Martha Woodmansee

Foreword

By Edo Segal

The sentence I almost skipped was about a vegetable.

Edward Young, 1759: the original work "rises spontaneously from the vital root of genius; it grows, it is not made." I was deep in the research for this book, reading about eighteenth-century German copyright disputes — not exactly my natural habitat — when that line stopped me. Not because it was beautiful, though it is. Because I recognized it. I had been saying it my whole career, just in different words.

Every pitch deck I ever built rested on that vegetable metaphor. The founder as origin. The vision as something that grows from within. The product as organic expression of a singular mind. I have stood in rooms and said, essentially: this came from me, and that is why it will work. I believed it. The feeling was real.

Martha Woodmansee spent four decades demonstrating that the feeling, while real, rests on a construction. The idea that a text belongs to its author because it expresses something uniquely hers — that the writer is the source, the spring, the vital root — was invented. Not discovered. Invented, in the late eighteenth century, by people who needed a philosophical justification for a new economic arrangement. Writers needed to own their texts because patronage had collapsed and the marketplace required property rights. The concept of original genius gave them those rights. It was an economic strategy dressed as aesthetics. And it worked so well that within a generation, nobody remembered it was a strategy at all.

This matters now — urgently — because every debate about AI and creativity is trapped inside that construction. Can AI be an author? Who owns AI-generated text? Is using Claude cheating? These questions assume that authorship means what the Romantic myth says it means: solitary genius producing original expression. Woodmansee shows that before the myth, Western culture understood writing completely differently. The compiler, the encyclopedist, the arranger of traditional materials — these were honored roles, not failures of imagination. Creativity meant skillful handling of what already existed, not conjuring from nothing.

I wrote The Orange Pill with Claude. The discomfort I described in Chapter 7 — not knowing where my contribution ends and Claude's begins — turns out to be a historically provincial discomfort. A medieval scribe would have been baffled by it. A Renaissance humanist would have shrugged.

Woodmansee does not tell you what to think about AI. She does something harder. She shows you the invisible walls of the room you have been thinking inside, and once you see them, every question about authorship, originality, and ownership in the age of AI looks completely different.

— Edo Segal ^ Opus 4.6

About Martha Woodmansee

1944-present

Martha Woodmansee (1944–present) is an American literary scholar and intellectual property theorist who spent four decades at Case Western Reserve University investigating the historical construction of authorship. Her landmark 1984 essay "The Genius and the Copyright: Economic and Legal Conditions of the Emergence of the 'Author'" traced the modern concept of the solitary creative genius to the economic crisis of the collapsing patronage system in eighteenth-century Germany, demonstrating that what Western culture treats as a natural fact about creativity was in fact an invention driven by market pressures. Her 1992 essay "On the Author Effect: Recovering Collectivity" documented the collaborative modes of literary production — compilations, encyclopedias, editorial collectives — that the Romantic genius ideology displaced. With legal scholar Peter Jaszi, she co-edited The Construction of Authorship: Textual Appropriation in Law and Literature (1994), a foundational interdisciplinary volume arguing that copyright law's reliance on Romantic authorship concepts was increasingly inadequate to the realities of contemporary cultural production. Her work anticipated, by decades, the legal and philosophical crises that artificial intelligence would create around questions of originality, ownership, and the nature of creative work.

Chapter 1: Before the Author — Writing as Compilation in the Age of Algorithms

The most consequential idea in the history of Western literary culture is one that most people have never examined, because it operates with the invisibility of an assumption so deeply held it feels like a fact of nature. The idea is this: that a written text originates in the mind of an individual author, that the text expresses something unique about that individual, and that the individual therefore owns the text in a legally and morally meaningful sense. This idea governs copyright law in every Western jurisdiction. It structures the publishing industry, the university, the literary prize, the plagiarism tribunal, and the school essay. It determines who gets paid for writing, how much they get paid, and what happens to those who copy without permission. It is, by any measure, one of the most powerful ideas in modern civilization.

It is also, as Martha Woodmansee demonstrated over the course of a career spanning four decades at Case Western Reserve University, an invention. Not a discovery. Not a recognition of something that was always true. An invention — constructed in the late eighteenth century, under specific economic pressures, to serve specific commercial and legal purposes, and then so thoroughly naturalized that the construction became invisible. The water in the fishbowl.

Before this construction took hold, Western culture understood writing very differently. The medieval scribe did not think of himself as an author. The concept would have been unintelligible to him. His task was transmission — the faithful copying of authoritative texts from deteriorating parchment to fresh vellum, from one generation's library to the next. His value lay not in what he added to the text but in the accuracy of what he preserved. An error was not a creative choice. It was a failure of duty. The text belonged to its source, understood as divine, classical, or traditional, and the scribe's role was to serve that source with the steady hand and careful eye of a craftsman whose craft was fidelity.

The Renaissance humanist operated under a different but structurally related framework. The great humanist encyclopedias and commonplace books were acts of curation, not creation in the Romantic sense. The compiler's art was selection and arrangement: which passages to excerpt from the classical authors, how to organize them by topic, what connecting commentary would render the collection useful. The compiler was valued for learning, judgment, and organizational intelligence — not for originality. Originality was not yet a literary virtue. Petrarch was celebrated not because he invented new forms but because he achieved new perfections of established ones, working within conventions that centuries of practice had refined.

Even as late as the seventeenth century, the dominant understanding of textual production remained fundamentally collaborative and tradition-dependent. Writers responded to previous writers. They adapted existing materials. They improved upon established forms. The metaphor that governed creative practice was not the fountain — the Romantic image of genius welling up from within — but the bee, gathering nectar from many flowers and transforming it into honey. The transformation was valued. The gathering was acknowledged. No one pretended that the honey materialized from nothing.

Woodmansee's scholarship established this pre-Romantic landscape with the precision of someone who built arguments from primary sources rather than theoretical assertion. Her 1984 essay "The Genius and the Copyright," published in Eighteenth-Century Studies, traced the emergence of the modern author concept through the specific language of German aesthetic theory, showing how the metaphysical vocabulary of genius was developed in direct response to the economic crisis of the collapsing patronage system. Her 1992 essay "On the Author Effect: Recovering Collectivity" documented the collaborative modes of literary production that the genius ideology displaced — the compilation, the encyclopedia, the commonplace book, the editorial collective — and argued that these modes were not primitive precursors to "real" authorship but legitimate forms of intellectual work rendered invisible by the triumph of the Romantic model.

The resonance of this historical analysis with the present moment is not incidental. It is structural. When a large language model generates text, it does something that bears a striking resemblance to what the medieval compiler did: it draws on a vast body of existing material, identifies patterns and connections within that material, and produces outputs that are consistent with the tradition while not being merely reproductive of it. The scale is different. The mechanism is different. The cultural context is incommensurable in many respects. But the structural position — the relationship between the individual producer and the accumulated body of material from which production draws — is analogous in ways that the contemporary discourse about AI authorship has not yet recognized, because the discourse operates within the Romantic framework and cannot see past its boundaries.

Woodmansee herself noted, as early as 1992, that electronic communication was already beginning to erode the authorship construct. Writing about the emerging culture of email and digital text, she observed that "the prose of electronic messages is almost as casual as conversation," and that the reader of one message could send off a reply "even incorporating part of the original message, blurring the distinction between their own text and the text to which they are responding." Her conclusion was pointed: "In a variety of ways, electronic communication seems to be assaulting the distinction between mine and thine that the modern authorship construct was designed to enforce."

That was 1992. The assault she identified has since become a siege. Large language models do not merely blur the distinction between one writer's text and another's. They dissolve it. A model trained on the entire digitized corpus of human writing produces outputs that cannot be traced to any specific source, that bear the statistical imprint of millions of authors without being attributable to any one of them, that are simultaneously derivative of everything and original in the narrow sense that no specific text in the training data matches the output. The Romantic framework can only classify this as imitation, reproduction, or fraud, because the framework has no category for creative production that is not original in the Romantic sense. The pre-Romantic framework, by contrast, has exactly the right category: skilled compilation, the creative arrangement of traditional materials into new and useful forms.

The Orange Pill provides a specific, documented case study of what this dissolution looks like in practice. The book's author describes working with Claude in a mode that would have been perfectly legible to a Renaissance humanist: posing questions, evaluating responses, rejecting what did not meet his standard, shaping the whole into a coherent argument that reflected his specific engagement with the world. He describes moments when Claude made connections he had not anticipated, when the collaboration produced insights that neither participant could have generated alone. And he describes the discomfort of not knowing where his contribution ends and Claude's begins — the specific unease of a person trained in the Romantic ideology of singular authorship discovering that his creative process is more collaborative than the ideology allows him to comfortably acknowledge.

The discomfort is genuine, and it deserves to be taken seriously. But it is a historically conditioned discomfort, not an existential one. A medieval writer would not have felt it, because the medieval writer did not operate under the assumption that the text should originate in a single consciousness. A Renaissance compiler would not have felt it, because the compiler's art was precisely the art of working with materials that came from elsewhere. The discomfort arises specifically from the collision between the Romantic ideology, which says the text should be the unique expression of an individual mind, and the lived experience of AI collaboration, which reveals a process that is irreducibly distributed.

Woodmansee's historical analysis provides the conceptual tools to understand this discomfort — to see it not as a sign that something natural is being violated but as a sign that a two-hundred-fifty-year-old construction is being exposed. The exposure is uncomfortable because the construction has been so thoroughly naturalized that challenging it feels like challenging reality itself. But the challenge is not to reality. It is to an arrangement — an arrangement that was made by specific people in a specific century to serve specific purposes, and that can be remade by different people in a different century to serve different purposes.

The implications extend directly into domains where the stakes are immediate and practical. Consider the contemporary panic about AI in education. Schools implement detection software. Teachers design assignments intended to be AI-proof. Institutions punish students who use AI tools in their writing, treating AI assistance as a form of academic dishonesty analogous to plagiarism. The assumption beneath these measures is that writing produced with AI assistance is not genuine writing — that it violates the norms of individual authorship that education exists to inculcate.

But the norms of individual authorship are themselves historical constructions, as Woodmansee's research demonstrates. The educational system teaches students to write as individuals because the Romantic ideology defines writing as individual expression. If the ideology is contingent rather than natural, then the educational norms it supports are contingent as well. The question becomes not whether the student's essay is "genuinely original" — a standard that Woodmansee's analysis reveals to be historically provincial — but whether the process of producing it developed the student's capacity for judgment, discrimination, and critical evaluation. These are different questions. They require different frameworks. And the historical framework that Woodmansee provides is the one best equipped to formulate them, because it sees past the Romantic assumption that has trapped the educational discourse in a debate about authenticity when the real issue is cognitive development.

The same reframing applies to the creative industries, where writers' guilds negotiate AI protections, publishers require disclosures, and studios restrict the use of AI in scriptwriting. These measures protect legitimate economic interests, and the protection is necessary. But the protection is being pursued through the wrong framework — the framework of individual authorship and original expression — and Woodmansee's analysis reveals why this framework will eventually prove inadequate. A framework grounded in the pre-Romantic understanding of creative production as collaborative, compilatory, and tradition-dependent would provide a more stable foundation for protecting creative workers, because it would not rest on a fiction about the solitary origin of the work. It would recognize that the creative worker's contribution is judgment, taste, and the specific quality of engagement that comes from lived experience — and it would anchor economic rights in the value of these contributions rather than in the mystification of the creative process.

None of this was visible before Woodmansee's intervention. The Romantic ideology had operated for so long as invisible infrastructure that challenging it required the specific tools of intellectual history — the capacity to recover lost frameworks, to read primary sources in their original contexts, to demonstrate that what seems natural is in fact constructed. These tools are not fashionable in the technology discourse, which tends to assume that the relevant history began with the transistor. But the relevant history began much earlier, in the scriptoria of medieval monasteries and the printing houses of Renaissance Venice and the debating chambers of eighteenth-century Germany, where the concept of the author was assembled, piece by piece, from economic necessity and philosophical ambition.

The scribe would have recognized Claude as a fellow worker in the tradition of transmission and compilation. He would not have been troubled by the collaboration. He would have been puzzled — genuinely, deeply puzzled — by the anxiety. Why does anyone think the text needs to come from a single mind? Why does anyone think the writer's subjectivity is the source of the text's value?

These remain the right questions. They are the questions the Romantic ideology was constructed to answer, and they are the questions that AI is now reopening with a force that the ideology's architects could not have anticipated. The chapters that follow will trace how those questions were answered in the eighteenth century, what consequences the answers produced, and what new answers the present moment demands.

---

Chapter 2: The Invention of Originality — An Economic Strategy Disguised as Aesthetics

In 1759, an aging English poet named Edward Young published a short treatise that would reshape the Western understanding of creativity for the next quarter millennium. Conjectures on Original Composition, addressed in the form of a letter to Samuel Richardson, argued that the true writer does not imitate established models but originates — that genius springs from the writer's own nature rather than from the tradition, that the original work grows organically from the writer's mind like a plant from a seed, while the merely imitative work is assembled mechanically, a dead thing made of borrowed parts.

Young's vegetable metaphor was not accidental. It carried the full weight of his argument. A plant grows from within. Its form is determined by its own nature, not by external arrangement. It cannot be assembled from components, because it is not an assembly. It is an organism, and its organic unity is the sign of its authenticity. The imitative work, by contrast, is a machine — constructed, derivative, lifeless. Young's distinction between the organic and the mechanical, the original and the imitative, the living and the dead, would become the foundation of Romantic aesthetic theory and, through that theory, the philosophical justification for the entire modern regime of intellectual property.

What is less often recognized — and what Woodmansee's scholarship placed at the center of the analysis — is that Young's argument emerged from a specific economic context and served specific economic purposes. The distinction between original and imitative creation was not a disinterested philosophical observation. It was an intervention in an ongoing battle over who could claim ownership of texts in a rapidly changing literary marketplace.

Before the eighteenth century, the Western literary tradition valued not originality but skill in handling received materials. The classical doctrine of imitatio — the creative imitation and improvement of established models — was not a failure of imagination but a discipline of mastery. Horace's Ars Poetica instructed the poet to follow established models. Quintilian's rhetorical training was built on the systematic imitation of admired predecessors. The good poet was the poet who could take a theme from Virgil, reshape it in the idiom of his own age, and demonstrate mastery of the tradition by improving upon it. Departure from established models was not genius. It was incompetence — a failure to command the forms that centuries of practice had refined.

Alexander Pope's dictum captured the classical position with epigrammatic precision: "True Wit is Nature to advantage dress'd, / What oft was thought, but ne'er so well express'd." The emphasis fell not on thinking something new but on expressing something known in a more perfect form. The measure of excellence was not departure from the tradition but perfection within it. Originality, in the modern sense, was not merely absent from the evaluative framework. It would have been disqualifying. The poet who departed too far from established models revealed not creative power but creative failure — an inability to work within the forms that embodied the accumulated wisdom of the literary tradition.

The shift from this framework to Young's celebration of original genius was driven, as Woodmansee demonstrated through meticulous archival research, by the economics of literary production. The crucial transformation was the collapse of the patronage system. For centuries, writers had been maintained by aristocratic patrons who commissioned works, supported writers in their households, and provided financial sustenance in exchange for literary production that served the patron's interests. The patronage system was hierarchical, often exploitative, and frequently degrading — writers were dependents, and their dependence shaped both what they wrote and how they understood the social function of writing. But the patronage system did not require the writer to own his text. The text was produced for the patron. Ownership was not the mechanism through which writers were compensated.

When the patronage system collapsed over the course of the eighteenth century, replaced by the literary marketplace in which writers sold works to publishers who sold them to readers, a fundamental problem emerged. The writer needed to claim ownership of his text, because ownership was now the basis of compensation. But what could the writer claim to own? If writing was compilation — the skilled arrangement of received materials into useful forms — then what, exactly, was the writer's property? The materials were common property. The language was common property. The tradition was common property. If the writer merely rearranged what everyone already shared, then the rearrangement was itself common property, and the writer's claim to exclusive ownership was incoherent.

The concept of original genius solved this problem with an elegance that was simultaneously philosophical and commercial. If the text was understood not as a rearrangement of common materials but as the unique expression of an individual consciousness, then the writer could claim ownership on the grounds that the text was inseparable from its creator. The text bore the imprint of its creator's genius the way a face bears the features of its owner: uniquely, inalienably, intrinsically. To copy it without permission was not merely to reproduce words. It was to appropriate something personal — to steal, in effect, a piece of the writer's identity.

Woodmansee traced this reconceptualization through the specific language of German aesthetic theory, where the philosophical foundations were laid with particular rigor. The German Romantics developed the concept of Geist — spirit, mind, intellect — as the source of creative production, and argued that the work of art was the objectification of the artist's Geist, the external form of an internal reality. This was not merely an aesthetic claim. It was a property claim. If the text was the external form of the writer's Geist, then the text belonged to the writer in the deepest possible sense — not as a possession that could be alienated but as an extension of selfhood that was by nature inalienable.

The German legal tradition formalized this connection. The concept of Geistiges Eigentum — intellectual property conceived as the property of the spirit — merged Romantic aesthetics with legal theory in a synthesis so compelling that it was absorbed into every subsequent copyright regime. The writer owned the text because the text was the product of her Geist. To reproduce it without permission was not theft of a commodity but a violation of spiritual integrity.

The construction was not cynical. Young, Herder, the Schlegel brothers, and the other architects of the genius ideology genuinely believed in the vision they articulated. The creative act does feel like expression — like something welling up from within rather than being assembled from without. The phenomenology is real. What is not real — what Woodmansee's historical analysis exposes — is the leap from phenomenology to ontology, from the experience of creation as expression to the conclusion that creation is expression, that the feeling of originality is evidence of actual originality, that the sense of individual authorship corresponds to the reality of individual authorship.

Large language models expose this leap with uncomfortable precision. A model trained on the accumulated corpus of human writing produces text that has all the formal characteristics of individually authored work — coherence, structure, argumentative development, stylistic consistency — without any individual consciousness behind it. The text has form, in exactly the sense that Fichte meant when he distinguished between the ideas in a text (common property) and the form of the text (the author's property). But the form was not produced by an individual Geist. It was produced by a statistical process operating on patterns extracted from millions of texts, none of which it "owns" in any meaningful sense.

This does not mean that AI-generated text is the same as human-authored text. The differences are real and important, and subsequent chapters will examine what survives the Romantic framework. But the formal indistinguishability — the fact that competent AI output cannot be reliably distinguished from human output by readers, critics, or even the copyright office — reveals that form is not, in fact, the unique and inalienable expression of an individual consciousness. Form is a pattern. Patterns can be learned. And the learning does not require a Geist.

The implication is not that the Romantic ideology was worthless. It was enormously productive. It enabled copyright law, which enabled the publishing industry, which enabled the mass distribution of knowledge on a scale that the patronage system could never have achieved. The ideology created the conditions for a flourishing literary culture, and the culture it created is the one we inhabit. The critique is not that the ideology produced bad outcomes. The critique is that the ideology mistook a historically specific economic solution for a universal truth about human creativity — and that the mistake is now producing confusion, because a technology has arrived that makes the economic solution visible as a solution rather than as a truth.

The confusion is everywhere. It is in the copyright office struggling to determine whether AI-assisted texts qualify for protection. It is in the university debating whether AI-assisted essays constitute plagiarism. It is in the publishing house inserting AI-disclosure clauses into contracts drafted on the assumption that a named author is the sole originator of the text. In every case, institutions built on the Romantic ideology are encountering a technology that the ideology cannot accommodate, and the institutional response is defensive — an attempt to enforce the old categories rather than to develop new ones.

Woodmansee's work provides the historical perspective that these institutional responses lack. The categories are not eternal. They were constructed for specific purposes at a specific historical moment. The moment has changed. The purposes can be served by different constructions. And the first step toward constructing alternatives is recognizing that the existing framework is a framework — not a mirror of nature but a human artifact, as contingent and revisable as any other artifact that humans have made.

---

Chapter 3: The Economic Origins of Genius — Copyright, Commodification, and the Literary Marketplace

The Statute of Anne, enacted by the British Parliament in 1710, is conventionally described as the first modern copyright law. The description is accurate but insufficient. The statute was not merely a legal instrument. It was a conceptual revolution — a transformation in the understanding of what a text is, who it belongs to, and on what grounds ownership can be claimed. And the revolution, as Woodmansee's analysis reveals, required an ideology that the statute itself could not supply. The law needed the concept of the author as original creator in order to function, but the concept did not yet exist in the form the law required. The legal structure and the Romantic ideology grew up together, each providing what the other needed, and the entanglement that resulted has shaped every subsequent debate about creativity, ownership, and the relationship between them — including the debate about AI that is now consuming the technology discourse with an urgency that the historical perspective makes both more comprehensible and more tractable.

Before the Statute of Anne, the book trade in England operated under a system of royal privilege. The Crown granted monopolies to members of the Stationers' Company — the guild of London publishers — giving them exclusive rights to print and sell particular texts. The writer's position in this arrangement was marginal. The writer produced the manuscript and sold it, usually for a one-time fee, to the publisher, who then held the exclusive right to print and distribute the work. The writer retained no continuing interest in the text after the initial sale. The economic relationship was straightforward: the writer sold a commodity, the publisher acquired it, and the commodity belonged to the publisher in the way that a bolt of cloth, once purchased from the weaver, belonged to the merchant.

The Statute of Anne disrupted this arrangement by introducing a time-limited period of exclusive ownership vested in the author rather than the publisher. The shift was motivated not by a sudden parliamentary concern for writers' welfare but by a political desire to break the Stationers' monopoly and to restructure the book trade along competitive lines. But the shift created a conceptual problem that the statute's language could not resolve. If the writer's ownership was merely the ownership of a physical manuscript, a manufactured object, then the writer's rights were exhausted by the sale of that object, and the publisher could do as he wished with it. For the writer to retain ongoing rights in the work — to own not just the manuscript but the text, the intellectual content expressed in the manuscript — the work had to be understood as something more than a physical object. It had to be understood as an expression of the writer's personality, inseparable from its creator in the way a thought is inseparable from the mind that thinks it.

Johann Gottlieb Fichte provided the philosophical architecture that the legal regime required. In his 1793 essay "Proof of the Illegality of Reprinting," Fichte drew a distinction that would become the cornerstone of copyright law in every Western jurisdiction and that remains, in essentially unchanged form, the cornerstone today. Fichte distinguished between the material content of a text — the ideas and information it conveys, which he conceded were common property once published — and the form of the text, the specific arrangement and expression that the writer gave to those ideas. The form, Fichte argued, was inseparable from the writer's intellect. It bore the imprint of the writer's unique cognitive process. No two minds, processing the same ideas, would produce the same form, because form was determined not by the ideas themselves but by the specific architecture of the mind that expressed them. To reproduce the form without permission was therefore to appropriate something that belonged inalienably to the writer — the writer's intellectual fingerprint.

Fichte's distinction was philosophically elegant and economically indispensable. It solved the ownership problem by anchoring property rights in expressive individuality while conceding that ideas themselves could not be owned. The distinction between idea and expression became the doctrinal foundation of copyright: you cannot copyright a fact, a concept, or an argument, but you can copyright the specific language in which the fact is stated, the concept is developed, and the argument is made. The protection attaches to the form, not the content, and the form is protected because it is the unique expression of an individual mind.

Woodmansee's contribution to this analysis was to show that Fichte's philosophical distinction was not a disinterested discovery about the nature of texts but a conceptual tool developed in direct response to the economic crisis of the German book trade. German publishers in the late eighteenth century faced rampant unauthorized reprinting — what we would now call piracy — and the lack of a unified German state meant that there was no national copyright regime to prevent it. The philosophical argument for authorial ownership served the interests of publishers as much as writers, because a text that belonged to its author could be exclusively licensed to a single publisher, and the exclusive license was the mechanism through which the publisher's investment in printing, marketing, and distribution could be recouped.

The economic interests were not hidden. They were openly acknowledged by the participants in the debate. But they were progressively obscured by the aesthetic ideology that was simultaneously developing — the ideology of original genius that Young had articulated in England and that the German Romantics were elaborating with philosophical sophistication. As the aesthetic doctrine grew more influential, the economic motives that had driven the development of the authorship concept receded from view. By the early nineteenth century, the concept of the author as original genius had been so thoroughly naturalized that its economic origins were invisible. Copyright was no longer understood as a commercial arrangement. It was understood as the recognition of a natural right — the writer's inherent right to the products of her own genius.

This naturalization is what Woodmansee spent her career exposing, and it is what makes her work so urgently relevant to the AI moment. Every contemporary debate about AI and copyright — Can AI-generated works be copyrighted? Who owns a text produced through human-AI collaboration? What compensation is owed to the writers whose works trained the model? — takes place within the framework that Fichte constructed and that Romantic aesthetics naturalized. The framework assumes that creative production is the expression of individual consciousness, that the form of a text bears the imprint of its creator's unique mind, and that property rights attach to form because form is the externalization of individual genius.

Large language models violate every one of these assumptions. The form of an AI-generated text does not bear the imprint of an individual mind. It bears the statistical imprint of millions of minds, processed through a mathematical architecture that is not a mind at all. The "expression" is not the externalization of a unique consciousness but the output of a pattern-matching process that operates on the accumulated residue of human textual production. Fichte's idea/expression distinction, which has governed copyright for over two centuries, was designed for a world in which expression was the unique product of an individual intellect. In a world where expression can be produced by a statistical process operating on aggregate data, the distinction does not merely strain. It becomes incoherent.

The U.S. Copyright Office's current position — that works generated by AI "without human creative control" are ineligible for copyright registration — is an attempt to preserve the Romantic framework by exclusion. The reasoning follows Fichte's logic: copyright protects the expression of an individual human mind, AI does not possess an individual human mind, therefore AI-generated expression is not copyrightable. The reasoning is internally consistent. It is also increasingly untenable, because the boundary between AI-generated and AI-assisted work is not a line but a gradient, and the gradient is becoming more continuous with every advance in AI capability.

Woodmansee's analysis suggests that the untenable position of the Copyright Office is not a failure of legal reasoning but a consequence of the ideological framework within which the reasoning operates. The framework was designed for a world of individual authors producing original texts. The world now contains a technology that produces texts through a process that is collaborative, derivative, and distributed — a process that resembles the pre-Romantic modes of textual production that the genius ideology displaced. The framework cannot accommodate this process because the framework was designed to exclude it.

The training data problem makes the inadequacy of the existing framework especially stark. Large language models are trained on vast corpora of copyrighted texts — books, articles, essays, stories, poems — produced by millions of writers who were not consulted about the use of their work and who receive no compensation for the contribution their texts make to the model's capabilities. This is a genuine injustice, and it is a consequence of the same copyright framework that the Romantic ideology supports. The framework was designed to protect individual works against unauthorized reproduction. It was not designed to address the use of works as training data for statistical models that learn patterns from the aggregate rather than reproducing any individual text. The gap between what the framework protects and what the technology requires is the space in which the injustice occurs.

Addressing this injustice requires moving beyond the Romantic framework rather than attempting to extend it. Collective licensing schemes, training data royalties, and other mechanisms for distributing the economic benefits of AI across the community of creators whose work constitutes the training corpus are all possibilities under active exploration. These mechanisms are difficult to implement, but they are conceptually coherent in a way that the extension of individual authorial rights to AI-generated works is not, because they do not depend on the fiction that the creative output can be traced to a single originating consciousness.

The economic origins of the genius construct reveal that copyright was never really about protecting genius. It was about solving a compensation problem — the problem of how to pay writers in a market economy. The solution worked brilliantly for two hundred and fifty years. It enabled a publishing industry, a knowledge economy, a cultural infrastructure of extraordinary productivity and reach. But the solution was specific to its moment. The compensation problem persists — writers still need to be paid, creative work still needs to be incentivized — but the specific mechanism that the Romantic ideology provided is no longer adequate to the conditions that now prevail.

The first step toward constructing adequate mechanisms is recognizing, as Woodmansee's scholarship makes possible, that the existing mechanism is a mechanism — not a law of nature but a human construction, made for a purpose, revisable when the purpose demands revision. The purpose demands revision now.

---

Chapter 4: The Institutional Machinery of the Romantic Author

The Romantic ideology of authorship would have remained a literary curiosity — an interesting chapter in the history of aesthetics — if it had not been absorbed into the institutional infrastructure of modern Western culture. What gives the ideology its power is not its philosophical elegance but its institutional embodiment. The concept of the solitary genius author is not merely believed. It is enforced — by copyright law, by publishing contracts, by university tenure committees, by plagiarism tribunals, by literary prizes, and by the entire apparatus of cultural evaluation that determines which texts are valued, which writers are rewarded, and which forms of creative production are recognized as legitimate. The institutions do not merely reflect the ideology. They reproduce it, generating the conditions under which the ideology appears self-evidently true.

Woodmansee recognized this institutional dimension as essential to understanding why the Romantic authorship construct persists even when its intellectual foundations are exposed. In The Construction of Authorship, the 1994 volume she co-edited with Peter Jaszi, she argued that the examination of authorship could not be confined to aesthetic theory. It had to address "the way that authorship is defined" across legal, commercial, and educational institutions, because "contemporary copyright law, rooted in a nineteenth-century Romantic understanding of the author as a solitary creative genius, may be inapposite to the realities of cultural production." The cautious academic phrasing — "may be inapposite" — concealed a radical claim: that the institutional machinery of authorship was built for a world that no longer existed and was being maintained not by its continued accuracy but by the inertia of the institutions that depended on it.

Consider the modern research university. The humanities are organized around the concept of the individual scholar producing original research. The single-authored monograph — a sustained original argument developed over years of solitary research — remains the gold standard of scholarly achievement. Tenure decisions, hiring committees, and promotion reviews hinge on whether a scholar has produced work sufficiently original to merit the institution's most consequential reward: a permanent position. The peer review system that evaluates this work operates on the assumption that the scholar is the sole originator of the argument, that the ideas in the monograph bear the imprint of a unique scholarly mind, and that the quality of the work can be assessed by evaluating the originality of an individual's contribution to knowledge.

The system has been productive. The monograph tradition has enabled deep, sustained engagement with complex questions. Peer review, for all its documented failures, has maintained a general standard of rigor. But the system is built on the Romantic ideology, and the cracks that AI introduces into the ideology extend directly into the institutional structure. When scholars use AI to generate literature reviews, identify patterns in data, draft arguments, or discover connections between disparate bodies of evidence, the fiction of the sole-authored original contribution becomes increasingly difficult to maintain. The scholar who uses Claude to identify a connection between two theoretical frameworks and then builds an argument around that connection has produced something valuable — but not a sole-authored original contribution in the sense that the tenure committee assumes.

The institutional response has followed the pattern that Woodmansee's analysis would predict: defensive enforcement of the existing framework. Universities issue policies restricting or prohibiting AI in scholarly work, treating AI assistance as academic dishonesty. The reasoning is Romantic: if the scholar did not produce the work from the resources of her own mind, the work is not genuinely hers, and presenting it as hers is fraud. The response is understandable — the institutional machinery requires the fiction of sole authorship to function — but it is ultimately unsustainable, because the technology is already integrated into scholarly practice in ways that prohibition cannot reverse, and because the prohibitory approach creates the perverse incentive of rewarding concealment over transparency.

Woodmansee observed a version of this dynamic three decades ago, noting that "as creative production becomes more corporate, collective and collaborative, the law invokes the Romantic author all the more insistently." The observation has proven prophetic. The U.S. Copyright Office enforces the requirement of human authorship with unprecedented rigidity precisely at the moment when authorship is becoming most distributed and algorithmically mediated. Universities tighten their plagiarism policies precisely when the tools available to students make the boundary between individual and collaborative production most porous. The institutions double down on the Romantic construct not because the construct is becoming more accurate but because it is becoming less so, and the institutional dependence on the construct makes the erosion feel existential.

The publishing industry embodies a parallel dependence. The author's name on the spine of a book is not primarily a philosophical claim about the origin of the text. It is a market instrument — a mechanism for organizing consumer expectations, creating brand loyalty, and enabling commercial exchange. The literary marketplace depends on named authors the way the stock market depends on ticker symbols: not because the symbol captures the full complexity of the underlying entity but because the market requires a shorthand that enables sorting, comparison, and transaction.

Michel Foucault's concept of the author-function, articulated in his 1969 lecture "What Is an Author?", provides the analytical framework for understanding how the author's name operates in the marketplace. Foucault argued that the name does not simply refer to the person who held the pen. It performs classificatory, organizational, and evaluative functions irreducible to the biological individual. The author-function groups texts into oeuvres, establishes interpretive frameworks, confers authority, and organizes the cultural hierarchy. A novel published under the name of a recognized author activates a different set of expectations, a different level of critical attention, and a different commercial trajectory than the identical novel published under an unknown name.

AI challenges the author-function not by eliminating the need for market sorting — the need persists and may even intensify in an environment of content abundance — but by revealing the distance between the sorting function and the Romantic ideology that legitimates it. The author's name sorts. The Romantic ideology says the name sorts because it identifies a genius. AI reveals that the name can sort perfectly well without the genius — that the brand can function regardless of whether the text behind it was produced by a solitary consciousness, a human-AI collaboration, or a team of ghostwriters operating under a single authorial persona, as has been common practice in commercial publishing for decades.

The revelation is uncomfortable for the publishing industry, which has invested heavily in the Romantic mythology of the author as the origin and guarantee of the text's value. The author tour, the literary interview, the social media presence, the curated image of the writer at work — all of this marketing apparatus depends on the reader's belief that the named individual is the source of the text's quality. AI does not destroy this apparatus. The apparatus has survived ghostwriters, collaborative pen names, and heavily edited manuscripts for centuries. But AI makes the apparatus visible as apparatus, as a commercial strategy rather than a reflection of creative reality, and the visibility is unsettling for an industry that has prospered by treating the strategy as transparent.

Literary prizes illustrate the institutional tension with particular clarity. The Booker Prize, the Pulitzer, the Nobel — each rests on the assumption that literary excellence is the achievement of an individual, that the prize recognizes not just a text but a person, a consciousness, a sustained creative career. When the judges award a prize, they are not merely saying "this book is excellent." They are saying "this person's creative vision, sustained over a career, has produced excellence." The award is personal. It attaches to a biography, a body of work, a trajectory of development that can only be traversed by an individual human being living in time.

AI problematizes this personal dimension without eliminating it. A book produced through human-AI collaboration may be excellent by every aesthetic standard the judges apply, but the excellence cannot be attributed to a single creative trajectory in the way the prize assumes. The human collaborator exercised judgment, maintained a vision, and accepted responsibility for the final product — capacities that subsequent chapters will argue are the elements of authorship that survive the Romantic framework. But the prize is not designed to recognize judgment. It is designed to recognize genius. And the gap between what the prize recognizes and what the creative process now involves is the institutional expression of the broader gap between the Romantic ideology and the reality of AI-mediated production.

The plagiarism tribunal is perhaps the starkest institutional expression of the Romantic construct. Plagiarism is defined, in virtually every academic context, as the presentation of someone else's work as your own. The definition assumes a clean boundary between "your own" work and "someone else's" — a boundary that the Romantic ideology draws clearly but that the reality of creative production has always blurred. Writers absorb influences. They internalize phrases. They develop ideas in conversation with others and then present those ideas in their own names. The boundary between "my" ideas and the ideas I have absorbed from my intellectual environment is not a line but a gradient, and the Romantic ideology maintains the fiction of the line by policing it through institutional mechanisms — citation requirements, plagiarism detection software, honor codes — designed to enforce a distinction that is more conceptual than real.

AI makes the enforcement of this distinction effectively impossible, not because AI encourages dishonesty but because AI dissolves the boundary on which the distinction depends. When a student converses with Claude about an essay topic, absorbs Claude's suggestions, and produces a text that integrates those suggestions with her own thinking, the resulting text occupies a zone that the plagiarism framework cannot classify. It is not "her own work" in the Romantic sense, because the ideas and their expression were shaped by a collaboration. It is not "someone else's work," because Claude is not a someone, and the specific text was not copied from any identifiable source. The framework requires a line. The practice produces a gradient. The gap between them generates the institutional anxiety that universities are currently experiencing.

Woodmansee and Jaszi anticipated this institutional crisis. Writing in 1994, they noted that the expansion of the concept of "text" and the transformation of the technology for producing texts required a reconsideration of "the ways 'authorship' is defined." Thirty years later, the reconsideration they called for has not occurred within the institutions themselves. The institutions have instead responded with defensive enforcement — tighter plagiarism policies, AI detection software, disclosure requirements — all designed to preserve the Romantic framework against a technology that the framework cannot accommodate.

The defensive response will fail, not because the institutions lack the power to enforce their rules, but because the rules are no longer aligned with the reality of how texts are produced. The misalignment will generate increasing friction — between institutions and their members, between official policy and actual practice, between the ideology that justifies the institution and the experience of the people who work within it. This friction is already visible in the growing gap between university AI policies and actual student and faculty behavior, in the tension between publishers' disclosure requirements and the ubiquity of AI in the writing process, and in the discomfort of creative workers who use AI tools daily but feel compelled to conceal or minimize their use because the institutional framework stigmatizes it.

The resolution will not come from within the Romantic framework. It will come from the recognition that the framework is a framework — a historical construction that served specific purposes and that can be reconstructed to serve different ones. Woodmansee's scholarship provides the essential first step: the demonstration that the institutional machinery of authorship was built for a specific historical moment, that the moment has passed, and that the machinery can be rebuilt without abandoning the legitimate functions it serves. The university's commitment to rigorous scholarship, the publishing industry's commitment to literary quality, the legal system's commitment to fair compensation for creative work — these commitments are valuable and worth preserving. The Romantic ideology that currently supports them is not the only possible foundation, and the sooner the institutions recognize this, the sooner they can begin constructing foundations adequate to the world they actually inhabit.

Chapter 5: The Collaborative Reality That Genius Conceals

The mythology of solitary genius has always required a feat of selective attention. To see the author as the sole originator of a text, one must not see the editor who restructured the argument, the publisher who insisted on a different opening, the colleague who suggested the central metaphor over dinner, the research assistant who found the crucial source, the spouse who read every draft and said "this part is not working." The mythology does not deny that these people exist. It renders them invisible — absorbs their contributions into the author's achievement the way a river absorbs its tributaries, so that by the time the water reaches the sea, no one can distinguish which drops came from which source.

Woodmansee's 1992 essay "On the Author Effect: Recovering Collectivity" was an explicit attempt to make the tributaries visible again. The essay documented modes of literary production that the genius ideology had displaced and rendered illegitimate: the compilation, the encyclopedia, the edited collection, the editorial collective, the collaborative revision. These were not primitive precursors to "real" authorship. They were sophisticated forms of intellectual work that produced some of the most consequential texts in Western history — texts that shaped law, science, theology, and political theory. The Encyclopédie of Diderot and d'Alembert, the collaborative product of over a hundred contributors working under editorial direction, was arguably the most influential publication of the Enlightenment. Under the Romantic framework, it is difficult to classify. It has no author in the singular sense. It has editors, contributors, revisers, arrangers — a network of intellectual workers whose individual contributions cannot be cleanly separated from the collective achievement. The genius ideology has no vocabulary for this kind of production. It can celebrate Diderot as an individual genius, but it cannot account for the Encyclopédie as an intellectual achievement, because the achievement is distributed across a network rather than concentrated in a person.

The concealment of collaboration is not limited to the pre-Romantic period. It operates with full force in the contemporary literary marketplace. The production of a novel involves agents who shape the project's commercial positioning, developmental editors who restructure narrative architecture, line editors who revise prose at the sentence level, copyeditors who enforce consistency, fact-checkers who verify claims, sensitivity readers who flag cultural missteps, designers who construct the visual identity, and marketing teams who determine how the book will be presented to the world. Each of these contributions shapes the final text. Some of them shape it fundamentally — a developmental edit that restructures a novel's chronology is not a minor intervention but a reconception of the work's narrative logic. Yet the conventions of attribution render all of these contributions invisible. The author's name alone appears on the spine. The single-author fiction is maintained because the publishing industry, the literary marketplace, and the cultural apparatus of reputation all depend on it.

The concealment extends into the academy with equal thoroughness. Scholarly work is presented as individually authored even when the conditions of its production are collaborative in ways that the conventions of citation cannot fully represent. A scholar develops an argument in conversation with colleagues over years of seminar discussion, conference presentations, and informal exchange. The argument is shaped by peer review — reviewers who identify weaknesses, suggest alternative framings, and direct the scholar toward evidence she might have missed. Research assistants contribute labor that is essential to the project's completion. The resulting monograph bears a single name and is evaluated as a single person's original contribution, because the institutional machinery of tenure and promotion requires this fiction in order to function.

Woodmansee's point was not that individual contributions do not exist or do not matter. The compiler exercises judgment. The editor makes choices. The scholar synthesizes. These are genuine intellectual acts, and they deserve recognition. Her point was that the Romantic framework systematically overstates the individual's contribution and systematically understates the contributions of everyone else — and that this systematic distortion is not an innocent error but a functional requirement of the economic and institutional structures that depend on the authorship construct.

The canonical case of Romantic authorship itself illustrates the concealment. William Wordsworth, whose 1800 Preface to Lyrical Ballads defined poetry as "the spontaneous overflow of powerful feelings," produced his most celebrated work in a collaboration so intimate that scholars have spent two centuries trying to untangle the respective contributions. Wordsworth and Samuel Taylor Coleridge conceived Lyrical Ballads together, wrote poems for the collection in each other's presence, and discussed the theoretical framework that would govern the project in conversations that neither fully recorded. Dorothy Wordsworth, William's sister, kept the journals from which he drew imagery, phrasing, and even entire passages — her description of the daffodils at Gowbarrow Park is the documentary source for one of the most famous poems in the English language, and the debt is not fully acknowledged in the poem itself.

Coleridge's intellectual debts were even more extensive. His concept of the "primary imagination," the theoretical cornerstone of Romantic poetics, drew heavily on the philosophy of Friedrich Wilhelm Joseph Schelling — a debt that Coleridge was reluctant to acknowledge and that scholars have documented with increasing precision. Thomas De Quincey accused Coleridge of plagiarism. The accusation was probably too strong, but the underlying observation was accurate: Coleridge's most original theoretical contributions were deeply collaborative, drawing on German philosophy, on conversations with Wordsworth, on the intellectual environment of early nineteenth-century England, in ways that the mythology of individual genius cannot accommodate without distortion.

None of this diminishes Wordsworth's or Coleridge's achievement. The synthesis was genuinely theirs — the specific angle of vision, the particular quality of attention, the capacity to draw from diverse sources and produce something that cohered. Woodmansee's analysis does not deny that individuals contribute something irreplaceable to the collaborative process. It denies that the individual's contribution is the whole of the process, and it insists that the systematic concealment of the rest has consequences — consequences for how the culture understands creativity, for how institutions evaluate creative work, and for how creative workers understand themselves.

The Orange Pill provides a contemporary case study that makes the concealment visible by refusing to participate in it. The book's author describes his collaboration with Claude with a level of specificity that has no precedent in the conventions of literary attribution. He identifies moments when Claude suggested a structure that changed the direction of the argument, moments when the collaboration produced connections that neither participant anticipated, and moments when Claude produced confident wrongness that required human judgment to detect and correct. The disclosure is itself a challenge to the Romantic framework, because the framework provides no mechanism for this kind of transparency. The conventions of attribution assume a sole author. There is no established way to credit an AI collaborator, no standard footnote, no conventional acknowledgment. The disclosure must be invented, and the invention is itself an act of authorship — a decision about how to represent a process that the existing conventions cannot represent.

The disclosure also reveals, by contrast, the extent of the concealment that the conventions normally enforce. Every book has collaborators whose contributions shape the final text. The difference is that most books do not name them, or name them only in an acknowledgments page that conventions dictate readers skip. The Orange Pill's transparency about Claude's role makes visible a collaborative reality that is usually hidden — not because previous authors deliberately concealed their collaborations, but because the institutional and cultural framework in which they operated provided no vocabulary for acknowledging them.

Woodmansee argued that the recovery of collectivity — the recognition that textual production has always been collaborative — was not a threat to the concept of authorship but a corrective. The corrective does not eliminate the author. It repositions the author within a network of contributions, distinguishing between the author's specific role — direction, judgment, synthesis, responsibility — and the contributions of others that the Romantic framework absorbs into the author's achievement. The repositioning is more honest than the concealment, and honesty about the conditions of production is a precondition for developing institutional and legal frameworks adequate to those conditions.

AI forces the repositioning by introducing a collaborator that cannot be absorbed. Previous collaborators — editors, research assistants, colleagues — were human, and their contributions could be rendered invisible by the conventions of attribution without producing visible contradiction. The editor who restructured a novel's chronology did not appear on the cover, and no one objected, because the conventions dictated that such contributions were subordinate to the author's creative vision. AI's contributions resist this subordination, because they are too substantial, too visibly constitutive of the final text, to be concealed by a convention designed for a different kind of collaboration.

The resistance is productive. It forces institutions to confront what Woodmansee spent her career demonstrating: that the solitary genius was always a partial truth elevated to a total ideology, and that the elevation served economic and institutional purposes that can be served — perhaps better served — by a more honest account of how texts are actually produced. The collaborative reality that genius conceals is not a diminishment of creative achievement. It is a more accurate description of creative achievement — one that honors the individual's contribution without mystifying it, that recognizes the network without dissolving the individual into it, and that provides a foundation for institutional and legal frameworks that can accommodate the full complexity of contemporary creative production, including the participation of machines.

---

Chapter 6: AI and the Return of the Pre-Romantic

There is a persistent assumption in the technology discourse that AI-generated text represents something unprecedented — a rupture in the history of human culture so fundamental that no historical framework can contain it. The assumption flatters the technology industry, which profits from the perception of radical novelty, and it flatters the cultural commentators who profit from the perception of radical crisis. But the assumption is wrong, and Woodmansee's historical scholarship reveals why.

What AI does to textual production is structurally analogous to what the medieval compiler, the Renaissance humanist, and the Enlightenment encyclopedist did: it draws on a body of existing material, identifies patterns and connections within that material, and produces outputs that are consistent with the tradition without being merely reproductive of it. The scale is incomparably larger. The mechanism is fundamentally different. The speed is orders of magnitude greater. But the structural relationship between the producing agent and the tradition from which it draws — the relationship that determines the character of the output — is recognizable. The pre-Romantic writer worked within a tradition and produced texts through the skilled arrangement of traditional materials. The large language model works within a training corpus and produces texts through the statistical arrangement of patterns extracted from that corpus. In neither case does the output originate in a singular consciousness creating from nothing. In both cases, the output is a synthesis of received material, shaped by a process that is simultaneously creative and derivative.

Woodmansee's recovery of the pre-Romantic understanding of textual production provides the conceptual vocabulary that the technology discourse lacks. The discourse can describe AI-generated text as "creative" or "not creative," "original" or "derivative," "authored" or "unauthored" — all binary categories inherited from the Romantic framework. What the discourse cannot do, operating within the Romantic framework, is describe AI-generated text as skillful compilation — as the creative arrangement of traditional materials into new and useful forms, produced by a process that is neither original in the Romantic sense nor merely imitative in the pejorative sense. This is precisely the vocabulary that the pre-Romantic tradition provides, and it is the vocabulary that the present moment needs.

The classical doctrine of imitatio was not, as the Romantic ideology retroactively characterized it, a failure of creative ambition. It was a discipline of mastery. The poet who imitated Virgil was not conceding defeat to a superior predecessor. She was engaging with the tradition at its highest level — taking the best that had been produced, studying its formal properties, and attempting to achieve a new perfection of established forms. The measure of success was not departure from the model but improvement upon it. The tradition set the standard. The individual's task was to meet and, if possible, exceed the standard through superior skill in handling the materials the tradition provided.

This framework evaluates creative work on the basis of quality — the skill with which traditional materials are handled — rather than on the basis of origin. The question is not "who created this?" or "is this original?" but "is this good?" — is the arrangement skillful, the variation interesting, the result effective? These are evaluative criteria that can be applied to AI-generated text without the conceptual contortions that the Romantic framework requires. A text produced by Claude can be evaluated on the basis of its argumentative rigor, its clarity of expression, its usefulness to the reader, without requiring a determination about whether the text is "original" or whether it has an "author" in the Romantic sense. The pre-Romantic framework dissolves questions that the Romantic framework cannot answer, not by ignoring them but by revealing them as artifacts of a historically specific ideology rather than as genuine features of the creative landscape.

The return to the pre-Romantic is not a regression. Woodmansee's analysis does not suggest that the Romantic ideology should be replaced by a simple restoration of classical imitatio. The Romantic period contributed something genuine to the understanding of creativity: the recognition that the individual matters, that the specific configuration of experiences, sensibilities, and commitments that a particular person brings to the creative process is irreplaceable, and that the products of that specific configuration are genuinely unique. This insight survives the critique of the genius ideology, because it does not depend on the ideology. The individual's contribution is real even when the mythology that surrounds it is not.

What the Romantic ideology got wrong was the leap from the individual's contribution to the individual's self-sufficiency — the claim that the text originates in the individual alone, that the tradition is merely the background against which genius operates rather than the medium through which it creates. AI corrects this error not by eliminating the individual but by making visible the medium. When Claude generates text, the medium — the entire accumulated corpus of human writing on which the model was trained — is undeniably present. The individual who collaborates with Claude cannot pretend to be self-sufficient, because the collaborator's dependence on the tradition is not merely implicit but computationally explicit. The tradition is not a vague background influence. It is a training dataset, and the dataset's contents directly determine the range and character of the output.

The Orange Pill documents this dynamic through the specific experience of a builder working with Claude on a sustained creative project. The author describes a process that would have been legible to a Renaissance humanist: posing questions, evaluating responses, selecting the most promising material, arranging it into a coherent structure, and shaping the whole to serve a specific purpose. The process is not imitation in the pejorative sense. It is compilation in the honorific sense — the skilled arrangement of materials drawn from a tradition too vast for any individual consciousness to contain, directed by the specific judgment and vision that only the individual can provide.

The analogy illuminates what the technology discourse has struggled to articulate: the sense in which AI-assisted creation is simultaneously new and traditional, simultaneously creative and derivative, simultaneously the product of individual vision and collaborative process. The pre-Romantic framework accommodated this simultaneity without difficulty, because the framework never assumed that creativity required originality in the first place. The Romantic framework cannot accommodate it, because the framework defines creativity as originality and therefore must classify anything that is not original as not creative. This classificatory failure is what produces the cultural anxiety about AI authorship — an anxiety that is real but historically provincial, rooted not in the nature of creativity but in a specific ideology of creativity that is barely two and a half centuries old.

Woodmansee's recovery of the pre-Romantic does not resolve the practical problems that AI creates — the legal problems of copyright, the economic problems of compensation, the institutional problems of evaluation. These problems require specific solutions that historical analogy alone cannot provide. But the recovery provides something that the specific solutions cannot provide on their own: a conceptual framework adequate to the reality of AI-mediated creative production. A framework that does not require the fiction of individual originality to account for creative value. A framework that can evaluate creative work on the basis of skill, judgment, and effectiveness rather than on the basis of the origin of the raw material. A framework, in short, that sees compilation not as the failure of creativity but as its most fundamental mode — the mode that the Romantic interlude of the last two centuries temporarily obscured but could not permanently replace.

The anxiety about AI authorship is, in this light, the anxiety of a culture discovering that its dominant model of creativity was always an anomaly rather than a norm — a two-hundred-fifty-year deviation from three thousand years of literary practice in which the skilled arrangement of traditional materials was not merely tolerated but celebrated as the highest form of creative achievement. The deviation served its purposes. It enabled copyright, the literary marketplace, and the institutional infrastructure of modern cultural production. But the deviation was always a deviation, and the technology that is now forcing the culture to confront this fact is not destroying creativity. It is returning creativity to its oldest and most fundamental self-understanding.

---

Chapter 7: What Survives the Author-Function

If the Romantic construct of authorship is a historical invention — assembled from economic necessity and philosophical ambition, maintained by institutional machinery, and now exposed by a technology that refuses its premises — then the question that matters most is what remains when the construct is stripped away. Not everything in the Romantic framework is ideology. Not everything is a fiction designed to serve commercial purposes. Some elements correspond to genuine features of creative practice, features that persist regardless of the technology through which creation occurs. Identifying these elements is the essential task, because any framework adequate to AI-mediated production must be built on what is true rather than on what is merely traditional.

The first element that survives is judgment. The Romantic ideology located creative value in the writer's unique subjectivity — in the genius that produced original expression. Woodmansee's analysis challenges the claim about originality but leaves intact the claim about judgment. The capacity to evaluate, to discriminate, to choose wisely among possibilities — this capacity is not an artifact of the Romantic construct. It is the operational core of every mode of textual production that Woodmansee documented, from the medieval compiler who chose which passages to preserve to the Renaissance humanist who decided how to arrange an encyclopedia to the Enlightenment editor who determined which contributions to the Encyclopédie met the project's standard. In each case, the creative contribution was not the generation of raw material but the exercise of judgment about which materials to include, how to arrange them, and what standard of quality to enforce.

AI makes judgment more necessary, not less. When the generation of raw material is cheap — when Claude can produce ten competent paragraphs on any topic in seconds — the scarce resource is no longer the capacity to produce but the capacity to evaluate production. The person who can distinguish between the paragraph that is merely competent and the paragraph that is genuinely illuminating, between the argument that is plausible and the argument that is true, between the structure that works and the structure that sings — that person's judgment has become the most valuable element in the creative process, precisely because everything else has become abundant.

The Orange Pill provides a specific illustration. Its author describes the discipline of rejecting Claude's output when it "sounds better than it thinks" — when the prose is polished but the idea underneath is hollow. He describes retreating from the screen to a coffee shop notebook, writing by hand until he found the version of an argument that was his own — rougher, more qualified, more honest about what he did not know. The retreat is an exercise of judgment: the deliberate interruption of the collaborative flow to test output against the standard of one's own understanding. The version that survives the test is not the version Claude produced. It is the version that the author's judgment authorized. And authorization — the acceptance of responsibility for the quality and truth of the output — is itself a creative act, perhaps the most consequential creative act in an environment where generation is automated and evaluation is not.

The second element that survives is voice. The Romantic ideology mystified voice, treating it as the expression of an ineffable inner essence that only genius possesses. Woodmansee's historicization reveals the mystification: voice is not essence but biography. It is the specific quality that accumulates in a person's prose through years of reading, writing, thinking, and living — the texture of engagement with language shaped by everything the person has experienced. Toni Morrison's prose does not sound like Don DeLillo's, and the difference is not a difference of essence but of biography — of what each writer read, where each lived, what questions each found urgent, what rhythms each absorbed from the specific cultural environments that formed them.

Voice in this demystified sense is not threatened by AI. It is clarified by AI. When the mechanical aspects of writing — grammar, structure, research, organization — are handled by a machine, what remains is precisely the element that the machine cannot handle: the biographical specificity of the writer's engagement with language. AI can produce competent prose on any subject. What it cannot produce is the specific gravity that comes from a person who has lived inside the ideas, who has felt the consequences of arguments in their own experience, who brings to the arrangement of words a weight that is not stylistic but existential.

This distinction between style and voice is crucial. Style can be mimicked. A large language model can produce prose in the style of Morrison or DeLillo or any writer whose work appears in sufficient quantity in the training data. The mimicry can be convincing at the surface level. But style is not voice. Style is a set of formal features — sentence length, vocabulary, syntactic patterns, figurative habits. Voice is the residue of a life lived in language. It manifests not in the formal features that can be catalogued and replicated but in the weight behind the features, the sense that the words carry the experience of the person who chose them. The machine can replicate the features. It cannot replicate the weight, because the weight comes from living, and the machine has not lived.

Woodmansee's historical analysis supports this distinction without articulating it in these terms. The pre-Romantic compiler exercised judgment in selecting and arranging materials, and the quality of the compilation reflected the compiler's specific expertise, taste, and intellectual commitments. The compilation bore the imprint of its compiler not because the compiler created from nothing but because the compiler's specific angle of vision determined which materials were selected, how they were arranged, and what connecting commentary shaped the reader's experience. The imprint was biographical — the product of a specific intellectual life — rather than essential. It was voice in the demystified sense: the trace of a particular person's engagement with a body of material, detectable in the choices made rather than in the raw materials chosen.

The third element that survives is responsibility. The Romantic ideology tied responsibility to ownership: the author is responsible for the text because the text is the author's property, the expression of the author's genius. Woodmansee's critique dissolves the ownership claim but does not dissolve the responsibility claim. The person who publishes a text — who decides that the text is ready for the world, who puts their name on it, who accepts the consequences of its reception — is responsible for that text regardless of how it was produced. The responsibility is not diminished by the collaborative nature of the production. If anything, it is intensified, because the more distributed the production process, the more important it becomes that someone accepts ultimate accountability for the result.

Responsibility in this sense is not a Romantic concept. It is an ethical one. The medieval compiler was responsible for the accuracy of his compilation. The Renaissance editor was responsible for the quality of his encyclopedia. The Enlightenment translator was responsible for the fidelity of her translation. In each case, the responsibility attached not to the claim of original creation but to the act of releasing a text into the world — the decision that the text was good enough, accurate enough, useful enough to deserve a readership. The same responsibility attaches to the person who collaborates with AI and publishes the result. The collaboration distributes the production. It does not distribute the responsibility. Someone must stand behind the text, and the someone is the person who exercised judgment, not the machine that generated options.

These three elements — judgment, voice, responsibility — constitute what survives the Romantic framework. They are not less valuable than genius. They are more honestly described, more precisely identifiable, and more practically cultivable. Genius cannot be taught, because genius, in the Romantic framework, is a gift — an innate capacity that either exists in a person or does not. Judgment can be taught, because judgment is a skill developed through practice, feedback, and the accumulation of experience. Voice can be cultivated, because voice is the product of a specific life lived attentively, and the specificity deepens with every year of reading, writing, and engagement with the world. Responsibility can be accepted, because responsibility is a choice — a decision to stand behind one's work and accept the consequences of its effects.

The post-Romantic author — the author who works after the myth has been exposed — is not a diminished figure. She is a more honest one. She does not claim to create from nothing. She claims to create from everything — to synthesize materials drawn from a tradition larger than any individual, directed by the specific judgment that only her particular biography can produce, shaped by the voice that her specific life has deposited in her language, and released into the world under the responsibility that only a person with stakes in the world can bear.

Woodmansee's scholarship provides the historical ground on which this reconceived author stands. The ground is not new. It is the ground that writers occupied for three thousand years before the Romantic interlude — the ground of the compiler, the arranger, the skilled handler of traditional materials. The Romantic interlude elevated the author above this ground, onto the pedestal of genius. AI is returning the author to the ground. The return is not a fall. It is a homecoming.

---

Chapter 8: Writing After the Myth

Woodmansee did not set out to destroy the concept of authorship. She set out to historicize it — to demonstrate that an idea the culture treated as natural was in fact constructed, that the construction occurred at a specific historical moment under specific economic pressures, and that understanding the construction was a prerequisite for navigating the challenges that new technologies of textual production were creating. The demonstration was successful. The construction has been exposed. The question now is what to build on the cleared ground.

The question is practical before it is philosophical. Writers are working with AI now, not in some speculative future but in the daily reality of contemporary creative practice. Scholars are using AI to research, draft, and revise. Students are using AI to learn, to write, and to think through problems. Publishers, law firms, marketing agencies, and newsrooms are integrating AI into workflows that were designed for human-only production. The integration is not waiting for the philosophical questions to be resolved. It is happening in advance of resolution, and the people who are doing the integrating need frameworks that are adequate to their experience, not reassurances that the old framework will eventually be patched to accommodate the new reality.

The first element of such a framework is transparency about the conditions of production. Writing after the myth means acknowledging collaboration rather than concealing it. Woodmansee's recovery of the collaborative reality that genius conceals is not merely a historical correction. It is a guide for contemporary practice. The concealment that the Romantic framework required — the invisibility of editors, collaborators, and institutional contributors — was functional in a world where the conventions of attribution served the market's need for a single named author. In a world where AI collaboration is ubiquitous and increasingly detectable, concealment becomes both dishonest and unsustainable. Transparency about the conditions of production is not merely an ethical obligation. It is a practical necessity, because the alternative — the pretense of sole authorship in the face of a technology that makes the pretense implausible — generates the institutional dysfunction that Chapter 4 described: covert use, inconsistent enforcement, and a growing gap between official policy and actual practice.

The second element is the relocation of evaluative criteria from origin to quality. The Romantic framework evaluates creative work by asking who produced it and whether it is original. The pre-Romantic framework, as Woodmansee's scholarship recovers it, evaluated creative work by asking whether it is good — whether the arrangement is skillful, the argument rigorous, the expression effective, the result useful. This evaluative shift is not a lowering of standards. It is a clarification of what standards actually measure. The question "is this original?" is a question about provenance. The question "is this good?" is a question about value. Provenance and value are related but not identical, and the Romantic framework's conflation of the two — its assumption that original work is valuable because it is original — is precisely the conflation that AI exposes as untenable.

A text produced through human-AI collaboration may fail to meet the Romantic standard of originality. It cannot claim to have originated in a single consciousness. It cannot claim the organic unity that Young's vegetable metaphor demanded. But it can meet the standard of quality — it can be rigorous, illuminating, well-structured, and useful — and the standard of quality is the one that actually serves readers, institutions, and the broader culture. The shift from origin to quality as the primary evaluative criterion is the single most consequential change that the post-Romantic framework requires, and it is a change that Woodmansee's historical analysis makes possible by demonstrating that the origin criterion was never natural but always constructed — a product of specific historical circumstances that no longer obtain.

The third element is the reconstruction of intellectual property on foundations adequate to the reality of AI-mediated production. The current framework, as Chapter 3 documented, anchors copyright in Fichte's idea/expression distinction — the claim that the form of a text is the unique expression of an individual mind and therefore the author's property. AI dissolves this anchor by producing form without an individual mind. The response cannot be to extend the existing framework to cover AI-generated works, which would require attributing authorship to entities that are not authors in any historically meaningful sense. Nor can it be to exclude AI-generated works entirely, which would leave an increasingly large domain of cultural production without legal protection or economic structure. The response must be to develop new forms of protection that are anchored not in the fiction of individual expression but in the reality of creative investment — the investment of judgment, care, and responsibility that the human participant brings to the collaborative process.

Such a framework would protect the human collaborator's contribution without requiring the fiction that the contribution is the whole of the work. It would recognize that creative value in AI-mediated production is generated through the interaction between human judgment and machine capability, and it would develop mechanisms for compensating both the human collaborator and the broader community of creators whose accumulated work constitutes the training corpus from which the machine's capability derives. Collective licensing schemes, training data royalties, and new categories of collaborative copyright are all possibilities that are currently under exploration. None of them is a complete solution. All of them are more adequate to the reality of AI-mediated production than the extension of a framework designed for a world of individual authors and printing presses.

The fourth element is the cultivation of the capacities that survive the Romantic framework — the judgment, voice, and responsibility that Chapter 7 identified as the enduring core of authorial practice. If the Romantic myth of genius is exposed, the question becomes how to develop the capacities that the myth obscured. Judgment is cultivated through practice — through the sustained exercise of discrimination in evaluating creative output, through the development of taste that comes from wide reading and deep engagement with the tradition, through the accumulation of the specific expertise that allows a person to distinguish between the plausible and the true. Voice is cultivated through living — through the accumulation of experience that gives language its specific gravity, through the development of the biographical specificity that no machine can replicate. Responsibility is cultivated through commitment — through the acceptance of accountability for the consequences of one's creative decisions, through the refusal to hide behind the collaboration when the decisions turn out badly.

These capacities cannot be cultivated by AI. They can be cultivated alongside AI, in a practice of creative work that uses the machine's capabilities to extend the human's reach while maintaining the human's judgment, voice, and responsibility as the irreducible core of the authorial function. The practice requires discipline — the discipline to reject plausible output in favor of true output, to maintain one's voice against the homogenizing pressure of machine-generated prose, to accept responsibility for a collaborative product that could not have been produced alone. The discipline is harder than the Romantic mythology suggested, because the mythology promised that genius would do the work. The reality is that judgment, voice, and responsibility require constant cultivation, and the cultivation is a practice, not a gift.

Woodmansee's scholarship ends, as good historical scholarship does, not with a prescription but with a clearing. The Romantic myth has been historicized. Its economic origins have been exposed. Its institutional embodiments have been identified. Its concealment of collaborative reality has been documented. What remains is the ground on which new structures can be built — structures that are honest about the conditions of contemporary creative production, that evaluate work on the basis of quality rather than origin, that protect creative investment without mystifying it, and that cultivate the human capacities that survive the dissolution of the myth.

The structures are not yet built. The clearing is recent, and the building materials are still being assembled. But the clearing itself is a contribution of the first order, because nothing adequate can be built on mystified ground. The Romantic myth had to be exposed before the post-Romantic practice could be conceived, and Woodmansee's career-long project of exposure is the precondition for everything that follows. What follows — the specific legal reforms, the institutional adaptations, the educational transformations, the creative practices — will be the work of many hands, many minds, and many years. The work begins where the myth ends: with the recognition that authorship was always a construction, that constructions can be reconstructed, and that the reconstruction, however difficult, is both necessary and possible.

The author after the myth is not a lesser figure than the Romantic genius. She is a more honestly described one — a person who exercises judgment, maintains a distinctive voice, and accepts responsibility for the creative decisions that bear her name. She does not claim to create from nothing. She claims to create from everything available to her, including the capabilities of machines that extend her reach without replacing her judgment. And she stands, as writers have always stood, at the intersection of a tradition too vast for any individual to contain and a specific life too particular for any tradition to replicate — producing, through the collision of the two, something that neither could have produced alone.

Chapter 9: The Training Corpus as Commons — Enclosure, Compensation, and the Unfinished Problem

The most consequential legal question of the AI era is not whether machines can be authors. It is who owns the tradition.

Every large language model is trained on a corpus. The corpus is not a database in the conventional sense — not a structured collection of discrete records that can be individually retrieved. It is something more like a landscape: the accumulated deposit of human textual production, layer upon layer, from which the model extracts statistical patterns that enable it to generate new text. The corpus includes novels, academic papers, journalism, legal briefs, technical documentation, poetry, forum posts, instruction manuals, and the entire miscellaneous output of a civilization that has been writing things down for five thousand years. The model does not memorize individual texts. It learns the patterns that connect them — the regularities of syntax, argument, narrative, and expression that constitute the deep structure of human written communication.

The legal question is whether the use of copyrighted works as training data constitutes infringement — whether the extraction of patterns from a copyrighted text, without reproducing the text itself, violates the author's exclusive rights. The question has no settled answer. Courts in multiple jurisdictions are considering it. The outcomes will shape the economics of AI for decades. And the question cannot be adequately analyzed within the Romantic framework that currently governs copyright, because the framework was designed for a world in which the primary economic threat was unauthorized reproduction — one publisher copying another publisher's book — and the training of a statistical model on the patterns extracted from millions of texts is not reproduction in any sense that the framework's architects contemplated.

Woodmansee's historical analysis provides a lens that the legal discourse has not yet employed with sufficient rigor. The eighteenth-century debates about literary property that she documented were, at their core, debates about the boundary between private ownership and common heritage. The question that Fichte's 1793 essay attempted to answer — What does the writer own? — was a question about where the commons ends and private property begins. Fichte's answer was that the writer owns the form but not the ideas, the expression but not the content. The answer drew a line that separated the writer's property from the common inheritance of language, knowledge, and cultural tradition that the writer drew upon. The line was useful. It enabled the literary marketplace. But the line was also, as Woodmansee demonstrated, a construction — drawn at a specific historical moment, under specific economic pressures, to serve specific commercial purposes.

The training data question requires the drawing of a new line, and the historical precedent suggests that the line will be drawn not where philosophy dictates but where economics demands. The eighteenth-century line was drawn to protect publishers' investments in printing and distribution. The twenty-first-century line will be drawn to balance the interests of the creators whose works constitute the training corpus against the interests of the companies that build and deploy the models, and against the broader social interest in the continued development of AI capabilities that serve the public good.

The analogy to enclosure is instructive. In England between the fifteenth and nineteenth centuries, common lands that had been used by entire communities for grazing, gathering, and subsistence were progressively enclosed — fenced off, privatized, converted from shared resource to individual property. Enclosure increased agricultural productivity. It also displaced millions of people who had depended on the commons for survival, concentrating wealth in the hands of landowners and creating the propertyless labor force that would fuel the industrial revolution. The productivity gains were real. The distributional consequences were devastating. And the legal instruments that enabled enclosure — the Enclosure Acts passed by a Parliament composed largely of landowners — were not neutral arbitrations of competing interests but interventions that served the interests of those who had the political power to shape the law.

The training of AI models on copyrighted works without compensation is a form of enclosure. The commons being enclosed is not physical land but the accumulated textual heritage of human civilization — the corpus of writing that represents millennia of intellectual labor, creative effort, and cultural investment. The models extract enormous value from this corpus. The value accrues primarily to the companies that build and deploy the models. The creators whose works constitute the corpus receive nothing, because the legal framework that governs their rights was designed for a different kind of use and does not clearly cover the extraction of statistical patterns from copyrighted texts.

The injustice is real, and Woodmansee's framework explains why the existing legal instruments are inadequate to address it. Copyright, as she demonstrated, was designed to protect the individual author's expression against unauthorized reproduction. It was not designed to address the collective value of the tradition — the value that emerges not from any individual work but from the aggregate patterns that connect millions of works. The training corpus has value precisely because of its aggregation. No individual text in the corpus is sufficient to train a capable model. The capability emerges from the statistical patterns that span the entire collection. This aggregate value is not captured by the individual property rights that copyright creates, because individual property rights attach to individual works, and the value at issue is a property of the collection rather than of any element within it.

The New York Times lawsuit against OpenAI illustrates the tension. The Times alleges that OpenAI's models can reproduce substantial portions of Times articles, which would constitute straightforward infringement under existing law. But the more fundamental question — whether the use of Times articles as training data, even when the model does not reproduce them, constitutes a compensable appropriation of value — presses against the boundaries of the existing framework in ways that the framework cannot easily accommodate. If the answer is yes, then every copyrighted work in the training corpus represents a potential claim, and the aggregate liability would be astronomical. If the answer is no, then the creators whose accumulated work makes AI possible receive no share of the value their work creates, and the enclosure of the textual commons is complete.

Neither outcome is satisfactory, and the dissatisfaction points toward the need for mechanisms that operate at the level of the commons rather than at the level of the individual work. Collective licensing schemes — agreements between AI companies and organizations representing broad communities of creators — offer one path. Training data royalties, calculated on the basis of a work's contribution to the model's capability rather than on the basis of individual reproduction, offer another. Both mechanisms acknowledge what the existing copyright framework does not: that the value of the training corpus is collective, that the collective value should be collectively compensated, and that the instruments of individual authorial ownership are inadequate to the task.

Woodmansee and Jaszi noted in 1994 that copyright law was "rooted in a nineteenth-century Romantic understanding of the author as a solitary creative genius" and that this understanding "may be inapposite to the realities of cultural production." The training data problem is the starkest confirmation of their warning. The realities of AI-mediated cultural production are collective at their foundation — dependent on the accumulated textual heritage of the entire species — and the individual property rights that the Romantic framework creates are inadequate to govern the collective processes that the technology employs.

The resolution will not come from within the Romantic framework. It will require the kind of institutional construction that Woodmansee's scholarship makes conceivable: the development of new legal categories that recognize collective creative value alongside individual creative expression, that distribute the economic benefits of AI across the broad community whose accumulated work makes AI possible, and that do so without either eliminating individual rights or pretending that individual rights are sufficient to govern a fundamentally collective process. The construction is urgent. The enclosure is already underway. And the people whose work is being enclosed — the millions of writers, scholars, journalists, and creators whose texts constitute the commons from which AI extracts its capability — deserve a framework that protects their interests with the same specificity that the Romantic framework once provided for a simpler technological world.

---

Chapter 10: The Author After the Author — Toward a Post-Romantic Creative Practice

Martha Woodmansee retired from active scholarship before the large language model arrived. She built the theoretical framework for understanding the AI authorship crisis decades before the crisis materialized, then stepped back from the arena before the framework's relevance became impossible to ignore. There is something appropriate in this timing. Intellectual historians do not typically live to see their analyses confirmed by technological revolutions. The confirmation, when it comes, is usually the work of a later generation that discovers in the archive the concepts it needs to make sense of its own predicament.

The predicament is this: the Romantic construct of authorship — the solitary genius producing original works from the depths of individual consciousness — has been the organizing principle of Western cultural production for two and a half centuries. It has governed law, commerce, education, and the self-understanding of every person who has ever written a sentence and felt it was theirs. And it is now being challenged by a technology that violates every premise on which the construct rests. The text produced through AI collaboration does not originate in a single consciousness. It does not bear the unique imprint of an individual mind. It cannot claim originality in the Romantic sense, because its production is irreducibly collaborative, distributed, and tradition-dependent. The construct cannot accommodate this reality, and the reality is not going away.

Woodmansee's contribution to navigating this predicament is not a prescription. It is a clearing — the removal of the ideological undergrowth that prevents clear thinking about the conditions of contemporary creative production. The Romantic myth has been historicized. Its economic origins have been documented. Its institutional embodiments have been identified. Its concealment of collaborative reality has been exposed. Its inadequacy as a framework for the training data question has been demonstrated. What remains is the ground on which new structures must be built.

The building requires, first, the acceptance of what is lost. The Romantic construct provided something that the post-Romantic landscape does not automatically replace: a stable foundation for creative identity. If the writer's value rests on her unique genius, then her identity is secure regardless of technological change, because genius is inherent and inalienable. The exposure of genius as a construction removes this security. The writer who accepts Woodmansee's analysis must accept that her value does not rest on an inherent gift but on a set of practices — judgment, voice, responsibility — that must be continuously cultivated and that are not guaranteed to retain their market value as the technology evolves. This acceptance is uncomfortable, and the discomfort deserves acknowledgment rather than dismissal. The Romantic myth was a comfort. Demystification is not.

But the acceptance also opens possibilities that the myth foreclosed. The myth imposed an impossible standard — the standard of creation from nothing, of originality that owes nothing to the tradition, of the text as the pure emanation of individual consciousness. No writer has ever met this standard. The myth merely concealed the failure. The writer who accepts the collaborative reality of creative production is freed from the impossible standard and can focus on what she actually does: select, arrange, evaluate, and shape materials drawn from a tradition larger than any individual, directed by the specific judgment that only her particular life has produced.

The building requires, second, the reconstruction of legal and institutional frameworks on foundations that Woodmansee's analysis makes visible. Copyright must be reconceived — not abolished, but reconceived — on the basis of creative investment rather than original expression. The university must revise its evaluative criteria to accommodate collaborative production without abandoning its commitment to scholarly rigor. The publishing industry must develop new mechanisms of attribution that acknowledge the reality of AI collaboration without destroying the market-sorting function that the author's name provides. The educational system must shift its emphasis from the production of original work — a standard that AI has rendered meaningless — to the cultivation of the judgment that distinguishes good work from competent work, true claims from plausible ones, illuminating arguments from impressive-sounding emptiness.

None of these reconstructions is simple. Each involves the revision of institutional structures that have been organized around the Romantic construct for generations, and institutions resist revision with a tenacity proportional to their dependence on the arrangements being revised. But the revision is coming regardless of institutional resistance, because the technology that necessitates it is already integrated into the practices the institutions govern, and the gap between institutional assumptions and actual practice cannot widen indefinitely without producing the kind of dysfunction that forces change.

The building requires, third, the development of what might be called a phenomenology of honest authorship — an account of what the creative experience actually feels like when the Romantic mystifications are set aside. The Romantic account says creation feels like expression — like something welling up from within, flowing from the writer's unique genius onto the page. The phenomenology is real, as even Woodmansee's historicization acknowledges. The creative act does feel like expression. The question is what the feeling means — whether it is evidence that the text originates in the individual consciousness, as the Romantic framework claims, or whether it is the subjective experience of a process that is actually more complex, more collaborative, and more tradition-dependent than the feeling suggests.

The strongest version of the counter-argument to Woodmansee's historicization rests on this phenomenological ground. The argument goes: even if the Romantic construct is historically contingent, even if genius was invented to solve an economic problem, even if the collaborative reality is more accurate than the mythology — the experience of authorship remains real. The writer who labors over a sentence until it sounds right, who feels the specific satisfaction of finding the word she was reaching for, who experiences the text as an extension of herself — this writer is not deluded. She is experiencing something genuine about the relationship between a consciousness and its expression. The constructionist analysis cannot explain this experience away, and any post-Romantic framework that ignores it will fail to account for the actual phenomenology of creative work.

The counter-argument is serious, and it deserves a serious response. The response is not that the experience is illusory but that the experience is compatible with multiple explanatory frameworks. The feeling of creative ownership — the sense that the text is "mine" — is real. But the feeling does not require the Romantic explanation. It is equally well explained by the investment of judgment, the exercise of voice, and the acceptance of responsibility that the post-Romantic framework identifies as the surviving elements of authorship. The writer who feels the text is hers feels this because she has invested her judgment in it, because the prose bears the imprint of her specific life, and because she has accepted responsibility for its quality and its effects. These are genuine grounds for the feeling of ownership, and they do not require the additional claim that the text originated in her consciousness alone.

The phenomenology of honest authorship is the phenomenology of the compiler who cares — the person who works with materials drawn from a vast tradition, who shapes those materials with the specific judgment and voice that only her biography can produce, and who releases the result into the world under the full weight of personal responsibility. The phenomenology is not less rich than the Romantic phenomenology. It is differently rich — honest about the collaborative nature of the process while preserving the intensity of the individual's engagement with it.

Woodmansee's scholarship terminates in the historical demonstration. It does not extend to the prescriptive question of how to write, evaluate, or compensate creative work in the post-Romantic landscape. That extension is the work of the present generation — a generation that has inherited both the Romantic construct and the tools that expose its contingency, and that must decide how to navigate the space between them.

The navigation will not be easy. It will require the simultaneous maintenance of two capacities that feel contradictory: the capacity to see the authorship construct as a construction, contingent and revisable, and the capacity to take creative work seriously despite its constructed foundations. The difficulty is real but not unprecedented. Humans have always lived within constructions — language, law, money, nation, identity — that are simultaneously constructed and consequential, simultaneously artificial and real. The authorship construct is one more such construction. Understanding its constructedness does not require abandoning it. It requires holding it with the specific combination of commitment and irony that all human constructions demand: building on the construct while knowing it was built, maintaining it while knowing it could be otherwise, and revising it when the conditions that produced it change beyond its capacity to accommodate.

The conditions have changed. The revision has begun. And the ground on which the revision proceeds — cleared by four decades of Woodmansee's scholarship, enriched by three thousand years of pre-Romantic creative practice, and complicated by a technology that neither the Romantics nor the pre-Romantics could have imagined — is more solid than the ground the Romantic myth provided, because it rests not on a fiction about the nature of genius but on the observable reality of how creative work is actually done: collaboratively, within traditions, with judgment, with voice, and with the specific human responsibility that no machine, however capable, can bear.

---

Epilogue

The sentence I underlined was not about technology. It was about vegetables.

Edward Young, 1759: the original work "rises spontaneously from the vital root of genius; it grows, it is not made." The genius as plant. The text as organism. Creation as something that happens from the inside out, requiring no external input, owing nothing to the soil it grows from.

I read that sentence during the research for this book and felt something click — not agreement but recognition. Recognition of the mythology I had been living inside without seeing it. Because that vegetable metaphor is not just an eighteenth-century literary conceit. It is the operating assumption of every pitch deck I have ever written, every team I have built, every product I have shipped. The founder as genius. The vision as something that grows from within. The company as the organic expression of a singular creative mind.

I have stood in boardrooms and said, in essentially those words, this is my vision, and it came from me, and that is why it will work. The Romantic authorship construct, dressed in Silicon Valley vocabulary. Young would have recognized it instantly.

What Woodmansee's scholarship exposed — and what this book has traced through law, economics, institutions, and the lived practice of creative work — is that the vegetable metaphor was always wrong. Not wrong about the phenomenology. Creation does feel like something growing from within. The feeling is real. What is wrong is the conclusion drawn from the feeling: that because creation feels individual, it is individual; that because the text feels like mine, it is mine in the absolute sense; that the tradition I draw from is mere background rather than the medium through which my work exists.

When I wrote The Orange Pill with Claude, I experienced both the feeling and its exposure simultaneously. The feeling: this is my book, my argument, my voice. The exposure: these connections I did not make alone, these structures I did not build unassisted, these moments when the collaboration produced something that neither participant predicted. Woodmansee gave me the vocabulary to hold both without collapsing into either — to take the work seriously without mystifying it, to acknowledge the collaboration without diminishing the judgment that shaped it.

The part of her analysis I find most difficult — and most necessary — is the part about what is lost. The Romantic myth was a comfort. It told the writer, the builder, the founder: your value is inherent. Your genius is yours. No technology can take it from you because it lives inside you, growing from the vital root of who you are. Woodmansee demonstrates that this comfort rests on a fiction. The fiction was productive — it enabled copyright, the literary marketplace, the entire infrastructure of creative compensation. But it was a fiction. And the moment you see it as a fiction, the comfort is gone.

What replaces the comfort is something harder and, I think, more honest: the knowledge that your value is not inherent but practiced. Not a gift but a discipline. Not a root but a relationship — between your judgment and the materials you judge, between your voice and the tradition it speaks from, between your responsibility and the world that receives what you make.

My children will not live inside the Romantic myth. The technology they are growing up with has already dissolved the boundaries between individual and collaborative production, between human and machine capability, between "mine" and "ours." They will need a framework for understanding creative work that does not depend on the fiction of solitary genius. Woodmansee built that framework. She built it by looking backward — into the scriptoria and printing houses and debating chambers of eighteenth-century Europe — and what she found there turns out to be exactly what the twenty-first century needs: a way of understanding creativity that honors the individual without mystifying her, that acknowledges the tradition without dissolving the individual into it, and that takes the work seriously enough to tell the truth about how it is actually done.

The author is not dead. The author was never quite what we thought she was. She was always a compiler, an arranger, a judge — working with materials she did not create, shaping them with a specificity that only her particular life could produce. The myth said she was a god, creating from nothing. The history says she was something more interesting: a person, creating from everything, responsible for the result.

That is harder than being a god. It is also more honest. And in this moment, honesty about how creation works is not a luxury. It is the foundation on which everything adequate to our situation must be built.

— Edo Segal