What we want to explore this week is not only the complexities of semiotics, basically the study of signs, but also some of its applications. Though Pier Paolo Pasolini had written a fascinating essay called the 'Cinema of Poetry' in the mid-sixties, which combined semiotic ideas with the Bazinian respect for the real, semiotics was first theorized, really, by Christian Metz, whose background was in linguistics. What semiotics wanted to do was counter the realist notion of the image that had come to predominate. Where Bazin believed cinema was close to reality, Metz, Peter Wollen and other key figures on the fringes of cinematic analysis, like Roland Barthes and Umberto Eco, wanted to point up that if it was not a language as such, it was at least like a language. Metz wasn't interested in how close or how far from reality a film was, but wanted to understand its code. As James Monaco says in How to Read a Film, "Semiology attempts to describe the codes and systems of structure that operate in cultural phenomena. It does so by using a linguistic model; that is, the semiology of film describes film as a "language"".
Let us take for example Barthes' short essay in Mythologies on Julius Caesar, written long before semiotic film theory, but relevant to it. Barthes is not interested in the truth of the film, of its authenticity in one form or another, but what its sign system says. "All the faces sweat constantly...And close ups are so frequent that evidently sweating here is an attribute with a purpose. Everyone is sweating because everyone is debating something within himself." The only person who isn't sweating is Julius Caesar: "...the object of the crime, remains dry since he does not know, he does not think..." Umberto Eco, in an amusing de-coding of the Western, and the Indians' roles as oblivious baddies, offers a twenty eight point analysis of its clichs, including "if possible, appear prominently in small groups on the surrounding hills. Set up sentinels on totally isolated peaks," and "attack by circling the wagons, but never narrow your circle, so that you and your companions can be picked off one by one." Codes and clichs are often closely linked, as Barthes and Eco point up how filmmakers use short-hand to play out their stories. Here, the complexity of character motivation, or the Indians' attempt at self-definition, is deemed irrelevant next to a sign system that makes things easy on the viewer.
Now there are three sets of terms here that might be useful to offer up before we go any further. The first is the signifier and the signified, the second denotation and connotation, and the third, the syntagmatic and paradigmatic. The first differentiates the signifier from what it signifies. In language this is categorical: there is a clear gap between the word rose and an actual rose. It could be called something else again and if we all agreed that a rose was a tulip that would be fine: there is no intrinsic relationship between the signifier and what it signifies, at least in languages based on writing, no matter the odd exception in words that are onomatopoeic. But semioticians point out this isn't the same in cinema: the image of the rose and an actual rose is much more obviously linked: the gap between the signifier and what it signifies is much smaller, and where Saussure talks of the signifier as the sound-image and the signified as the concept and that they together make up the sign, cinema is closer to what Andre Bazin, anticipating semiotics in film, called the 'image fact' and the semiotically inclined Pasolini called the "written language of reality".
The second set of terms would be denotation and connotation. This is really the difference between what something is and what something implies. A rose is a flower that grows in nature, but it also has numerous connotations: when we see someone walking along the road with a red rose in his hand, we are less likely to think that it is a flower he has picked from a garden, than that he is going to meet his lover. Think back to our quotation from Barthes. If he finds all those sweaty faces so absurd, it lies in their connotative function over their denotative purposefulness. Would all these Romans really be sweating so profusely under a temperature they would be well used to; or is it the film's heavy-handed way of connoting their wracked consciences, as Barthes implies?
The third set of terms is the paradigmatic and syntagmatic, and these are useful in understanding the nature of film form. When we see a shot, however consciously or unconsciously, we are aware of how the shot could be very different. If we are offered a low-angled shot of the villain as he chases our heroine through the streets, we are aware that there are other choices available to the filmmaker. He could have used a standard close-up, a high angled shot, or a point of view of the heroine as she turns to face him. This is different from the syntagmatic, where its significance resides in the shot chosen in relation to all the other shots chosen. Loosely, this is the difference between mise-en-scene on the one hand and editing on the other. The filmmaker must choose how to shoot a particular scene (the paradigmatic), but he must also work out how it will fit into the film's overall scheme (the syntagmatic). As James Monaco says in How to Read a Film: "what choices to make: the paradigmatic...how to edit it: the syntagmatic how to film it".
The terms paradigmatic and syntagmatic are extremely useful in making sense of form, and denotation and connotation in making sense of content. When Barthes talks of Julius Caesar he is chiefly commenting on content, and to interpret a film's content it is sometimes helpful to keep in mind how films connote meaning, no matter cinema's denotative capacity that lends it so well to realism. Semioticians have borrowed terms from philosophers (like C. S. Peirce), linguists (like De Saussure) and from literary theory more generally to make sense of how film functions like a language. Here are some of them: the index, the icon, synecdoche, metonymy and the symbol. These are not always easy to differentiate and some are overlapping, but their applicability is more important than their definability.
Take for example metonymy and the way filmmakers may use it in relation to colour. Metonymy means how an attribute stands for the whole, and linguist Roman Jakobson differentiates metonymy from metaphor by saying that metaphor requires substitution; metonymy contexture. Cinema is, generally, too immediate a form to work with metaphor, lends itself well to metonymy, and filmmakers will frequently use a colour scheme that will give a broader context to the characters' feelings whilst still remaining within the plausibility of the diegesis - the story. Far From Heaven, for example, creates an active metonymy of colour as Julianne Moore's character and her husband go through various crises. If we identify with, or can't quite escape from, David Thewlis's pessimism in Naked, is it partly because director Mike Leigh insisted on a drab colour scheme where red, for example, was removed from the palette? This is metonymised identification, where the mise-en-scene represents an aspect of character without decoupling it from diegetic plausibility.
The index can be a very useful way for a filmmaker who wants to work with great subtlety, or is working on a low-budget and wants to avoid expensively filmed drama, to achieve his or her aims. We can notice an example of the indexical near the beginning of Three Colours: Blue as the car crash is presented chiefly by off-screen sound and through a young man's reaction to the event. It is also very often used in Robert Bresson's work, as in L'Argent, where Bresson holds the shot on the hand that has pushed a man over, and we listen to the off-screen sound of the man crashing into a table. In the metonymic example there is an enriching of our understanding of the film if we choose to comprehend the film's use of colour, but in our examples of the indexical the device is central to our comprehension of the film's events - in Naked and Far From Heaven the colours simply adds to the film's thematic texture.
The synecdoche is often used interchangeably with the metonymic, but can be usefully utilised in relation to the equivalent of the fetish object or the personal characteristic. It is the part that can signify the whole, and often defines character in little more than a detail. The film noir is often fascinated with synecdoche: think of the ankle bracelet Barbara Stanwyck wears in Double Indemnity, indicative of sexual possibility, or the cigarettes Rita Hayworth and Orson Welles smoke to indicate bored frustration on Hayworth's part, and gnawing worry on Welles's in Lady from Shanghai. In Eisenstein's Strike, the comfortable smoke cigars in a semiotic gesture that of course has its roots in reality, but can also become useful short-hand for filmmakers seeking to illustrate complacent power.
Perhaps certain genres lend themselves to certain signs better than others. The Western would seem to utilise the icon through the iconographic. All we need to see is an image of Monument Valley in John Ford's The Searchers and numerous other films, and we will likely assume we are in a western. The genre is full of iconic images, like the lone rider on horse back, the Stetson hat, the homestead. If we often believe the western is the most iconographic of genres, does it reside in the number of images it possesses that locates us in its world? Thus, like the synecdoche, but in a different way, the icon indicates the milieu, the environment the filmmaker places us in. Think of how many films will show the Eiffel Tower and Notre Dame to illustrate that the location is Paris, or use red buses and Big Ben to ground us immediately in London. An icon frequently sums up a city by only showing a part of it, and thus of course in many ways segues into the metonym, and seems like a broader, less intimate form of the synecdoche.
But where does the symbol fit into all this? In Apocalypse Now, the slaughtering of the water buffalo would seem to symbolise Colonel Kurtz's slaying that it is intercut with, but narratively we have been given more than enough information telling us that the American military believes Kurtz is a liability and must be sacrificed for the needs of military command: we don't really need the symbol, and it can often seem superfluous, no matter the brilliance of its use in Coppola's film. If it isn't a vital narrative element to the film, it can be a detail that brings out thematic richness, but can just as easily seem over-emphatic. Some filmmakers and critics are suspicious of symbolism. When someone asked Andrei Tarkovsky what certain images symbolised, the hapless enquirer was told not to be so stupid. Eric Rohmer insisted he was against symbolism on principle, and the great philosopher on cinema, Gilles Deleuze, says in an interview published in Two Regimes of Madness he is "especially hostile to different levels" of reading films.
If the symbol becomes lumberingly significant, does it represent an over-determination by the filmmaker? Perhaps in merely filming a man walking down the street with a rose in his hand where we assume he will be meeting his lover will be symbolism enough: it is part of our connotative relationship with the world at large. But when a filmmaker cuts to crashing waves to symbolize the lovers' passion, or cuts to heavy rain after lovers part, can't the filmmaker simply film the passion and the pain without symbolizing it as well? Yet all semiotic devices, finally, can lend themselves to clich. Do we want to see another Parisian film introduced to us with a shot of the Eiffel Tower, or a businessman presented to us with a fat cigar? To sum up, if semiotics can be useful, it is to make us aware of how well or badly a filmmaker utilises his image structure, his use of the frame and editing procedures.
© Tony McKibbin