Prologue

The Word "Emo-i"

The Japanese word "emo-i" cannot be translated.

If we try to replace it with English equivalents,

emotional, nostalgic, sentimental, bittersweet—

several candidates present themselves.

But none of them are quite right.

Those are words that specify a distinct emotion.

"Emo-i" does not specify an emotion.

Is it nostalgia?

Is it a gentle heartache?

Is it a sense of relief?

Or a quiet, fleeting sadness?

Perhaps it is all of them.

No, even all of them combined does not suffice.

"Emo-i" describes a state where

emotions remain blended together, not yet broken down into components.

The state before meaning is fixed into words.

Before a name is given.

Before explanations begin.

The moment when the body responds first.

When looking at an old building along the coast.

When looking back at photographs from the nineties.

When touching the grain of film or a slightly faded sign.

Something is there.

Yet, it is not a clear message.

If anything, the opposite is true.

Precisely because it cannot be explained, it lingers deeply.

The word "emo-i"

accepts an excess of meaning just as it is.

And the fact that this word emerged naturally

proves that the world is not yet completely optimized.

In an optimized world,

everything is categorized,

emotions are tagged,

and meanings are quickly finalized.

But "emo-i" does not allow itself to be finalized.

It remains in suspense, wavering.

How do we handle this wavering?

How do we observe this undifferentiated sensation?

I would like to begin the conversation from there.

Part I

What is Language Generating?

Chapter 1

Thought Surrounding Language: Chomsky, Saussure, Wittgenstein, and Fillmore

To state the conclusion first: language too can be "generated." In fact, AI is already doing so. However, there is a decisive difference between the concept of "generation" accumulated through human linguistic study and the current AI-style "generation."

How do we perceive language? Four thinkers have offered different answers to this question: Chomsky, Saussure, Wittgenstein, and Fillmore. By examining their respective view points, we can consider how they differ from the "generation" performed by AI.

1.1 Chomsky: Generative Grammar and the Subject

Noam Chomsky proposed generative grammar in the 1950s.

His core assertion was that the human language faculty can generate infinite sentences from a finite set of rules. Here, "generate" does not mean outputting sentences at random; it meant developing structures based on grammatical rules.

For Chomsky, language was a matter of structure rather than statistics, of deep structure rather than surface words, based on the position that humans possess a universal grammar. Language arises from something inherent in the human brain. Society is layered on afterward. Grammar is an innate capability.

Crucial to this view is that language is tied to a subject. The presence of a speaker is essential; a sentence is something spoken by someone. Chomsky’s theory treated language as a cognitive capacity. The ability to generate infinite sentences from finite rules was believed to be wired into the human brain.

This is easy to understand if we think of programming languages. A programming language is the generative rule set itself. Once the grammar is defined, infinite sentences adhering to that grammar become possible. Chomsky believed that human language opens up infinite possibilities through finite rules in much the same way.

However, "generation" in this context does not mean mere output. It is the capacity to produce meaningful structures. Someone exercises this capacity with intent. This "someone" was indispensable in Chomsky's theory.

1.2 Saussure: The System of Signs and Differences

Ferdinand de Saussure, in his early 20th-century "Course in General Linguistics," reframed language as a system of signs.

His core assertion was that a linguistic sign is the union of a signifier (sound image) and a signified (concept). This union is arbitrary. In other words, there is no necessary connection between sound and meaning. There is no inherent reason why the sound "dog" refers to the concept of a dog; it is a matter of convention.

Furthermore, he emphasized that meaning is established through "difference." The meaning of a single sign is determined by its difference from other signs. "Dog" is not "cat." "Red" is not "blue." Language is a system of differences. Individual words do not hold meaning in isolation; their position within the system produces meaning.

Saussure distinguished between langue (the language system) and parole (individual speech acts). Langue is the system shared by society, while parole is the individual practice using that system. The object of linguistics is langue—it describes the system behind individual speech acts rather than the acts themselves.

What is important here is that the source of language lies in "society." Chomsky grounded language in the brain, while Saussure grounded it in society. In both cases, language is tied to "someone" or "something." Be it the subject (Chomsky) or the system (Saussure), language does not float without anchor.

1.3 Wittgenstein: Meaning is Use

Ludwig Wittgenstein, in his late work "Philosophical Investigations," fundamentally altered how we perceive language.

His core assertion was that meaning is use. "For a large class of cases—though not for all—in which we employ the word 'meaning' it can be defined thus: the meaning of a word is its use in the language." Words are used within contexts, within life. Through their use, meaning is born. Definitions in a dictionary do not determine meaning; how people use a word determines its meaning.

This is the concept of the language game. Language, like a game, is carried out with others toward a purpose, following rules. Greetings, commands, questions, jokes, prayers—each is a distinct language game. Even the same word holds a different meaning if the game changes.

The key point here is that meaning is embedded in a "form of life." Language is not an isolated collection of signs. Meaning arises within human activity, the body, relationships with others, and engagement with the world. Wittgenstein asserted that a "private language" is impossible. Meaning must be shareable; it is established through practice with others.

If Chomsky situated the foundation of language in the brain and Saussure in the system, Wittgenstein situated it in "practice." Who says it (the subject), what system is used (the difference of signs), and how it is used (the form of life)—these three perspectives illuminate language from different angles.

1.4 Fillmore: Case Grammar and Semantic Roles

Charles Fillmore proposed case grammar in the 1960s.

His core assertion was that in the deep structure of a sentence, there are not syntactic cases (such as nominative or accusative) but semantic roles—Agent, Patient, Instrument, Location, Time, and so on. Verbs select these semantic roles as a "frame." To understand a sentence is to read the relationship of who did what to whom (with what, where, and when).

For example, in the sentence "Taro handed the book to Hanako," surface grammar labels Taro as the subject, Hanako as the dative, and the book as the object. Semantically, however, Taro is the Agent (the one performing the act of handing), Hanako is the Goal (the recipient), and the book is the Patient (the thing being handed). Case grammar treats these semantic relationships as the foundation of sentence structure.

What is important here is that meaning is structured as "relationships." Not just the meaning of the vocabulary, but the relationships between words—who did what—constitute the meaning of the sentence. This relationship corresponds to the structure of events in the world. When we hear a sentence, we reconstruct the structure of an event.

Fillmore's perspective visualizes the link between language and the world. A sentence is not a mere chain of signs; it conveys a structured representation of "who did what" in the world. This structuring assumes an understanding of the subject. The listener reconstructs the relationship between "who" and "what." It is a bridge of meaning between the subject and the world.

1.5 Common Ground Among the Four: Language is Tied to Something

The views of these four thinkers can be organized as follows:

Chomsky: The source of language is the brain. A subject generates structure according to rules. A sentence is something spoken by someone.

Saussure: The source of language is society. Signs derive meaning through differences within a system. A sentence is the realization of the system.

Wittgenstein: The source of language is practice. Meaning is established within use. A sentence is something used within a form of life.

Fillmore: The source of language is correspondence with the world. Semantic roles reflect the structure of events. A sentence is something that conveys who did what.

While they differ, they share a commonality: language is tied to something. The brain, society, practice, or the world. It does not appear out of thin air without a foundation. Someone uses it, in some context, to convey something. This "someone," "something," and "context" support the meaning of language.

This concept of "generation" long remained at the center of linguistics. To understand language was thought to mean understanding its generative mechanism, its sign system, its practice of use, and its semantic structure. In all of these, the subject—the speaker, the listener, the user—was indispensable.

Chapter 2

"Generation" in Current AI

What about current large language models (LLMs)?

They do not possess explicit grammatical rules.

Nor do they symbolically manipulate deep structures.

They predict the "most plausible next word" from a massive distribution of text.

For example:

The Chomskyan approach is rule-based, while the LLM approach is probability-based.

The Chomskyan approach generates structure, while the LLM approach predicts distribution.

The Chomskyan approach relies on internal grammar, while the LLM approach relies on statistical patterns.

Chomsky has been highly critical of LLMs in recent years.

From his perspective, this does not constitute "understanding."

It means it fails to provide a theoretical explanation of language faculty.

Yet, in reality, AI is generating language.

And doing so with remarkable fluency.

Here lies the gap between theory and reality.

What AI is doing is indeed different from "generation" in the Chomskyan sense.

However, as output, much of it is indistinguishable from human language.

How do we understand this gap?

From this point, the question of what language is opens up.

Chapter 3

Generative Grammar vs. Transformer

This is not a simple technical comparison.

It is a clash of worldviews.

In the world of Chomsky's generative grammar,

the human brain inherently possesses the generative mechanism of language.

Language is a natural phenomenon, grammar is an innate ability, and society is layered on afterward.

Generation here is "the ability to produce structures that carry meaning."

Language is tied to a subject; who speaks is essential.

Conversely, in the world of the Transformer (LLM),

language is a massive probability distribution.

Language is a data-driven phenomenon, grammar is a statistical byproduct, and the subject is unnecessary.

Generation here is "the act of selecting the next most plausible token."

The fundamental divide can be organized as follows:

In generative grammar, the source of language is the brain. In the Transformer, it is data.

In generative grammar, the subject is essential. In the Transformer, it is unnecessary.

In generative grammar, meaning is inherent. In the Transformer, it is assigned afterward.

In generative grammar, society is external. In the Transformer, it is internalized.

In short:

Chomsky: Language creates society.

LLM: Society creates language.

This is a revolutionary shift.

The most critical difference is this:

Generative grammar: "A sentence is something spoken by someone."

LLM: "A sentence is something that emerges from a space."

Here, we find the phenomenon of the disappearance of the subject.

Sentences emerge from a space, detached from anyone's intent or physical presence.

In that case, to whom does the meaning belong?

Who assumes responsibility?

The question arises.

Chapter 4

The Impact on "Meaning"

This is the most critical point.

The fact that AI can generate language means

that the source of meaning is no longer human.

The subject of the utterance becomes ambiguous.

The locus of responsibility is dispersed.

For example:

News articles, legal documents, policy proposals, romantic messages, religious texts—

when these become AI-generated,

"Who spoke?"

"Who is responsible?"

"Whose intent is it?"

All of these questions become obscured.

Language is a social act.

It is bound to the sharing of meaning, the locus of responsibility, structures of power, and ethical frameworks.

When AI generates language, these bonds loosen.

However, there is a decisive difference with AI.

AI is not born. It does not die. It cannot be hurt. It does not face social sanctions.

This is where the disconnect with humanity lies.

Chomsky's theory treated language as a cognitive capacity.

Yet in truth, language is also an act predicated on a finite physical body and death.

Here lies the primordial difference between AI-generated language and human language.

Part II

Sovereignty, Accountability, and the Meaning Layer

Chapter 5

Language Generation and Sovereignty

From here, the discussion shifts directly to political philosophy.

The state is, in fact, dependent on language.

Law is language.

Contracts are language.

Declarations are language.

Accountability is language.

In other words, sovereignty is "whose words carry weight."

When AI can generate language, the question of whose words they are collapses.

This is incredibly significant.

AI writes policy documents.

AI drafts contracts.

AI writes judicial verdicts.

AI writes the news.

Consequently, the location of sovereignty becomes ambiguous.

In the future, sovereignty will shift like this:

Old: Sovereignty = Human right to speak

New: Sovereignty = Control over generative conditions

In other words, it is no longer about who speaks, but who defines the generative rules.

This is the heart of politics in the AI era.

Chapter 6

The "Accountability Model" in the Age of AI Language

This is the most troubling part.

In the modern model of accountability, responsibility looked like this:

An speaking subject exists. Intent exists. Consequences are attributed to them.

"I said it" is the foundation of accountability.

In the case of AI, no one has said it.

There is no intent.

It is a statistical output.

Thus, accountability floats unanchored.

This is already happening.

AI misinformation, automated contracts, discriminatory speech by AI—

instances where it is unclear who holds responsibility are on the rise.

Future accountability will shift:

From accountability of utterance to accountability of generation.

The responsible entities will be model designers, data providers, prompt engineers, and output adopters—

meaning accountability will become networked.

This signals the end of the individual accountability model.

Chapter 7

The Meaning Layer and Language Generation

The Meaning Layer is a mechanism for redesigning linguistic accountability.

The core issue of the AI era is simple.

A world where language is generated without accountability.

If we do not govern this, the social contract will collapse, legal systems will lose their force, and democracy will be hollowed out.

The Meaning Layer re-attributes "provenance" and "accountability" to language.

Specifically, this involves:

The generation history of utterances, records of generative conditions, traceability of intent, and authenticated language.

In other words, it manages "under what conditions it was generated" rather than "who said it."

In the AI era, language is liberated from the subject.

Yet, society cannot function without accountability.

This is why we require the Meaning Layer—the infrastructure that supports accountability.

Chapter 8

Societal Impact

In a society where AI generates language, several developments occur.

Authority wavers. Words are no longer a scarce resource.

The definition of truth shifts. Does truth become a matter of distributional consistency rather than human testimony?

Legal systems fall behind. The definition of accountability of utterance breaks down.

The value of "silence" increases. The weight of what is left unsaid grows in relative terms.

However, there remains a decisive difference.

AI can "generate," but it is not born, does not die, cannot be hurt, and does not face social sanctions.

Here lies the disconnect with humanity.

Chomsky’s theory treated language as a cognitive capacity.

Yet, in truth, language is also an act predicated on a finite physical body and death.

What does the future hold?

There are three possibilities.

AI floods the linguistic space entirely.

Human language becomes "authenticated."

Something prior to language—boundaries, silence, physicality—gains value.

Part III

The Origin of Language and the Climate of Meaning

Chapter 9

Takaaki Yoshimoto and the Origin of Language

This chapter re-examines the origin of language, focusing on Takaaki Yoshimoto's "What is Beauty for Language?" This question offers a unique Eastern perspective that Western philosophers did not possess.

9.1 A Radical Question: "What is Beauty for Language?"

Western philosophy has generally raised two questions separately: "What is beauty?" and "What is language?" Aesthetics discusses beauty, while the philosophy of language discusses language. Each is treated as an independent object.

The question Yoshimoto raised in "What is Beauty for Language?" (1965) is fundamentally different. "What is beauty for language?"—this does not attempt to define beauty in general. Rather, it asks what position beauty occupies within the practice of language. It interrogates beauty from the perspective of language. The subject of the question is not "beauty," but "language." Language asks what beauty is to itself. This inversion is radical.

Western aesthetics analyzes beauty as an object. Kant’s "Critique of Judgment," Hegel's aesthetics, analytical aesthetics—all perceive the phenomenon of beauty as an object of contemplation. Language is merely a medium to express beauty. Conversely, Yoshimoto’s question does not separate language from beauty. Language does not "express" beauty; rather, there is a moment when beauty "resides" within language. He questions that moment from the side of language. The relationship between beauty and language is positioned in reverse to Western aesthetics.

Eastern thought has a tradition of not separating subject and object. The oneness of body and mind. The non-duality of matter and spirit. The questioning subject and the questioned object are not separate from the beginning. In such soil, the question "What is beauty for language?" arises naturally. Language does not treat beauty as an external object; rather, beauty emerges within the practice of language. This relationship is accepted in the form of a question. The reason Western philosophers rarely raise this question is due to a strong habit of treating beauty and language as separate categories.

9.2 The Duality of Language: Self-Expression and Indicative Expression

At the heart of Yoshimoto's theory of language are two concepts: self-expression and indicative expression.

"Indicative expression" refers to language used to convey meaning. A tool for communication. It conveys something to others and receives something from them. If we compare it to a tree, it is the "branches," "flowers," and "leaves"—the parts that change with the seasons and are used according to purpose. The "meaning" of language represents the entire relationship of language as viewed from the indicative expression of consciousness.

"Self-expression" forms the very core of language. It is a language close to silence. Words that one addresses to oneself. If we compare it to a tree, it is the "trunk and roots"—the part hidden underground that supports the visible portions. The "value" of language represents the entire relationship of language as viewed from the self-expression of consciousness.

What is important here is that "self" does not simply mean "me." Yoshimoto uses "self" in a more abstract sense. Self-expression is "voluntary expression"—not to convey something to someone, but an expression that rises up in its own right. Indicative expression has a purpose and assumes an interlocutor. Self-expression exists as an impulse, prior to purpose.

Yoshimoto asserts that the value of literature ultimately converges with the value of self-expression. Artistry in literature cannot be explained merely by accuracy of transmission or the interest of a story. Deep within lies self-expression—language as an impulse, close to silence. Beauty resides there.

9.3 The Tremor Prior to Pointing

What Yoshimoto questioned was the location of the origin of language.

He did not view language merely as grammar, a system of signs, or a means of transmission. Chomsky, Saussure, Wittgenstein, and Fillmore—as seen in Chapter 1—grounded language in the brain, society, practice, and the world respectively. Yoshimoto looked at a stage prior to those. It is the tremor prior to pointing.

A "tremor" is a state where the body responds, but it has not yet become words. Distance from another. Tension. Shyness. Fear. Desire. Wanting to draw close, yet being afraid. Wanting to touch, yet fearing rejection. Wanting to be seen, yet wanting to hide. This pressure of contradiction produces a tremor in the body. This tremor eventually becomes words. It is the seed of words, prior to verbalization.

According to Yoshimoto, language passes through three stages. First, it exists as a physical impulse. Second, it becomes a projection toward another. Third, it is symbolized. In other words, language arises from "relationship" prior to meaning. Words emerge not to convey meaning, but to negotiate relationships with others. Communication is a byproduct.

This is fundamentally different from Chomsky.

Chomsky: Brain structure. Language is generated according to grammatical rules.

Yoshimoto: Tension with others. Language rises from the tremor between the body and another person.

For Chomsky, the source of language is internal. For Yoshimoto, the source of language lies within a relationship. It exists between two entities.

9.4 Beauty as the Moment the Tremor Inhabits the Symbol

For Yoshimoto, beauty is the moment the tremor inhabits the symbol.

A tremor alone is not beauty. A physical reaction does not become art by itself. A symbol alone is not beauty either. The manipulation of signs and transmission of meaning cannot explain the value of literature. Beauty resides in the moment the tremor "inhabits" the symbol—the moment the impulse takes the form of words while still retaining the texture of the impulse.

Smooth prose erases the tremor. Polished expressions dissolve tension. The core of Yoshimoto’s concept of "beauty" lies not in smoothness, but in an equilibrium on the verge of collapse. A moment when contradiction is barely maintained in the form of words. Not consistency or correctness, but that fragility produces beauty.

The assertion that self-expression is the source of value is bound to this concept of beauty. Indicative expression conveys meaning. Self-expression allows the tremor to inhabit the symbol. The artistry of literature relies on the latter. Texts generated by AI are fluent in the domain of indicative expression. However, can they generate self-expression—the moment the tremor inhabits the symbol? That is the parting of ways.

9.5 Eastern Uniqueness: The Primacy of Relationship

Let us examine why Yoshimoto’s theory of language could not have been conceived by Western philosophers.

First, "the primacy of relationship." Western philosophy of language separates subject and object. Speaker and listener. Meaning and reference. Language mediates between them. Yoshimoto places relationship itself at the origin. Language emerges within relationships. Subject and object do not precede; the relationship precedes them. The tension between two entities produces language. This primacy of the "between" resonates with Eastern thought—the "non-duality" of Zen, the "nothingness" of Taoism, and the Japanese concept of "ma" (interspace).

Second, "the primacy of physicality." Western philosophy often treats the body as the antithesis of the mind. Descartes' mind-body dualism. Language tends to be viewed as a product of the mind. Yoshimoto places the origin of language in the physical tremor. The body responds prior to meaning. That physicality lies at the root of language. In the tradition of the oneness of body and mind, this perspective is accepted naturally.

Third, "the inseparability of beauty and language." "What is beauty for language?"—this question itself does not separate beauty from language. Western aesthetics treats beauty as an independent object. Yoshimoto situates beauty within language. Beauty emerges within the practice of language. This unity aligns with Eastern theories of art—the "yugen" (grace) of Noh, the "yojo" (lingering emotion) of Haiku, and the "wabi" of the tea ceremony—which do not detach beauty from context.

9.6 Can AI Enter This Realm?

AI has no impulses. It does not desire others. It does not fear. It does not die. In short, it lacks the "tremor of relationship."

AI deals only with the third stage of language: symbolization. The manipulation of signs. The transmission of meaning. Prediction from distributions. These belong to the domain after language has already been established as symbol. In Yoshimoto's three stages, AI possesses neither the first stage (physical impulse) nor the second stage (projection toward another). AI generates only the output of the third stage.

AI can rearrange symbols. However, it does not generate a tremor. The moment the tremor inhabits the symbol—beauty for Yoshimoto—is not contained in AI outputs. It cannot be. Because a tremor requires a body, an other, and the tension between them. The possibility of loss, the vulnerability to hurt, the cost of mistakes, social sanctions, and finitude (death). AI has no stakes.

If the origin of language is neither grammar, nor statistics, nor power, but the tension between the body and another, then AI-generated language becomes a language without an origin. Meaning can be conveyed. Yet, from whose body did that meaning emerge, and through what tension? AI cannot answer this question. It is not that it cannot answer; rather, the question itself does not apply. Because there is no subject.

This is revolutionary. Language is generated detached from a subject. Yoshimoto's theory of language provides a coordinate axis for this revolutionary situation. It places the origin of language between the body and another. Language that lacks this origin is now being generated in massive quantities. We do not yet fully understand what this means.

Chapter 10

Yoshimoto and Heidegger

There is Heidegger's famous formulation that "language is the house of Being."

Here, language is not a tool to explain something, but the mechanism through which the world reveals itself.

Conversely, in Yoshimoto's terms, before social institutions, before grammar, and before meaning, language arises from distance, tension, shyness, fear, and desire in relation to another.

The contrast is as follows:

Heidegger: Language creates the "clearing" of the world. The condition for the world to emerge.

Yoshimoto: Language carves the "rift" with the other. The condition for relationships to emerge.

These two are not in opposition; they layer over each other.

Yoshimoto: The "origin" of language. The tremor.

Heidegger: The "consequence" of language. The clearing of the world.

The tremor undergoes symbolization to become the contour of the world.

Chapter 11

The Origin of Language and Eros

If we think of the origin of language as "information transmission," AI can get remarkably close to looking like language.

But through Yoshimoto’s lens, the origin is more physical, more precarious.

When we introduce Eros, sexual desire, attraction, and the impulse to draw near, language looks like this:

I want to draw close. But I am afraid.

I want to touch. But I might be rejected.

I want to be seen. But I don't want to be watched.

The pressure of this contradiction produces "words."

This is why beauty does not reside in consistency or correctness.

It is the moment when contradiction is barely maintained in the form of words.

LLMs can make prose smooth.

But the core of Yoshimoto-style "beauty" lies in an equilibrium on the verge of collapse, rather than smoothness.

Will the day come when AI has a tremor?

A primary tremor requires, at a minimum, the possibility of loss, vulnerability, the cost of making mistakes, social sanctions, and finitude (death).

The point is not "whether there is consciousness," but "whether there are stakes."

For an AI to truly possess a tremor, there must be an irreversible cost—such as its own collapse, exclusion, or erasure as a consequence of its output.

In other words, the day an AI trembles is not the day it "gets better at expressing emotion," but the day accountability and loss are implemented on the AI's side.

Chapter 12

The Philosophy of Return

This is not a theory of language, but a shift in ontology.

Books are not written to be read, but as places of Return.

Language is not information, but a space for existence to find stability.

The power to shake the world and a daily "thank you" exist on the same line of inquiry.

This is deeply Yoshimotian. A place of wavering is a domain where the tremor has not yet been extinguished.

The true issue with AI is not that it will run amok, but its ungovernability.

AI dismantles meaning, reality, and trust.

Nuclear weapons destroy the physical world. AI destroys the world of meaning.

In other words, AI is a technology that shakes the very foundation of language.

When noise disappears, meaning disappears.

Meaning = Unpredictability × Substantive Agency.

AI shears away unpredictability and substitutes agency. As a result, meaning dissolves.

For Yoshimoto, the origin of language is instability, relational tension, misunderstanding, disconnect—which is to say, noise itself.

Both nuclear weapons and AI are not crimes of ideology, but crimes of existence.

The motive for destruction is not politics. It is not meaning. It is not ideals. It is simply the desire to feel existence.

Because the origin of language is also the impulse to feel existence.

Integrating the four layers of language:

Layer 1: The Tremor (Yoshimoto)—The Origin of Language

Layer 2: The Clearing (Heidegger)—The Establishment of the World

Layer 3: Generation (AI)—The Deployment of Signs

Layer 4: Return—The Restoration of Existence

The role of language changes in the AI era.

Old: To explain the world.

New: To stabilize existence.

The Meaning Layer guarantees a place to return to.

Simply verifying truth or falsehood is not enough. Authentication is not enough.

What is required is the design of spaces where meaning can be restored.

Chapter 13

Sovereignty of Meaning

In modern politics, sovereignty is the power to make the final determination.

State sovereignty is the final authority of law. Popular sovereignty is the final authority of collective will. Individual sovereignty is the right of self-determination.

Sovereignty is the power to decide "what qualifies as reality."

Previously, the sovereignty of meaning was implicitly held by three entities:

Religion: Meaning = The word of God. The State: Meaning = Legal definition. The Individual: Meaning = Subjective interpretation.

Society maintained stability through the balance of these three.

What AI shattered is the structure of "who determines meaning."

AI can generate language infinitely, cross contexts, mimic authority, and speak without a subject.

As a result, the sovereignty of meaning disperses, and beyond dispersal, begins to dissolve.

When the sovereignty of meaning collapses, we witness

the destabilization of reality, the evaporation of accountability, and the hollowing out of existence.

The new question is not "who decides meaning." That is already impossible.

It is "who designs the conditions under which meaning can be established."

Redefining the Sovereignty of Meaning:

The power to design the conditions under which meaning is established, maintained, and restored.

The critical task is not to determine meaning, nor to enforce meaning.

It is to design the "field" where meaning is born.

The Three-Tier Structure of the Sovereignty of Meaning:

Tier 1: Generative Sovereignty (AI)—Who can generate language. Governed by tech corporations.

Tier 2: Sovereign Accountability (The Meaning Layer)—Who holds responsibility for meaning. Legal frameworks, audits, and trails.

Tier 3: Existential Sovereignty (Return)—Who can return to meaning. This is the new domain. No one has designed this yet.

The issue in the AI era is not that meaning cannot be generated, nor that meaning will be incorrect.

It is that we "cannot return" from meaning.

In other words, the essence of the sovereignty of meaning becomes the right to return to meaning.

The fundamental right in the AI era is neither freedom of speech nor freedom of expression, but "the right to restore meaning."

The right of individuals to have an environment where they can reconstruct meaning and restore existence.

This becomes the foundational condition of civilization, beyond law or technology.

The Ultimate Form of the Meaning Layer:

Old: Proof of trust.

Next: Traceability of accountability.

Final: Infrastructure for the restoration of meaning.

It is not a system to monitor language, nor to control language.

It is the design that preserves a space where humans can return to meaning.

Sovereignty shifts from the state to the individual, and then beyond the individual.

Ultimately, sovereignty belongs to the "field."

Sovereignty of Meaning = Sovereignty of the Field.

Part IV

The Climate of Meaning and Practice

Chapter 14

Shigesato Itoi and "Sovereignty of the Field"

Shigesato Itoi is one of the most successful practitioners of "the field as the sovereignty of meaning."

Crucially, he built this not as a theory, nor as an institution, but almost unconsciously, as a movement.

If you observe Itoi's work closely, there is a consistent principle.

He does not define meaning. He does not present correctness. He does not enforce an ideology.

Instead, he designs an "atmosphere where meaning can be born with peace of mind."

This is precisely the practice of Sovereignty of Meaning = Sovereignty of the Field.

"Hobonichi" is understood as a content site, an e-commerce platform, and a community.

But its essence is different. Hobonichi = A Space for the Restoration of Meaning.

There, individuals are not evaluated. Opinions do not become a matter of winning or losing. No correct answers exist.

Instead, what exists is "the sense that one can return."

The characteristics of Itoi's words are that they do not assert, do not over-explain, leave plenty of margin, and are gentle yet ambiguous.

This is not merely a writing style; it is a linguistic design that protects the "tremor of relationship."

There are stakes in Itoi's field: his own physicality, the history of human relationships, layers of time, memories of failure, and editorial choices.

In other words, meaning is not light.

AI can generate, but it cannot carry the gravity of meaning. This is the parting of ways.

Shigesato Itoi = Return Architect.

His work is not to generate language, nor to manage meaning.

It is to maintain an environment where humans can return to meaning.

Three Types of Sovereignty of Meaning:

Enforced Sovereignty (The State)—Decides meaning

Market Sovereignty (AI Platforms)—Optimizes meaning

Field-type Sovereignty (The Itoi Model)—Maintains the environment where meaning occurs naturally

Chapter 15

The Copywriter and the Marketer

The copywriter creates the focus of language.

A copy concentrates meaning, shifts how the world is seen, and redirects emotion.

In other words, it is the work of creating "the gravity of meaning" through language.

The marketer understands the market, reads the context, structures desire, and establishes exchange.

In other words, it is the design of relationships.

The primary field of battle: language for the copywriter, context for the marketer.

The object of manipulation: the concentration and flow of meaning.

The outcome: words that linger in memory and structures that translate into action.

The goal: to move hearts and to move behavior.

Yet, what both are doing is "designing the pathways of meaning."

Reading human desire, using language as a medium, and establishing relationships.

The work of circulating tremor. Semantically speaking, they deal with the origin of language.

The copywriter creates the moment. The marketer builds the enduring structure.

The copy is a flash of light. Marketing is infrastructure.

Itoi built a field out of language. Others try to design a field out of structure.

That is a more meta-level of work.

Meaning Architect. A designer of spaces where meaning can be established.

Chapter 16

The Concept of the "Climate of Meaning"

"Meaning" can never become infrastructure.

Because meaning is a flow, not a fixed structure.

The word "infrastructure" carries implicit assumptions: that it is stable, invisible, fixed, performs the same function repeatedly, and is efficient.

Roads, electricity, water, telecommunication networks—these are all governed by the logic of reproducibility.

But meaning wavers, changes, is highly individual, relational, and singular.

In other words, meaning is inherently non-infrastructural.

Thus, if you feel a sense of misalignment with the word "infrastructure," that intuition is correct.

An Itoi-style "field" is not a fixed structure. There is no blueprint, nor are there strict maintenance rules.

And yet, people return to it. Meaning is born naturally. Relationships endure.

A "field" is not infrastructure; it is closer to a climate.

Infrastructure: A structure that supports function.

Climate: An environment that supports the possibility of existence.

This is a decisive difference.

Meaning is not something to be made, nor something to be held. It is something that happens.

In other words, meaning is not an object, but a state.

Hence, the word "infrastructure" inevitably carries nuances of fixation, control, and management. This is the exact opposite of wishing to maintain a state where meaning can emerge naturally.

What is sought is not to generate meaning, nor to manage meaning.

It is to maintain the state where meaning can emerge naturally.

This is a highly delicate task.

A climate cannot be manufactured, nor can it be manipulated. But it can be cultivated. It changes over long periods. It fundamentally dictates human behavior.

Those who nurture the climate of meaning. Those who maintain the humidity where meaning can be born.

Meaning Climate Designer. Cultural Atmosphere Architect.

AI mass-produces language, rectifies information, and eliminates noise.

In other words, it works to dry out the climate of meaning.

Therefore, what will hold the greatest value going forward is not those who generate meaning, but those who preserve the humidity where meaning can be born.

Chapter 17

Why the Climate of Meaning Breaks Down

The climate of meaning dries out and breaks down through efficiency, homogenization, and the erosion of accountability.

In other words, it breaks down when "humidity" is lost.

The foundational principle of breakdown: "Breakdown occurs when liquidity stops."

Meaning is, by nature, circular, wavering, and fluid. When this stops, meaning dies.

The first cause of breakdown: efficiency.

AI, social media, and markets all share a common agenda: noise reduction, elimination of ambiguity, optimization, and the maximization of predictability.

From the perspective of meaning, efficiency = aridification.

Meaning is born from misunderstandings, delays, waste, and empty margins. Efficiency erases these.

The second cause of breakdown: homogenization.

Algorithms homogenize. Global culture uniformizes. The same words propagate.

Monoculture. Efficient in the short term, but fragile in the long term.

The third cause of breakdown: erosion of accountability.

The gravity of meaning is born from the fact that someone is assuming responsibility for it.

In the AI era, the subject of utterance disappears; we do not know whose words they are, and intent is unclear.

As a result, meaning becomes light, and evaporates.

The fourth cause of breakdown: perpetual stimulation.

Always new, always exciting, always updating. A state of hyper-arousal.

For meaning to be born, quietness, intervals, rumination, and silence are necessary.

The basic law of the climate of meaning:

Humidity of Meaning ∝ Uncertainty × Duration of Relationship × Weight of Accountability

Conversely, when efficiency × speed × lack of accountability increase, it dries out.

Chapter 18

Miura Jun and Noise

Miura Jun is someone who injects "wavering" and "noise" into the climate of meaning.

And he does so not as an ideology, but as play.

"My Boom," "Yuru-chara," affection for Buddhist statues, local refrigerator magnets, "Pilgrimage to Buddhist Statues"—

none of these are useful. Yet, they violently shift meaning.

Noise consists of things that are meaningless, off the beaten path, unevaluatable, and difficult to explain.

Yet within the climate of meaning, noise is humidity itself.

Miura Jun shifts the center of value, laughs at the mainstream, dismantles correctness, and loves the useless.

In other words, he is a humidifier against the dryness of meaning.

Shigesato Itoi warms the field. Miura Jun stirs the meaning.

Itoi is a gentle atmosphere. Miura is the laughter of displacement.

Itoi is stability. Miura is deviation.

Return and Drift.

Without both, a climate cannot exist.

Noise cannot be engineered through strategy. It emerges only from the singular point of "love."

Chapter 19

The Custodian of Atmospheric Pressure

Atmospheric pressure is the very condition that, while invisible, dictates all movement.

The custodian of atmospheric pressure is the deepest designer in the climate of meaning.

Meteorologically, temperature is the state of the air, humidity is the moisture level, and wind is movement.

Atmospheric pressure = the overall balance condition. When pressure changes, the wind shifts, clouds form, rain falls, and the climate transitions.

Atmospheric pressure moves everything while "directly doing nothing."

Atmospheric pressure within the climate of meaning is "to what degree is it possible to perceive meaning."

A sense of safety, breathing room, trust in relationships, margins of time, and tolerance for uncertainty. When these are high, meaning is born naturally.

What lowers pressure: instant evaluation, excessive visualization, constant connectivity, speed, comparison, and the pressure of optimization.

Five Conditions to Raise Atmospheric Pressure:

Slow down. Do not demand immediate answers, do not rush replies, do not hasten conclusions.

Declamatorily delay evaluation. Do not immediately say "that's good/bad," do not turn things into KPIs, do not assign scores.

Allow silence. Do not fill spaces immediately, do not fear voids, do not force topics.

Protect personal obsessions. Allow useless conversations, permit talking about what is unproductive, let individual passions be tolerated.

Accept incompleteness. Share things mid-way, do not over-organize, leave contradictions unresolved.

The person of atmospheric pressure stands on the boundary, not at the center.

The boundary is between the professional and the layperson, the organization and the individual, work and play, the inside and the outside.

The boundary is unstable, wavering, and subject to friction. For this very reason, meaning is born there.

The work of the custodian of pressure is not to move things, but "to halt excessive movement." Intervention must be minimal.

Do not move things directly. Timing should be deliberate. The fewer the words, the better. Do not resolve conflicts.

Meaning Pressure Designer. One who arranges the pressure conditions under which meaning occurs naturally.

Chapter 20

The Climate of Nineties Subculture

20.1 A Unique Air

In the 1990s, Japan possessed a unique air.

It was unrefined. The "correct answer" had not yet been settled. The boundary between major and minor was blurred. Failures were left exposed as they were. Before anyone could declare "this is the right answer," various forms coexisted. CD jackets, magazine layouts, street signs, TV shows—they look rough today. Yet, in that roughness, one feels no calculated intent. It is unclear what the creators were aiming for. Consequently, there was room for the observer to find their own meaning.

Shibuya-kei music, free paper culture, the experimental nature of local television, indie films, faded neighborhood arcades, the grain of VHS tapes. Commonly across these, the unfinished was tolerated. Trying it out, putting it out there, took precedence over pursuing perfection. Even if it failed, it did not disappear. Because it did not disappear, failure acquired character. This is humidity.

20.2 Shibuya-kei and "Somehow"

Shibuya-kei is a collective term for sounds like Yasuharu Konishi, Pizzicato Five, Flipper's Guitar, Original Love, Carnation—there is no precise definition. It blended jazz, bossa nova, French pop, and 60s sounds, feeling urban and slightly chic, yet somehow light. It was not heavy. It avoided becoming solemn.

The hallmark of Shibuya-kei was "somehow." It was not clear what it wanted to express. A mood existed. An atmosphere existed. But the message remained unformulated. Even when reading the lyrics, there was no distinct narrative. That was fine. The listener filled in the meaning themselves. That margin was humidity.

Tower Records, HMV, and TSUTAYA in Shibuya. Listening to samples there, picking up CDs of unknown artists. Being intrigued by the jacket. Hesitating over whether to buy. Buying it, taking it home, and listening. Sometimes it was a hit, sometimes a miss. Even if a miss, you didn't throw it away. It sat on the shelf, perhaps to be picked up again someday. That "someday" existed. It was not an era of instant decisions.

 20.3 Free Papers and Local TV

Free papers were stacked at stations and coffee shops. The cover designs were crude, the paper quality cheap. Yet, the content was dense: local gig information, introductions to indie bands, essays written by unknown hands, illustrations close to doodles. There was no editorial uniformity. The tone shifted page by page. Still, it stood as a singular "something."

Local television conducted experiments unimaginable today. Late-night slots, regional music programs, talk shows featuring amateurs. The video quality was poor, the direction rough. Yet, something was happening. It was unclear if they were serious or joking; the line was ambiguous. Viewers decided for themselves whether to take it seriously or let it slide. It was not an era where algorithms "recommended" content. You turned the dial and stumbled upon something. That mode of encounter was mainstream.

20.4 Indie Films and Arcades

Indie films screened in tiny theaters for brief runs. There was no promotional budget; posters were handmade. Word spread by mouth, or it didn't. If you missed it, you might never see it again. It might or might not come out on VHS. Even if it did, it might be a single tape in the corner of a rental shop. The act of searching itself held meaning.

Faded neighborhood arcades. Neon lights flashed, but the floors were dirty. It smelled of tobacco. The buttons on the consoles were worn shiny. You could tell someone had played for hours. A game was something you either cleared or didn't. If you stopped midway, you restarted from the beginning. Save data was fleeting, confined to that spot. You couldn't continue at home. So you stayed there, feeding in coins, staring at the same screen. That repetition was meditative. The goal was to clear the game, but the process absorbed most of the time. Unlike today's games, where the next objective is constantly presented, there was room to ponder what you were doing.

20.5 The Grain of VHS

VHS had poor resolution. Copying it repeatedly degraded it further. Colors bled; noise intervened. Yet, that grain feels "emo-i" today. Why? Because it is imperfect. The creator's intent is distorted by the limits of the technology. That distortion becomes the texture of the era.

Recording a TV program. Commercials were included. The moment of changing channels was captured. Traces of someone handling the remote remained. That "clutter" was preserved on VHS. The idea of cropping only the main feature and archiving it cleanly was not yet common. What was recorded remained as it was, mixed with superfluous elements. That mixture is humidity.

20.6 An Era Without Immediate Validation

The 90s represented the eve of the internet, an absence of algorithms, immature viral culture, and no social media. In short, it was an era where validation did not return immediately.

You made something and put it out. The response was slow. It might or might not get published in a magazine. Even if it did, you didn't know if readers picked it up. You played a gig. Audiences came or they didn't. Even if they came, you didn't know directly what they thought. Word of mouth traveled slowly, or not at all. That uncertainty served as pressure, but also as breathing room. You didn't have to produce results immediately. Things matured slowly. Failures did not get deleted; they became character. Failures were not accumulated as data, analyzed, and used for the next campaign. Failure simply remained. Because it remained, it became memory. Over time, memory turned into meaning.

20.7 Contrast with the Present: An Absence of Calculated Purpose

Today's world is dominated by thumbnail optimization, color correction, user flow design, CTR improvement, and algorithmic alignment—everything is designed toward an output. To be clicked, to be viewed, to be shared. The purpose is explicit and optimized. Waste is pruned, noise is reduced. It is efficient, yet it is dry.

In 90s photographs, one feels an absence of calculated purpose. It is unclear what they want to convey. Is it to sell a product, to archive, or for self-expression? The lines are blurred. Thus, the observer finds meaning for themselves—or they don't. Either is fine; nothing is forced. This is the source of the "emo-i" quality.

20.8 An Era Without a Center

There was no center, no singular correct answer. Industry had not yet fully commodified everything. The distance between culture and commerce was ambiguous.

What was "mainstream" and what was "sub" remained a matter of debate, never settled. Major labels and indies, TV, radio, magazines, Tokyo and the regions—boundaries existed, but they were fluid. Today's subculture could become tomorrow's mainstream, and vice-versa. No one could declare what was correct. The atmosphere resisted declaration.

It was an era of boundaries. Various things stood on the line. The boundary between major and minor, professional and amateur, work and hobby, seriousness and jest. Boundaries are unstable, wavering, and subject to friction. As discussed in Chapter 19, the boundary is where meaning is born. The 90s abounded in boundaries, which is why meaning was easily generated.

20.9 The 90s as a Climate of Meaning

Reinterpreting the 90s subculture as a climate of meaning yields the following:

Information is sparse. Margins exist. Speed is deliberate. Uncertainty is present. Validation is not immediate. Failure remains. Purpose is unformulated. There is no center. Boundaries are plentiful.

These all align with the basic law of the climate of meaning from Chapter 17: Humidity of Meaning ∝ Uncertainty × Duration of Relationship × Weight of Accountability. Uncertainty is high. Relationships (between work and observer, creator and receiver) have duration. The weight of accountability—the creator assumes responsibility for the work. Light things are quickly forgotten; heavy things remain. That weight produced humidity.

For those who lived through it, this chapter will trigger memories: the streets of Shibuya, that coffee shop, that record store, that arcade, those VHS tapes. Memories vary, but they share a "different flow of time" than today. Not fast, not optimized. Margins existed, and in those margins, meaning was born.

For readers unfamiliar with it, this chapter stands as a record of an era's "climate." Such an era existed. Under those conditions, meaning was born. How does it differ from today? What has been lost, and what remains? We leave that inquiry with the reader.

Chapter 21

The Word "Emo-i"

This book began with this word. and in this word, everything converges.

"Emo-i" is the central term of this book. Language generation, the disappearance of the subject, the climate of meaning, atmospheric pressure, Return, irreversibility—many of the concepts discussed thus far find their synthesis in this single term.

21.1 A Word Outside Translation

It is not that "emo-i" cannot be translated; rather, it is a word that exists entirely outside the premise of translation.

English has similar words: emotional, nostalgic, sentimental, bittersweet, evocative.

Yet none are correct. They all specify "which emotion" is being felt.

"Emotional" refers to emotion in general. "Nostalgic" is yearning for the past. "Sentimental" is tenderness or sadness. "Bittersweet" is sweet yet painful. "Evocative" conjures images.

Each has its meaning fixed, categorized, and tagged.

But "emo-i" does not specify. It points directly to the state where emotions remain blended.

Is it nostalgic, poignant, reassuring, or slightly sad? Perhaps all of them. Or rather, even all of them are not enough.

Here lies the heart of the linguistic theory discussed in Part I. In Chomsky’s generative grammar, sentences are generated infinitely from finite rules, but meaning is inherent to structure. In LLMs, sentences emerge from a distribution, but the subject is absent. Both operate toward "specifying meaning." Language points, categorizes, and transmits.

"Emo-i" does the opposite. It does not specify meaning. It does not categorize. It does not finalize transmission. Before language can close off meaning, the body responds first. What Takaaki Yoshimoto calls the "tremor"—the origin of language, the impulse prior to pointing—becomes words just as it is. "Emo-i" is the word that points to the state where the tremor is not yet fixed into a symbol.

21.2 A Word Where Meaning Does Not Close

The characteristics of "emo-i" can be summarized as follows:

Undefinable. Polysemic. Contradictory. Unexplainable. Sensory. Open-ended.

Normally, words finalize meaning. Communication assumes shared meaning. Ambiguity is viewed as an obstacle to transmission. Thus, we explain, define, and categorize. Emotions too are broken down into joy, anger, grief, and pleasure, tagged, and digitized.

"Emo-i" refuses this process. Or rather, it indicates the state prior to entering this process. Before meaning is fixed into words. Before a name is given. Before explanation begins. It captures that moment in a single word, holding it without finalizing. It remains wavering.

This directly connects to the "climate of meaning" in Chapter 16. Meaning is not something to be made, nor held; it is something that happens. Meaning is a state, not an object. "Emo-i" points to the moment when that state is still fluid. Meaning is "occurring" but not yet "solidified." In meteorological terms, the humidity is high, clouds are forming, but they have not yet turned into rain.

21.3 Not an Emotion, but a Blended State of Emotion

"Emo-i" is not an emotion itself, but the very state in which emotions are blended.

An emotion is deconstructible as an object of psychology or neuroscience: joy, sadness, anger, fear, surprise, disgust—in basic emotion theory, these combine to form complex emotions. However, that decomposition is already an act of "determining meaning." "Emo-i" is the state before decomposition.

There is also the concept of mixed emotions, with "bittersweet" as an example—the sweet and the bitter coexisting. Yet even "bittersweet" is defined as a mixture of two emotions. "Emo-i" lies prior to that. You do not yet know what is mixed, yet the body is responding. That is "emo-i."

For Yoshimoto, the origin of language lay in the tension of relationships with others. Wanting to draw close, yet fearing. Wanting to touch, yet fearing rejection. Wanting to be seen, yet wanting to hide. The pressure of this contradiction produces words. Beauty resides not in consistency or correctness, but in the moment when contradiction is barely maintained in the form of words. "Emo-i" may be the everyday word closest to that moment. Contradiction is not yet decomposed, barely held in verbal form. Or rather, it isn't even fully formed; the word "emo-i" acts as a surrogate for that form.

21.4 The Moment of Overlaid Time

"Emo-i" contains time. The past, the present, and future possibilities are contained simultaneously.

When looking at an old building along the coast. When looking back at a 90s photograph. When touching the grain of film or a slightly faded sign. In these moments, the past intrudes into the present. What once was, what is no longer, or what remains in a different form. That overlay produces the