The Death of the Author

Barthes’s title is intentionally startling and provocative. What author? How did the author die? It sounds more like the title of a murder mystery than a piece of literary criticism. The essay’s overall structure is also important to note. It is made of up seven individual sections. In some of the original printings of the essay, asterisks or ellipses appear between the seven sections to call the reader’s attention to the fact that each section is itself a separate reflection, like a mini-essay unto itself. The sections are not to be read strictly as a logical or linear sequence of paragraphs.

The essay is intentionally (Barthes would likely say necessarily, given his thesis) fragmentary and difficult, thereby living up to the French root of the word “essay”; essayer means “to try” or “to attempt.” Many of Barthes’s sentences are extraordinarily long and complex, which has the effect of continually postponing any final “payoff” the reader might get by arriving at their end to grasp some final meaning. This can be frustrating, but Barthes intends to frustrate his readers’ attempts to “nail down” a fixed, unambiguous meaning of his text. This deferral of any final interpretation of meaning is Barthes’s central theme.

Even very advanced readers will find the essay a constant challenge to unpack—not only because of Barthes’s impressive literary erudition and dense writing style, but also because he uses familiar words in unfamiliar ways and provides no neat, step-by-step, linear path through his ideas. Instead, he circles around his theme from multiple vantage points. As noted earlier, Barthes uses the words “writing” and “text” in special ways. He prefers the word “scriptor” to “writer,” and he would like to do away with “literature,” replacing it with “writing.” Even though Barthes’s writing is often quite graceful and eloquent, the essay makes the reader work hard. This is fitting, given the essay’s thesis that there is no meaning in any text that can be accessed by referring to its author’s intention(s) or any other context. Barthes thus uses style to try to break readers’ mental habit of assuming that a discrete, finalized meaning resides at some location in the text where an author has deposited it, like a letter that can be understood simply by opening the containing envelope.

In his first section, Barthes deepens the reader’s curiosity about this “death” by referring to a complex scene in Sarrasine, a novella by Honoré de Balzac. Barthes strategically chooses this scene in at least three ways. First, the main story in the novella is that of the title character, Ernest-Jean Sarrasine. Sarrasine is a man who witnesses the singing performance of a castrato, an adult male singer castrated before puberty, which enables him to sing in the higher registers usually only reachable by female singers. Additionally, Sarrasine’s story is related, sometimes in first person by a narrator speaking to a woman acquaintance many years after the main action. Furthermore, this narrator is himself a writer who begins the novella by describing his own act of writing one evening while looking out his window, simultaneous overhearing conversations in the Parisian street below and contemplating the contrasting images of life and death that he has just written about. Sarrasine thus involves a multilayered story-within-a-story, told in multiple overlapping voices. Barthes devoted an entire book, titled S/Z, to a line-by-line analysis of Balzac’s novella, which brings out many nuances in Barthes’s decision to open this essay with this scene. In short, Sarrasine is a perfect vehicle for conveying the essay’s idea that a text is not created by an author but has an existence in language unto itself.

Second, when Sarrasine hears what he thinks is a natural female voice, he effusively exclaims how that voice embodies and represents the essential nature of all womankind. He recites a list of very artificial characteristics associated with women (very 19th-century French, male-defined stereotypes). Similarly, Barthes wants his readers to question their “natural” sense of what is being communicated, and by whom, when we read any piece of writing. In Sarrasine, is the line about womanhood spoken by Balzac the man, relating his lived experience? Or Balzac the literary writer, creating a story using established literary conventions? Or is it the novella’s writer-narrator channeling French cultural stereotypes? Is it Sarrasine? Is there some voice of origin that naturally resides in the text? Barthes’s answer is that we cannot know.

Third, this scene centers on the voice of a castrato named Zambinella, a man whose voice has been permanently altered through the violent act of castration. (This horrific practice, carried out in the 16th through 18th centuries in Europe, was first done when the Catholic Church banned women from performing on stage, literally silencing them.) Sarrasine mistakes Zambinella’s voice and costumed body as indicating a woman, and he falls in love with “her.” Later, when Sarrasine is scandalized to admit to himself that Zambinella is indeed a man (or at least not a woman), Sarrasine plots to murder the singer but is himself is murdered by others. Even the revelation that the castrato is not a woman is unreliable and ambiguous, because Zambinella in some sense is neither a man nor a woman. The “manhood” of the castrato has been forever removed, his original identity forever erased. Barthes strategically does not name the singer (Zambinella) in his essay, referring only to “a castrato.” The so-called author of any text, Barthes is already arguing here, is as unknowable as a fictional character whose identity has been permanently erased through castration, in a story relayed by a narrator (in a tale framed by an account of his own writing about death) who describes experience of a character who is murdered. If this seems disorienting, that is the point.

In using this scene to illuminate the complexity of the “death of the author,” Barthes also suggests that accepting the idea that meaning can be imposed on a text by its author or anyone else is metaphorically an act of violence that disfigures the real nature of language, torturing it to do what it cannot. Later in the essay, in Section 3, Barthes makes this analogy explicit, referring to “the castrating objectivity of the realist novelist” (143). He means that the pretense of conveying some objective reality in writing metaphorically “castrates” language, maiming and incapacitating it.

Barthes’s style throughout the essay is significant, but nowhere more so than in the first section. The quoted line in Sarrasine expresses the purportedly true characteristics of “woman” in a series of six short parallel clauses. Barthes craftily echoes these six bold assertions of supposed truth in a parallel string of six questions about the speaker’s identity, each of which increasingly undermines the certainty conveyed by the text’s enunciated line about ideal womanhood. These questions, designed to shake the readers’ confidence about who is speaking, are thus “disguised” in the style of the confident speaker, just as the castrato is disguised as a woman in the novella, and just as castration “disguises” the male singer’s voice. This is typical of Barthes’s playful but difficult writing style.

Some of his cleverness is lost in translation. For example, this English translation describes writing as “the negative where all identity is lost” (142). A more literal translation would be “the black-and-white [noir-et-blanc] where all identity is lost.” Most people assign unambiguous certainty to things said to be “black and white,” but Barthes paradoxically indicates that that which is black-and-white (writing, supposedly) is always masking an underlying uncertainty. What readers think is predetermined and thus safely taken for granted in writing (such as who is speaking) is completely uncertain. Barthes’s term noir-et-blanc cleverly points to both the black-and-white of letters and words on the printed page, which work by visually contrasting a positive figure (black) against negative ground (white), as well as to black-and-white images that, despite their name, are actually composed of many shades of gray. (Barthes was also very interested in and wrote about the rhetoric of images.)

When Barthes states his thesis that “writing is the destruction of every voice, of every point of origin,” we should realize that he is using the word “writing” in a specialized sense. The French word for writing, écriture, is more closely related to the English word “script.” (Both come from the Latin root scriptura, and “script” came into English via French; “writing” in English has a different, old Germanic root). Where English-speaking readers encounter the word “writing” in translations of Barthes, it is helpful to think of writing more as transcription from a mysterious linguistic source, almost like a fortune-teller channeling a spirit (as long as we recognize that it would be the spirit that gives life to the fortune-teller, not the other way around). In the third section, Barthes replaces the word “author” with “scriptor” to help make this point.

Barthes notes in Section 2 that this “dead” author is not really a new phenomenon, suggesting that in more primitive cultures it was well understood by listeners that a shaman, telling a mythical story, was not the story’s author. In French he refers not to primitive societies (though this is used in some translations) but to “ethnographic societies,” as Heath’s translation has it. This is an intentionally unusual turn of phrase. Ethnography is a method of investigation employed in the social sciences, such as anthropology, whereby the scientist immerses him or herself in another culture and writes detailed descriptions of the society under study, thereby revealing some original truth about that society. “Ethnography” literally means “writing about culture.” In referring to “ethnographic societies,” Barthes highlights the fact that ethnographic writing is not really done by the investigator as author. Far from revealing any truth about its subject matter (the primitive society at hand), Barthes suggests ethnographic writing (like all writing), only disguises meaning in scientific language while making a false claim to authoritative knowledge.

In Section 2, Barthes begins to use strong metaphorical language to condemn the idea that authors make decipherable meaning in texts. This idea has made readers focus “tyrannically” on the author. With such language, he reveals an oppositional stance toward the cultural sources that he believes have led to the undue crowning of the “author” as a commonplace idea. Thus, when Barthes notes that this illusion is due to dominant ideas of the Enlightenment, such as English empiricism (the 17th- and 18th-century philosophy foregrounding individual sensory experience as the source of knowledge), French rationalism (philosophy of the same period emphasizing reason as the only reliable mode of knowing), and the “personal faith” of the Reformation (the Protestant idea that religious faith is centered in the individual, not directed by Church authority), readers should apprehend that Barthes does not regard these as benign influences on modern society. They all seem to traffic in a “big lie” about writing and language. Likewise, he cites “capitalist ideology” as also according undue primacy and importance to the author. Thus, Barthes’s essay is not just out to revise a commonplace misunderstanding of what an author is and does; he is sharply critiquing the dominant culture that has led to this misunderstanding. Barthes continues using this kind of critical language throughout the essay. In the translation used here, for example, he refers to “the sway of the author” (143). In French, however, he refers to “the empire of the author,” whose grip certain modern French writers have sought to “loosen” (in French he uses even stronger language; the verb ébranler means not so much “to loosen” as “to shake” or “to topple.”) In democratic society generally, and French culture specifically, “empire” is a term loaded with negative connotations.

In Section 3, Barthes surveys some writers who have nevertheless grasped the “death of the author” thesis. The section’s structure is rhetorically significant. Barthes begins by referring to a poet, Stéphane Mallarmé, who was one of the most prominent symbolist poets. Mallarmé embodied many of the characteristics associated with the symbolist movement, particularly the use of extravagant metaphor, disorienting syntax, and ambiguous imagery, which made his work highly resistant to interpretation. Next Barthes refers to the poet Paul Valéry, a follower of Mallarmé. Readers familiar with these poets would know that Valéry was known for a period called “the long silence,” when he wrote nothing for nearly two decades after Mallarmé’s death. So far there is a sequence of three references to symbolist poets (Barthes mentioned Charles Baudelaire in the previous section), one barely mentioned as being misinterpreted, the next well known for resisting interpretation, the third who went silent after the death of his mentor (leaving nothing from that period to be interpreted). Thus, Barthes presents a fitting series of literary references in an essay that seeks to challenge the idea that authorship is a key to interpretation.

This trio of poets then yields to a novelist, Marcel Proust. Prose fiction is generally thought to be more transparent than poetry, but Proust, equally aware of the ambiguity of authorship, intentionally blurred the line between author and character, constructing a massive, multivolume account (some 3,500 pages!) of the deep psychological experiences of an anonymous character who plans to write the very novel at hand, and who only commences his writing at the end of Proust’s novel. Barthes then discusses surrealism, an artistic movement that sought to subvert the conventions of art though could only create “an illusory subversion” (i.e., it failed) because the rules (or codes) of convention can only be tinkered with from within. The surrealists made progress on a course toward “secularizing” the Author, however.

The next step in this progress is, finally, “outside literature itself”: the science of modern linguistics. Once again, readers are apt to think of scientific nonfiction as the most transparent writing (compared to poetry and fiction), something that can stand objectively “outside” the thing it studies. But once again, linguistics points only to an evanescent miragelike “source” for any writing. The writer is not a person but only a “subject”—in a sense, a passive receiving point for language and culture—who “holds together” language. However, in so doing, the writer also “exhausts” or uses language up, like an engine that runs by burning fuel. This section superficially mimics a progression from obscurity and dubious interpretation to greater clarity and certainty, but carries out instead a regression to greater and greater obscurity. The section begins by talking about “the empire of the Author,” but ends in an empty land from which the emperor has vanished. Barthes’s paradoxical playfulness has led some critics to read “The Death of the Author” as a kind of satire of ordinary essay writing.

The last words of Barthes’s essay are ambiguous enough to be considered as possible satire. They are often quoted partially: “the birth of the reader must be at the cost of the death of the Author” (148). This seems to point to the liberation of the reader as an ultimate authority and final determiner of meaning. But when taken in full, this last line is fittingly ambiguous: “[W]e know that to give writing its future, it is necessary to overthrow the myth: the birth of the reader must be at the cost of the death of the Author” (148). Is the birth of the reader-as-author the outcome of the death of the author, or is the birth of the reader-as-author part of the myth that must be smashed? Barthes raises this question, but unlike the question he raised at the beginning while discussing Sarrasine, he does not provide an answer, at least not directly. If the essay provides any answer, it is this: The reader-as-interpreter is a tenuous thing that must never become a godlike Reader-as-Author. Readers are free to do one thing: read.