R. Barthes, ‘The Death of the Author’ (London, 1977), pp. 142-148
Summary of thoughts on “Death of the Author”
- The author’s intended meaning is irrelevant
- The only meaning resides in the reader
- His argument is related to various works of linguistic philosophy, e.g. Wittgenstein’s Philosophical Investigation, Blue and Brown Books and W.V.O. Quine’s Word and Object.
- How could you relate the argument of the text to an artwork?
- See notes on Manet’s Olympia below.
- Are there problems with his approach?
- If we all bring our own meaning there is no communication from author to reader or reader to reader. Culture (in the broadest sense) creates an area of common understanding (otherwise we could not communicate) between reader and reader. Barthes does emphasis the problem of trying to see a work as it was seen at the time (or the impossibility?).
The book Semiotics: The Basics by Daniel Chandler (a lecturer at the University of Wales) is very useful. Also, consider the example from W.V.O. Quine’s Word and Object – you are out walking on an island with a native who does not speak English, a white rabbit jumps out and he points and shouts “gavagai”, what does the word mean? It could mean white, or food, or four legged animal, or movement, or animal suitable for shooting, or even “spirit of my dead uncle”, the point is the list is endless and we can only get nearer the meaning by experiment and questions, in a sense we never finally know, even as native speakers, we hone the possibilities down. So the meaning of all language is created by every person individually, there is no absolute connection to “reality”. Meaning is decoded in the mind of the receiver (“the reader”), which is not to deny it is first coded by the author.
To some extent this is stating the obvious as we typically read a text or look at a picture without any consideration of the author or the intention of the author or what the author “really” meant. For example, newspapers, novels, advertisements and so on the text is transparent and the author designed to disappear or “die”. The author becomes more relevant when we want to look at the text more critically particularly when we are considering what some call a “work of art”. That is because we assume a work of art has hidden meanings beyond the surface appearance. This point of view has been criticized by some art historians as the “puzzle theory of art”. This says we privilege works that purport to present a puzzle with clues that require our intellectual engagement.
The approach suggested by Barthes is also related to the area of semiotics that started with Saussure and Peirce. One technique is the deconstruction of texts as advocated by the poststructuralists such as Derrida. This is typically applied to written texts rather than visual works although many art historians and semioticians have applied it to visual works such as advertisements, film and painting. Some of the concepts, such as marked and unmarked term, are harder to apply to works of art but the attempt is often interesting.
I think Barthes is saying the cult of the author is misleading us into paying to much attention to the intention, or even the lifestyle of the author. We are all floating in a culture full of words and images and the author pulls together ideas and phrases from this cultural environment. Some authors do pull together ideas and phrases that millions enjoy or feel are important to them. However, the interpretation is always made by the reader and to some extent this is stating the obvious. Semiotics is interesting as a technique for looking at texts and images in a novel way that can highlight implicit assumptions in our culture that we normally overlook.
Interestingly, the idea of pulling ideas from the culture ties in with Dawkins ideas of memes, thoughts or concepts that compete in a culture with the fittest surviving.
Mark Tansey (1949-), Mont Sainte-Victoire, 1987, Oil on canvas, 100 x 155 in., Zurich
Tansey Mont Saint-Victorie (inverted)
The Tansey painting presents men reflected in a pool as women but it is in danger of over labouring its point. Tansey’s work presents the reflections as the opposite; referring to the idea of the marked term (e.g. female), generally regarded as negative and the unmarked term (e.g. male), generally regarded as positive. However, any work that makes its point in such a laboured, heavy-handed way is in danger of being trivial.
When we deconstruct a piece we draw attention to the implicit assumptions and the creation of an expectation through the marked and unmarked terms used (of course this means a purposeful activity, perhaps unconscious on the part of the author). A simplistic interpretation of this is “all works are plagiarized, all works are propaganda”. What do I mean by this? All works are plagiarized in so far as they cannot stray too far from conventional comment or representation otherwise they will not be understood. All works are propaganda in that they are conveying a message that is promoting a cause or at least a known point of view.
The obvious comparison is with Vasari as in some ways, for art, it was Vasari that started the cult of the artist genius. Barthes is reacting against this model by saying that the artist has no importance. What is important is the viewer’s understanding of the text. This may be overstating the point but it does serve to redress the balance.
Panofsky (page 60) compares the interpretation of two pictures, one with a floating baby and the other a floating city. In one the author (van der Weyden) intended the baby Jesus to be a vision in the other the artistic convention c.1000 meant that the city was just shown in a flat two dimensional space and was not intended to be floating. Meaning depends on our understanding of the context and what the author intended given the culture conventions of the time. Without knowing these the reader could interpret both as visions.
Barthes would argue that the culture of the period led to the signs being constructed in that way and the reader will understand the pictures differently depending on their knowledge. This implies the more we know of the conventions of the period the more we interpret the pictures in the way they would have been interpreted at the time and in the way the author intended.
Of course, the artist of c.1000 could not choose to have represented meaning in the same way as van der Weyden and in this sense there is change, which could be described as progress depending on the cultural assumptions of the person using the word “progress”. The culture directs the author to use certain conventions and the author’s ability to choose is limited to choosing between narrow variations in representation in order to achieve certain given plan objectives (see Baxandall’s Patterns of Intention).
(Read David Chandlers Introduction to Semiotics on the University of Wales website at and as a book.)
The work of Levi-Strauss who saw the myths of a culture as variations on a limited number of basic themes suggested that there may be a “myth of pictures”, i.e. a set of simple common themes that underlie many pictures? Do pictures fall into certain story telling types or patterns? There is “the nude” and “the landscape” but are these the right category? Take Venus of Urbino (see below), is it a nude? Or a representation of marital fidelity? Or is the picture part of a male dominated society representing in various ways the of power of man over woman? In this case is this one type of picture and are all nudes of this type?
Giorgione (finished by Titian?): Fete Champetre, c. 1510, oil on canvas, 105 x 136 cm, Louvre, Paris
Is Giorgione/Titian Fete Champetre a different type as the women seem to be above and beyond and more significant than the men, but is this interpretation undermined by the phallic flute? Are the men actually the solid grounded centre of the meaning and the women peripheral and subservient?
Take a seminal work such as Manet’s Olympia. Applying a deconstructive approach we must look for key concepts and their unstated relation to absent signifiers and privileged terms. We must also decide if the concept of marked terms has any sense for visual images. An unmarked term is the basic form and is regarded as more positive and the marked is the same term changed in some way to signify the opposite (e.g. male – unmarked, female – marked). But does this mean that a picture of a woman is marked and a man unmarked? This seems doubtful. It is possible that certain artistic conventions (such as the nude) signify a marked term but one would have to explain male nudes.
It could perhaps be argued that a naturalistic (photographic) nude is unmarked and the marked term is then signified by removing blemishes and body hair, a perfect marble skin, ideal features and so on. These conventions would then constitute the mark and it would need to be argues that the unmarked term, the naturalistic nude, was privileged. It might be found that in the nineteenth century more male nudes were unmarked and more female nudes marked in this way, thus demonstrating a gender bias.
Edouard Manet, Olympia, 1863, oil on canvas, 130.5 x 190 cm (51 3/8 x 74 3/4 in), Musee d’Orsay, Paris
We must also consider a paradigmatic analysis of a visual image. One interpretation of this is to look for objects that correspond to textual terms, such as female, flowers, cat, maid, bed and so on. Paradigmatically we need to consider other images that could be substituted in each position in the overall image. For example, a naked man on the bed, a dog rather than a cat, a man rather than a maid. Each substitution highlights aspects of the decisions made and aspects of our interpretation of the image as presented.
Absent signifiers could be implicate terms and visually could be the similarity to other works, in this case female nudes lying on a bed, such as Titian’s Venus of Urbino and various Danae works.
Titian, Venus of Urbino, 1538, oil on canvas, 119 x 165 cm, Uffizi, Florence
Titian, Danae, 1545, oil on canvas, Museo del Prado, Madrid
With this interpretation we are translating the two dimensional image into a type of two dimensional picture of textual terms. There are dangers in this approach as the image is worth “a thousand words”. In other words we must substitute a long description for each aspect rather than a single word. However, substituting a word or phrase does create a form of analysis that can be pursued with interest, Manet did paint a female nude even though it is of a particular type. In fact, the particular features, such as solid black outline, flat tone and so on could themselves be substituted for the visual image to create a two dimensional word map that could then be deconstructed. Of course, the translation of the image into a word map is itself an act of authorship, is it a servant, a maid, a black women, an African, a friend, a Madame, a pimp, a lesbian client. Each interpretation throws up various objections and thoughts that reflect on our cultural assumptions and the act of interpretation from visual image to verbal description. A verbal text avoids one level of interpretation as the words are as stated, however, any text about the given text involves words that themselves are subject to authorship, choice and further interpretation.
The power of Barthes idea is that it allows us the freedom to pursue an interpretation that the author may not have intended and it allows us to ignore any appeal to the authors claims for the work. For example, is Olympia really about the exploitation of black people, and the way they were forced into peripheral and supportive roles to the main subject. The key person under this interpretation is the black woman. This is supported by the use of a black cat and the flat, pale and sickly appearance of the white woman.
Or it may be a positive comment on black women. The black woman holds the flowers, maybe they are hers? The black woman is shown as active, as just entering the room, as engaging in a process. The white woman is passive, resting, unengaged. The black woman is about to speak, to take action, to change the situation, the power implicit in the players is focused on the black woman and supported by the form. Initially she appears submissive as she is behind the bed and so partly hidden but this also gives her strength, she is able to move forward topple the bed to take control to exert her power, to become the central player.
Manet’s reference to the Venus of Urbino in the composition of the painting supports this interpretation. The lying figure is more plastic, less central. The servants in the Venus turn away, they are not players. Manet has fixed the woman on the bed as a stylized, forever immobile cardboard cut-out of a classic nude. The central figure is the black woman who replaces the passive servants and becomes an active participant, full of energy and on the point of taking the main role. She doesn’t look at us as we are irrelevant, she looks at the woman she will replace. She looks diffident as she considers the magnitude of the cultural change about to take place. The woman on the bed starts to look worried, maybe she suspects the intentions of the woman who has just entered.
Seeing A Work As It Was Seen At The Time
This seems impossible. We can learn a common language (e.g. of Gothic) as it is understood and debated today but this is not how it was seen then. Consider the difficulty of interpreting a recent period, such as the 1960s. I was a teenager in the 60s but find it impossible to communicate what was understood at the time by widely used terms, such as “love” and “peace”. Even today we have the difficulty of communicating between different groups, for example, the middle aged and the young. These represent fundamental problems that can be mitigated by communication when both groups are alive but this becomes impossible as time passes.
“Common humanity” is frequently called upon to imply a core understanding between humans. After all we are all human. But how far does this get us. In other words what is nature, what nurture? At the level of culture and language this may be a meaningless question.
A fundamental question is whether semiotics can be applied to pictures in a meaningful way. Semiotics is the study of signs and these include icons and symbols. However, it is more questionable about whether the four tropes of metaphor, metonymy, synecdoche and irony can be applied to a picture. It raises the question of whether a picture can lie. It can certainly mislead, a road sign pointing the wrong way could be said to lie, a black person being treated well in former apartheid South Africa, a well known store advertising free goods (although this would need to use text) and so on could all be said to be pictorial representations of lies. This would imply a picture could be ironic, for example, some political cartoons, but the range of possibilities seems very limited and even more limited for the other tropes. Another possibility is that explored by the surrealists, such as ceci n’est pas une pipe, but again this uses text, or a train coming out of a fireplace, although this seems like a dream sequence rather than a lie.
Leaving aside the fact that some words have multiple meanings, the simplistic notion of a signifier and a signified having a one-to-one relationship does not exist for even the simplest word. We must also avoid considering the signified as an “ideal”. For example, the signifier “game” is not associated with a single concept or the ideal game. We use words in different ways, they have overlapping multiple, even conflicting meanings. Even a word such as “chair” we might imagine as referring to some idealised four-legged, high backed dining room chair but a little thought shows this is too simplistic. A word is a tool we use to communicate certain ideas that depend on the context, there is no ideal chair just a complex word play. Consider, “She was asked to be the chair.” (although it could be argued this is a different sense of the word) or a balans (kneeling) chair, is a throne a chair or a toilet?
We also employ tropes such as metaphor and metonymy all the time. Language is a tool we use for multiple purposes, convince, impress, convey emotion and so on. Language is an activity humans engage in not a formal system for proving propositions.
Art is a human activity and human beings have a common understanding of the world and common feelings and interests otherwise we could never use language to communicate in the first place. It is not true that all things are possible and all interpretations possible otherwise there would be no common ground a child would have no basis on which they could build their language skills. When someone points we look where they point not at their finger.
This means that any approach that is purely focused on language will fail to include the humanity that is essential to understanding any work of art.
A series of lectures on Mythologies , Tony McNeill, The University of Sunderland, Last Update 15-Apr-96
Semiotic analysis of images , Ross Woodrow, The University of Newcastle, created Dec. 1996, last modified & Links verified and updated – 21 Feb. 2004