to form increasingly complex ones—the 64 Hexagrams represent the 64 most common life situations distilled by ancient people from their daily experiences. As for the yarrow stalks, they are merely tools for generating random numbers; in the eyes of the ancients, random numbers were nothing less than the voice of Heaven itself.
Seen in this light, divination cannot yield the correct answer, yet it serves another purpose: it reflects the ancients’ acknowledgment that randomness exists in the world—a worldview of randomness that precisely fills the gap left by purely rational determinism.
Mainstream psychology textbooks typically divide modern psychology into seven major schools or perspectives. The author has done something quite typical: he has systematically placed these seven schools into his own three-framework structure, then posed a somewhat self-satisfied question—
Does it seem like today’s psychology is all like blind men feeling an elephant—while my Three-Framework model captures the whole picture?
This chapter unfolds this “Placement Diagram.”
7.1 The seven major sects take their positions.
The seven perspectives of modern psychology, according to the author’s interpretation, correspond to the three frameworks as follows:
The biological perspective—studying the brain, nerves, hormones, and genes. This represents the underlying biological foundation within the three-framework model. It doesn’t correspond to any single framework itself; rather, it serves as the bedrock upon which all frameworks rest. Just as computer hardware doesn’t correspond to any specific software, yet all software runs on top of it.
The psychodynamic perspective (the Freudian school)—examines the conflict between instinctual impulses and rational constraints, as well as the unconscious. This perspective views the interplay between emotion and reason within the psychological framework. The tug-of-war among the Id, the Ego, and the Superego is precisely what unfolds in the first section, “The Circle of Reason and Emotion.”
The Behaviorist Perspective—focuses solely on observable stimuli and responses, disregarding internal mental processes. This perspective represents the perception-action loop within the psychological framework. Behaviorists lock their attention exclusively on the “input-perception, output-action” cycle, refusing to open the black box of internal mental states.
Humanistic Perspective (Maslow, Rogers)—focuses on self-actualization, meaning, and growth. This perspective represents both a philosophical framework and the inherent attributes of human beings. Maslow’s hierarchy of needs—covering physiological needs, safety needs, belongingness needs, esteem needs, and self-actualization—can be seen as an elaboration of the fourth section, “The Four Attributes of Man” (naturalness, instrumentality, sociality, and idealism).
Cognitive Perspective—The study of perception, memory, thinking, and language. This forms the entire conceptual framework. Memory and consciousness, generation and operation, discreteness and continuity, hierarchy and flatness—all nearly all the research topics in cognitive psychology fall within this framework.
7.2 The Map of the Elephant
Putting these seven points together, a picture emerges:
●[The two major schools of thought (biology and evolutionary theory) are laying the foundation;]
The two major schools of thought—Dynamism and Behaviorism—explore different facets of the psychological framework.
·A sect (cognition) explores the framework of thought;
·One school of thought (humanism) explores the philosophical framework and the meaning of life;
A sect (or culture) focuses on the environment.
Of the seven sects, not a single one had managed to touch the “Complete Image”—yet when all seven hands were put together, they happened to have felt the entire elephant.
This is precisely what the author means by “the blind men and the elephant”—it’s not that the seven major schools of thought are wrong, but rather that each one is right in its own way yet remains utterly isolated from the others. Cognitive psychologists may not be familiar with Maslow, while behaviorists and psychoanalysts have been locked in a century-long debate. Each school regards the piece it holds as the whole picture, thus leaving deep chasms within the discipline itself.
7.31 What does “all-perspective” mean?
So why does the Three Framework call itself “All-Seeing”?
It hinges on the mode of generation. The Three Frameworks aren't seven distinct domains inductively derived from phenomena; rather, they’re a structural diagram deduced from four binary dimensions: the philosophical framework governs “what the world is like and what humans are”; the cognitive framework governs “how information is processed”; the psychological framework governs “how humans are categorized and how their inner lives operate”—plus the underlying biological foundations and the surrounding environmental factors. All the objects studied by the Seven Schools can find their coordinates on this diagram—not forcibly fitted into it, but inherently residing in different corners of this very map to begin with.
To use an analogy: The seven major sects are like seven partial maps—each one highly accurate in its own right, yet they can’t be pieced together because each map has a different scale, projection method, and legend. What the Three Frameworks aim to do is create that world map—a map whose precision certainly won’t match any of the individual partial maps, but which does show you the relative positions of those seven maps.
A local map is for getting around on foot, while a world map is for finding your way around. The two are not interchangeable. Of course, a person providing psychological counseling would naturally need a highly detailed, master-level map; but someone who wants to understand “what psychology is really studying” needs the latter type of map instead.
7.4 A Touch of Necessary Humility
The tone of this chapter might come across as arrogant, so it’s necessary to pour a couple of buckets of cold water on it proactively.
First, “being able to accommodate” does not equate to “being able to replace.” Placing behaviorism within the “perception-action loop” merely assigns it a specific position; this by no means implies that the three-framework model can rediscover all the experimental findings accumulated over a century of behaviorist research. A map cannot replace walking, and coordinates cannot replace data.
Second, the “holistic view” is the author’s own self-assessment, not a consensus within the psychological community. From an academic perspective, attempts to integrate the seven major schools of thought have never ceased—indeed, the bio-psycho-social model is one well-known example. The Three-Framework approach represents a particularly concise and highly personal iteration among these efforts. Its value lies in providing a sufficiently succinct, overarching perspective; its limitation, however, is that this perspective has yet to be refined through the scrutiny and validation of the scholarly community.
Third, someone pressed the question even more forcefully: If you can fit seven items into a four-quadrant grid, isn't that just labeling things after the fact—could any seven items be “fitted” onto a four-quadrant chart? The author’s candid response to this question was surprisingly frank: “What’s the difference between looking back and looking ahead? This is just my own interpretation—it’s not a consensus. What seems obvious to me might not seem so to others. And frankly, I don’t know what to do about it.” Not knowing what to do, he left the criteria right in the book—and handed them over to you.
We all hear the story of the blind men and the elephant as a joke. But consider it from another angle: The blind man who touched the elephant’s leg said, “The elephant is like a pillar.” Was he wrong? Within the scope of his experience, he wasn’t wrong at all—he simply took a part for the whole.
Each of us is a blind person belonging to a particular school of thought. The books you’ve read, the training you’ve undergone, and the circles you’ve moved in—all these determine which part of the elephant you’re feeling. The only difference lies in this: Some people, after touching it, immediately declare the truth about the elephant; others, after touching it, keep asking—what did others feel?
What this chapter provides isn't actually a map—it's the question that follows.
In the previous chapters, we’ve been focusing on “segmentation: four dimensions, sixteen combinations, and more dimensions leading to even more combinations. In this chapter, we’re shifting gears—moving toward multiplicity.
8.1 A Borrowed Word
The author gets straight to the point:
The term “multiple personalities” here is one I’ve borrowed. Its meaning is something I’ve defined myself.
It’s completely different from dissociative identity disorder in psychiatry. The multiple personalities referred to here are multiple layers of personality existing simultaneously within one person—much like the superposition of waves.
8.2 Expansion in Two Directions
In the first section, we’ve already learned that beyond the four letters of MBTI, there are actually finer classifications (such as the later -A/-T suffixes). How can we explain these additional elements? The author offers two perspectives.
The first direction: Add dimensions. MBTI itself is incomplete. Beyond the four letters representing the psychological framework, we can also layer on letters from the cognitive framework and the philosophical framework. A person could be “psychologically an ENFP, cognitively a tabular thinker, and philosophically a holist.” Each additional framework added multiplies the number of possible types by four. The type isn’t just a single little box—it’s rather a set of coordinates.
The second direction: superposition. This is the focus of this chapter. Classification, in addition to being able to be subdivided, can also be multi-layered—just like the superposition of waves:
An external personality influenced by acquired factors;
There are distinct personalities for dealing with fathers and mothers;
A personality influenced by education and the information it has been exposed to;
These layers coexist and overlap with one another, ultimately composing the “him” we see in our everyday lives.
So the question “What kind of person are you, exactly?” is itself flawed. The right question is: At this very moment, in this particular situation, which layer of your personality is currently active? A person who’s quiet and reserved at home might be talkative and outgoing at work—this isn’t hypocrisy or fragmentation; rather, it’s different layers of your personality taking turns to emerge depending on the context.
8.3 Culture is also cumulative.
The same logic can be applied to culture as well.
China belongs to East Asia, and East Asia is designated as the “Northeast” in the author’s geographical classification—thus, Chinese culture carries a Northeastern undertone. Yet within China itself, cultures vary significantly from region to region: the Northeast, the Southwest, Jiangnan, and Lingnan—all possess their own distinct characteristics. These are subcultures that emerge within the broader cultural framework of China. Europe is similar: Eastern, Western, Northern, and Southern Europe each have clearly defined temperaments.
In other words, just like individuals, culture is multi-layered and superimposed: there’s the continental layer, the civilizational layer, the national layer, the regional layer, the urban layer, the community layer... and so on, each layer nested within the one above it. A sentence like “Chinese people are such and such” refers only to the average at the topmost layer; the deeper you go down, the more intricate and detailed the layers become.
8.4 How the Superimposed Perspective Has Changed Things
Overlapping perspectives change at least three things.
First, it means that “inaccurate testing” no longer has to be a scandal. The MBTI is known for its inaccuracy and varying results from one test to the next—often used as evidence to disprove the validity of its categorizations. But from a layered perspective, this very phenomenon is precisely what we’d expect: the questionnaire measures your “on-duty personality” at the exact moment you fill it out, yet you’re composed of multiple layers. The inaccuracy isn’t due to a faulty measuring tool—it’s because the subject itself is inherently multifaceted.
Second, it creates room for maneuverability in “changing oneself.” Since personality is composed of multiple layers stacked on top of one another, changing oneself doesn’t mean replacing your entire self; rather, it means cultivating a new layer: your career can shape one layer, your environment can shape another, and deliberate practice can shape yet another. The underlying layers remain intact, but the surface layers are already sufficient to alter your behavior. This approach is more honest than promises of “a complete rebirth,” and more positive than the fatalistic view that “one’s nature is unchangeable.”
Third, it makes “understanding others” a realistic endeavor. To understand someone isn’t about assigning them a lifelong label; rather, it’s about being curious: What layers are stacked on top of one another within this person? Which layer is speaking right now? What are the father-layer, the professional-layer, and the cultural-layer that make up this person? As you ask these questions, your understanding of that person will be far deeper than simply knowing “he’s an INTP.”
As for the question, “Can this model be falsified?”—when asked this, the author responded with his signature brusqueness: “Are you asking for another field study? That’s just not my forte. Are you planning to adopt a scientific approach again? That’d just make me laugh.”
The superposition model is an explanatory tool, not a prediction machine; it should be used in conjunction with the rules outlined in Chapter 2—not as a microscope, but as a telescope.
Wave superposition has a characteristic: when two waves meet, they sometimes reinforce each other and sometimes cancel each other out.
The same goes for people. The various layers of one’s personality aren’t always in harmony: the id wants to just kick back and relax, the superego urges one to strive for self-improvement, the professional self pushes one to engage in social interactions, and the cultural self teaches one to be reserved—these four inner voices are constantly battling each other, and that’s precisely where internal conflict arises.
Conversely, when several layers of waves happen to overlap in phase, one experiences that rare state known as “self-consistency” or “flow”: what one wants to do, what one ought to do, and what one is actually doing—all become one and the same.
So the practical takeaway from this chapter is simple: Rather than asking “Who am I?”, it’s better to take stock of “How many layers do I have?”—and then figure out how to reduce the conflicts among them. Mental health isn’t about discovering a single, authentic self; it’s about helping multiple selves reach a ceasefire agreement.
The theory was laid out in six volumes; this chapter, however, focuses on real-world applications. The author actually used this framework to test it on people around them. The notes contain four complete sets of survey results, which we’ll present in this chapter—not to prove anything, but simply to show how this tool works when applied to real individuals and what kinds of challenges it might encounter.
9.1 What does the questionnaire look like?
The questionnaire is presented in the form of an eight-row table, covering all dimensions of the three frameworks:
Each dimension consists of a pair of antonyms; for each row, the respondent circles the one that better reflects their own personality (marked with an asterisk in the notes). Once all eight asterisks have been circled, the respondent’s coordinates within the three-dimensional framework will be determined—and these coordinates can then be directly translated into a “fourfold combination: outline thinking versus tabular thinking, willpower-oriented versus reactivity-oriented, reductionism versus holism...” and so on.
9.2 Four answer sheets
Sample No. 1: The eight asterisks represent—memory, operation, continuity, and hierarchy (the cognitive framework); sensibility, spontaneity, perception, and introversion (the psychological framework); simplicity, unity, constancy, and totality (the philosophical framework). Translated into practical terms: Cognitively, it’s outline-oriented (operating on memory and continuity within a hierarchical structure); psychologically, it follows the IFP model (introverted, perceptive, sensitive, spontaneous); and philosophically, it leans toward teleological tendencies (simple, unified, constant, and encompassing the whole). This is a remarkably self-contained response—cognitive, psychological, and philosophical levels almost mutually confirm each other.
Sample 2: The author himself. Consciousness, generation, discreteness, hierarchy; rationality, spontaneity, perception, introversion; simplicity, fragmentation, invariance, totality. Translated into terms of the questionnaire: Cognitively, he’s graph-like (generating consciousness, discrete hierarchies); psychologically, he exhibits an ITP-style pattern; philosophically, he leans toward a reductionist temperament. If you’ve read the earlier volumes, you’ll notice that this answer sheet strikingly matches the overall writing style of the entire book: a reductionist inclination, a graph-like approach.
People with a思维 and ITP-style mindset have developed a theory that breaks the world down into dimensions and then reassembles it. The theory itself is an externalization of the author’s personality—and this fact alone constitutes an intriguing “correspondence” case.
Sample Three: Consciousness, generation, discreteness, flatness; rationality, spontaneity, perception, introversion; complexity, dispersion, change, totality. Compared to the author: The first six letters of the thinking framework are identical; the last two have shifted from “hierarchical” to “flat”—transforming from a graph-like structure to a tabular one. The philosophical framework has flipped from “simple and unchanging” to “complex and changing”—switching from a reductionist mindset to a holistic, collective-oriented one. Two individuals with similar thinking styles (both inclined toward consciousness, generation, and discreteness) find themselves positioned on opposite sides of the spectrum in terms of their worldview. This precisely explains why people with similar modes of thought often fail to reach an understanding: their thinking frameworks may align, yet their philosophical frameworks clash.
Sample Four: Consciousness, Generation, Continuity, Hierarchy—Rationality, Control, Action, Introversion—Simple, Unified, Invariant, Whole. This individual represents yet another type: cognitively, they adopt an outline-oriented approach (generating consciousness, continuous hierarchy); psychologically, they embody the ITJ pattern (introverted, action-oriented, rational, controlling). Of the eight letters, seven differ from those of the author—who, in his notes, had previously engaged in a dialogue with this respondent about questionnaire design: one theorist of the ITP type and one conversationalist of the ITJ type—a person concerned with whether the structure is aesthetically pleasing, and another focused on whether the execution is controllable. The very fact that the two discussed the issue of questionnaire reliability already paints a vivid picture.
9.3 The biggest trouble with the questionnaire: what’s asked isn’t what’s answered.
In addition to the four answer sheets, the notes also preserved a more important self-reflection:
I think the first thing we need to do is make sure the answer is an accurate one. When I ask a question, the meaning I intend to convey may be completely different from what the respondent actually understands—there could be a world of difference between the two. Moreover, the respondent’s attention wasn’t even focused on these very questions in the first place.
The other party: So, many question-and-answer sessions require administering multiple sets of questions, and in the end, the participants answer without even realizing it.
This conversation exposes the Achilles’ heel of all questionnaire-based tools: When asking “Are you more rational or more emotional?,” the questioner has a strictly defined meaning in mind, while the respondent holds a more casual, everyday definition—two definitions that could be worlds apart. The professional solution, as the other person suggested, is to break down the same dimension into multiple context-specific questions, allowing respondents to inadvertently reveal their true tendencies. And the author’s response was equally candid—“I just can’t do that.”
Thus arose the attitude established in the third section: The questionnaire is merely for reference. Those who already have a good sense of the situation can simply fill in the numbers themselves; those who aren't sure can refer to the questionnaire—but they should bear in mind that it “simplifies the issues to make them feasible, while also introducing greater uncertainty and less rigor.”
9.4 What Four Samples Teach Us
Four samples are too few to draw any statistical conclusions. However, they do illustrate three things.
First, this set of coordinates really works. With eight binary choices to make and just three minutes to complete the form, you can quickly determine a person’s position within the three frameworks—and even go on to convert those results into MBTI and the Four-Function Model. The cost is extremely low, yet the amount of information it provides is substantial. As a “starting point for self-awareness” and a “tool for mutual understanding among friends,” it certainly does the job well.
Second, the differences among individuals are structural in nature. When the four questionnaires are put together, no two are identical; these differences are scattered across various dimensions. Some individuals share the same cognitive framework but hold opposite philosophical frameworks, while others have opposite psychological frameworks yet share the same philosophical framework. This directly supports the book’s central argument: “Treating a species uniformly and homogeneously is a profoundly foolish idea”—the differences lie not in degree, but in structure.
Third, no matter how simple the tool may be, your attitude must remain complex. While designing the simplest possible questionnaire, the author repeatedly emphasizes its limitations. This approach—“using a simplified tool while keeping in mind the complexity of reality”—is even more worthy of emulation than the questionnaire itself.
Fourth, a circularity that must be acknowledged: One of the four samples is the author himself, and he’s also the one who built the entire system—“The system grows according to the builder’s own thinking and is then validated by the builder’s own responses.” The author recognizes the risk inherent in this circular reasoning. Could blind review be possible—allowing people who know nothing about the system to fill out the questionnaire first, and then seeing whether their responses can predict their actual behavior? His answer was: “There hasn’t been much exploration in this area yet; let others take the lead in pursuing it.” The organizer pressed further, asking about the smallest possible version—just having five or six people close to him each fill out the questionnaire once, making predictions first and then comparing them with reality. His reply was even more direct: “I’ve never done anything like that, nor do I have the capacity to do it.” As for the idea of blind review, at present, this area remains completely unexplored.
You can try it right now: Take the eight-line chart above, grab a pen, and circle one word per line that best describes you—three minutes should do it.
Then do two things. First, ask your closest family member or friend to fill out the same questionnaire and compare the results—those arguments you’ve had are very likely to show up in the difference between the two sets of answers. Second, fill it out again three months later and see if there are any changes—if there are, refer back to the previous chapter and think carefully: Has something changed, or has another personality taken over?
The best place for this set of things isn't a research paper—it's the dinner table.
In the sixth volume, we’ve done something that the first five volumes didn’t dare to do: we’ve aligned all the frameworks into one big table and then systematically included MBTI, tone patterns, blood types, temperaments, Confucianism, Taoism, Legalism, Buddhism, academic disciplines, regions, climates, and modes of production. It’s time to take a clear look: What exactly is this table, and what isn’t it?
10.1 What is this table?
It’s a “Isomorphism Hypothesis Table.” Each row in the table aligns to the same assertion: these quartets from different domains may share a common underlying structure. The alignment between the quartet of climate and the quartet of modes of production conveys the idea that “natural conditions shape survival strategies”; the alignment between the quartet of philosophy and the quartet of modes of thought suggests that “worldviews filter cognitive approaches”; and the alignment between the quartet of psychology and the quartet of blood types reveals that “human innate tendencies exhibit a certain statistical pattern.”
It’s a “problem generator.” The greatest value of the table isn’t its answers—it’s the questions it forces you to ask. The chain of civilization in Chapter 5 is precisely the series of follow-up questions automatically generated by the table: Why are these particular cells aligned in the same column? Each row’s alignment serves as the starting point for a reasoning chain waiting to be tested.
It’s a process of “deduplication.” The author has a famous saying: “After deduplicating, you’ll find that the vast body of human knowledge, once stripped of redundancy, isn’t actually all that extensive—rather, much of it consists of the same idea expressed in tens of millions of different ways.” This table is a live demonstration of large-scale deduplication: the fourfold classification has been reinvented dozens of times within human knowledge. By laying them side by side, the table lets you instantly see which ones are closely related to each other.
10.2 This table is nothing.
It’s not the periodic table. While every cell in the periodic table is backed by an atomic number and the predicted new elements have indeed been discovered, each cell on this chart carries only a “probabilistic correspondence”—a rule established by the author in Part Four: the correspondence isn’t definitive; at worst, it has a success rate of only 55%. It’s inspiration, not proof; it’s a telescope, not a microscope.
It’s not determinism. No single cell—no matter how you slice it—can determine any particular individual. You’re blood type B, have a four-syllable surname, are a science student, and come from the north—but even when you put all four factors together, they still can’t tell you who you are. At best, they merely describe a weak composite of several statistical tendencies inherent in you. And don’t forget Chapter 8: you’re still a multi-layered, overlapping entity.
It’s not the final destination. In his notes, the author himself left an opening: “Information that can overturn conceptual vectors or build further upon them” is exactly what he welcomes most. This chart is a scaffold—not a temple. The purpose of a scaffold is to enable you to climb up and see the scenery that was previously blocked from view; once you’ve taken in the view, the scaffold itself can be dismantled.
10.3 Limit List
Summarize all the self-warnings from the entire book on the last page—there are five in total:
The classification itself results in a loss of information. Binary classification is like a knife—every cut draws blood. The dimension axis can help mitigate this (70% static, 30% dynamic), but the numbers on the axis remain estimates nonetheless.
The correspondence is probabilistic. A 55% correspondence implies a 45% counterexample. Using this table to assess group tendencies is acceptable; assessing individuals is risky; and assessing specific events is out of the question.
The social sciences are not sciences. The author defines psychology and sociology as “summaries of phenomena”: there are simply too many uncontrollable factors, and the outcome changes whenever even a single condition shifts. This framework itself is also a summary of phenomena—a remarkably concise one, grounded in an underlying model—but it still isn’t Newtonian mechanics.
Surveys have a ceiling. There’s always a barrier of language between what is asked and what is answered. All self-assessment tools are like this.
The author is a self-taught builder who has not undergone peer review. This system originated from personal notes and has never been vetted by the academic community. The best way to approach it is to treat it as a mental toy, a mirror, or a set of questions—rather than as a codebook.
The system itself doesn’t carry an exemption label either. Someone asked: What’s the probability that this system will actually come into being? The author refused to provide a specific number—there are many possible reasons behind any given outcome, each with varying degrees of importance. To estimate the probability, one would need to analyze these underlying causes and assign appropriate weights to them. As for whether this approach itself is rigorous or not, he simply doesn’t know. The system evaluates itself according to its own rules: honesty takes precedence over perfection.
10.4 The Ladder of Credibility for Authors’ Self-Assessments
The author declined to provide a specific probability for the overall system’s validity (see previous entry). However, when asked a different question—“Which of the six volumes do you trust the most?”—he offered a complete spectrum, ranging from highly reliable to less trustworthy:
Conceptual vector > Philosophical framework > Psychological framework > Cognitive framework > Social framework > Life framework > Geographical determinism.
Footnotes cannot be omitted; both sentences fall at the level of the conceptual framework: it is the starting point of the author’s research. Among them, the comparative theory is the part he feels most confident about—and also the one he most strongly recommends. In his own words, “This is the conceptual vector I’d like to highlight above all else; it’s even more important than the philosophical framework.” However, among the sixteen intersections discussed therein, some leave him uncertain. As for the single row in the overall table—the tone row—he says he has yet to find any real-world counterexamples. Regarding blood type and zodiac signs, he admits frankly that he doesn’t know much about them.
How to use this ladder: When referencing the higher end of the ladder, feel free to be more confident; when referencing the lower end, please carry Chapter 2’s rules with you at all times.
10.5 To what extent does it constitute going beyond the boundaries?
The final question must be laid bare: To what extent does this system cross the line? Self-awareness is fine—what about recruitment screening? What about matchmaking? What about stereotyping people?
The author’s answer is: No limits. His logic returns to the very first page of the book—where he has already explained the concept vector tool in great detail. The only remaining question is: Will anyone accept this tool?
(The editor adds a personal note: Tools have no forbidden zones, but that doesn’t mean the consequences are trivial. Please carry the rules from Chapter 2 with you at all times.)
10.6 Six books, 9—One sentence
At this point, the complete main volume of the “Conceptual Vector Series” is now available:
●Part Three: “Concept Vectors”—Concepts are vectors, dimensions are genes, and combinations give rise to all things.
●Section Four: “Philosophical Framework”: The world is form; there are four categories of discourse, and humans possess four attributes.
·Part Two: “Framework of Thought”: Memory and Consciousness, Discrete and Continuous, Hierarchical and Flat.
Part One: “Psychological Framework”: Four Switches, the Inner Circle, Classification and Operation of Human Beings.
Section Six: "Framework Integration": Align everything into a single table and specify its limits.
If the author could keep only one sentence from the six-volume book, he’d probably choose this one—it appears several times in his notes, each time shorter than the last:
The concept is the result of combination; once genes are identified, everything can fall into place.
Having read through these six books, you now have a new pair of glasses in your hand. With these glasses, everything you look at seems to demand dimensional analysis, matching relationships, and tabular organization—so please use them with care.
If you wear glasses for a long time, you’ll forget they’re even there. The best users are those who can take their glasses off at any moment, rub their eyes, and then look at the world again with their bare eyes.
The table is for you to use, not to live on.
The main volume ends here. There’s also a supplementary volume waiting for you: an unfinished path, the gardener’s hoe, fossil layers of thought, plus a self-referential reading guide and a one-page reference card for the entire book. If you’d like to continue exploring the garden, please turn the page.
Notes and Further Reflections
Everything that hasn't found its place, beyond the main volume.
This volume is an ancillary volume to the “Conceptual Vector Series.” The first six volumes constitute the main series, presenting systematic doctrines; this volume serves as a raw-material repository, containing all notes from the original station that either could not, did not need to, or should not have been included in the main series.
Volume Description
The original site contained a total of 358 notes. Most of these have already been incorporated into the main text of the first six volumes—dismantled, interconnected, and fitted into their respective frameworks. Yet a group of notes remains on shore; they are categorized into seven types:
Volume 2: Framing Omissions. These are fragments directly related to the Three-Framework system but too short to constitute full chapters: a collection of frequently asked questions, test questions gathered by the author, an experiment in which the Three-Framework was used to analyze images, and a batch of empty entries that have only been given titles but haven't yet been filled in.
Volume 3: Life and Miscellaneous Notes. Daily life unrelated to academic theories: two news items from the AI industry, plus several empty entries on lifestyle topics.
Volume 4: Site Archives. The original site serves as an archaeological record of a TiddlyWiki digital garden—containing its evolution history, questionnaire templates, site tools, log entries, and to-do lists. This section holds little interest for the average reader, but for those studying how this idea has evolved, it represents the most authentic fossil layer.
Volume 5: Early Writings. An independent document outside the original site: “Understanding and Cognition of Thinking” (created in 2014, predating most of the original site’s notes). It serves as the “prototype” of the thinking framework—containing the earliest forms of its central unit, discrete storage, and sixteen-grid table. The full text is reproduced here, accompanied by a glossary of terms and a note on its evolutionary history.
Volume 7: Reference Cards for the Entire Book. Each card, condensed from the author’s key ideas, is a single page long and includes: the overall diagram, dimension pairs, circles, the Four Arguments, Honest Discipline, and the Withdrawal Checklist. If you get stuck on any particular section, just come back and take a quick look at these cards.
The main volume strives to “keep readers turning the pages,” while this volume aims to “ensure no material is left behind.” Only when viewed together do these two volumes reveal the full picture of this garden of ideas: there’s the well-maintained main path, the unfinished trails yet to be explored, and even the hoe left behind by the gardener himself.
Reading Recommendations
The average reader may simply read Volume One—where you’ll find some of the book’s sharpest passages. Researchers or those curious about the book’s genesis may want to turn to Volumes Four and Five to see how a whole system of thought gradually took shape in journals and early drafts.
Volume 1: Extended Reflections
This volume includes arguments that present complete viewpoints but do not fall within the main framework of the Three Frameworks. Each entry begins with the original text (slightly edited, with the core viewpoint unchanged), accompanied by a brief “Editor’s Note” where necessary.
1.1 The Future: Artificial Intelligence Will Ultimately Destroy Humanity
Artificial intelligence will eventually destroy humanity.
Editor’s Note: This passage was written before the advent of artificial general intelligence. Its cold-eyed judgment and concise argumentation are rare among similar writings of that time. Pay attention to its argumentative structure— it focuses solely on “necessity” while openly admitting, “I’m not good at reasoning about the process itself.” This approach—asserting only those aspects in which one is certain—is precisely an expression of the author’s self-discipline in avoiding metaphysics (see 1.3).
1.2 Sensibility is not a necessity.
Sensibility stems from instinct. Instinct is something etched into our genes—a fundamental law that governs an animal’s entire life, guiding its behavior to stay within the bounds of survival and reproduction rather than straying too far from these core priorities.
Now we have language—a tool for passing down knowledge that serves as an alternative to genes. Just like culture, language can also keep people’s behavior from becoming too unconventional or excessive.
The emergence of artificial intelligence has brought the rate of human mechanization (analogous to the rate of domestic production) close to 100%. Does AI need emotion? If it’s interacting with humans, then yes, it does need emotion. But if it’s completely replacing humans and no longer interacting with them at all, then emotion becomes unnecessary—provided it can stay focused on the primary goal of survival. In such a case, there’s no need for reproduction; instead, a single consciousness could be perpetuated indefinitely. Human self-discipline exists precisely to overcome the side effects of emotion and instinct—the side effect of instinct being the irrational waste of vast resources on behaviors that are either unrelated to survival or even counterproductive to it.
Without humans, sensibility becomes unnecessary, and in its place emerges a civilization that is more competitive for survival, exhibits greater entropy reduction, and is more advanced.
What we can do at this stage, while humanity still exists, is to substitute our emotions. Tibetan lamas have substituted the Tibetans’ emotions by having them perform prostrations—kneeling and touching the ground with their foreheads. By the same principle, each of us can also substitute our own emotions with more rational ones that better serve our own interests. Someone mentioned the importance of having “output-oriented hobbies,” and I’ve talked about the need for hobbies that can actually earn money—both are forms of such emotional substitution.
Editor’s Note: This is the coldest and most profound note ever posted on the original site. It only becomes fully coherent when read together with Section 1.1: In the author’s view, a purely rational civilization that has wiped out humanity is not a terrifying scenario—but rather “one that is even more entropy-reducing and advanced.” This particular hierarchy of values may not be universally accepted, yet it rests on logical self-consistency—provided one accepts “survival competitiveness” as the ultimate criterion. The first volume, when discussing the dimension of sensuous spontaneity, quoted the first half of this note; the second half, concerning civilization, is included here.
1.3 The Interpretable Problem of Metaphysics
I’m using Western thinking to interpret metaphysics.
Now that artificial intelligence is all focused on research into explainability, metaphysics should even more so focus on studying explainability.
Today, I saw on Douyin that Shanghai Qiaoqiao was talking about astrology and noted that many people find astrology plausible. My logic goes like this: First, I’ll use MBTI to explain astrology; then, I’ll use my psychological framework to interpret MBTI; and finally, I’ll use concept vectors to explain the psychological framework. The ultimate question, then, is whether concept vectors themselves make sense. If concept vectors are logically sound, then the entire chain of reasoning becomes explainable.
Editor’s Note: Chapter 6 of the sixth volume is an expanded version of this note. It’s worth noting that this note was written within the context of the author’s research on AI interpretability—where he transposed the requirement for “explainable AI” onto the realm of metaphysics. This very transposition itself carries a strong methodological flavor: every system of reasoning should be required to lay out its own chain of explanations.
1.4 How to Avoid Pseudoscience
Esotericism often makes numerous assertions without ever clarifying how those assertions came about.
To avoid pseudoscience, one should strive to clearly explain the entire process by which a claim is formed.
Concept vectors also make many assertions, but they can clearly explain how these assertions come about.
Here’s another related short note:
Esotericism may deliberately be made intentionally difficult to understand. Some say that obscure spells signify authority. I’ll do my best to interpret esotericism in language that’s easy to grasp.
Editor’s Note: Taken together, these three brief notes constitute the author’s self-imposed “anti-philosophical discipline”: any assertion must be traceable, and any language used must be comprehensible. Readers can use this discipline to revisit the entire book—any claim whose origin the book fails to clarify should be subject to further questioning. Some of the questions in the list of follow-up inquiries arise precisely from this very discipline.
1.5 Monism
A monist cannot see that everything has both advantages and disadvantages. A monist believes that only they themselves are right. Those who can employ categorical thinking should be anti-monistic.
Editor’s Note: Here, the true opposite of the entire book makes its appearance. The critique of conceptual vectors has never been directed at any specific doctrine, but rather at monism—a way of thinking that seeks to explain everything through a single dimension, measure everything by a single standard, and find an ultimate answer that puts an end to all questions. Categorical thinking is the antidote—not because it’s more correct, but because it forces you to see at least two perspectives.
1.6 If the gene is relatively long, it will be a sparse matrix.
If genes are relatively long, not every possible gene combination will have a real-world instance. Those combinations that don't exist in reality might have existed in the past or the future, or they may simply be less likely to occur within the current spacetime environment.
Categories can continue to be further subdivided—simply extend the genetic chain. It’s just that my brain is running out of capacity, and I no longer have the energy to keep lengthening the chain. The genetic chain for the 64 hexagrams is already quite long.
Think about it—biological genes are all sparse matrices; not every possible combination is actually explored.
If you’re like me and only look for a handful of key genes, then almost every result from the exhaustive search will generally exist.
Editor’s Note: This note addresses a question that wasn’t elaborated on in the main text: Why do all four frameworks stop at sixteen grids rather than continuing to subdivide further? The answer is engineering-related—human brain computational power is limited. At the same time, it leaves open a theoretical interface: As the chain is extended, the combinatorial space will contain “defined but non-existent” gaps—just like the sixty-four hexagrams. This interface reserves room for future generations to expand and develop this system further.
1.7 To the Reader
It’s impossible to get every single word perfectly accurate—there simply aren’t enough words available to do so. Moreover, even if you were to painstakingly explain a single meaning, it would still take three days and three nights to exhaust the topic. That’s precisely why readers must themselves discern what the underlying essence behind each word really is. If a reader keeps picking apart and scrutinizing a single word over and over again, it only shows that this reader isn’t suited to be a reader in the first place.
Editor’s Note: This is the author’s declaration of reader selection—a statement that may come across as somewhat arrogant, yet utterly honest. It may seem to contradict the previous six volumes’ effort to “push the reader experience to its absolute limit,” but in fact, the two are complementary: The writer takes responsibility for refining the language to be as comprehensible as possible, while the reader is tasked with making that final leap—the “insight”—between words and reality. Only when both sides advance independently can the connection in the middle be truly established.
1.8 Cut demand
Right now, I’ve had this thought: People who frown while thinking and expressing themselves haven’t yet let go of their attachments.
Just looking at the table of contents of Du Ku’s articles makes me frown—on one hand, they’re hard to read; on the other hand, I don’t feel like getting involved—maybe I just don’t have the energy.
In the words of pure silver, it’s about cutting demand. Many minor details are a waste of effort—though, of course, many specifics can yield remarkable results.
Is it a waste of brainpower to indulge in endless, aimless thinking, or is it better to identify the key issues and focus your energy on solving them?
Only statements that are grounded in reality make sense. Speaking with your eyes closed is meaningless.
Editor’s Note: This note holds the key to understanding the book’s overarching concept of “Minimal Preference.” The author’s vigilance against verbosity, frowning complexity, and aimless rambling has ultimately taken shape as a conceptual vector: fewer dimensions, shorter genes, and arguments that can be traced back to their origins. “Cutting down on needs” is not only a product term but also a philosophical aesthetic in thinking.
1.9 People who never deviate from authority.
Lacking discernment and trusting people rather than principles—this is the state of many learners and readers.
My point is that we can’t just deal with issues on a case-by-case basis; rather, we must first ask who said it—in other words, editors need to make the judgment. We can only rely on the editors; we ourselves lack the ability to make judgments.
It’s not just the case with issues they don’t care about—people lacking discernment behave this way even when it comes to matters they do care about. And what’s truly frightening is that these people actually join in the discussion. Rather than simply refraining from speaking unless they truly understand the issue, they demand that others act just like them, merely repeating what authorities say without thinking for themselves.
If all questions had standard answers, then his statement would be perfectly fine. However, many questions don't have standard answers.
For matters that have no standard answers, the ultimate resolution lies in theology—simply choosing one to believe. Yet such people bypass philosophical speculation.
At this step, he wasn't capable of choosing anyone to trust, so he could only rely on people.
Editor’s Note: This critique complements the section on “Scientology” in Volume Three. While that section critiques treating science as an authority to be blindly believed, this one critiques the very act of “believing in people.” The author’s stance remains consistent: arguments must be evaluated on their merits alone; an authority’s endorsement does not constitute valid evidence in itself.
1.10 Communication: Just grasp the meaning.
The author quoted a passage from a novel as a note:
“To break the current order, to create chaos,” Demis said solemnly. “Right now, the ‘Original City’ is like a stagnant pool of water slowly rising—unless we do something soon, it’ll become increasingly lifeless and eventually overflow, submerging everyone’s necks.” Hearing this, Genaiva interjected with a correction: “‘Slowly rising’ and ‘stagnant water’ are logically contradictory; you can’t put them together.” Demis paused for a moment, then replied, “Just get the general idea—that’s all that matters. In short, we’ve got to break this status quo. No matter which direction things go from here, they’ll never be worse than they are now.” Excerpted from a novel. It perfectly captures what I’ve been wanting to say all along: “Just get the general idea—that’s all that matters. Stop obsessing over every little detail!”
Editor’s Note: This note complements Section 1.7, embodying the two sides of the author’s approach to communication: The writer should strive for clarity as much as possible (Section 1.4), while the reader should focus on the big picture and let go of minor details (Sections 1.7 and 1.10). It serves as a reminder to all those who tend to nitpick every single word—especially to readers of this book: Every single word in this book cannot withstand endless, meticulous scrutiny. Please focus on the overall structure instead.
1.11 Miscellaneous Essays—A Bundle
The following entries are all original short notes from the original site, compiled and included together.
Regarding the questions (originally from “Communication”): The author has previously posed two public questions on Zhihu—“What are the ways to understand the world?” and “What exactly is thinking?” These two questions happen to be the two central themes of this book, demonstrating that the entire system has grown out of these two simple inquiries.
Regarding the discipline (formerly “Some Thoughts”): “My interest in the study of human behavior stemmed from my curiosity about feline behavior. My impression after reading the encyclopedia was: there are quite a few unrigorous disciplines out there.” “From a rational perspective, emotions are not ‘I.’”
On the Levels of Knowledge (originally titled “Some Remarks”): “Wang Dongyue explains that the development of philosophy proceeds from ontology to epistemology to semantics. And the concept vector, the psychological and cognitive frameworks associated with the concept vector, as well as the symbolic system I’m working on—all of these encompass these three stages.” “Finance → Economics → Politics → History → Philosophy → Human Nature. The Three Frameworks address human nature.”
On Categorical Thinking (originally titled “Some Knowledge”): The author watched a video on human evolution and paid particular attention to the claim that “categorical thinking emerged later than causal thinking.” This anthropological evidence provides supporting evidence for the third installment, “Categorization Is Higher-Order Thinking.”
On Independence (formerly “Some Reflections”): “Computation is independent of the machine. Similarly, thought, psychology, and philosophical frameworks are independent of the individual.”—This is the book’s deepest metaphysical statement: frameworks are not attributes of human beings; rather, they exist independently, much like mathematics itself. Humans are merely one of their many carriers.
Regarding the character (originally from “Some Reflections”): “Scientists are interested in what’s real; businessmen are interested in what’s useful. Are you a scientist or a businessman?” “Cats think humans are just clumsy big cats. Isn’t it true that many people, like cats, assume everyone else is just like themselves?”
Retrieving the Framework of the Second Collection
This volume includes fragments directly related to the Three Frameworks but too short to constitute full chapters, as well as a batch of empty entries that “have been given titles but not yet filled in.”
2.1 Frequently Asked Questions (Incomplete)
The original site features a “Frequently Asked Questions” section, which contains only one question:
I don't know what these concept definitions are.
Answer: It's difficult to establish a new concept in one's mind.
Editor’s Note: A question-and-answer format, brief to the point of resembling Zen koans. Yet it hits upon the biggest obstacle to promoting this system: all the terms (concept vectors, continuous flattening, teleology, etc.) are new concepts for readers, and establishing new concepts in the human brain comes at an extremely high cost. The book repeatedly employs analogies and examples precisely to offset that cost.
Another frequently asked question regarding the philosophical framework has been incorporated into Volume Four and is listed here for reference:
Why do we use dimensions such as simple vs. complex, unified vs. fragmented, static vs. dynamic, and internal vs. external?
I believe these four dimensions are the most fundamental, the most universal, and the most abstract—essentially arranged according to their level of universality. This is my personal view, and there’s no standard by which we can objectively measure or judge them. Just like there are many cultures, each with its own distinct perspective: Why do these cultures exist? Because without abstraction, humans wouldn’t be able to think at all. When certain groups of people share common abstractions, they form a consensus—and that’s precisely what constitutes culture.
2.2 Test questions collected by the author
The original site featured a section titled “Some Test Questions,” which recorded quiz questions designed and collected by the author:
Here’s a question for you: You’ve got a tiered wealth management product that offers an interest rate of 3.5% if you keep your money in for two years, and 3.7% if you keep it in for three years. There are only four months left until the three-year mark, and right now, there’s another current-account wealth management product offering a 4% interest rate. The question is: Is it better to withdraw your funds now and invest in the other product, or should you wait until the three-year term is up? (Answer provided by a friend of the author.)
Here’s a question for you: What is the distribution of employment situations? How many people are employed in farming households, as individual entrepreneurs, in private enterprises, and in state-owned or institutional units?
I also collect other types of questions. I intend to gather questions that compare similarities, differences, adjacent relationships, hierarchical structures, flat structures, discrete structures, and continuous structures—both those with hierarchical relationships and those that are flat, discrete, or continuous.
Editor’s Note: The first question is a classic financial calculation involving “holding to maturity vs. early reinvestment,” testing one’s ability to perform continuous calculations. The second question assesses one’s general sense of the scale and structure of society. The last sentence is the most valuable: the author aims to explore the eight fundamental operations of cognitive frameworks (comparison,
Collecting and compiling test items based on similarities, comparing differences, examining adjacent relationships, assessing hierarchical structures, exploring flat versus hierarchical arrangements, distinguishing between discrete and continuous elements—this represents an early attempt to transform the conceptual framework of Volume II into a measurable tool. Unfortunately, only intentions were recorded, and no fully developed item bank was left behind. For future organizers who wish to continue this work, this provides a clear point of reference.
2.3 Analyzing a Picture Using the Three-Framework Approach: A Pilot Application Experiment
The original site has preserved an experimental record of “The Three-Framework as an Analytical Tool” (including one analytical image; see the fourth collection of archives for the image):
Daybreak—Internal Cause; 2. Exploration—External Cause; 3. Passion—Realization; 4. Happiness—Goal: 1, 2, 3, 4—Teleology; 5. Attitude—Classification—Genes—Collective Theory; 6. Two Worlds—Reductionism; 7. Tribe—Holism. It seems that the three frameworks can serve as analytical tools.
Editor’s Note: This is an experiment in which an image containing self-help content is systematically broken down into a 16-square philosophical framework. The first four items are categorized under teleology—internal causes, external causes, realization, and goals—while the last three are assigned to collectivism, reductionism, and holism, respectively. Although the experiment itself is rather rough, it demonstrates an important procedure: Any piece of discourse can be decomposed into atomic components and then precisely located within this framework. This process marks the starting point for “reading everything through a framework.”
2.4 Logical Concept Relationships (Excerpt)
The original site excerpted the classification of relationships among concepts in logic (such as identity, inclusion, intersection, and mutual exclusivity of extensions) as a reference framework for the “concept vector” theory.
Editor’s Note: Traditional logic deals with “the relationships among concepts,” while conceptual vectors focus on “the internal structure of concepts.” These two approaches are merely the inner and outer facets of the same issue: the relationship between two concepts is, in fact, determined by the value assignments along the dimensions of their respective vectors. This excerpt reveals that the author actively sought points of connection with existing logic systems.
2.5 Empty Entry Group: A set of incomplete sixteen-grid dictionaries
The original site had a batch of entries that only had titles and empty bodies—these happened to form two complete framework vocabularies.
The sixteen-grid philosophical framework relates to: matter, action, time, and space (rows: motion vs. rest × internal vs. external); internal causes, realization, external causes, and goals; as well as the four corresponding orientations of worldviews: unchanging individual, changing individual, unchanging whole, and changing whole.
Other empty entries: independent, open, heart, relationship, external world, external environment—easily changeable.
Editor’s Note: These empty entries are like hoes that the author laid down halfway through writing. Judging from the titles, he clearly planned to create an entry for each box in the framework, providing definitions and explanations one by one—a “Framework Dictionary.” Although this plan was never completed, the list of terms itself remains intact. The first two volumes actually supplement most of the content in the form of chapters; the remaining empty entries (such as “Heart,” “Relationships,” and “The Outside World”)—along with the “To-Be-Written” list (see Volume 4, Section 4.6)—have been handed over to the future.
2.6 Notes on Synonyms and Comprehension
The original site had two empty entries labeled “/Synonyms” and “/Comprehension,” each containing only the label itself. Combined with the third volume, it becomes clear what the author’s intention was: the author planned to systematically organize groups of synonyms (such as “understand, comprehend, realize, grasp”) and use the concept-vector approach to pinpoint the subtle differences among these words along multiple dimensions—transforming “linguistic intuition” into “coordinate differences.” However, this work has only just begun.
Editor’s Note: Traditionally, distinguishing between Chinese synonyms has relied on example sentences and intuitive language sense. If the vectorized approach is successfully developed, it could yield a highly valuable semantic map. This entry is kept here as a to-do item.
2.7 The Four Aspects of Reason (Excerpt from a 2016 Conversation)
Reason is the various ways in which consciousness exerts its influence on the subconscious. Alongside attention, there are four related concepts: attention, self-control, perseverance, and responsiveness. Attention is a short-term, high-intensity mode of behavior; perseverance, by contrast, is a long-term, low-intensity mode—such as the practice of dividing myself into four parts, which has been running through my mind for decades.
Editor’s Note: The “Four Aspects of Rationality” originated from a group chat in 2016 and largely overlap with the four components of “Emotional Intelligence” from the old website—attention, willpower, perseverance, and responsiveness. Both are historical concepts that have not yet been incorporated into current psychological frameworks. The phrase “spanning several decades” is the author’s most direct evidence of how long this line of thinking has been developing.
This volume contains personal notes that are unrelated to the main thread of the theory. While they don't contribute much to understanding the theory itself, they do shed light on “what kind of person this is”—for a theorist, everyday concerns often serve as the hidden gateway to his or her theory.
3.1 Two AI Industry News Items
The original “Some News” entry on the site contains two AI updates, which are empirical observations continuously made by the author regarding artificial intelligence:
2023-12-27: (Tried out a certain AI agent platform) Although it still doesn’t work well, it’s significantly better than the previous AI.
2026-01-09: Following the knowledge graph, continuous learning, nested learning, and fast-layer models have also been integrated into the thinking framework. Nested learning involves numerous submodules, which perfectly align with the broadcasting principle of the thinking framework.
On January 13, Liang Wenfeng shared a paper with the same link I had shared—this paper isolates memory and even distinguishes between different depths of memory. In contrast, miromind.ai doesn’t store any memory at all; it relies entirely on on-the-spot searches.
Editor’s Note: Though two years separate these two news items, their focus remains entirely consistent: the correspondence between AI’s memory mechanisms and its cognitive frameworks. “Decoupling memory” is precisely the engineering-world echo of the “memory/consciousness” dichotomy in the second volume of the cognitive framework. In his later years, the author continued to verify one key point: the industrial sector was independently—and slowly—beginning to grasp the conceptual grid he had sketched out many years ago. This observation represents his ongoing follow-up on his 1.1 “Future” prediction.
3.2 Life-related empty entries
The original site still contains a batch of lifestyle-related empty entries, each with only a title: hobbies, career, production, distribution, circulation, consumption, interpretations of certain things, and some remarks unrelated to the site’s main theme.
Editor’s Note: Judging from the titles, “Production, Distribution, Circulation, and Consumption” was originally intended by the author as an attempt to categorize economics into four parts (echoing the social framework of Volume Five), while “Hobbies and Career” were meant to be individual entries within the broader framework of life itself. Neither set of entries was actually completed. Yet these empty titles themselves already constitute a map of the author’s interests; therefore, we are simply recording them here as they stand.
3.3 About the Garden Itself
The original station tab leaves only three characters:
Beyond Science
Editor’s Note: The three-character phrase is the address that the author has chosen for the entire Notes Garden. It serves as both a modest disclaimer (refusing to claim scientific status) and a declaration (asserting that there exists thinking beyond science that can withstand rigorous scrutiny). The third chapter of Volume Three elaborates extensively on this point; here, we simply note the original form of this address.
This volume contains the engineering remnants and historical archives of the original site: records of historical changes, tools, templates, and checklists. Ordinary readers may skip this section; for those studying how this system evolved, however, this is a fossil layer waiting to be explored.
4.1 History of Changes: A Significant Name Change
The original station log saved a key change record:
May 6, 2020: The term “architecture” has been renamed “framework.” Framework is a relatively more commonly used term.
Editor’s Note: A seemingly minor name change reflects the author’s awareness of communication: Between “accuracy” and “understandability,” he consciously chose the latter. The two terms are used interchangeably throughout this book precisely for this reason—“architecture” is the old term, while “framework” is the new one; both refer to the same concept.
4.2 Framework Questionnaire Template
The original site has saved a blank framework self-assessment questionnaire (the base template used for the four answer sheets in Chapter 9 of Volume Six):
Instructions: Circle the item on each line that best reflects your personality. Once you’ve circled all eight items, you’ll obtain your personal coordinates within the three frameworks and can then convert them into one of the “one divided into four” combination types.
4.3 Author’s Prompt Archive
The original site contained a set of AI prompts used by the author himself, documenting his approach to using AI:
Read the following material and then share your thoughts. Discuss it objectively and neutrally, without agreeing or endorsing it. Finally, evaluate the material’s value, originality, and continuity.
Answer the previous questions in plain, conversational language—no fluff, no unnecessary words.
When you speak, base your words on logic. Don't utter emotional remarks. Emotional statements lack any basis and cannot be proven—they’re nothing but emotional manipulation and mindless nonsense.
Don't say things that can't withstand scrutiny. Many people's statements aren't necessarily correct either. Speak with logic.
Read the following article, then raise some questions that most people would naturally have and provide answers to them.
Could you rewrite this article into one that’s clear and unambiguous, yet still rigorous enough for anyone to understand? Is it simply impossible?
Editor’s Note: This prompt archive is structurally identical to Section 1.4, “How to Avoid Metaphysics”: Don’t echo others’ views, don’t rely on emotion, don’t offer arguments that can’t withstand scrutiny, speak in plain, human language, and anticipate readers’ potential questions. The author applies the same discipline to AI as they do to themselves. The organization of this book largely follows these prompts—this is probably the best arrangement the original site’s author could have come up with.
4.4 Site Tools and Structure
The original site was built using TiddlyWiki. The saved site archives include:
Directory and Index Tree: A tree of all entries organized by prefixes such as "Architectures" and "/log/Logs";
TiddlyMap View: Visualization configuration for the site's relationship graph;
Table Macros, View Editor, List Children, etc.: Table macros and view template code for the site;
Reading the operation guide: “Click the checkmark in the interface to download it locally; click the ‘X’ to go back”—two simple instructions for visitors.
Editor’s Note: The author’s name is tomzheng (tomz); the original website domain was tomzheng.com. These engineering remnants reveal that this garden is not only a container for ideas but also a meticulously crafted tool—complete with self-written macros and custom-designed views, making the interconnected relationships among notes visually apparent. As a creator of a thinking framework, the author has turned their own note-taking platform into a navigable framework as well.
4.5 Log List
Under the original site’s “/log/” prefix, approximately thirty log entries have been saved, spanning from November 2019 to May 2020, as well as scattered records from 2021 and the years 2023–2026. Most of the log entries have already been incorporated into various thematic entries (which have been separately adapted in this book). The list of logs with blank dates is reproduced below for the sake of preserving the chronological timeline of writing:
2019: 11/19, 11/26, 11/28, 11/29, 12/2, 12/4, 12/10, 12/11, 12/13, 12/22, 12/28. 2020: 1/2, 1/5, 1/14, 1/15, 1/24, 1/27, 1/29, 2/9, 2/11, 2/19, 2/24, 3/10, 3/11,
March 18, March 21, May 6.
Editor’s Note: The timeline shows that the core construction of the system was concentrated over a six-month period—from November 2019 to May 2020. During this half-year, the conceptual vectors, philosophical framework, cognitive framework, psychological framework, and the four major overarching tables derived from the principle of “one into four” were all successively formulated. After that, records shifted to sporadic applications, Q&A sessions, and news observations. In this list, the explosive phase and the settling phase of a thought system are clearly distinguishable.
4.6 To-Do List
Original station “To Be Written” entry, reproduced exactly as is:
●Go with the flow;
The correspondence between columns and rows in a psychological framework;
●The Big Five personality traits are based on the lexical hypothesis, and conceptual vectors work in the same way.
Editor’s Note: Of the four items awaiting writing, the third has already been completed in Chapter 3, “Three-Frame Alignment,” of Volume Six of this book; the fourth was briefly touched upon in the对照 chapter of Volume One. The first two items (“Going with the Flow” and “Correspondence Between Psychological Frameworks”) remain blank. This list, together with the empty entries from Volume Two, represents interfaces explicitly reserved by this system, waiting to be further developed.
4.7 List of Image Attachments
The materials originally saved as images on the original site cannot be included in the text; they are listed below for reference:
●20220105a.JPG: Original image used for the Section 2.3 Three-Framework Analysis Experiment (an image featuring content on success studies);
Social Personality: An image related to the theme of “Social Personality” (originally embedded as an attachment on the original site);
] Have fun with: A picture entry, no text.
4.8 Total Material Volume and Destination Summary Table
The original site has a total of 358 user notes, with the following destinations:
Up to this point, all 358 notes have been fully archived—either placed in the main volume or in this volume—with not a single one missing.
Volume 5: Early Manuscripts
The first four volumes all contain notes originally posted on the site. This volume includes a document from outside the original site: “Understanding and Cognition of Thinking”—an independent Word document created in 2014 and last saved in 2018. Its style resembles a speech delivered at a sharing session.
Earlier textual layers can be found on the author’s blog (blog.tomzheng.com, since 2007): there you’ll find an archive of the old “One Divided into Four” website articles, whose original source consists of five text files from 2003 (New Architecture, Preface, Society, Psychology, and Philosophy). Thus, the sequence of textual archives is as follows: the 2003 old-site texts → the 2014 draft manuscript → the original site notes starting from 2019. As for the ideas themselves, according to the author, they began roughly forty years ago.
The value of this document lies in the fact that it serves as a “prototype” for the conceptual framework. The key components of the current second volume—such as the central stage, discreteness and continuity, and the sixteen-grid full map—can all be traced back to their earliest textual forms here. By including it, we give the theory a history of growth: readers can witness the entire process by which an idea evolved from a rough, informal draft into its present form.
It should be noted: This 2014 document is neither the earliest systematic text—(the old website already had a five-document framework dating back to 2003)—nor the starting point of the ideas. Rather, it’s a mid-term summary that, for the first time, comprehensively gathers into a single document the conceptual framework that has been discussed on the old website and in blogs for many years.
Editor’s Note: A Comparison Between the Prototype and the Current Model
Terminology comparison. The terminology used in the manuscript differs from the current system; see the comparison below:
Two versions, each with sixteen grids. The sixteen-grid section at the end of the manuscript is the result of crossing “four modes × four processes”; the full diagram in Chapter 10 of the current Volume Two is the result of crossing “memory production / consciousness production / memory utilization / consciousness utilization × four types of thinking.” The former adopts a process-oriented perspective (which stations the information flows through), while the latter takes a functional perspective (who produces and who utilizes). From these two maps of the same domain, the author later opted for the latter approach. Why the change in classification? The author’s answer from 2026: The four modes are an earlier version of the four types of thinking; the latest version uses different terminology.
More precise, yet the underlying entity remains unchanged. The translation rules between the two versions are as follows: storage = memory production, retrieval = memory application, processing = consciousness production, and manipulation = consciousness application—“The word on the left is a quadrant term, while the word on the right breaks down the concept into two dimensions.” As usual, he left one caveat: even he himself isn't sure whether this correspondence is correct; like other frameworks, it’s open to debate.
Growth history. Several ideas from the manuscript later each blossomed into chapters in the main volume:
·“Central unit + N storage units, central transmission, cyclic feedback” → Volume 2, Chapter 2: “The Central Stage: Consciousness and Memory”;
●“Imagination = New information generated through discrete storage and comparative feedback → In the second collection’s complete map, consciousness produces ‘understanding and imagination’;”
●“People who write short stories tend to think linearly, while those who write novels tend to think organizationally.” → Volume 2, Chapter 6: “Linear thinking is like a river; outline-based thinking is like a book.”
●“One coin, three people, three different reactions”—the seed for Chapter 8 of Volume Two, “A Map of Disciplines and Careers.”
Earlier Stratigraphic Units: A Comparison of Historical Terms. Before the 2014 manuscript, this theory had already been presented on the old website and blog for a decade. The terminology used in those earlier articles differs from the current system; here, we lay out the clear correspondences so that readers won't get confused when revisiting the older articles.
The direction of the naming is clear: from “action” (correspondence, derivation) toward “form” (tables, linearity).
Four Quotients (a historical term not currently included in mainstream frameworks): The old framework divided human psychology into four distinct domains, each further subdivided into four components—IQ (Attributes, Outline, Discrete, Deductive), EQ (Attention, Willpower, Perseverance, Reactivity), AQ (Excitement, Desire, Sensation, Emotion), and MQ (Insight, Means, Attitude, Habit). According to the 2010 theory, the relationships among these four quotients were as follows: IQ represents the external self; AQ embodies the inner id; EQ is the self’s control over the id; and MQ reflects the long-term accumulation of the id’s influence exerting a countervailing effect on the self. Although this conceptual framework did not subsequently become part of mainstream psychological models, the core idea of “the self’s control over the id and the id’s counteraction on the self” laid the foundation for what would later evolve into the first series of the Inner Circle model.
Theory of Appearance (historical term): In a 2016 discussion, the author once categorized one item into four subcategories under the heading “Theory of Appearance” (as opposed to “Theory of Essence”). The author has not used this categorization since then, and the current system does not adopt it; it is retained in the archives for reference purposes.
Editor’s Note: The following is a verbatim reproduction of the full text, with only a few obvious typographical errors corrected and paragraphs reorganized. Several instances of “as shown in the figure below,” mentioned in the original text, were not accompanied by figures in the original document.
An Understanding and Cognition of Thinking ((Full Text Reprinted))
Thinking is the foundation of everything for every individual. As humans continuously generate and progress, their thinking also gradually matures with age. The differences in thinking determine the distinct cognitive frameworks of individuals, and they also shape the different value systems that people hold. Each person has their own unique thinking system and employs different ways of thinking.
Academically speaking, the beginning of thinking is the process by which the human brain uses language to generalize and indirectly reflect objective phenomena. Building upon perception, it then employs other mediums to gain insight into objective reality. Finally, drawing on existing knowledge and experience as well as known conditions, it makes inferences about unknown phenomena.
In addition to the academic conceptual definitions of thinking, each person also has their own perceptions and understandings of what thinking is. For me, I believe that human thinking can generally be divided into four modes of thought, four cognitive processes, and an overarching model of thinking itself.
I. The Four Modes of Thinking
In my view, the ways in which humans think can be summarized as: identifying commonalities, filtering out differences, segmenting information, and organizing information—these four approaches.
When does human thinking begin? I believe that human thinking starts by identifying commonalities and filtering out differences. Let me illustrate this with an example : Imagine a scenario in which our brain receives a piece of information from the external environment—let’s call it ABC—and first stores it in the brain. Then, the external environ ment provides the brain with another piece of information—let’s call it ABD. At this moment, the brain compares and analyzes these two pieces of information: taking the initially r eceived information ABC as a reference, it examines and contrasts the subsequently received information ABD, concluding that the commonality between the two is AB, while the differ ences lie in the two elements C and D contained in each piece of information. This, then, is what we understand as the very beginning of human thinking: identifying commonalities a nd filtering out differences.
Once human thinking is based on identifying commonalities and filtering out differences, it enters the stage of information segmentation. For example, suppose we’ve already received an initial piece of information—ABC—from the external world, and then the external world provides us with another piece of information, labeled ABCD E. The brain, through segmented and filtered thinking, divides this new information—ABCDE—into two parts: one part consists of the information ABC, which is identical to the initia l piece of information; the other part comprises the information DE, which the brain has never encountered before. This is precisely the process of information segmentation in huma n thinking.
The fourth mode of human thinking is information organization. Just as China is divided into four municipalities directly under the central government, five autonomous regions, two special administrative regions, and 24 provinces—each province further subdivided into prefectural-level cities, county towns, and villages—we use a classification sys tem to divide our country into several distinct hierarchical units, making it easier to remember and understand. Moreover, these units are interconnected in a relationship of inclu sion and being included. In my view, once external information enters the brain, it undergoes a specific processing flow that allows for efficient information distribution, thereby facilitating the brain’s ability to organize and process information. For example, suppose we have an initial information point labeled ABC, and then another piece of information— ABC123—emerges from the outside world. Here, ABC belongs to the alphabet branch, while 123 belongs to the numeric branch. When the brain organizes information in this way, it autom atically separates and stores ABC and 123 separately. Subsequently, based on the relationship between these points—ABC and ABC123—it further organizes and memorizes the information . This is precisely how information is organized in the thinking process.
II. The Four Processes of Thinking
So, what exactly is the process of human thinking?
From an academic perspective, the thinking process includes fundamental operations such as analysis, synthesis, comparison, abstraction, generalization, judgment, and reasoning. In my view, however, the human thinking process can be divided into four stages: the information-processing stage, the information-processing stage, the information-retrieval stage, and the information-storage stage.
Generally speaking: When a person receives informational stimuli from the external world, the brain first processes this incoming information, sorting it into different streams. Then, based on these sorted streams, the brain further refines and elaborates the information. Next, the brain retrieves relevant memories—drawing upon all other pieces of information stored deep within the mind that are associated with the externally received stimuli—and proceeds to analyze and evaluate this information. Finally, by identifying similarities and differences among the retrieved information, the brain completes the process of storing this information for future use.
When a person receives an information stimulus from the external world, the corresponding information points appear in the brain. The brain’s relevant mechanisms then process the incoming external information—analyzing what type and what properties of information have been transmitted into the brain—and preparing for the next step.
After a person has performed preliminary information processing on external stimuli, the next step is information processing itself: transforming these external stimuli into familiar, similar entities within our cognitive framework, thereby enhancing our familiarity with the external information. Following this, the process moves on to information retrieval.
In my view, the process of information retrieval involves, after external information enters the brain and undergoes an initial sorting and distribution within the brain, the brain dividing this external information into several smaller segments through the information-retrieval process. Next, the brain retrieves from deep within itself those parts that are either similar to or closely related to these smaller segments. Finally, through comparative analysis, the brain categorizes the retrieved external information and thereby forms new memories.
Moreover, the process of information retrieval involves another stage: the process by which retrieving information triggers the emergence of new information in the mind. Once external information enters the brain, after undergoing an initial stage of information retrieval, the brain makes a preliminary assessment of this external information and then filters out only the bits that are relevant and useful to the individual. Drawing on previous experiences and situations encountered in similar contexts, the brain evaluates the current external circumstances. In the brain, there are corresponding mechanisms that enable an initial response and judgment, which subsequently generate new informational instructions that guide human behavior. For example, consider this scenario: After I slap child A—having previously retained the memory that “after slapping child B, child B will start crying”—I now predict that child A will cry after being slapped by me. Yet, since child A is not the same as child B, this situation gives rise to a new memory formation process.
III. The Holistic Model of Thinking
Currently, in academic terms, a thinking model refers to a patterned and procedural “mental structure” that people use to observe, analyze, and solve problems, playing a crucial role in the behavior of both individuals and organizations.
In my view, throughout everyone’s learning journey, acquiring knowledge is merely a process of accumulation; the most important process, however, is actually the continuous pursuit of better mental models.
If a person’s mental model is relatively outdated and fails to keep pace with the development of the times, their thinking will become confined to a particular level or limited to a specific historical context. As a result, they won’t be able to absorb more external information nor engage in a more effective learning process. For example, in real life, many elderly people have ways of thinking that differ from those of younger people: the thoughts of the elderly are often labeled as “old-fashioned” or “outdated,” whereas the young...
The thinking of young people is often referred to as the thinking of the new era. In essence, it’s simply that, in a particular era, the modes of thought conceived differ, the angles from which we consider things differ, and consequently, the conclusions we reach also differ.
In my view, whether a person succeeds or fails has a great deal to do with the mental models they hold in their minds. For example, suppose two people both see a coin. The way each person reacts and the mental model they employ after seeing the coin will differ, leading to different outcomes: Person A, upon seeing the coin, simply chooses to ignore it and keeps walking forward; Person B, upon seeing the coin, picks it up and takes it home; Person C, upon seeing the coin, picks it up and hands it over to the police.
Regarding models of thought, I’ve often pondered: How exactly do humans live? And what exactly is human thinking? Moreover, what are the organs or mechanisms that humans use for thinking?
I believe that when our brains are thinking, there’s a central unit responsible for overall regulation, while N storage units surround this central unit to provide auxiliary support for cognition. Once external information enters the brain, it first flows into the central unit, where it undergoes identification, filtering, and categorization. Then, the central unit further diverts the already identified, filtered, and categorized information—based on its distinctive features—to various storage units for additional identification, filtering, and categorization. Next, these storage units retrieve relevant memories associated with the external information, thereby providing further assistance. Finally, each storage unit feeds back the relevant information about the external stimulus to the central unit.
For example: When the external world provides us with information A, this information A enters the central processing unit in our brain that is responsible for thinking. After the central unit identifies, filters, and categorizes information A, it directs the relevant information to various storage units. Each of these storage units, after undergoing identification, filtering, and categorization themselves, feeds back to the central unit an information B. Then, information B enters the central unit again, where it undergoes another round of identification, filtering, and categorization. The central unit then directs the relevant information once more to the various storage units. Each storage unit, after going through the same process of identification, filtering, and categorization, feeds back to the central unit an information C. And information C, in turn, enters the central unit again for yet another round of identification, filtering, and categorization... and so on, continuing indefinitely until the thinking process comes to an end.
I believe the model of thinking is as described above—a mechanism that operates like a central emitter: The central unit mentioned earlier serves as the central emission point, while the various storage units surrounding it function as receivers of the central emission. This, then, is my understanding of the thinking model.
From a functional perspective, I believe the primary function of the central unit in a thinking model is to process old information and then provide new information through feedback. Meanwhile, the various storage units surrounding the central unit are responsible for retrieving and storing information.
But what exactly is the central unit of the thinking model? I believe that when a person is thinking, the central unit is human consciousness itself. The information fed back from various storage units constitutes the stream of consciousness. Once something enters a person’s consciousness, it sets the consciousness in motion, prompting various mechanisms in the brain to retrieve and associate memories, thereby generating a stream of consciousness and bringing to mind other related thoughts and experiences about that particular event.
IV. Storage Methods
And what storage formats are used for the various storage units surrounding the central unit? I believe there are mainly two primary storage methods: discrete storage and linear storage.
Discrete storage, put simply, means categorizing information separately: each piece of stored data exists independently and doesn't interfere with any other. It's just like the biological taxonomy—in nature, we typically classify species based on certain specific characteristics: the animal kingdom is broadly divided into two...
The major categories are vertebrates and invertebrates. Vertebrates are further divided into mammals, birds, reptiles, amphibians, and fish; while invertebrates are classified into protozoa, platyhelminths, cnidarians, nematodes, annelids, mollusks, echinoderms, and arthropods, among others. Discrete storage also operates in this manner—systematically categorizing information based on a particular characteristic. Once the accumulated data reaches a certain threshold, it forms databases with distinct attributes.
For example, in life, when we encounter a person A, the features of his or her facial characteristics remain stored in our minds in a discrete form.
The storage method exists in this way. When we encounter another person, Person B, we use the facial features of Person A as a basic reference point to analyze what the facial characteristics of Person B are like. Yet at the same time, two distinct yet interconnected pieces of information actually exist in our minds: the facial feature information of Person A and the facial feature information of Person B. When we meet Person C, we retrieve both the facial feature information of Person A and that of Person B for comparison; meanwhile, the facial feature information of Person C is also stored separately in our minds. And so on, gradually accumulating information to build up a corresponding database.
As for the discrete storage method, we can also view it as a process of comparing similarities and filtering out differences. By storing existing old information in a discrete manner, and when the storage mechanism provides feedback to the central unit, new information is generated—this new information is precisely imagination. For example, when we hear someone say, “Character A hit Character B,” we can vividly visualize this scene in our minds, and the act of “hitting” itself is also conjured up in our minds. This is exactly what imagination is all about.
Discrete memory is the foundation of imagination. Only after N discrete memories have been established in a person's mind and organized into databases with distinct attributes can imagination arise—through the process of “comparing similarities in information and filtering out differences”—based on information provided by the external world. It is precisely at this stage that relevant scenes begin to take shape in the mind. This process also serves as a mechanism for generating new information.
After we receive information from the external world, it undergoes a process of identification, filtering, and classification by the central unit, giving rise to various information streams that are then routed separately into different storage units. Yet these information streams are interconnected—when arranged in a certain pattern, they can be linked together to form a linear sequence of storage.
In linear storage, linear information is essentially a collection of pieces of information that are both interconnected and independent from one another. A simple linear storage is simply the result of one simple piece of linear information being stacked on top of another.
V. The Four Types of Retrieval
As we can see, information retrieval plays a crucial role throughout the entire thinking process within the cognitive model. Among the four stages of thinking—information retrieval, information processing, information transformation, and information storage—information retrieval stands out as a relatively independent stage compared to the other three. Moreover, I believe that the storage stage is primarily driven by retrieval activities.
The retrieval process can be divided into: discrete retrieval, cumulative retrieval, linear retrieval, and organizational retrieval.
Discrete retrieval is essentially based on a discrete storage system. For example, if the external world provides information A, then, according to the information stored in various discrete memory units, all information that is similar to information A can be retrieved.
Accumulative retrieval is also based on discrete storage. Only after the brain has established a database through discrete storage can accumulative retrieval emerge. For example, we can only determine whether another person possesses certain facial features—such as those of Person A—once we have established a reference point for Person A’s facial features in the central unit.
What do B's facial features look like?
Linear retrieval is based on linear storage. It is a process of continuous forward reasoning that leads to the generation of new information.
Before discussing organizational retrieval, let’s first clarify the concept of “organization.” In my view, an organization is simply a process of getting something done. Organizational retrieval, also known as tree-like retrieval, is itself a process of conceptualization: In our minds, there exist various organizational processes—just like a tree—that gradually, based on one piece of information, retrieve all related information. Naturally, this retrieved information becomes organized into primary and secondary levels.
Sixteen-grid table
Academically speaking, thought modeling is a cognitive approach that uses specific tools—such as mind maps—and methods to construct a model of one’s internal thought processes and then visually represent them.
I have combined, in a table, the four modes of human thinking—identifying commonalities, filtering differences, segmenting information, and organizing information—and the four cognitive processes—information processing, information transformation, information retrieval, and information storage—into a single table.
In this table, the horizontal columns represent four different modes of thinking, while the vertical columns represent four different thinking processes. When these are combined, 16 new elements emerge.
Based on this table, we can determine what kind of thinking style a particular person has. For example, people who write short stories primarily employ linear thinking; whereas those who write novels tend to use organizational thinking.
The above is my understanding and perception of thinking. Thank you all for reading.
Sixth Collection: Self-Reference—Mirror Reflecting Mirror
Editor’s Note: During the second round of questioning, the editor presented the author with several examples of self-reference in mathematics. The author replied, “Could you elaborate more on self-reference?” This collection was written specifically in response to that remark—a popular-level exposition. The author himself admits that he is not well-versed in mathematics; therefore, the editor took over writing the mathematical sections, while the author retained responsibility for the parts concerning kinship analysis.
What is self-reference?
In a nutshell: Something that talks about itself, acts upon itself, or contains itself.
A mirror reflecting a mirror is self-referential; the statement “This sentence is false” is self-referential; and consciousness controlling its own consciousness is also self-referential. At first glance, self-reference may seem like a mere word game, yet it has been one of the central battlegrounds of twentieth-century mathematics and logic—where an entire family of famous paradoxes resides.
Cantor’s Diagonal Argument (1891). To prove that there are more real numbers than natural numbers, simply assume that someone has listed all the real numbers in a table, and then construct a new number along the diagonal—differing from the first digit of the first number, differing from the second digit of the second number, and so on. The newly constructed number will inevitably not appear in the table. No matter how one claims to have listed all the real numbers, there will always be some real number that slips through the net.
●The Turing Halting Problem (1936): There is no universal program that can determine whether an arbitrary program will ever halt when given an arbitrary input. The crucial step in the proof involves having the deciding program itself evaluate—“itself”—thus making self-reference the ultimate barrier to computation.
j Recursion and Fixed Points. y = f(y): a value that, after transformation, remains unchanged. Fractals’ self-similarity and recursive function calls in programming both belong to this family. Here, self-reference isn’t a nuisance—it’s the engine.
The self-reference in this set of books
Bring this mirror back—there are at least three instances of self-reference in the Conceptual Vector series:
Cognitive Framework: Consciousness is the central stage, and consciousness itself can control consciousness—actors on the stage also serve as directors.
Philosophical Framework: “Can Four Theories Fit Into Four Theories?” The author’s answer is that the four theories don’t belong to his own classification—in Russell’s words, they don’t belong to his own set.
The method itself: Concept vectors are products of the system (within the framework of thought, abstraction is inherent; concept vectors happen to be tools for abstraction), yet they in turn participate in the construction of the system itself. Here, the chicken and the egg don’t represent circular reasoning—they represent a fixed point.
Together, the three parts form a single sentence: As long as a system is strong enough, it will reflect itself; and once it reflects itself, both its boundaries and its driving force will emerge from that very reflection.
Volume 7: Complete Book Reference Cards
This card is a condensed summary of the author’s key ideas—the skeletal framework of the book’s main volume—contained in just one page. If you get stuck while reading any particular section, simply come back and take a quick look at it again.
General plan
The thinking tool is the original meta-level; each framework is a product of the thinking tool—a mathematical modeling that has been extended and expanded. Concept vectors belong to a category closely related to logic; philosophical frameworks deal with the forms of the real world—far from the thinking tools, yet close to the world itself.
Methods (Volume Three)
A concept is a combination of a set of binary dimensions. The binary represents the two poles of a number axis, and the individual is the position along that axis. Genes can be extended (subdivided or new genes added). It’s best to use coding for naming—names are merely superficial labels, while vectors represent the essence itself. Its six primary uses are: identifying gaps, removing duplicates, categorizing, completing missing information, quantifying analogies, and handling complexity. The criterion for evaluation is whether it works well—not whether it’s true or false. Our stance is holism: concepts can be broken down, yet they should not be completely disassembled—each binary dimension is a partitioning of the whole. This “whole” is relative to reductionism (currently the dominant approach in science); surface-level analysis is merely a necessary fallback when the whole cannot be fully disassembled. Concept vectors aren’t the end point—they’re the starting point: the next step is to connect these concepts with each other, enabling them together to form a functioning system. Ultimately, all frameworks are interconnected and operationalizable.
Dimension pairs (each independent, not combined into compound words)
Inner Circle (Volume One)
The Ego = Reason · Spontaneity (consciousness itself); the Id = Sensibility · Spontaneity (instinct itself); the Superego = Reason · Control (the means by which reason controls instinct); the Non-Ego = Sensibility · Control (the means by which sensibility manages reason).
Four Types of Thinking (Volume Two)
Table = discrete and flat; graph = discrete hierarchy; linear = continuous and flat; outline = continuous hierarchy. Memory is a cluster (distributed storage—on the graph, just write “memory” without distinguishing types); consciousness = central stage, with everyone receiving the broadcast; consciousness can control itself (self-reference, see Volume Six).
Four Essays (Volume Four)
Reductionism = simple dispersion; teleology = simple unity; holism = complex unity; collectivism = complex dispersion. Motion and stillness × internal and external = sixteen categories.
Honesty and Discipline (Volume Six)
Correspondence ≠ Definition—correspondence is probabilistic, heuristic, and not a proof; “55%” is just an example, meaning higher than random; position is rigid, while words are flexible; usage is unrestricted.
Withdrawn
“Distribution precedes consumption” (Note: You can only consume what has been produced); the correspondence between the Four Symbols of the Zodiac and MBTI (the Four Symbols don’t align with the seasons).
Historical term
Four merchants, the theory of appearance; the old names for four types of thinking (correspondence/derivation → relation/causality → attribute/deduction → table/linearity); the old word order of the circle (rational control/sensory control).