Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Literature. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 27 February 2024



  • John FrowJohn FrowDepartment of English, The University of Syndey


Questions of authorship bring into play many of the central questions of literary theory: questions as to what constitutes the unity and coherence of texts, the interpretive relevance of authorial intention, the relation of oral to literate cultures, the regulation of writing by church and state, the legal underpinnings of literary property, the significance of forgery and plagiarism, and so on. At the heart of many of these questions is a distinction between two different orders of phenomena. Writers are not necessarily authors: authorship requires recognition and attribution, and these depend on institutional processes of publication, textual stabilization, criticism, education, and appropriate legal, regulatory, and economic conditions. Those processes and conditions vary from culture to culture, as do the particular historical forms that authorship takes. In the contemporary world authorship tends to be cast as though it were directly expressive of a personality, an inner core of selfhood, that underwrites the coherence of the texts attributed to it; the commercialization of that form gives rise to a cult of the author in both academic and popular culture.


  • Literary Theory

The author still reigns in histories of literature, biographies of writers, interviews, magazines, as in the very consciousness of men of letters anxious to unite their person and their work through diaries and memoirs. The image of literature to be found in ordinary culture is tyrannically centred on the author, his person, his life, his tastes, his passions, while criticism still consists for the most part in saying that Baudelaire’s work is the failure of Baudelaire the man, Van Gogh’s his madness, Tchaikovsky’s his vice. The explanation of the work is always sought in the man or woman who produced it, as if it were always in the end, through the more or less transparent allegory of the fiction, the voice of a single person, the author “confiding” in us.

Roland Barthes1

Author and Owner

In the winter of 1930 Bertolt Brecht and his collaborators took to court the company that had produced a film of The Threepenny Opera and that had engaged Brecht’s services in the preparation of the screenplay. The details of the contractual disagreements are complicated and technical, but the gist is that, contrary to the usual terms of employment of authors and scriptwriters, Brecht had insisted on having chief responsibility for the preparation of the script and on a final power of veto over it. On receiving a draft of the screenplay the company had taken fright and tried to buy him off. Once the film, directed by G. W. Pabst, was completed (but before its release in 1931), Brecht began his “sociological experiment.”2

There is a certain amount of (quite characteristic) ex post facto rationalization in Brecht’s account of his motives in initiating the trial and, more crucially, in coming to a settlement with the film company after the case had been lost. But the court case did achieve the goal of forcing into the open the question of intellectual private property in film production and of its grounding in the concept of authorship.

In this case, however, the question of authorship is a somewhat difficult one. The work is a loose adaptation of John Gay’s The Beggar’s Opera (1728), and the first playbill acknowledged Brecht only as the “adapter,” noting that there were interpolated ballads by Villon and Kipling. In the first publication of the libretto, however, Brecht was credited as its author.3 There are several problems with this. First, it seems likely that Brecht’s collaborator and lover Elisabeth Hauptmann “was responsible for as much as 80 or even 90 per cent of the published text of The Threepenny Opera.”4 Second, Brecht took over almost verbatim and without acknowledgment an existing translation of the Villon poems he used; when taxed with plagiarism his rather offhand defense was his own “fundamental laxity in questions of literary property.”5 Yet that laxity seems to have been opportunistic. Erwin Piscator reported in 1929 that Brecht “had repeatedly shouted at the Piscator Collective, ‘My name is a brand name, and whoever uses this brand name has to pay for it.’” As John Fuegi says in citing this report, “Brecht’s basic laxity was with the property of others, but he would allow no such laxity by others in using the work sold under his brand name.”6

When the case came to court the film company’s first move was a striking one: to posit an ideal of authorship so transcendent that Brecht could never meet it, and then to brand him as a mercenary for having sold the script to them in the first place. “Honour the writer who on principle does not allow his works to be filmed,” wrote the company’s lawyer, Dr Frankfurter. “He must decide whether he will not permit his work to be filmed, which would be his free, artistically very respectable will or, if he wants to have the financial earnings, he must once and for all make do with what he gets.”7

Brecht’s response is to demonstrate the ways in which the production of film as a mass commodity, involving the application of elaborate technical resources and a heavy investment of capital, breaks down the opposition of the creative to the mechanical, the artistic to the commercial, the authentic to the inauthentic. The film apparatus, he writes, “can be used better than almost anything else to supersede the traditionally untechnical, anti-technical, ‘transcendent’ ‘art’ associated with religion.”8 The model of authorship as pure disinterested creativity is destroyed by the apparatus of capital, although it continues to lead a ghostly and at times forceful afterlife.

The company’s primary argument, however, is that “a film is such a major economic undertaking, represents such an aggregate of risk, capital and film labour, that it must not be endangered by the moods of prima donnas, lack of knowledge about the needs of the cinema, or – in the case of Brecht – even political tendentiousness.”9 The court agreed, with the judge considering only the extent to which the work of the author supported or hindered production; the attempt to produce a different kind of film counted as hindering. Legal contracts are to be strictly subordinated to their role in the production process, and any contractual claim by the author to a say in determining the film’s quality is to be treated as proof of sabotage.

Let me single out two of the lessons that Brecht draws from his experiment. The first is that the collaborative nature of film production undermines the categories of individual authorship and of organic form, together with the causal connection between them, and in principle points the way toward new, socialist forms of artistic production. If the film’s production team had, for example—like the team Brecht gathered around him—had specific didactic objectives, it would of itself form an organic body. But in practice the film collective is a false one, dominated by the representatives of finance and predicated upon a division rather than a sharing of knowledge and technical skills.10 Against this argument, however, we might counterpose the question of Brecht’s own sexually and financially exploitative practice of collaborative authorship; in Fuegi’s words, the group that Brecht had formed around him was “a ‘collective’ where Brecht overwhelmingly collected and where others contributed to his fame and financial welfare.”11

The second lesson has to do with the use of authorship as an instrument of branding. Capitalism is indiscriminate, Brecht says; it will sell the “author’s” name, regardless of who wrote the screenplay or how many people were involved; the reputation of a politically radical author like Brecht can be used even in the absence of any radical political content. Authorship, he argues, does not designate the intentions of a writer. It designates a property right, and the case of film makes it clear that the right belongs to capital, not to the author—or rather, that the role of author is here played by capital. The author is not the one who produces the “raw material” (whether a script, a piece of software, or a genetic cell line) but the person or the corporation that can transform it into an exploitable commodity.12

Moses, Confucius, Homer, Shakespeare

Consider the following authors: Moses, Homer, Confucius, Shakespeare.

(1) Moses is traditionally the author of the Pentateuch—that is, the first five books of the Old Testament, the Jewish Torah. Although that authorship is nowhere claimed in the texts themselves, Moses is represented on a number of occasions as writing in accord with the Lord’s commandment: in Exodus, Numbers, and Deuteronomy he records a battle, the travels of the Israelites out of Egypt, and above all the Laws that are passed on through him to the people of Israel. This is a complex mode of authorship in that he acts as a medium for the word of God; it is only in the Hellenistic period that this becomes a claim for his full, inspired authorship of the whole of the Pentateuch, a claim that is consolidated in the 12th century ce in the eighth of Maimonides’ Thirteen Principles: “I believe with perfect faith that the entire Torah presently in our possession is the one given to Moses.” In the Christian church Moses’ authorship of the Pentateuch is an unchallenged principle until the 17th century, remains a core belief well into the 20th, and is still an article of faith for a number of Evangelical churches.

That belief begins to be shaken by Thomas Hobbes who points in Leviathan to inconsistencies in the biblical text, and, more radically, by Baruch Spinoza, in the Tractatus Theologico-Politicus, where it is challenged directly by a close reading of anachronisms. A key scholarly advance was made in the mid-18th century by Jean Astruc when, observing the use of different names for God (Yahweh and elohim) in passages marked by stylistic and thematic differences, he posits that Moses composed the Torah by drawing on discrepant documentary sources. From then on a long tradition of source analysis develops a form of textual analysis that looks at internal textual inconsistencies—later supplemented by comparative analysis of the Mesopotamian texts that were discovered and deciphered in the 19th century and more broadly of archaic religious beliefs and practices, together with emerging archaeological evidence—to try to understand the patterns of authorship that make up the Pentateuch and the other texts of the biblical canon.

The key figure here is the German scholar Julius Wellhausen, who in the Prolegomena to the History of Ancient Israel (1878) built on the work of Kuenen and other scholars to formulate the Documentary Hypothesis that posits the existence of four distinct strands or “documents” of authorship in the Pentateuch; he labeled them J, E, D, and P, documents written in a succession of stages marking the passage from a tribal worship of one God among many, to a centralized monotheism in the monarchical period, to the late-exilic hegemony of the priestly caste. The J or Yahwist source is characterized by its consistent use of the name Yahweh, its description of this God through “vivid anthropomorphisms,” and its location of many of the genealogical narratives in the southern kingdom of Judah.13 The E or Elohist source, which uses the name elohim consistently through Genesis until the revelation of the name Yahweh to Moses in Exodus 3, describes a more remote deity, “typically revealing himself indirectly through dreams . . . divine messengers . . . and prophets,” and its narrative is often set in the north of Israel, a separate kingdom from the late 10th to the late 8th centuries bce.14 The D source is found largely in Deuteronomy, is hortatory in style, and probably also originated in the northern kingdom; it contains elements that are clearly archaic. The P or Priestly source, finally, is characterized by its use of fixed formulae and repetitions and by its emphasis on religious observance and ritual, and it seems to reflect the interests of the priestly caste during or shortly after its Babylonian exile.

The description and dating of these sources, the relations between them, the modality of authorship they embody, the relation between oral and scribal cultures, and the kind of authority they carry have all been matters of intense dispute among biblical scholars.15 Many now reject the postulate of a separate E source, and there is little consensus on much else. Coogan gives the following conservative guess at the process by which the separate sources or documents were combined:

Using some earlier traditions, and reflecting their own perspectives as well, J and E were independently written, the former in Judah, probably during the tenth century BCE, and the latter in the northern kingdom of Israel in the ninth, or perhaps a century or so later. When the northern kingdom fell to the Assyrians in 722 BCE, refugees from there brought with them to Jerusalem the E source, which was combined with J in Jerusalem, but in such a way that while J remained intact, E was used as a kind of supplement; this accounts for its fragmentary character. Finally, in the sixth century BCE, P shaped these sources along with its own material and D into what became the Pentateuch.16

John Van Seters, by contrast, surmises that

The first Pentateuchal source was Deuteronomy (seventh century), which was then used as an introduction to a larger historical work, DtrH [the broader Deuteronomistic History, which includes sections of Joshua, Samuel, Judges, and Kings] (early exilic period). . . . This was expanded in the late exilic period by a history of the people’s origins (J) from creation to the death of Moses. . . . This combined D+J work was then supplemented in the postexilic period by a Priestly writer with his own distinctive traditional material and his own ideological interests.17

For all their disagreement, these scholars concur in their rejection of a single author—whether or not his name is Moses—of the Pentateuch. Paradoxically, however, the discovery of a multiplicity of authorial or editorial strands only strengthens the understanding of authorship as a principle of textual unity. As Van Seters puts it, “A particular text-unit or a larger ‘source’ is the work of a single author only if it manifests a fairly high degree of consistency. . . . Blatant contradiction, therefore, suggests that a second hand is at work . . . the very idea of authorship demands consistency.”18 If the Pentateuch is not the work of a single, inspired author then it is the work of an unknown number of single authors, each of whom represents a self-contained act of intention.

(2) Confucius (the Latinized version of Kong Fuzi) is the name associated with the five Confucian classics: the Odes (or Songs; Shijing), the Documents (Shangshu), the Rites (Yili, Liji, and Zhouli), the Changes (I Ching), and the Spring and Autumn Annals (Chunqiu). The term “Confucian” should be thought of as an honorific: Confucius is often regarded as either the author (especially of the Spring and Autumn Annals) or the editor or compiler of these texts, but his connection with them may more broadly be construed in terms of their being incorporated into the tradition of his teachings. “Most of the texts were evolving in oral as well as written forms for centuries before they acquired the designation ‘classic’ or ‘Confucian,’” writes Michael Nylan:

The classics can properly be called Confucian in only two senses: Confucius and his followers may have used some—but not all—of them as templates for moral instruction, much as the Greek pedagogues once used Homer. And early traditions ascribe to Confucius the tasks of compiling, editing, and in some few cases composing the separate parts in this repository of wisdom texts, although modern scholarship generally disputes those pious legends.19

Confucius can plausibly be deemed their “author” in the sense that the spirit of his teachings is thought to pervade these texts—a result of the long history of Confucian commentary on the ethical and political lessons to be drawn from them. They are thus, by a metalepsis, taken to be expressive of his innermost character:

In the Odes we can see the mind of the Master Confucius. From the Documents we can know his judgments. With the Rites we can shed light on his models. With the Music we can grasp his virtue. With the Changes we can examine his character. And with the Spring and Autumn we can preserve his purpose.20

Another text, the Analects (Lunyu), is traditionally thought to have a more direct connection to Confucius. It is ostensibly a collection of sayings and ideas spoken by Confucius to the disciples who wrote them down, although the book was probably compiled and edited over a period of several hundred years during the Warring States period (476–221 bce) before being canonized during the Han dynasty (206 BCE–220 ce) when Confucian doctrine became a key dimension of the state ideology of the Han ruling house. Like all of the Classics, then, the Analects is “a ‘sedimented’ text, an accretion of divergent textual materials that coalesced over a very long period,” and its attribution to Confucius is a matter of authorization rather than of an actual reality of composition.21 In any case, although Confucius describes himself in the Analects as one who “transmits yet does not create,” there are many, writes Hanmo Zhang, “who still believe that . . . every word in the Lunyu ‘was decided by the sage himself’ . . . and ‘sealed with Confucius’s approval.’”22 The text constructs an image of the Master, certifying his name as that of a really existing and knowable person, and that image in turn enhances the authority of the text and makes it a reliable source of biographical information. Finally, the same procedures of textual authentication as are employed to establish biblical authorship are used to resolve contradictions in the text of the Analects, including the identification of inconsistent passages as later additions or interpolations.23

(3) Homer is the name attributed to the author of two Greek epic texts, the Iliad and the Odyssey, as well as a number of other texts including the Homeric Hymns. A large number of biographies of Homer circulated in the world of antiquity, all dating from the Alexandrian period onward and offering often quite detailed information about the poet’s life (his birth on the banks of the river Meles, his blindness, his wanderings, his stays on Chios and Ios, his death as a result of his failure to answer a riddle proposed by fishermen, his burial on Ios, and so on).24

Modern scholarship, influenced in particular by the ethnographic work of Milman Parry and Albert Lord among Serbian oral epic poets that they subsequently applied to the formulaic substitutions and the narrative set pieces that form much of the architecture of the Homeric texts, has largely accepted that the Greek epic texts were composed and recomposed by generations of oral performers before being stabilized in written form at a certain time (the proposed dates range from the 8th to the 6th centuries bce) and then performed without recomposition at the great pan-Hellenic festivals, in particular the Panathenaia, by a specialized class of rhapsodes called the Homēridai. This thesis leaves open the question of the extent of the contribution of the stabilizing poet, who may or may not have been the same person as the writer or transcriber of the text.

Gregory Nagy has proposed a model of how the attribution of authorship to a poet called “Homer” might have come about in this shift from creative to reproductive performance. At some point in the tradition of performance when there is still continuing improvisation around a traditional repertoire of stories, a certain performer—let’s call him L—appropriates a particular act of recomposition as his own work. A subsequent performer—M—grants recognition to this appropriated work as though it had indeed been composed solely by his predecessor, and thus seeks to stabilize the flow of transmission. The successors of M then continue this attribution of the still evolving poem to L, and future performers come to be seen simply as repeating the work of this now-mythical figure.25

The originating author, “Homer,” that is to say, is thus “produced retrospectively, as a back-formation, through the performers’ own differentiation of themselves from the imagined originator of the song.”26 And this imagined originator is then revered as the singular source whose intentions are the point of reference and authorization for all interpretation of the texts.27

(4) Shakespeare is the name of a person whose existence is attested by contemporary mentions and documents and by the attachment of his name to both the authorized and unauthorized Quarto and the Folio publications of his plays and poems. The information we have about him is scant (and there are ongoing disputes about the authenticity of Shakespeare’s authorship of those texts), but there is no shortage of posthumous biographies—the first is Nicholas Rowe’s Preface to his 1709 edition of Shakespeare’s Works—none of which is connected to the direct testimony of his peers.28 The later and more comprehensive biographies tend to be offered more tentatively in the modalized form “Shakespeare may have, is thought to have, probably, almost certainly . . .” They are one manifestation of a cult of the divine Bard—already Milton and Dryden speak of him in these terms—which reaches its apogee in David Garrick’s organization of the Stratford Jubilee in 1769, where the full apotheosis was performed.29 One of the many acts of worship was the singing of a ballad written by Garrick and offered in exchange for the mayor’s gift of a cup made from a mulberry tree supposedly planted by Shakespeare’s own hand; the first stanza reads:

Behold this fair goblet, ’twas carv’d from the tree,Which, O my sweet Shakespeare, was planted by thee;As a relick I kiss it, and bow at the shrine,What comes from thy hand must be ever divine!

That event, with its “barely secularized version of the Eucharist,” is the first of many that enact a ritual performance of the cult of the author.30 David Davídházi cites, for example,

the ceremony that took place in 1936 when a replica of the Globe theatre was opened in Dallas: lumps of earth taken from Shakespeare’s garden and water from the Avon were sprinkled over the new stage. The earth was sent to America in a box made of the wood of the Stratford Memorial Theatre that was burned down in 1926, so even the box symbolized an authenticating continuity, the miraculous survival of things Shakespearean.31

And in Victorian England, when the cult takes on more explicitly nationalist (and cultural-imperialist) forms, the volume of Shakespeare’s plays stands in the bookcase next to, and in some cases instead of, the Bible. “By the middle of the eighteenth century,” writes Mark Rose,

the name “Shakespeare” had become in England the standard by which literary value was measured, the authorizing sign at the center of the entire galaxy of literary commodities. And underwriting the name “Shakespeare” was the notion that there had once lived a human being so extraordinary that all the value radiating from that galactic center was in turn the sign of his personal worth.

This “divine personality of the author of the plays” is now quite separate from “the human specificity of the actor-playwright-shareholder William Shakespeare,” and it bears little relationship to the essentially collaborative nature of that corpus of plays.32

Moses, Confucius, Homer, and Shakespeare are extreme examples of the construction of authorship; each of them involves the attribution to a mythical author of a body of work that is then unified around his name—the Iliad and Odyssey, for example, being accepted as parts of a single oeuvre. But in another sense they are not at all exceptional: authorship works in all these cases as a principle of textual coherence, and it has the force of an institution, a consolidated belief reinforced by a complex social apparatus, rather than representing the intentions of a subject of writing or composition.

The Author Function

The terms of the contemporary discussion of authorship have been set above all by Michel Foucault’s 1969 essay “What Is an Author?” Tacitly a response to Roland Barthes’s earlier pronouncement of the author’s death, Foucault’s essay picks up on Barthes’s “qui parle?,” who is speaking here?—a question with which Barthes defines the author (here in a story by Balzac) as an unidentifiable voice, no longer a singular point of origin but rather a textual space made up of the plethora of voices and of other, more or less identifiable texts that wash through any piece of writing.33 Quoting Samuel Beckett Foucault poses what looks like a very similar question: “What does it matter who is speaking?” But then this statement is repeated as a reported statement: “‘What does it matter who is speaking,’ someone said, ‘what does it matter who is speaking.’”34 Behind every indifference toward voice is a voice voicing that indifference.

Foucault’s point is neither to undermine nor to repeat Barthes’s polemic against the fetishization of the author in contemporary criticism and popular culture, but to ask a different set of questions that suppose, not that the author is dead but that to be an author is to inhabit a particular kind of social space; authorship is not the characteristic of a person but a function defined by historically specific legal, theological, technological, and economic conditions of existence. Brecht, to take my example, is the author of the libretto of The Threepenny Opera not because he wrote it but because it is attributed to him within a certain system of legal and cultural recognition. That attribution could change and Elisabeth Hauptmann could come to be recognized as the author; these are not matters of indifference, but they are also not a question of who the author really is: they are matters of social acknowledgment and of the conditions under which it is granted.

The category of author is symbiotically related to that of the work (oeuvre), and one way of figuring out the parameters of authorship is to ask what counts and what does not count as a work. Even in the case of an accepted author like Nietzsche, Foucault writes, we must decide where to draw the line. Is a rough draft a part of the oeuvre? A deleted passage? A laundry list? There are only certain kinds of text to which an author’s name is likely to be attached: “A private letter may well have a signer – it does not have an author; a contract may well have a guarantor – it does not have an author. An anonymous text posted on a wall probably has an editor – but not an author.”35 To write, to be a writer, is not sufficient to make one an author; something else is required to set up that circular relation between authorship and work.

An author is a named and recognized entity, and part of the function of the author’s name is to help classify the work as a coherent and unified text or body of texts. Even in the case of a mythical author, “the fact that several texts have been placed under the same name indicates that there has been established among them a relationship of homogeneity, filiation, authentication of some texts by the use of others, reciprocal explication, or concomitant utilization.”36 The name of the author designates a point and a moment of origin, an intention, an act of will that brings the work into being, and it guarantees an overall level of quality, a conceptual coherence, a stylistic unity, and a temporal homogeneity. The principles used by contemporary literary criticism to establish those forms of coherence are “directly derived from the manner in which Christian tradition authenticated (or rejected) the texts at its disposal.”37 In their post-Romantic form, which is the dominant mode of our critical and popular culture, they lead us endlessly to ask the questions: “Who really spoke? Is it really he and not someone else? With what authenticity or originality? And what part of his deepest self did he express in his discourse?”38

In summary, Foucault sets out four key traits of the author function:

(1) the author function is linked to the juridical and institutional system that encompasses, determines, and articulates the universe of discourses; (2) it does not affect all discourses in the same way at all times and in all types of civilization; (3) it is not defined by the spontaneous attribution of a discourse to its producer but, rather, by a series of specific and complex operations; (4) it does not refer purely and simply to a real individual, since it can give rise simultaneously to several selves, to several subjects – positions that can be occupied by different classes of individuals.39

These traits are the logical scaffolding of authorship and they take on a specific articulation in any historical period and in relation to particular genres of text; “we do not construct a ‘philosophical author’ as we do a ‘poet’, just as in the eighteenth century one did not construct a novelist as we do today.”40 Foucault does not attempt that detailed work of historical articulation, but it is the necessary consequence of his theorization of authorship as a function that takes on different forms under different circumstances, and a growing body of scholarship has sought to do so.41

The Logic of Authorship

Before moving on to examine some of those historically differing forms, however, let me sketch out a tentative definition of the complex of elements that make up the logical framework of the author function.

Authorship is an attribute of texts. It endows them with a source, and a legitimacy as the emanations of that source.42 All textual interpretation, and the prior establishment of what counts as a text, relies on such an attribution of a unified authorial intention.

Authorship is thus a historically specific institution: a set of established and repeatable beliefs about the origin of and responsibility for a text or a body of texts.

Authorship may therefore not coincide with itsefficient cause,” the writer or creator of a text.43 Theodore Sorenson wrote the book published under the name John F. Kennedy as Profiles in Courage, but the author of that book, and the person who was awarded the 1957 Pulitzer Prize for Biography, was Kennedy. The writers in Alexandre Dumas’s novel factory, or those who wrote the hundreds of Nancy Drew books published under the name “Carolyn Keene,” were not the authors of the books they created.

Writer and author are different (but intersecting) kinds of entity: a historically actual person, and an ascribed property of texts. The name of the writer (Henri Beyle, Mary Ann Evans, John Cornwell) refers to a person external to a body of texts; the name of the author (Stendhal, George Eliot, John le Carré) refers to no properties other than the authorship of those texts.

In contemporary culture authorship tends to assume the form of the person, as though it were directly expressive of a personality, an inner core of selfhood, that underwrites the coherence of the texts attributed to it. The commercialization of that form gives rise to a cult of the author in both academic and popular culture.

In other cultures the casting of the author function in the form of a mythical originary figure may be an integral part of a centralized state or religious ideology.

Authorship is a function: it performs certain kinds of work, the primary one being that of authorization of texts in relation to the figure of the writer.

Authorship bounds a body of work and forms a symbiotic relationship in which the author and the work mutually define each other.

Authorship operates as a principle of unity and coherence within a text and between texts attributed to the same author; it classifies and contextualizes a text in relation to other texts by the same author and by other authors.

Authorship generates credit and recognition for the writer or performer to whom it is attributed; it is a source of symbolic capital. In some regimes of authorship it is, in addition, the locus of a property right.

Finally, authorship works as a brand, certifying the quality and, where this is relevant, the truthfulness of the text or oeuvre. That brand may have commercial value.

In what follows I seek to recognize and briefly describe the diversity of modalities of authorization across different periods and cultures. These modalities are not, however, reducible to a linear account of periods, since models of authorship in very different cultures may in some respects quite strongly coincide. The notion of an author as “someone who speaks in writing, with authority” was already present in an annotated bibliography, the Seven Résumés, compiled in the Western Han period by Liu Xiang (77–76 bce), which included detailed biographical information in its catalog; Chinese authorship from very early times is understood in affective and expressive ways that are strongly reminiscent of Romantic authorship; and some authors in Western classical antiquity—Euripides, Catullus, or Propertius, for example—seem in their apparent freedom of invention to be very close to modern modes of authorship, although their relation to their work is shaped by an ethos of credit rather than of proprietary rights.44 And named authors exist in the medieval period, even in the absence of an expressive and individualist model of writing. The history of authorship is characterized by recurrence and by continuities as well as by the kinds of discontinuity that are at the heart of historical and cultural difference.

Modalities of Authorization

Oral and Early Scribal Cultures

Three of the textual complexes examined in the section “Moses, Confucius, Homer, Shakespeare”—those of the Pentateuch, the Confucian classics, and the Homeric texts—were strongly marked by their formation in a culture of oral performance that included, or that was in a process of transition to, a dimension of scribal authorship; and the plays attributed wholly or in part to Shakespeare were of course written for and shaped by their oral performance in the theater.

Oral performance in traditional societies works with material—songs, stories, genealogies, songs of praise, and so on—that is authorless and that is embedded in religious ritual and social hierarchy. A West African griot may be famous as a performer but has no rights of individual ownership over his materials. Authorship comes into being only, if at all, under quite particular circumstances.

The early history of poetry in the world of Archaic and Classical Greece, to take a familiar example, is that of an emergent distinction between song and poetry, each of which was performed to a musical accompaniment on the lyre or the flute, and neither of which was dependent on writing. Its religious origins are indicated by the early lack of differentiation between the terms aōidos, singer, and mantis, prophet or seer; after a process of differentiation the word aōidos came to designate a general category and mantis a sub-category within it. By the Classical period a new category, poiētēs, emerged to underpin a distinction between song, which is still tied to the sacral realm of prophecy and the inspiration of the muses, and poetry, which belongs to a desacralized realm; “The poiētēs was a professional; he was a master of tekhnē, the work of an artisan.”45 That differentiation is tied to a loss of the immediacy of interactions between performer and audience and to an increasing cosmopolitanism, as local traditions and cults were incorporated into a pan-Hellenic culture where “a progressively restricted series of recompositions” took place “in ever-widening circles of diffusion, with the streamlining of convergent local traditions happening at the expense of divergent ones.”46 In this process texts became stabilized, helped at a certain point in the process by being written down, and poets began to make explicit claims of authorship—for example “by including a name or other identifying information within the poem, so that any re-performance would name the absent author.”47

Auctor and Auctoritas

The fusion of the roles of poet and seer eventually becomes a literary convention. Beginning with Virgil, the term vates, “a ‘herald’ of prophecies encountered in Roman divination practice,” comes to be “employed affirmatively for authorial self- descriptions of being a ‘seer’ who has privileged access to truth and the gods by way of their and the Muses’ inspiration.”48 This is one of a range of strategies for building recognition as an author. Another is association with a high-ranking patron, and a third is self-designation through an internal “seal,” a sphragis: that is, usually, a short self-description, which may include the writer’s name or a common epithet, as well as his place and date of birth. One of the functions of the sphragis is to authorize an oeuvre, as Virgil does at the end of the Georgics, where a sphragis mentions the earlier Eclogues in such a way as to indicate a continuity between the two texts. This continuity of authorship is built into the conjunctive formula ille ego, “I am the one who,” used most famously in the spurious preproem of Vergil’s Aeneid: “I am that man who once sang on a slender reed and coming out of the woods forced the neighboring fields to obey their owner, however greedy for gain, a work pleasing to farmers, but now of Mars’ bristling arms I sing and the man . . .”: a statement that imposes a retrospective coherence on the Virgilian oeuvre, while also distinguishing the epic ego of the Aeneid from the ille of the author of the two earlier works.49

The Hellenistic philologists of the 3rd and 2nd centuries bce played a major role in forming a canon of the works of antiquity through their establishment of stable and authoritative texts and their commentaries on Aristotle, Homer, Demosthenes, and others. They also helped form authorial repute by writing biographies based on information that was less historical than prescriptive and exemplary. Similar operations of formation and maintenance of repute were carried out by the system of Greek paideia and by the Roman imperial education system, which placed Homer and Virgil, respectively, at the heart of an emulative practice of moral and civic education. The Catholic Church, finally, labored over many centuries to construct both a doctrinal canon and the authority of the Church Fathers who acted as its guarantors.

That authority forms the heart of the institution of authorship in the European Middle Ages. The medieval auctor is directly identified with authority, and a person or text “becomes authoritative only if what he brings into being becomes an exemplar, fashioning rules for language, form, or belief, or offering expertise, which medieval writers generally tried to incorporate rather than surpass.”50 The great auctores—Aristotle, Cicero, Ptolemy, Clement, Jerome, Boethius, and so on—stood at the head of the various disciplines of knowledge.

Since these auctores were unsurpassed exemplars in their respective fields, however, they left little space for more modern writers; and they themselves occupied a very different space from that of the more humble activities of writing, in its medieval sense that does not distinguish between the physical act of writing and the act of composing a text. Hence St. Bonaventure’s famous distinction between the one who writes out the words of others (scriptor), the one who selects and compiles the words of others (compilator), the one who subordinates his own words of exegesis to the words of others (commentator), and the one who mixes his own with the words of others “but with his own work in principal place,” the only one to be called an author (auctor).51 To occupy the role of author is difficult when “authority, almost by definition, [was] located in the past” and the field of writing seemed to be so fully occupied by the bearers of authority.52 In particular, the field excluded vernacular writers, who either did not need to be named because their manuscripts were distributed only in their own community, or whose name had little import if it was disseminated outside it.53 Dante is a rare precursor poet here (and in Britain Chaucer, although he deprecates his authorship as mere compilation), but it is only gradually that late-medieval writers begin to be able to distinguish their modernity from the authority of the Ancients.

Humanist Authorship

Roger Chartier traces this shift to the 14th century, when the word author (acteur and then autheur in French) began to endow contemporary writers previously thought of as compilers or commentators with “the authority traditionally reserved to the ancient auctores”; the word writer (escripvain) “began to designate the person who composes a work as well as one who copies a book”; and “invention came to mean an original creation rather than solely the discovery of what God had produced.”54 An author such as Petrarch can now exercise a much stricter control over the production and distribution of his books (which come increasingly to be single works rather than collections of several texts or components of an anthology); he oversees the copying of manuscripts from an autograph copy, these subsequent copies being designed for limited circulation and instituting “a direct and authentic relation between the author and the reader,” thus reflecting “one of the major expressions of the author-function, the possibility of deciphering in the forms of a book the intentions that lay behind the creation of the text.”55

Changes in manuscript production thus anticipate some of the transformations brought about by print; but that transition is a slow and by no means linear one. Susan Stewart notes that “at the time of the emergence of printing, there were in Paris and Orléans alone more than 10,000 copyists,” and the scribal medium remains an often preferred option for authors such as Donne, in part because of the so-called “stigma of print,” in part because of a desire to control possession of copies of the works or because of affiliation with a scribal community, and in part because of the need to restrict access to material that might be considered “indecent, heterodox, seditious, or simply too far ahead of its time for a rising statesman or cleric.”56 Censorship by church and state is a major constraint on print in the early modern period, and authorship as a principle of legal responsibility is “an essential weapon in the battle against the diffusion of texts suspected of heterodoxy.”57

The Professional Author

Writers are many but authors are few; it was rare in early modernity that the writers who hired out their pens achieved recognition as the authors of coherent bodies of work. The major source of income for writers in that period, directly or indirectly, remained patronage, although one might note the way someone like Ben Jonson, who published his collected Works with a title claiming a status usually reserved for classical authors, made his living by way not only of patronage but of “publication, fees from theatrical companies for composing plays, [and] payment for collaborating on court masques.”58 The theater companies of Shakespeare and Molière similarly depended on royal or aristocratic patronage for the protection that would allow them to ply their trade.

The Grub Street writers of the late 17th and early 18th centuries were part of a process of professionalization that stretches through the entire print period. Daniel Defoe barely emerges as an author from this anonymous mass, and it is only, symbolically, with Samuel Johnson’s rejection in 1755 of Lord Chesterfield’s belated offer to act as patron of the Dictionary that a widely recognized author could be seen to break free of the demands of patronage in order to accept an existence fully in the marketplace. The tension between the marketplace and an aristocratic disdain for it “persisted through the eighteenth century, to be caught up with an emerging discourse of original genius that gave it new potency in the latter decades of the period.”59

Romantic Authorship

The concept of originality, in its distinctive modern sense designating “works that were their own origins – works that had origins nowhere beyond the author’s mind,” begins to be articulated in the mid-18th century; the locus classicus is Edward Young’s Conjectures on Original Composition, which was widely influential on early German Romanticism.60 The concept of the genius who creates an original work of art with the resources of his (sic) imagination is intimately bound up with the notion of immaterial property: “The author is conceived as the originator and therefore the owner of a special kind of commodity, the work. . . . The author and the work. The autonomous creator and the distinct literary object, unitary, closed, and caught up in relations of ownership.”61 It is from this nexus, connected by the idea of value (the value of the person and the value of property), that the protection of authorial works by copyright evolved in the course of the 18th century.

That history has been thoroughly rehearsed, in particular (from a literary-historical perspective) by Martha Woodmansee and Mark Rose, and I won’t follow it through in any detail here. In brief, the history is one of a struggle, carried out through the parliament and the courts, between the London booksellers who, until the breakdown of the licensing system in 1695, had had a virtual monopoly on publishing; authors, who were initially little more than a front for the booksellers but who by the end of the century had come to have a strong sense of having property interests in their own right; and the provincial booksellers who sought to challenge the monopoly privileges of the London guild. The 1710 Statute of Anne established a time-limited statutory right of authors in their own work, in place of the perpetual common-law right that the booksellers sought to continue. That statutory right was finally confirmed in the case of Donaldson v. Becket, or rather in the overruling by the House of Lords of the court’s decision in favor of a common-law right.

This shift from what Rose calls a regime of regulation to a regime of property derives from two rather different but interconnected principles, both articulated in their classic form by John Locke.62 The first says that property rights are grounded in a prior right in one’s own person. The second says that they are grounded in the labor that transforms commonly occurring raw materials into a private possession.63 Together they combine to conceptualize property through the topos of the author’s imprinting of his personality on the “work of original authorship.”64

That component of creative originality differentiates art both from mechanical invention and from craft. At the same time, the concept of the work becomes increasingly immaterial, increasingly separate both from its physical incarnations and from its content. The property that is protected by copyright is not the material book or the letter or the picture (known as the “copy”), which can all be owned by someone other than the author or painter, and not the “ideas” they may contain, but the expression of those ideas in a work. The distinction preserves a division between those things that are held in common and cannot be privately owned (ideas) and the unique expression that, at least for a limited period, can.

In principle every expression is distinct insofar as it emanates from a unique intentional act on the part of the author. If it can be demonstrated that an independent act of creation has taken place, then two identical works may each be entitled to copyright protection. Hence Judge Learned Hand’s dictum:

Borrowed the work must indeed not be, for a plagiarist is not himself pro tanto an “author”; but if by some magic a man who had never known it were to compose anew Keats’s Ode on a Grecian Urn, he would be an “author,” and, if he copyrighted it, others may not copy that poem, though they might of course copy Keats’s.65

Authorship as origin is, then, the most fundamental category of copyright law, in relation to which all other categories are secondary. It is the principle that founds both the work and the copy in their respective modes of being, and the act and form of expression of an idea that constitutes the work. A work can always be traced back to an origin, and it is this that defines its specificity. That connection holds particularly true for moral rights (the rights of the author with respect to the integrity of the work), which are more prevalent in the Continental than in the Anglo-American legal world; but the two systems have coevolved over the last two centuries, so that the latter is permeated with the reasoning of moral rights, just as the former is by compromises with the recognition of economic interests. It is sometimes objected that the concept of originality was rarely invoked during 18th-century copyright debates, and that many of the doctrinal structures of contemporary copyright law are at odds with the model of Romantic authorship; but, as Peter Jaszi argues, the model of expressive authorship, “when introduced into English law in 1710, was a charged receptacle, prepared to collect content over the next century . . . the term took on a life of its own as individualistic notions of creativity, originality, and inspiration were poured into it”—even to the extent that it might at times impede rather than facilitate commodification.66

The figure of the Romantic author morphs seamlessly into that of the modernist making it new or the performance artist transforming his body by means of digital prostheses. The imperative of originality, of the imprint of personality upon the work, marks Eliot’s espousal of impersonality as much as it does that of the surrealists or John Cage or the L=A=N=G=U=A=G=E poets. The discourse of originality shared by literature and the law goes as deep as their mutual commitment to property in the person and its corollary, the emanation of the work from whatever is most unique about its creator and his or her consequent right to dispose of it freely. Even the output of Brecht’s collective is marked by the imprint of the authorial brand, and Brecht too seeks to preserve the integrity of the work in its passage through the vicissitudes of the market.

Let me briefly mention two further dimensions of the Romantic author-function, one of them negative, one positive. As Learned Hand stated, a plagiarist is not an author; but plagiarism and forgery both pay tribute to the notion of originality, and indeed two exemplary works of British Romanticism, Thomas Chatterton’s “Rowley” and James Macpherson’s “Ossian” poems, are forgeries. The 18th-century crisis of authenticity resulting from these works and from the scandal of the decontextualized and aestheticizing reproduction of “folk” ballads must be understood, Stewart argues, “within a history of the establishment and legislation of spheres of originality and accountability for writing. . . . It is not so much that the ballad scandals of the 18th century were the products of rules regarding forgery, authenticity, plagiarism, and originality as that the ballad scandals helped produce such rules.”67

The “positive” dimension of the Romantic author-function is that of attribution studies, the purest form taken by the doctrine of expressive originality. Like the industry of attribution and authentication in art history, the contemporary, quasi-scientific study of literary attribution is based in a belief in the indelibility of the traces of authorial personality in the text and thus in “a fundamental concern for the integrity of the individual signature.”68 Seeking authenticity, it finds the author at the very heart of textual corruption.

Scientific Authorship

The signature that authenticates a scientific text is an important parallel to the signature attached to literary texts (insofar as these categories are ever fully distinct). The history is a long and complex one. The oldest European scientific texts are either unsigned or employ a corporate signature: Euclid’s Elements were published anonymously, and medical writers in the Hippocratic tradition transmitted their work under the signature “Hippocrates.”69 The name here designates something like “the authority of the body of medical knowledge.” If anonymity or corporate pseudonymity act as a form of deference to that authority, this attitude continues to inform later uses of the signature in scientific publications, where deference to the impersonal institution of science is frequently in tension with a claim to individual or shared credit for named individuals.

Foucault overgeneralizes in arguing that in the Middle Ages texts we would now call scientific, unlike the majority of literary texts, required the signature of their author in order to be valorized and accepted as true, but that a reversal occurred in the 17th or 18th century such that scientific texts came to be underwritten by “their membership in a systematic ensemble” while “literary discourses came to be accepted only when endowed with the author function.”70 Chartier points out that medieval and Renaissance “scientific” discourse in fact tended to be anonymous and collective, whereas the authentication of experiments conducted in the course of the scientific revolution of the 17th century “required the guarantee given by an authority progressively displaced from princely or aristocratic power to scientific authorship.”71 But this author function was qualified by a display of deference to aristocratic patrons or to an institution such as the Royal Society or the Paris Academy of Sciences, and by a reluctance to engage in a transaction seen to be economically motivated.72

The more recent rise of the large-scale multiple authorship of scientific papers (sometimes running into many hundreds of named contributors) reinforces the tension between a claim to individual or joint credit and what Foucault calls “the anonymity of an established or always redemonstrable truth . . . membership in a systematic ensemble.”73 And it generates problems for the definition of authorship as a system of accreditation. When large numbers of co-authors are involved there may be little correlation between responsibility for the research and for authorship of a paper, and that may mean an inflation of credit. Increasingly too there are problems with “guest” authors who make little substantial contribution (for example, with supervisors whose name is added to a paper written by their graduate student, or lab directors who have had little or no contact with a particular piece of research). At the limit there may be a refusal of responsibility in cases of fraud by a co-author, or scientists may lend their name to reviews of new drugs ghostwritten for a pharmaceutical company.74

Some of the issues to do with large-scale multiple authorship arise in an even more extreme form in the case of a discipline such as particle physics, where the whole pool of scientists might belong to a small handful of massive teams and where there is pressure for every member of the team to be listed as an author on major publications, including people on leave of absence—something that makes perfect sense, Mario Biagioli argues, if you think of authorship in terms not of origination or of direct responsibility but of “credit for accumulated labor.”75 In such cases, where there is a “powerful drive to diffuse authorship around the entirety of the collaboration,” any withdrawal of names from the author list appears “tantamount to epistemic subversion.”76 It is a short step from naming the team or the institution as the real author of publications arising from them to saying that it is science itself, the impersonal body of knowledge, that authors those publications.

Anonymous and Collaborative Authorship

Anonymous publication has been an important converse of the authorial signature across all literary periods. Stephen Owen writes that if authorship is considered as a property of Classical Chinese texts then in many cases it is something that is added by inference in the same way as titles often are. The ascription of authorship is always an act of valuation, and this means that in early Chinese poetry, where texts may oscillate between being assigned to an author and described as anonymous, the particular value being invoked by anonymity is that of being early: being of such antiquity as to be venerable. We can thus understand how anonymity can function “as a value in its own right. A poetic text that simply circulates without a name does not become ‘anonymous’ until it enters a literary regimen where names are assigned, decided, and arranged in chronological order.”77 The operation of anonymity varies considerably between cultures, and even within cultures there will be a diversity of motives for the choice of anonymity or its assignation to a text, including, as Robert Griffin argues, “an aristocratic or a gendered reticence, religious self-effacement, anxiety over public exposure, fear of prosecution, hope of an unprejudiced reception, and the desire to deceive.”78 But anonymity is never simply an absence of the authorial function: “The Author of Waverley,” for example, indicates a filiation between all of those novels that we now know to have been written by Walter Scott—the coherence of an oeuvre; “By a young lady” valorizes social class, gender, and age (and perhaps, by implication, sexual attractiveness) over the individual name.

Collaboration, similarly, is not a distinct mode of authorship but something like the condition of all writing. Harold Love distinguishes a range of modes of collaboration, from teams of writers dividing up a work, to ghost-writing; writing from dictation; the composition of posthumously published works (in a line running from Aristotle to Saussure and Wittgenstein) from students’ lecture notes; to single authorship that nevertheless depends on vetting by friends, revisions, editing, the material construction of the document by printers, proofreaders, bookbinders, and publicity, as well as subsequent editorial work on the establishment of a “good” text. He cites the example of Yeats, who made use of his wife Georgie’s spirit writing, relied on others to punctuate his texts, and made collaborative decisions about the order of poems within a book.79

At the most fundamental level, writing is inherently collaborative to the extent that any writer works within the logic of a language “whose proper system, laws, and life his discourse by definition cannot dominate absolutely. He uses them only by letting himself, after a fashion and up to a point, be governed by the system.”80 Authorship gives a particular form—an authority and a unity—and a particular semblance of control to those limitations of the writerly will.

The Digital Author Function

Digital authorship is so recent, so complex, so rapidly evolving a phenomenon that it is difficult to do much more than map out some of its salient features. Much digital publication (one form of this encyclopedia, for example) mirrors print publication, and in many cases, such as much hypertext cyberfiction or most blogs, authorship is attributed in the traditional manner. Conversely, of course, many elaborately constructed digital texts—games, websites, applications—are authorless, as are computer-generated literary texts and most of the vast mass of materials and processes that have migrated the communication systems of Real Life online (interactions with governments or businesses or financial systems, among others).

Some of the more interesting transformations in textuality and authorship have to do with the immediacy, the fragmentation, and the interactivity of online publication and the malleability of online identity. New forms of publishing with a “virtually zero marginal cost” of reproduction, such as fanfiction and the plethora of forms of digital self-publishing, have rendered the business models of print publication, with its high marginal production and gatekeeping costs, increasingly obsolete, as has more generally the Web 2.0 paradigm of many-to-many communication, multimodality, and highly interactive communities of readers.81 Much online data, like the work of art in the age of its technological reproducibility, has become—in John Berger’s words—“ephemeral, ubiquitous, insubstantial, available, valueless, free” (although the notion that this paradigm shift has simply replaced previous models of authorship in a sweeping and democratizing revolution is frequently exaggerated).82

Perhaps the majority of online publications—social media posts, fanzine contributions, messages, emails—are transient and their authorship unstable. The digital universe facilitates anonymity, heteronymity, and pseudonymity; it is populated by the fake identities of trolls, fantasy identities on social media or in digital games, self-promoting digital celebrities and influencers, and bots that present themselves as human authors. And much online authorship is collaborative and cumulative: the wiki is the archetype, but we could also think of texts created by exchanges or accretions of material, as in Twitter threads or online reviews.

Despite its common base of freely circulated protocols and codes and what now looks like a fantasy of the liberation of information in the digital world, the Internet quite quickly became the site of novel and strictly patrolled property relations.83 These have differed from many other forms of intellectual property, particularly copyright, however, in not being primarily author-based: instead, the enforcement of property rights on the Internet tends to take place by way of digital locks (DRM, digital rights management), license agreements, the rental of access to databases, and the algorithmic harvesting and sale of personal data while nevertheless leaving that data available for further use and repurposing. The result is that, while these mechanisms create rarity and hence commercial value on the basis of digital abundance, authorship in the online world has lost much of its scarcity and prestige, and with it a considerable amount of its authority.

Author and Writer

Questions of authorship bring into play many of the central questions of literary theory: what it is that constitutes the coherence of texts, the interpretive relevance of authorial intention, the relation of oral to literate cultures, the regulation of writing by church and state, the legal underpinnings of literary property, the significance of forgery and plagiarism, and so on.

Traditionally the ascription of authorship has been a way of guiding the interpretation of texts: of ascertaining the author’s intentions in writing it, or of reading off from their biographical data a set of contextual cues. The problem with appealing to intentions is that they are not available for inspection and are difficult to recreate with any validity, so that appealing to them tends to be a projection on the part of the interpreter. But this is not to say that biographical narratives can or should contribute nothing to the reading of texts. They can serve, for example, to preclude anachronism: knowing that Montaigne wrote the Essais between 1571 and 1592, one can rule out any reference in his work to the historical ethos of the Edict of Nantes. Conversely, knowing that Jane Austen lived in the time of the slave trade in the Americas and of the Napoleonic wars (and had close personal connections to the latter) allows us to notice that those historical events are nearly completely absent in her novels. Biographical information is particularly important when an explicit or implicit claim depends on it: that a text conveys an authentic experience of gender or ethnicity, for example. The case can best be made negatively. Psalmanazar’s claim to be a native Formosan wrongly underpinned his phantasmatic ethnography, as did Carlos Castaneda’s claim to be a shaman. The Australian writers Streten Bozic (who passed as Banumbir Wongar), Leon Carmen (who passed as Wanda Koolmatrie), and Colin Johnson (who passed as Mudrooroo) claimed an authenticity of Indigenous experience to which they were not entitled, and they will be judged by that claim (although the spectrum runs from deliberate deception in the case of Carmen, who falsely claimed to be both a woman and a member of the Stolen Generation, to something much more complex in Johnson’s case).

These ascriptions have to do with writers, however—historical individuals who are extrinsic to the texts they produce—rather than with authors, who are defined only by their immanent relation to those texts. Alexander Nehamas draws the distinction with great clarity: if Henri Beyle was the same person as Stendhal, this equation “simply specifies the feature of having been the author of Stendhal’s works, and nothing more. . . . Stendhal is whoever can be understood as the author of these texts; it is these texts that point us to him.”84 Drawing this distinction allows us

to avoid the view that to understand a text is to re-create or replicate a state of mind which someone else has already undergone. . . . Such states of mind, whatever their relation to the meaning of the text, belong to writers but not to authors. Though instrumentally important, perhaps, they have of themselves no critical significance. Authors, not being persons, do not have psychological states that might determine in advance what a text means.85

The figure of the author is broader than that of the implied author, but both emerge through an interpretive process. That process may of course be hurried along, so to speak, by authors who thematize themselves in their work (in the autofiction of Rachel Cusk or Maggie Nelson, for example), or who vigorously brand themselves in the public realm. Authors are interpretive constructs, too, in the sense that biographies are themselves always interpretations of a life rather than simple factual records, and indeed they would have little value if they were.

But authors are not simply textual constructs, because they can never simply be differentiated from the writers with whom they cohabit. The best metaphor I can find for this relationship is Freud’s concept of anaclisis (Anlehnung), which he uses to describe the “propping” of the infantile sexual instincts on certain bodily functions.86 Authors are propped on writers: materially supported by them but distinct from and independent of them. A focus on the life of writers is neither correct nor incorrect; it merely serves a different kind of interpretive end from a focus on the institution of authorship. And understanding that institution can tell us a great deal about the workings and the organization of the literary system in all its historical complexity.

Discussion of the Literature

Rather than being interrogated as a theoretical category, authorship has until quite recently been deployed primarily as a taken-for-granted principle for organizing textual study, particularly in literary-historical narratives and in textual editing, or analyzed as an empirical phenomenon. The value of the essays that Barthes and Foucault published at the end of the 1960s was to generate an impetus for the study of authorship as a key systemic dimension of the order of writing and its historical transformations. A number of collections respond more or less directly and defensively to Barthes’s provocation, including Authorship: From Plato to the Postmodern; The Death and Resurrection of the Author?; and Rückkehr des Autors.87 Much excellent scholarly work on the conditions of existence of authorship continues to occupy the fields of literary history and textual editing: studies of the organization of writing in particular periods such as Alvin Kernan’s Printing Technology, Letters, and Samuel Johnson and Betty Schellenberg’s Literary Coteries and the Making of Modern Print Culture, 1740–1790, or descriptions of the scribal culture of early modern England by Harold Love and Arthur Marotti.88 For oral composition the seminal text remains Albert Lord’s The Singer of Tales.89 For the medieval period a key text is Alastair Minnis’s Medieval Theory of Authorship; for classical antiquity and ancient China Gregory Nagy’s and Stephen Owen’s work is illuminating, as is Alexander Beecroft’s Authorship and Cultural Identity in Early Greece and China, which spans the two cultures.90 Roger Chartier offers a more theoretically oriented account of the early-modern author function in The Order of Books.91 From the literature on plagiarism, forgery, and other transgressions of authorship let me single out Susan Stewart’s Crimes of Writing and Ken Ruthven’s Faking Literature.92

The topos of the (authorial) signature runs through all of Jacques Derrida’s work, but particularly perhaps Limited Inc. and Signéponge.93 On copyright, the standard texts with a literary focus would include Mark Rose’s Authors and Owners, Martha Woodmansee’s “The Genius and the Copyright,” and Peter Jaszi’s “Toward a Theory of Copyright.”94 James Boyle has pointed to the strikingly negative consequences of the operation of the Romantic author paradigm in intellectual property law, “with costs in everything from biodiversity and the production of new drugs to the shape of the international economy and the structure of the computer industry.”95

Studies of the attribution of authorship are rarely theoretically inclined, but exceptions include the work of Harold Love, and John Burrows and Hugh Craig; and a number of essays in Artist, Authorship and Legacy speak interestingly of the practice of authentication in the art world.96 Collaborative authorship is extensively described in Jack Stillinger’s Multiple Authorship and the Myth of Solitary Genius.97 Theories of Authorship, edited by John Caughie, is a useful collection on filmic auteur theory.98 Finally, The Cambridge Handbook of Literary Authorship provides an invaluably comprehensive coverage of the topic.

Further Reading

  • Barthes, Roland. “The Death of the Author.” In Image-Music-Text. Translated by Stephen Heath, 142–148. Glasgow: Fontana, 1977.
  • Biagioli, Mario, and Peter Galison, eds. Scientific Authorship: Credit and Intellectual Property in Science. New York: Routledge, 2003.
  • Bennett, Andrew. The Author. Abingdon and New York: Routledge, 2005.
  • Berensmeyer, Ingo, Gert Buelens, and Marysa Demoor, eds. The Cambridge Handbook of Literary Authorship. Cambridge, UK: Cambridge University Press, 2019.
  • Chartier, Roger. “Figures of the Author.” In The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Centuries. Translated by Lydia G. Cochrane, 25–59. Stanford, CA: Stanford University Press, 1994.
  • Foucault, Michel. “What Is an Author?” Translated by Josué V. Harari. In Aesthetics, Method, and Epistemology: Essential Works of Foucault. Vol. 2: 1954–1984. Edited by James D. Faubon, 205–222. New York: New Press, 1998.
  • Jaszi, Peter. “Toward a Theory of Copyright: The Metamorphoses of ‘Authorship.’” Duke Law Journal (1991): 455–502.
  • Nagy, Gregory. “Early Greek Views of Poets and Poetry.” In The Cambridge History of Literary Criticism. Vol. 1. Edited by George Kennedy, 1–77. Cambridge, UK: Cambridge University Press, 2008.
  • Nehamas, Alexander. “Writer, Text, Work, Author.” In The Death and Resurrection of the Author? Edited by William Irwin, 95–115. Westport, CT: Greenwood Press, 2002.
  • Rose, Mark. Authors and Owners: The Invention of Copyright. Cambridge, MA: Harvard University Press, 1993.
  • Stewart, Susan. Crimes of Writing: Problems in the Containment of Representation. Durham, NC: Duke University Press, 1994.
  • Woodmansee, Martha. “The Genius and the Copyright: Economic and Legal Conditions of the Emergence of the ‘Author.’” Eighteenth-Century Studies 17, no. 4 (1984): 425–448.


  • 1. Roland Barthes, “The Death of the Author,” in Image-Music-Text, trans. Stephen Heath (Glasgow: Fontana, 1977), 143.

  • 2. Bertolt Brecht, “The Threepenny Lawsuit,” in Bertolt Brecht on Film and Radio, ed. and trans. Marc Silberman (London: Bloomsbury, 2000), 147 (translations slightly modified).

  • 3. Stephen McNeff, “The Threepenny Opera,” in The Cambridge Companion to Brecht, ed. Peter Thomson and Glendyr Sacks (Cambridge, UK: Cambridge University Press, 1994), 62.

  • 4. Klaus Völker, cited in John Fuegi, The Life and Lies of Bertolt Brecht (London: Flamingo, 1995), 196.

  • 5. Stephen Hinton, ed., Kurt Weill: The Threepenny Opera, Cambridge Opera Handbooks (Cambridge, UK: Cambridge University Press, 2008), 11.

  • 6. John Fuegi “The Zelda Syndrome: Brecht and Elisabeth Hauptmann,” in The Cambridge Companion to Brecht, ed. Peter Thomson and Glendyr Sacks (Cambridge, UK: Cambridge University Press, 1994), 111–112.

  • 7. Brecht, “The Threepenny Lawsuit,” 160–161.

  • 8. Brecht, “The Threepenny Lawsuit,” 162.

  • 9. Brecht, “The Threepenny Lawsuit,” 169.

  • 10. Brecht, “The Threepenny Lawsuit,” 172.

  • 11. Fuegi, “The Zelda Syndrome,” 114.

  • 12. On the work-for-hire doctrine that describes the employer, not the employee who creates it, as the true begetter of a piece of software, cf. Peter Jaszi, “Toward a Theory of Copyright: The Metamorphoses of ‘Authorship,’” Duke Law Journal (1991): 488. On the case of John Moore, who was denied a property right in a cell line derived from his body, cf. James Boyle, Shamans, Software, and Spleens: Law and the Construction of the Information Society (Cambridge, MA: Harvard University Press, 1996); and John Frow, Time and Commodity Culture (Oxford: Clarendon Press, 1997), 154–161.

  • 13. Michael D. Coogan, The Old Testament: A Historical and Literary Introduction to the Hebrew Scriptures, 2nd ed. (New York: Oxford University Press, 2011), 52.

  • 14. Coogan, The Old Testament, 52.

  • 15. Cf. Philip R. Davies, “The Dissemination of Written Texts,” in Writing the Bible: Scribes, Scribalism and Script, ed. Philip R. Davies and Thomas Römer (London: Routledge, 2014), 35–46.

  • 16. Coogan, The Old Testament, 54.

  • 17. John Van Seters, The Pentateuch: A Social-Science Commentary (London: Bloomsbury T&T Clark, 1999), 55–56.

  • 18. Van Seters, The Pentateuch, 12.

  • 19. Michael Nylan, The Five “Confucian” Classics (New Haven, CT: Yale University Press, 2001), 2, 8.

  • 20. Ouyang Xiu (c. 1000 ce), quoted in Nylan, The Five “Confucian” Classics, 13.

  • 21. Alexander Beecroft, Authorship and Cultural Identity in Early Greece and China: Patterns of Literary Circulation (Cambridge, UK: Cambridge University Press, 2010), 23.

  • 22. Hanmo Zhang, Authorship and Text-Making in Early China (Boston and Berlin: de Gruyter, 2018), 93–94.

  • 23. Zhang, Authorship and Text-Making, 140.

  • 24. Beecroft, Authorship and Cultural Identity, 64–65.

  • 25. Gregory Nagy, “Early Greek Views of Poets and Poetry,” in The Cambridge History of Literary CriticismI, ed. George Kennedy (Cambridge, UK: Cambridge University Press, 2008), 38.

  • 26. Beecroft, Authorship and Cultural Identity, 34.

  • 27. Cf. Ruth Scodel, “Authorship in Archaic and Classical Greece,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 47.

  • 28. Cf. William Leahy, ed., Shakespeare and His Authors: Critical Perspectives on the Authorship Question (London: Continuum, 2010); and Samuel Schoenbaum, Shakespeare’s Lives (Oxford: Clarendon Press, 1991).

  • 29. Cf. Johanne M. Stochholm, Garrick’s Folly: The Shakespeare Jubilee of 1769 at Stratford and Drury Lane (London: Methuen, 1964); and Christian Deelman, The Great Shakespeare Jubilee (London: Michael Joseph, 1964).

  • 30. David Davídházi, The Romantic Cult of Shakespeare: Literary Reception in Anthropological Perspective (London: Macmillan, 1998), 16; and Michael Dobson, The Making of the National Poet: Shakespeare, Adaptation and Authorship, 1660–1769 (Oxford: Clarendon Press, 1992).

  • 31. Davídházi, The Romantic Cult of Shakespeare, 16.

  • 32. Mark Rose, Authors and Owners: The Invention of Copyright (Cambridge, MA: Harvard University Press, 1993), 123.

  • 33. Barthes, “The Death of the Author,” 142.

  • 34. Michel Foucault, “What Is an Author?,” trans. Josué V. Harari, in Aesthetics, Method, and Epistemology: Essential Works of Foucault, vol. 2: 1954–1984, ed. James D. Faubon (New York: New Press, 1998), 205. Samuel Beckett’s French original—from the third of the Textes pour rien (Paris: Éditions de Minuit, 1955)—reads: “Qu’importe qui parle, quelqu’un a dit qu’importe qui parle.” Michel Foucault, “Qu’est-ce qu’un auteur?,” in Dits et Écrits 1954–1988, vol. 1: 1954–1969, ed. Daniel Defert and François Ewald (Paris: Gallimard, 1994), 792.

  • 35. Foucault, “What Is an Author?,” 211.

  • 36. Foucault, “What Is an Author?,” 210–211.

  • 37. Foucault, “What Is an Author?,” 214.

  • 38. Foucault, “What Is an Author?,” 222.

  • 39. Foucault, “What Is an Author?,” 216.

  • 40. Foucault, “What Is an Author?,” 214.

  • 41. And he does, strangely, reintroduce the category of the transcendental author in his account of the “founders of discursivity” (Marx and Freud are his examples) and his tortuous struggle to distinguish them from the category of “transdiscursive” authors (Homer, Aristotle, the Church Fathers) and from the founders of sciences; Foucault, “What Is an Author?,” 217ff.

  • 42. Cf. Beecroft, Authorship and Cultural Identity, 16: “Authorship is a property ascribed to a literary text. It reflects an attempt to ground and contextualize that text by assigning its composition and/or performance to a specific individual, real or hypothetical. . . .”

  • 43. Alexander Nehamas, “Writer, Text, Work, Author,” in The Death and Resurrection of the Author? ed. William Irwin (Westport, CT: Greenwood Press, 2002), 99.

  • 44. Richard P. Smiraglia, Hur-Li Lee, and Hope A. Olson, “Epistemic Presumptions of Authorship,” in Proceedings of iConference 2011: Inspiration, Integrity, and Intrepidity (New York: ACM, 2011), 138–139.

  • 45. Nagy, “Early Greek Views,” 23.

  • 46. Nagy, “Early Greek Views,” 39.

  • 47. Scodel, “Authorship,” 53.

  • 48. Christian Badura and Melanie Möller, “Authorship in Classical Rome,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 65.

  • 49. Badura and Möller, “Authorship,” 70.

  • 50. Emily Steiner, “Authority,” in Middle English: Oxford Twenty-First Century Approaches to Literature, ed. Paul Strohm (Oxford: Oxford University Press, 2007), 143.

  • 51. Quoted in Elizabeth L. Eisenstein, The Printing Press as an Agent of Change: Communications and Cultural Transformations in Early Modern Europe (Cambridge, UK: Cambridge University Press, 1979), 121–122.

  • 52. Michael North, “Authorship and Autography,” PMLA 116, no. 5 (2001): 1380.

  • 53. Andrew Bennett, The Author (Abingdon and New York: Routledge, 2005), 40.

  • 54. Roger Chartier, “Foucault’s Chiasmus: Authorship between Science and Literature in the Seventeenth and Eighteenth Centuries,” in Scientific Authorship: Credit and Intellectual Property in Science, ed. Mario Biagioli and Peter Galison (New York: Routledge, 2003), 25.

  • 55. Roger Chartier, “Figures of the Author,” in The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Centuries, trans. Lydia G. Cochrane (Stanford, CA: Stanford University Press, 1994), 55.

  • 56. Susan Stewart, Crimes of Writing: Problems in the Containment of Representation (Durham, NC: Duke University Press, 1994), 10; J. W. Saunders, “The Stigma of Print: A Note on the Social Bases of Tudor Poetry,” Essays in Criticism 1 (1951): 31–48; and Harold Love, “Scribal Publication in Seventeenth-Century England,” Transactions of the Cambridge Bibliographical Society 9, no. 2 (1987): 141.

  • 57. Chartier, “Figures of the Author,” 50.

  • 58. Kevin Dunn, “Literary Authorship in the Traditions of Rhetoric and Poetics,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 246.

  • 59. Betty A. Schellenberg, “The Eighteenth Century: Print, Professionalization, and Defining the Author,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 137.

  • 60. Jack Lynch, “Plagiarism and Forgery,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 356; and cf. Martha Woodmansee, “The Genius and the Copyright: Economic and Legal Conditions of the Emergence of the ‘Author,’” Eighteenth-Century Studies 17, no. 4 (1984): 425–448.

  • 61. Rose, Authors and Owners, 1.

  • 62. Rose, Authors and Owners, 15.

  • 63. John Locke, “The Second Treatise of Government,” in Two Treatises of Government, ed. Peter Laslett (Cambridge, UK: Cambridge University Press, 1960), §27.

  • 64. Rose, Authors and Owners, 114.

  • 65. Sheldon v Metro-Goldwyn Pictures Corp., 81 F2d 49, 54 (1936, CA2 NY), cert den 298 US 669, 80 L Ed 1392, 56 S Ct 835.

  • 66. Erlend Lavik, “Romantic Authorship in Copyright Law and the Use of Aesthetics,” in The Work of Authorship, ed. Mireille van Eechoud (Amsterdam: Amsterdam University Press, 2014), 47; and Jaszi, “Toward a Theory of Copyright,” 471.

  • 67. Stewart, Crimes of Writing, 103.

  • 68. Bennett, The Author, 98.

  • 69. Scodel, “Authorship,” 58.

  • 70. Foucault, “What Is an Author?,” 212–213.

  • 71. Chartier, “Foucault’s Chiasmus,” 27–28.

  • 72. Chartier, “Foucault’s Chiasmus,” 22; and Mary Terrall, “The Uses of Anonymity in the Age of Reason,” in Scientific Authorship: Credit and Intellectual Property in Science, ed. Mario Biagioli and Peter Galison (New York: Routledge, 2003), 92.

  • 73. Foucault, “What Is an Author?,” 212–213.

  • 74. Drummond Rennie and Annette Flanagin, “Authorship! Authorship! Guests, Ghosts, Grafters, and the Two-Sided Coin,” JAMA 271, no. 6 (1994): 469–470.

  • 75. Mario Biagioli, “Rights or Rewards? Changing Frameworks of Scientific Authorship,” in Scientific Authorship: Credit and Intellectual Property in Science, ed. Mario Biagioli and Peter Galison (New York: Routledge, 2003), 270.

  • 76. Peter Galison, “The Collective Author,” in Scientific Authorship: Credit and Intellectual Property in Science, ed. Mario Biagioli and Peter Galison (New York: Routledge, 2003), 342, 345.

  • 77. Stephen Owen, The Making of Early Chinese Classical Poetry, Harvard East Asian Monographs (Cambridge, MA: Harvard University Asia Center, 2006), 7–8.

  • 78. Cf. Robert J. Griffin, “Anonymity and Authorship,” New Literary History 30, no. 4 (1999): 885.

  • 79. Harold Love, Attributing Authorship: An Introduction (Cambridge, UK: Cambridge University Press, 2002), 38.

  • 80. Jacques Derrida, Of Grammatology, trans. Gayatri Spivak (Baltimore: Johns Hopkins University Press, 1976), 158.

  • 81. Adriaan van der Weel, “Literary Authorship in the Digital Age,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 219.

  • 82. John Berger, Ways of Seeing (London: BBC, 1972), 32; Mark Poster, What’s the Matter with the Internet? (Minneapolis: University of Minnesota Press, 2001); and R. Lyle Skains, Digital Authorship: Publishing in the Attention Economy (Cambridge, UK: Cambridge University Press, 2019).

  • 83. Cf. Lawrence Lessig, Code: Version 2.0 (New York: Basic Books, 2006).

  • 84. Nehamas, “Writer, Text, Work, Author,” 99.

  • 85. Nehamas, “Writer, Text, Work, Author,” 110.

  • 86. Sigmund Freud, Three Essays on the Theory of Sexuality [1905], The Standard Edition of the Complete Psychological Works of Sigmund Freud, vol. 7, ed. James Strachey (London: Hogarth Press, 1953), 181.

  • 87. Seán Burke, Authorship: From Plato to the Postmodern; A Reader (Edinburgh: Edinburgh University Press, 1995); William Irwin, ed., The Death and Resurrection of the Author? (Westport, CT: Greenwood Press, 2002); and Fotis Jannidis, et al., eds., Rückkehr des Autors: Zur Erneuerung eines umstrittenen Begriffs (Tübingen: Niemeyer, 1999).

  • 88. Alvin B. Kernan, Printing Technology, Letters, and Samuel Johnson (Princeton, NJ: Princeton University Press, 1987); Betty A. Schellenberg, Literary Coteries and the Making of Modern Print Culture, 1740–1790 (Cambridge, UK: Cambridge University Press, 2016); Harold Love, Scribal Publication in Seventeenth-Century England (Oxford: Clarendon Press, 1993); and Arthur Marotti, John Donne, Coterie Poet (Madison: University of Wisconsin Press, 1986).

  • 89. Albert Bates Lord, The Singer of Tales, 2nd ed. (Cambridge, MA: Harvard University Press, 2000).

  • 90. Alastair Minnis, Medieval Theory of Authorship: Scholastic Literary Attitudes in the Later Middle Ages, 2nd ed. (Philadelphia: University of Pennsylvania Press, 2010); Nagy, “Early Greek Views”; Gregory Nagy, Homeric Questions (Austin: University of Texas Press, 1996); Owen, Making of Early Chinese Classical Poetry; and Beecroft, Authorship and Cultural Identity.

  • 91. Roger Chartier, The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Centuries, trans. Lydia G. Cochrane (Stanford, CA: Stanford University Press, 1994).

  • 92. Stewart, Crimes of Writing; and Ken K. Ruthven, Faking Literature (Cambridge, UK: Cambridge University Press, 2001).

  • 93. Jacques Derrida, Limited Inc. (Evanston, IL: Northwestern University Press, 1988); and Jacques Derrida, Signéponge = Signsponge, trans. Richard Rand (New York: Columbia University Press, 1984).

  • 94. Rose, Authors and Owners; Woodmansee, “The Genius and the Copyright”; and Jaszi, “Toward a Theory of Copyright.” Cf. also Martha Woodmansee and Peter Jaszi, eds., The Construction of Authorship: Textual Appropriation in Law and Literature (Durham, NC: Duke University Press, 1994).

  • 95. Boyle, Shamans, Software, and Spleens, xi.

  • 96. Love, Attributing Authorship; John Burrows and Hugh Craig, “Attribution,” in The Cambridge Handbook of Literary Authorship, ed. Ingo Berensmeyer, Gert Buelens, and Marysa Demoor (Cambridge, UK: Cambridge University Press, 2019), 325–340; and Daniel McClean, ed., Artist, Authorship and Legacy (London: Ridinghouse, 2018).

  • 97. Jack Stillinger, Multiple Authorship and the Myth of Solitary Genius (Oxford: Oxford University Press, 1991).

  • 98. John Caughie, ed., Theories of Authorship (London: Routledge, 1981).