Roland Barthes's famous essay "The Death of the Author" (1967) is a meditation on the rules of author and reader as mediated by the text. Barthes's essential argument is that the author has no sovereignty over his own words (or images, sounds, etc.) that belong to the reader who interprets them. When we encounter a literary text, says Barthes, we need not ask ourselves what the author intended in his words but what the words themselves actually say. Text employ symbols which are deciphered by readers, and since function of the text is to be read, the author and process of writing is irrelevant.
"The death of the author" notion means that meaning is not something retrieved or discovered, having been there all the while, but rather something spontaneously generated in the process of reading a text, which is an active rather than passive action. Barthes does not intend to suggest that the death of the author lets any reader read any text any way he or she like (though others aside from Barthes perused this line of thought). What Barthes is suggesting is that reading always involves at least a little bit of writing or rewriting of the text's meaning.
Barthes's "The Death of the Author" is an attack on traditional literary criticism that focused too much on trying to retrace the author's intentions and original meaning in mind. Instead Barthes asks us to adopt a more text oriented approach that focuses on the interaction of the reader, not the writer, with it. This means that the text is much more open to interpretation, much more fluid in its meaning than previously thought.
See additional summaries of works by Roland Barthes:
Roland Barthes / Rhetoric of the Image
Roland Barthes - Myth Today
This might also interest you:
Stanley Fish -Is There a Text in This Class?
In the fourth essay in his series on the French thinker, political theorist Andrew Robinson takes on one of Barthes' most famous (and controversial) proclamations: the death of the author.
By Andrew Robinson
Death of the Author
Many of Barthes’s works focus on literature. However, Barthes denied being a literary critic, because he did not assess and provide verdicts on works. Instead, he interpreted their semiotic significance. Barthes’s structuralist style of literary analysis has influenced cultural studies, to the chagrin of adherents of traditional literary approaches.
One notable point of controversy is Barthes’s proclamation of the ‘death of the author’. This ‘death’ is directed, not at the idea of writing, but at the specifically French image of the auteur as a creative genius expressing an inner vision. He is opposing a view of texts as expressing a distinct personality of the author.
Barthes vehemently opposes the view that authors consciously create masterpieces. He maintains that authors such as Racine and Balzac often reproduce emotional patterns about which they have no conscious knowledge. He opposes the view that authors should be interpreted in terms of what they think they’re doing. Their biographies have no more relevance to what they write than do those of scientists.
In ‘The Death of the Author’, Barthes argues that writing destroys every voice and point of origin. This is because it occurs within a functional process which is the practice of signification itself. Its real origin is language. A writer, therefore, does not have a special genius expressed in the text, but rather, is a kind of craftsman who is skilled in using a particular code. All writers are like copywriters or scribes, inscribing a particular zone of language.
The real origin of a text is not the author, but language. If the writer expresses something ‘inner’, it is only the dictionary s/he holds ready-formed. There is a special art of the storyteller to translate linguistic structures or codes into particular narratives or messages. Each text is composed of multiple writings brought into dialogue, with each code it refers to being extracted from a previous culture.
Barthes’s argument is directed against schools of literary criticism that seek to uncover the author’s meaning as a hidden referent which is the final meaning of the text. By refusing the ‘author’ (in the sense of a great writer expressing an inner brilliance), one refuses to assign an ultimate meaning to the text, and hence, one refuses to fix its meaning.
It becomes open to different readings. According to Barthes, the unity of a text lies in its destination not its origin. Its multiplicity is focused on the reader, as an absent point within the text, to whom it speaks. The writer and reader are linguistic persons, not psychological persons. Their role in the story is defined by their coded place in discourse, not their specific traits.
A text cannot have a single meaning, but rather, is composed of multiple systems through which it is constructed. In Barthes’s case, this means reading texts through the signs they use, both in their structure in the text, and in their wider meanings.
Literature does not represent something real, since what it refers to is not really there. For Barthes, it works by playing on the multiple systems of language-use and their infinite transcribability – their ability to be written in different ways.
The death of the author creates freedom for the reader to interpret the text. The reader can recreate the text through connecting to its meanings as they appear in different contexts.
In practice, Barthes’s literary works emphasise the practice of the craft of writing. For instance, Barthes’s structuralist analysis of Sade, Fourier and Loyola emphasises the structural characteristics of their work, such as their emphasis on counting and their locations in self-contained worlds. He views the three authors as founders of languages (logothetes).
The Structure of Narrative
In ‘Introduction to the Structural Analysis of Narratives’, Barthes explores the structure of narrative, or storytelling, from a structuralist perspective. Narrative consists of a wide variety of genres applied to a wide variety of substances – for example, theatre, film, novels, news stories, mimes, and even some paintings. We can see what Barthes terms ‘narrative’ whenever something is used to tell a story. People using this theory will often refer to the way people live their lives as narratives, and some will talk about a right to tell our own story.
Narrative is taken to be humanly universal – every social group has its own narratives. Barthes models the analysis of narrative on structuralist linguistics. The structure or organisation is what is most essential in any system of meaning.
The construction of a narrative from different statements is similar to the construction of a sentence from phonemes. Barthes argues that there are three levels of narrative: functions, actions, and narration. Each has meaning only in relation to the next level.
Functions refer to statements in narratives. Every statement or sentence in a novel, for example, has at least one function. Barthes gives examples like: ‘James Bond saw a man of about fifty’ and ‘Bond picked up one of the four receivers’.
For Barthes, every statement has a particular role in the narrative – there are no useless statements, no ‘noise’ in the information-theory sense.
But statements vary in their importance to the narrative, in how closely or loosely it is tied to the story. Some are functions in the full sense, playing a direct role in the story. For instance, a character buys a gun so s/he can use it later in the story. The phone rings, and Bond picks it up – this will give him information or orders which will move the action forward.
Others are ‘indices’ – they index something which establishes the context of the story. They might, for instance, convey a certain atmosphere. Or they might say something about the psychology or ‘character’ of an actor in the story. The ‘four receivers’ show that Bond is in a big, bureaucratic organisation, which shows that he is on the side of order. The ‘man of about fifty’ indicates an atmosphere of suspicion: Bond needs to establish who he is and which side he is on.
Among the former – the true functions – these can be central aspects of the narrative, on which it hinges (‘cardinal points’ or ‘nuclei’), or they can be complementary (catalysers). To be cardinal, a function needs to open or close a choice on which the development of the story depends. The phone ringing and Bond answering are cardinal, because the story would go differently if the phone didn’t ring or Bond didn’t answer.
But if Bond ‘moved towards the desk and answered the phone’, the phrase ‘moved towards the desk’ is a catalyser, because it does not affect the story whether he did this or not. Stories often contain catalysers to provide moments of rest from the risky decision-points.
Barthes sees true functions as forming pairs: one initiates a choice and the other closes it. These pairs can be close together, or spread out across a story. The choice is opened by the phone ringing, and closed by Bond answering it.
Indices are also divided into true indices, which index things like an actor’s character or an atmosphere, and informants, which simply identify something or situate it in time and space. A character’s age is an example of an informant. True indices are more important to the story than informants.
All moments of a narrative are functional, but some more so than others. Functions and indices are functional in different ways. Cardinal functions and true indices have greater functionality than catalysers and informants. At root, however, a narrative is structured through its nuclei. The other functional elements are always expansions on the nuclei. It is possible, as in folk-tales, to create a narrative consisting almost entirely of nuclei.
Functions are arranged into narratives by being attached to agents – characters in the story who engage in actions. Every narrative necessarily has agents. The actions of an agent connect the nuclei of the narrative to particular ‘articulations of praxis’ – desire, communication and struggle.
The third level, narration, occurs between the narrator (or writer) and the reader. The narrator compiles the narrative in a way which is addressed to the reader, and ‘produces’ the reader as a particular position in the narrative. The positions of narrator and reader are clearest when a writer addresses a factual statement directly to the reader: ‘Leo was the owner of the joint’. Narrator and reader are largely empty positions within the narrative.
Narratives also have a kind of logical time which is interior to them and is barely connected to real time. This logical time is constructed by the series of nuclei (which open and close choices), and their separation by other nuclei and by subsidiary elements. It is held together by the integration of the pairs of nuclei.
Narratives implicitly receive their meaning, however, from a wider social world. Barthes maintains that narratives obtain their meaning from the world beyond them – from social, economic and ideological systems.
Barthes criticises the narratives of his day for trying to disguise the process of coding involved in constructing a narrative. As in Mythologies, he again argues that this naturalisation of signs, and denial of the process of social construction of meaning, is specifically bourgeois. Both bourgeois society and its mass culture ‘demand signs which do not look like signs’. They are reluctant to declare their codes.
Narrative also contains other potentials. Like dreaming, it alters the familiar in ways which show different possibilities. Although what is ‘known’ or ‘experienced’ is constantly re-run through narratives, the narratives do not simply repeat what is re-run through them. They open a ‘process of becoming’. In other words, things can run differently when run through narrative. Narrative shows that other meanings are possible. Familiar things can be given different meanings.
What happens in narrative has no referent. It doesn’t refer to something in the real world. Rather, what happens in narrative is language itself – the celebration of its many possibilities. However, it is also closely connected to monologue (which follows in personal development from dialogue).
Barthes is highly critical of realist and naturalist views of writing. For Barthes, literature is built on emptiness: it represents something which is not really there. All the arts of fiction, including theatre, cinema and literature, are constructed based on signs. They function by the suspension of disbelief. They function by calling certain desires or structures into play, causing people to feel various emotions. They are not representations of reality, but rather, a way to induce feelings in the audience.
The attempt to convince the audience that the story is real is a way of reproducing the naturalisation of signs. A supposedly realistic or naturalistic art or literature never really ‘tells it like it is’. It represents through a set of conventional signs which stand for ‘reality’.
Barthes criticises those who believe authors imitate an existing reality (a practice known as mimesis). He is in favour of an emphasis on the creation of a discursive world (semiosis) rather than mimesis. Hence his interest in Sade, Fourier and Loyola. Instead of conventional views of the world, alternative presentations can denaturalise the present and provide utopian alternatives.
Barthes also criticises the idea of clarity in literature, for similar reasons. Clarity is simply conventional. It is relative to a particular regime of signs. It amounts to a criterion of familiarity. Therefore, it has conservative effects. Barthes views clarity as a class attribute of the bourgeoisie, used to signify membership of this class (this contrasts sharply with the more common claim in activist circles that speech should be clear so as to be working-class or inclusive).
However, this is not strictly an expressive view either. The actor or author doesn’t necessarily induce sympathy for their own feelings. Such an effect can amount to confusing art with reality. Instead, the actor, author and audience all know it’s fiction.
In some contexts, such as theatre, wrestling, and (in Barthes’s view) Japanese culture, performance or artifice is recognised for what it is. It is not taken to be natural or real. In these contexts, signs have no content. Their operation serves to show the existence and functioning of signs. It also allows an expressive use of signs, to stand for particular emotions.
In ‘Rhetoric of the Image’, Barthes discusses the different levels of meaning in a Panzani advert. Firstly, there’s a linguistic message, which has the usual denoted and connoted levels. Secondly, there’s a connotation, established by juxtaposition, associating the brand with freshness and home cooking. Thirdly, there’s the use of colours and fruits to signify ‘Italianicity’, the mythical essence of Italy. Fourthly, the processed product is presented as if equivalent to the surrounding unprocessed items. These signifiers carry ‘euphoric values’ connected to particular myths. According to Barthes, at least the third of these meanings is quasi-tautological.
The language of images is constructed in particular zones or ‘lexicons’. Each of the connoted meanings refers to a specific body of social practice which certain readers will receive, and others may not. For instance, it mobilises ideas from tourism (Italianicity) and art (the imitation of the style of a still life painting). Often the same signifieds are carried by text, images, acting and so on. These signifieds carry a particular dominant ideology. A rhetoric of the image deploys a number of connotative images to carry messages.
All images are ‘polysemous’ – they can be read in a number of ways. In an image such as this, language is used both explicitly and implicitly to guide the selection of meanings. The text directs the reader as to which meanings of the image to receive. Barthes thus suggests that texts have a repressive value relative to images: they limit what can be seen. It is in this limitation that ideology and morality function. Ideology chooses among multiple meanings which ones can be seen, and limits the shifting flow of signification which would otherwise happen.
Euphoria and Affect
Euphoria has both positive and negative meanings in Barthes’s work. As a negative term, it refers to the enjoyment of a closed system or familiar meaning which is induced by mythical signifiers. For instance, the fashion system is euphoric because its persistence as a system defies death. People can partake in a system of meanings which seems eternal, and thereby experience some of its illusory universality as euphoria. Myth provides euphoria because it provides a sense that something is absolutely clear. It aims for a euphoric security which comes with enclosing everything in a closed system. Tautology, for instance, gives someone the minor satisfaction of opting for a truth-claim without the risk of being wrong (because nothing substantive has been said). This can be compared to Negri’s argument in Time for Revolution that systemic closure yields a certain type of enjoyment.
On the other hand, it can also signify an experience of fullness arising from actually escaping the regime of myths. In ‘The Third Meaning’, Barthes analyses Sergei Eisenstein’s films, suggesting the presence of what he terms an ‘obtuse’ meaning alongside the explicit denotative and connotative meanings.
These images simply designate an emotion or disposition, setting in motion a drift in meaning. They don’t represent anything. They are momentary, without development or variants. They have a signifier without a signified. They thus escape the euphoria of closed systems, pointing to something beyond.
Indeed, an obtuse meaning is not necessarily visible to all readers. Its appearance is subjective. It is permanently empty or depleted (it remains unclear how this positive ’empty signifier’ relates either to the ‘mana-words’ of Mythologies, or to Laclau’s rather different use of the same term). It can also serve as part of mythical schemes. For instance, ,moral indignation can function as a pleasant emotion.
The obtuse meaning is not present in the system of language, though it is present in speech. It almost sneaks into speech, on the back of language. It appears as a rare and new practice counterposed to the majority practice of signification. It seems like a luxury: expenditure without exchange. And it seems to belong, not to today’s politics, but to tomorrow’s. Barthes sees such facets as undermining the integration of characters, turning them into nubs of facets. In other words, the ‘molar self’ of the character (who, in Mythologies, is connected to social decomposition and misrepresentation) is replaced by a different kind of connection which is, perhaps, directly lived and connected to the world, rather than projecting a literary figure onto it.
It has been read in terms of a moment of emotion prior to thought. I think it might be better linked to Deleuze’s idea of the ‘time-image’: the obtuse image is a momentary image which expresses the contingency of becoming. Barthes suggests that the obtuse image is carnivalesque, and that it turns the film into a ‘permutational unfolding’, a flow of becoming in the system of signs.
Writerly Reading: S/Z
In S/Z, a text devoted primarily to the study of Balzac’s short story Sarrasine, Barthes proposes a distinction between two types of texts.
A text is ‘writerly’ if it can be written or rewritten today. A ‘writerly’ text is constructed in such a way as to encourage readers to reuse and reapply it, bringing it into new combinations with their own meanings. It is celebrated because it makes the reader a producer, not a consumer, of a text. The ‘writerly’ value restores to each person the ‘magic of the signifier’. The writerly text is inseparable from the process of writing, as an open-ended flow which has not yet been stopped by any system (such as ideology or criticism).
It is necessarily plural. This is a kind of plurality distinguished from liberalism: it does not acknowledge partial truths in different positions, but insists on difference as such. Difference constantly returns through texts, which re-open the network of language at a different point.
Barthes counterposes this view to an essentialist or Platonic view in which all texts approximate a model. For Barthes, texts instead offer entrances into the network of language. They do not offer a norm or law. Rather, it offers a particular perspective constructed of particular voices, fragments of texts, and semiotic codes. Texts have only a contingent unity which is constantly rewritten through its composition in terms of codes. A writerly text should have many networks which interact without any of them dominating the others.
The ‘readerly’, in contrast, reduces a text to something serious, without pleasure, which can only be accepted or rejected. A ‘readerly’ text is so heavily attached to a particular system of meanings as to render the reader passive. It is a reactive distortion of the ‘writerly’ through its ideological closure.
Readerly texts must, however, contain a ‘limited’ or ‘modest plural’ in order to function. This limited plurality of the text is created through its connotations. There are also writerly and readerly styles of reading texts, depending whether one seeks predetermined meanings in it, or seeks instead to inscribe it in new ways.
Instead of treating a text as a single phenomenon which represents something, Barthes proposes to examine a text through the plural signs it brings together. Instead of giving a unified image of a text, it decomposes it into component parts. Such a reading uses digressions to show that the structures of which the text is woven can be reversed and rearranged.
Barthes calls this style of reading ‘starring’ of a text. It cuts the text up into blocks of signification, breaching its smooth surface and especially its appearance of naturalness. It interrupts the flow of the text so as to release the perspectives within it. Each block is treated as a zone, in which the movement of meanings can be traced. The goal of this exercise is to hear one of the voices of the text.
Readers should reconstitute texts as plural. Among other things, this means that forgetting meanings is a necessary part of reading. It ensures that multiple readings remain possible, and therefore, that signifiers are allowed to shift or move.
One can’t reduce all stories to a single structure, because each text carries a particular difference. This kind of difference is not an irreducible quality, but the constant flow of language into new combinations. Analysing the function of each text restores it to this flow of difference.
He also calls for re-reading, as a means to avoid repetition and to remove texts from linear time (before or after) and place them in mythical time. Re-reading is ‘no longer consumption, but play’, directed against both the disposability of texts and their distanced analysis, and towards the return of difference. It helps create an experience of plural texts.
In this text, Barthes criticises many of his earlier views. He now claims that connotation is ever-present in ‘readerly’ texts (though not in some modern texts). There is no underlying denotative layer. Denotation is simply the most naturalised layer of connotation.
Further, connotation carries voice into the text, weaving a particular voice into the code. The writer, here, has more of a role than Barthes previously allowed. Writing brings in historical context through connotation.
The text as expression for the reader is also criticised. Readers are also products of prior texts, which compose subjectivity as subject-positions in narratives. Reading is itself a ‘form of work’. The content of this work is to move, to shift between different systems or flows which have no ending-point.
The work is shown to exist only by its functioning: it has no definite outcome. To read is to find meanings within the endless flow of language. We might think of it as creating particular, temporary points or territories by finding resonances within a field which is like an ocean or a desert.