Do we learn to read television and film and do televisual and filmic codes constitute a kind of language?

I am prepared to agree that we have to learn a small amount to read film and television narratives. However I am not convinced that there is enough evidence to support the contention that filmic and television codes constitute a language. Certainly not a distinctive and separate language or code.

I propose that the codes we use to interpret signs and signifiers, exists and moves between many levels in humans. To simplify these levels into main headings (that I do not propose to expand on here) they could be described as being:

These levels do not seek to break down the codes themselves into sub headings, ‘perceptual, social, textual, or interpretive’. (Chandler 2002, 149,150) Whilst these categories are definable they exist at all the levels above. In practice all the levels interact and can therefore not be precisely separated. The age ranges are arbitrary because they are based on my own observations, I will therefore accept that these stages can occur later or earlier at every level with the exception of perhaps the first two, dependant of course on culture, environment, exposure, individuality, etc.

What evidence can I offer to support a level on the table above. Babies learn to smile very quickly, within levels 1 and 2 to immediate family or intimate contact but not usually with strangers. Rather when confronted with a stranger the baby will look toward the face it knows for a sign that it should smile at the new unknown face. There is almost a quizzical look on the baby’s face as it tries to decode the new face. In western culture there are not many thirteen year olds who cannot work a VCR or a computer but there are plenty of over fifty years olds who cannot do either. Despite the fact that computers and VCR’s have been with them for longer than they have been with the thirteen year olds, the thirteen year olds have cracked the code. This may well lend credence to the trope, ‘you can’t teach an old dog new tricks’ perhaps if that were related to semiotics the T shirt slogan might read ‘all signed out’. Consequently this may mean that there is optimum time within the human life span during which it is both natural and easy to learn code and/or language, semiotic or otherwise.

One only has to look at the newest form of communication, mobile phone text messaging, to see the development of new code in the form of signifiers (symbols and abbreviations) to signified (words) which are in turn signifiers of their own. A whole new set of code has been introduced, mutually inclusively, exclusively, to a narrow band of the population. All of them (who own mobile phones) are conversant with the code, understand it and have learnt it in a remarkably short space of time and will know it forever.

How does this all relate to the subject under discussion? We have to make meaning of what happens around us, to do that we constantly refer signifiers to an inbuilt index or glossary. If we can’t cross relate directly we go the nearest match, like a computer spell checker. Before television there was literature and film but it was more likely that a child was able to read before it was a regular cinemagoer. So perhaps a set of relevant code had already been learned from literature before that child was exposed to film. The fact that writing contains literary devices similar to film codes is not surprising really, since film and television narratives are almost always derived from writing and that writing usually describes images or actions. Interestingly though neither photographs or paintings are derivatives of writing. So is the code or ‘language’ of literature transferred to film or is a new code learned afresh to understand film and television? I will argue neither.

When one considers that the first people to apply ‘code’ to film were interpreting the written word and as Eisenstein points out ‘Dicken’s nearness to the characteristics of cinema in method, style, and especially in viewpoint and exposition, is indeed amazing.’ (Eisenstein 1977, 206) Similarly quoting from the same source when discussing Griffith’s work ‘cross cutting, close shots, flashbacks and even dissolves, all film methods, have equivalent literary parallels’ (Reisz and Millar 1972, 27). It does at first seem as though literary code is being transferred to film but this is not the case.

The parallels between film and literature are frequent and commonplace. The establishing shot can be the equivalent of a location or character description, the close up is the parallel of a literary description of a grimace or facial expression or a highly focused activity and so on. It is not difficult to see those codes transposed to film almost automatically based on ‘if it looks right then it is right’. Since one of the main considerations in any form of visual narrative is ‘continuity’ (Reisz and Millar 1972, 15; Thompson 1997, 41,48) one can imagine the early exponents of editing simply trying it out to see if it looked right and if it did (subconsciously cross referenced to known codes) then leave it.

What is interesting is the way we interpret information however it arrives. It has long been stated that the language one thinks in is one’s native language no matter how fluent in another language a person may be. So to hypothesise if a native Welsh speaker who also speaks English fluently, is learning Spanish delivered by a native French speaker teaching in English, we could assume the flow chart of interpretation to be as follows:

If that formula is applied to film from someone whose native codes are derived from literature, would the flow chart of interpretation be as follows?

When we read words or a combination of words (signifiers) we think an object, in the form of an image, or a perception. This is what is signified (from our cross referenced index). So would we convert a picture to a word to think an image when the picture is already there? If we need words to describe an image, as the original author did, yes we would convert but otherwise why bother?

Since the absolute indexical nature of the signifier, being a picture of the signified, is apparent and therefore the meaning denotative, what follows must be connotative. Interestingly Roland Barthes is quoted as not using the category ‘indexical’ but ‘sees the photographic print simply as ‘iconic’’. (Wollen 1972, 124)

In images, denotation is the first order of signification: the signifier is the image itself and the signified is the idea or concept – what is it a picture of. Connotation is a second order signifier that uses the first sign, (signifier and signified), as its signifier and attaches an additional meaning, another signified, to it.(Allen 1994, 39)

Well there are supporting arguments that this is precisely what happens that there is, “….. an unending chain of chain of sign production …..” (Allen 1992, 35) and “…. we are always translating signs into other signs…” (Allen 1992, 35). However whilst this may be the case relating to other forms of signifiers, I believe that is not necessarily the case in the film or moving image other than the context in which an image is presented or the juxtaposition of other images it is associated with.

I appreciate that I am creating a minefield here based on my limited knowledge and time scale for research. However I feel that this is an important question since I am discussing either the learning of, and/or the transference of codes/language inter-textually. Because if that transference did not happen that means that a whole new code/language had to be learned to enable us to understand film. Is that an option? I would argue that there is a whole set of signifiers that are already programmed in us, nothing to do with literature or film that, nevertheless, with a little tweaking, enables us to decode film. Furthermore those signifiers have everything to do with ‘real world’ perception.

Our brains are subtly coding and decoding our environment constantly. Our ears are always scanning for sound (the ears being more sensitive than the eyes, the eyes can’t work in the dark) even when we are asleep. It is well known that people who won’t necessarily be awakened by a loud familiar sound (father’s/baby crying) will be startled to be wide awake by an unfamiliar quiet sound. Our senses also act as filters, heightening and lowering perceptions around us, compensating for our situation. ‘In a loud environment we are anaesthetised, what was really loud when we walked into a room now, a few minutes later sounds OK…’ (Slaney 2002, 78) When danger is near things seem to go into slow motion, what was good peripheral vision a moment ago is now tunnel vision focused solely and entirely on the perceived danger. These processes are all natural and ‘invisible’.

When we watch film or television we immediately compare the tele-visual world with the real world our brains know. Our brains know that there is no perspective in flat objects, we have learnt that from mirrors and reflections. So we compensate for that knowing that the camera, whilst mimicking the eye, can only capture perspective, like a mirror, as a flat image.

Similarly we know that sounds do not exclusively emanate from two sources in front of us, or from five specific points around the room, they arrive from all around both directly and indirectly. The ear is excellent at discerning both direction and distance of sound sources at certain frequencies. These codes have not been learned from literature but from the real world. When we recount an experience we don’t tell it ad verbatim. We miss bits out unless they are story specific. Another thing we do is skip time and recount flashbacks. We didn’t learn that, it is just something most of us do. So perhaps the code or language of narrative film and television is not that far removed from our basic code which would appear to apply compression1 to all the ‘real world’ information we receive and to things we recount. Metz refers to code saying:

If a code is a code, it is because it provides a unified field of commutations, i.e., a (reconstructed) ‘domain’ within which the transformations of the signifier correspond to variations in the signified, … (Metz 1974, 28-29)

The reconstruction of the domain from the compressed information is what we do. For example if I were to complain about my garage not doing the work on my car and having it ready when they promised. I would not recount the tale ad verbatim in real time from when I took the car in, until I got it back. I would say, “Phoned the garage Friday they said the car would be ready that afternoon. When I phoned on Saturday it still wasn’t ready, I finally picked it up on Tuesday.” That simple, highly compressed, explanation provides all the relevant information required to reconstruct the ‘domain’. It involves cutting between scenes, a dissolve, a disrupted temporal sequence and is being recounted as a flashback. All ‘cinematic devices’. The code is as it would be in a visual representation of the action involved in the tale I am recounting, I have coded it that way for the telling.

A film applies that compression before we receive the information, so our brains don’t need to do it. This is where film departs from literature ‘a picture is worth a thousand words’ literally, so the codes we use to decipher literature do not apply to film. Rather, when a writer, writes, he describes or transfers images he sees in his imagination on to paper, so it is at that stage that the images are coded for the end user.

In the real world authority sits above us ‘the judge sits higher than the rest of the court’ (Pease 1986, 110) so in the world of film is it surprising that a low angle shot looking up denotes inferior to and a high angle shot looking down denotes superior to. Notice I immediately classify these are denotation not connotation, meaning we don’t have to extrapolate any more than what is shown to get the message. This is a real world code, as is a close up, either intimacy or danger, depending on the eyes, again a real word code. Actors and filmmakers are acting out real world situations and so duplicating real world code in doing so.

In editing the dissolve is a real world code. We close our eyes slowly to focus on a particular experience, pleasure, a memory, a daydream, we don’t blink quickly we close our eyes slowly, this is an instinctive reaction (from level 2. in the table above). Now when we see that device in a film we know what it denotes therefore we know the connotation but we didn’t have to learn that for film. The film showed us that in order that we understand the emotion conveyed. Similarly a blink is a cut, it is instinctive to blink when looking away at something else just as a film cuts between scenes.

If we read the word ‘tree’ (signifier) we (English speakers) would know what category of object is signified, a Japanese would not. If a picture (signifier) (photograph or film) of a tree is an internationally recognisable object, that would not need to be decoded. That being the case what is it that needs to be decoded? If the picture (signifier) showed the tree (signified) was green and lush and full (signifier) we may take it (connotation perhaps) that the healthy tree, (signified and signifier) (denoted) = (denoted or a connotation) fertile ground, spring or summer, temperate climate and so on.

The images in a film whilst being both signifier and signified can also be further engaged in signifying other things by virtue of how they are presented. What film is able to do, is to make any further connotation drawn from what is being denoted by the images on screen at any moment in time, associative and related to the image(s) that directly precede and succeed them ‘tricks of montage’ (Bazin 1967, 27) ‘Kuleshov effect’ (Fairservice 2001, 181). So if the image of a smiling woman holding her stomach in a mans arms was followed by the image of a fertile lush and healthy tree laden with fruit, the connotation could be that the woman is pregnant, with no other indicators. This is the only exception, I would argue, where there is anything that needs to be learnt, to read a film.

I don’t believe that there is a particular grammar or code to be learnt in order to read narrative film or television. I believe that the code is already learnt and that the code or language is a ‘real world’ code. There maybe a case for saying that connotation has to be learned but I think that since humans have associative imaginations based on experience only the most abstract links would be problematic to adults. I don’t believe that literal codes are transferred to film because I believe that the reverse is true, writers see images then describe them. Film pictures that description. However it is interesting to note the way that the ‘code’ of cinema is being subverted by younger filmmakers who have had more exposure to tele-visual media than their predecessors. Fulfilled (according to my chart) and therefore understanding the code intimately, they are able to subvert it, and so enhance it and move it on to another level.

There is an argument that ‘cinema is so different from language that we must be wary in applying linguistic theory to it.’ (Allen 1992, 45) and that images

are not reducible to smaller units; that they are already “texts” - that is, combinations of signs – and that they are governed by a code that is weak compared with the grammar rules that govern language. Weak codes are flexible, changeable, and can produce an unforeseeable number of individual signs. (Allen 1992, 45,46 after Eco Theory, p.214)

This second argument could be attributable to ‘One of the most powerful design features of language … called double articulation’ Chandler (2002, 10) which enables us to construct an infinite number of sentences from a relatively small subset of words, in turn constructing infinite meanings.

Whilst I would agree with the first argument, (Allen 1992, 45) though not for the same reason. I would disagree with the second argument about the code being weak and agree with Langer and I would further cite Langer in supporting the first argument:

Rather than dismissing ‘non discursive’ media for their limitations, however, Langer argues that they are more complex and subtle than verbal language… She argues that we should not seek to impose linguistic models on other media since the laws that govern their articulation ‘are altogether different from the laws of syntax that govern language’.(Chandler 2002, 11)

I believe that the argument (Allen 1992, 45,46) confuses the fact that because we find the code easy to read, that in turn must mean the code is weak. I would agree with Langer that ‘non discursive’ media is extremely complex, and subtle, like the real world but that we are good at reading ‘real world’ codes and therefore, complex codes.

This complexity of the code is demonstrable in the overriding ability of film to control us at a subconscious level, in spite of the fact that we are consciously aware that it is an artefact, that it is only a film, and there is no perspective. In any of the wrap around cinema sideshows that used to be popular at funfairs, the audience, as one, will rock around in response to the motions of the roller coaster film on the screen with the camera positioned to take an audience point of view shot. The only differences between this film and one seen in a normal theatre is the use throughout of the POV status accorded to the audience, the image is larger and the screen is curved so that the whole image fills the visual receptors. The audience knows it is standing up, it knows it is in a cinema, yet it is unable to detach itself from interacting with what it sees, so strong is the code. The code is so strong because it is so deeply embedded in us, because it is a ‘real world’ code.

It could be argued that if film uses specific code, ‘There are codes which are specific to one group of language system, including cinema, …’ (Metz 1974, 225) that is it a language system. My argument is that the code, no matter how specific it may seem, at times, to be to cinema, it is not. Because we use the word cinema code, as I have done in this essay, the implication is that the code belongs exclusively to cinema but it does not it is only more noticeable there. The most obvious cinematic device, montage/cross cutting, is not purely a cinematic code it is real world code which we all employ all the time as we go about everyday tasks whilst thinking about a variety of other things and cutting between them. As previously discussed the flashback, the slow motion replay are not solely cinematic devices, neither is the cut or dissolve, or indeed the close up, or even the camera angle.

One of the canons that supports the notion of film code, ‘invisible editing’, is the eye line match where the last shot is of someone looking towards something and the next shot happens to be the object of that gaze at the correct angle to indicate it as being the object of the gaze. In reality the eyes (when the conditions are right for them to work) are the most important semiotic organ (if there is such a thing) in the human body. The ‘windows of the soul’ convey so much information without apparently doing anything. So much so, we are all interested in the object of the gaze. Stand on the edge of the pavement looking up at an imaginary something and see how many passers by will do the same. The eye line match is, in my opinion, another real world code seconded into the world of film to denote, that’s all.

To summarise, I would argue that there is not a specific, distinctive language or coding system involved in reading film and television. I would however distinguish that conclusion, to apply only to the reading of television and film in a narrative sense, not to the reading of film and television in the way things are presented. I have not touched on this anywhere in this essay because there is more at work here, modality being one of the main issues.

1. Reducing the amount of information stored whilst retaining enough pertinent information so that the compressed information when reconstructed remains a faithful representative of the original. (my definition)

References