P:O.V. No.6 - The Art of Film Editing

Editing in the depth of the surface
A few basic principles of graphic editing

Lars Bo Kimergård

The history of film editing can be roughly divided into three major steps:

From the single shot of the very first films - the only 'editing' being the timing of the action to take place in the short time of approximately one minute between the beginning and the end of the take - to the 'photoplay' where several single shots of this kind were joined together, some of them with texts in between.

Secondly, the step towards different ways of establishing time and space, the so-called continuity principle, with the actual footage being shot discontinuously - in numerous pieces, from different angles and in varying scales. This principle evolved mainly in the United States in the first decade of this century, but soon was adapted by most film producing countries.

And finally the step towards the breaking up of these conventions, with an emphasis on the graphic qualities of the picture at the expense of clear time and space configurations.

It is, however, impossible to assign an exact date, year or even decade to these various steps. They somehow exist simultaneously. For instance, you still see films made in only one shot. Alfred Hitchcock's The Rope (1948) is a feature film example (with the well known fact to be added that the film of course does contain a few hidden cuts - each reel runs only about 11 minutes), but single shot films can be seen even on the MTV channel, one of the heaviest users and developers of the new editing style. The Massive Attack music video Unfinished Sympathy, from the album Blue Line (1991), is made as one long steadycam ride that follows the band as they stroll down the street.

Correspondingly, the breaking up of continuity conventions, the last step in our three step history of film editing, is far from new. Some of these new principles date back to the European avant-garde movements of the twenties, others to the French 'Nouvelle Vague' of the sixties, and documentary aesthetics, of course, have always been less compulsory and conservative, probably because this genre doesn't feel the tight limits of a narrative structure. And even though this third step is the latest, the continuity tradition is still in the best of health. The vast majority of films are still edited according to continuity principles[1], and the growing school of new editing will probably co-exist with classical continuity well into the future, maybe even forever.

When a film is edited according to the rules of continuity, you will know exactly where everybody is and how the different persons, locations and props are situated in relation to one another, and there are no explicit time lapses - no elliptic editing. It is important to remember, however, that even though the continuity system is meant to give the impression of a coherent time and space, it is - de facto - only an illusion. Even Hollywood continuity classics such as Casablanca (Curtiz, 1942) and The Maltese Falcon (Huston, 1941) manipulate time and space quite heavily. Likewise, if you try to make a map of the labyrinth in Orson Welles' Kafka adaptation The Trial (Welles, 1963), you'll see that this is not at all possible with real three-dimensional space as your yardstick.

The important thing, however, is that time and space relations appear to be solid. Cuts are not used explicitly, unless you have a dissolve or a wipe as a filmic punctuation between scenes, but implicitly, without attracting the spectator's attention to the cut itself. The spectator thus makes a kind of functional, cognitive map of filmic space and doesn't care that editing only makes the film look right and doesn't reflect a true space. And this holds true for the experience of time in film as well.

The development of a new editing style

to the top of the page
This article will present some of the editing principles after the third step. The questioning of the continuity style started off as avant-garde, but has now entered mainstream film and television 'language', first through MTV and other youth programmes. Then it entered the world of television jingles, mainly in sport programmes, and now it is used in widely distributed TV-series such as Homicide and The Kingdom, in features like Woody Allen's Husbands and Wives, Lars von Trier's Breaking the Waves, not to mention the films made as part of the Dogma 95 project. In the documentary genre, not only directors such as Jørgen Leth and Jacob Thuesen have had great success in employing the new style[2], also TV-documentaries and docu-soaps are now using this aesthetic approach.

As a first principle, this new kind of editing is based, not on explaining the spatial relations as is the case with the continuity system, but on using different visual and auditive tricks to make the audience relate to a two-dimensional picture surface, thus not missing the explanation of the third dimension.

The American film scholars David Bordwell and Kristin Thompson[3] list four different relations between two shots joined together:

  1. Graphic Relations
  2. Rhythmic Relations
  3. Spatial Relations
  4. Temporal Relations
According to Bordwell, graphic and rhythmic relations are present in any kind of editing, whereas spatial and temporal relations are irrelevant in more abstract forms of non-narrative film.

Another way of treating the four relations would be to suggest that the graphic and spatial relations both have to do with the picture, while the rhythmic and temporal relations have to do with time.

In traditional continuity editing, spatial and temporal relations serve to tell the story, explaining where we are and what is happening at any given moment. The two other sorts of relations are often thought of as a kind of polish, making the nice and meaningful flow of shots look even better.

Conversely, in the style of new editing, spatial and temporal relations are given a lower priority. Consequently, the graphic and rhythmic relations get more important, but even though the style to some people may appear more abstract, the films are still narrative.

Changing emphasis to the two-dimensionally based, graphic editing principles, new style editing allows itself to overrule two of the most basic continuity principles: it breaks the 180 degree rule and makes jump-cuts.

But this is not easy at all, it's not just something you do. The continuity convention is so established not only as a construction principle but in our conception of a film that when you chose to ignore it, it's important that you make your own contract with the audience. When using the continuity system it's enough to 'refer to the law', but when you don't recognize this basic law, you have to make a special contract with your audience - every time.

A guide to two-dimensional editing

to the top of the page
There are two basic ways to make the audience accept violations of the continuity system. Either you build up a whole new set of rules or you distract spectators to make them overlook that you are breaking well-worn continuity rules.

The principle of eye-scanning is by far the most important feature among graphic editing principles.

It's based on the fact that the human eye is pre-cognitively attracted to whatever 'it' finds interesting at a given moment. As in normal, minute-to-minute perception, when looking at a picture or a film, the eyes move in saccadic patterns, relatively consistent from individual to individual, even considered over time. This is a pre-cultural aspect of visual perception, a bottom-up process[4], and is thus something we do involuntarily.

If you show a picture of a human being to someone, his interest is most likely to lie in something like the question "who is this?", and the eye will travel directly to the face of the person. If it's a close-up, and thus already a face, the eye will go directly for the eyes and after that for the mouth, the ears or other recognizable facialelements. All the different elements that attract the eye are called eye-catchers.

One of the most powerful eye-catchers is movement. Other important eye-catchers are contrast, bright colours or objects with a clear-cut meaning that can be used in the construction of the narrative logic or in the description of characters.

Movement, as mentioned above, is one of the most potent seducers of the eye. In the 'animal part' of our brain, we turn to see every movement, to check whether it involves some kind of danger - a predator in the jungle, a car on the road. Scanning the film frame, the same thing happens. If a person suddenly makes a fast move with the hand, our eyes are glued to the movement until it stops or another stronger movement takes over. Accordingly, there are basically two ways of using movement in editing.

A sudden, but relatively short movement can move the spectator's eyes where you want them, and you can cut to a shot with the eye-catcher at approximately the same spot.

If the movement is longer, you have to consider its speed and direction - i.e. to get the movement to flow from one shot to the other.

An object always has a certain meaning, either for the narrative or in the description of a person or a location. When a man suddenly reaches for a gun, our attention obviously follows the hand because of the movement. But if we know that our hero has a gun, and he finds himself in a dangerous situation, our eyes check out the gun even without the movement.

Imagine an untidy nursery, with toys lying scattered on the floor. Between all the toys, there is a teddy bear, which a child got for Christmas two scenes ago. This is what we are looking at.

Eye-catching based on contrast is not only a question of light versus darkness. It's obvious that our eyes are attracted by the little black dot in the snow or the flashlight in the midst of the trees in the big dark forest. But contrast can also be applied more generally. If all but one of the elements in a picture are alike, the one sticking out - in size, colour, light, texture or any other quality - will surely catch our eye.
You turn your head when you see bright red or yellow, nature's own alarm colours. They are used to signify danger. Some animals or flowers use them to warn other animals, since if they are eaten, it doesn't help the brightly coloured entity that the predator dies afterwards.

These colours catch our eyes before the message reaches our consciousness, and only at a subsequent level, our cultural background will add a conventional, coded meaning such as red for love and yellow for cowardice.

But there are also other eye-catchers in colours. For instance, you can apply the principles of contrast to colour, when a brightly coloured object appears in a pastel-shaded environment. Or more extremely, only one colour in a black and white film, such as the little red girl in Schindler's List (Spielberg, 1993) or the red smoke in Kurosawa's High and Low (1963).

All these elements should be considered relatively. In a red room with red furniture, as in Cries and Whispers (Bergman, 1972), the alarm colour is white, since red has become the general colour backdrop. In a panicking crowd, a calm person is the one who attracts our attention.

If every picture element but one is out of focus, our eyes will be caught by the part in focus. Then if the focus changes so that a different part of the picture gets in focus, our eyes will migrate almost instantly to the new center of focus. If a new defocused object enters the screen, our eyes will try to focus on it, and if the camera doesn't try to do the same and the object remains in a central position, we get annoyed. In good films this is rare. Either the object (or person) is irrelevant and just passes through the picture, or it is meant to take over focus as soon as it enters the frame.

Tolerance in time

to the top of the page
But how do you decide when to apply the above mentioned rules? The answer is quite pragmatic. To find a strong eye-catcher at a certain point in a shot is relatively easy. There is little difference between the behavior of human eyes. All you have to do is watch the screen and notice where you have your eyes - everybody else's eyes will be there too.

There is always, most editors will claim, one specific frame - and nowhere else! - to place a cut if it's to be perfect. Of course, it depends on a lot of things, for instance what the next shot is like, and certainly also the rhythm of the whole sequence as such. But one thing is the perfect frame, the perfect split second for a smooth cut, another the tolerance for an unacceptably bad one. This tolerance is much greater within traditional continuity editing than within new style editing.

When you have looked at a picture for a while, your eyes get bored and start to move about to find new places of interest. This process is not very consistent from individual to individual. The eyes of the spectator will still concentrate on eyes and facial expressions, but now this activity is not so synchronized any more. The eye-catcher is becoming weaker. This is what you use for instance in shot/reverse-shot editing and eyeline matches, but it's not enough to carry over a jump-cut. Here you have to introduce a new and strong eye-catcher to divert attention from the jump-cut. Almost all eye-catchers start off strong and grow weaker, as spectators get used to their presence.

This means that tolerance towards editing on strong eye-catchers is quite small. In a case where your eye-catcher is a movement that has to continue in the next shot, your tolerance could be down to one single frame. Whereas a simple dialogue scene edited in shot/reverse-shot can be cut almost at any point without breaking the concentration on the dialogue. That a cut in this case will be acceptable anywhere is not to say that it cannot be better or worse, and getting the right rhythm into a shot/reverse-shot scene is quite an art of its own.


to the top of the page
The other basic way to cover up that you are not using the traditional continuity convention is to distract the spectator every time you break the rules. Distraction works almost like a magic trick. The magician attracts your attention to an innocent thing, while the 'magic' is going on in his other hand. In film making, you can make the viewer think of something else, while breaking the rules of continuity. The distraction can be visual or it can be auditive.

White flash editing
One of the most used visual features is called white-flash-editing. The distraction is a short white fade-in/fade-out or sometimes just one or two white frames between the shots. It is, as most tricks in graphic editing, not a new trick. But in traditional editing it is mostly used in environments where white flashes occur naturally: lightning in a thunderstorm or the flash of a photographer's camera, for instance in the beginning of Highlander (Mulcahy, 1986), where a helicopter ride around a boxing ring at the Madison Square Garden ends with a close-up of Christopher Lambert amidst the audience. To make the transition from the helicopter to the tripod, a flash from a boxing spectator's camera beside Lambert covers up what would have been a jump-cut.

In other films, especially music videos and commercials, a longer, more dynamic fade to white (and back) is used, referring to the over-exposed frames at the end of a shot that people mainly know from Super-8 home movies.

Making a very quick pan blurs the picture so that you lose any sense of place, giving you the opportunity to cut to a totally different location than where you started. If you look attentively at a piece of film where a violent pan starts, you'll notice that there's only one frame between the clear and the blurred picture. This means that it's possible to cut from anywhere in the blur to any picture, or from any picture to the blur. Sometimes the trick is made with a short dissolve between two blurry frames.

This is not a new trick either. One of the most famous places it's used is Some Like it Hot (Wilder, 1959). Marilyn Monroe is trying to seduce Tony Curtis at a yacht and at the same time Jack Lemmon is dancing rumba with the actual owner of the yacht in a restaurant ashore. A distance of several miles is covered only by panning the camera.

But there is a difference between this swish-pan and the use of swish-pan in for instance Lars von Trier's The Kingdom. In Some Like it Hot and Hollywood-like productions the swish-pan always moves left or right according to the continuity of space. The pan from the yacht to the dance hall is in the opposite direction than the earlier wipe from shore to yacht and thus perfectly in concordance with the shore/ship relation as it's explained to us.

In The Kingdom (Part 1) there is a confusing morning-conference where the camera is swish-panning from person to person. Here, the swish-pans are used to cover up the breaking of the 180 degree rule and there is even one special cut where two pans cut together move in opposite directions, one left-to-right, the other right-to-left.

The sound-bridge
Sound is, however, one of the most frequently used distractions. Obviously in MTV productions the music plays a very conspicuous part and loud music seemingly makes almost everything look good. To pick up on David Bordwell, you could say that the rhythmic relation takes over.

But there is also another, more specific use of sound: the sound-bridge. In traditional films, the slam of a door, someone's blowing his or her nose or the shot of a gun often carries a bad cut from one shot to the next. In new style editing, these sounds are added without any connection to the story. Cartoon-like sounds such as SSSWWWHISSS or WHOOOOWHH are added, almost as a kind of auditive white flashes.


to the top of the page
In many films based on two-dimensional editing, there is a tendency towards more cross-cutting than in most films. In continuity based films, cross-cutting is mainly used to show that two actions are taking place simultaneously, but in new style editing this is far from always true. To start in the extreme, music videos often have two, three or more layers from totally different worlds. One might be a narrative structure with actors, another the musicians on a moody location, all mixed up with documentary footage from a concert and so on, with an abstract connection only through the music and lyrics.

This type of woven structure has been taken over by some of the new style documentaries. For instance, in Heart and Soul, Tómas Gislason's portrait of Danish documentarist and poet Jørgen Leth, in one scene Leth is talking about the similarities between making films and writing poetry. There are shots from three different interviews intercut with Leth reading his own poetry and pictures of the carnival in Haiti. And the interviews are shot in different qualities, and, of course, intermingled with the rest of the film, as though it was one long plait - a 'plait structure' rather than the 'pearl-on-a-string structure' that most documentaries employ.

It's difficult to say whether it's the freedom from spatial relations that gives the possibility of making this structure or the wish to make a kind of structure that makes the style necessary...

In new style fiction, however, the narrative structure is normally quite ordinary. Fictional TV-series, of course, may have a plait-like structure, but this is only on a scene-to-scene level, which is a general rule rather than an exception in soaps and series.

But "why this new style?", a lot of people might ask. Isn't continuity editing good enough? These questions sound like the ones posed to the first modernist painters. Weren't realistic paintings based on the conventions of central perspective good enough?

Breaking down conventions gives a freedom to express feelings in different ways. Carl Th. Dreyer once wrote[5] that he was tired of the fact that the grass was always green. That reality in itself isn't art, only when it's made into a style. I don't think that the confusion at the morning-conference in The Kingdom would have been the same with continuity editing, nor the insecurity of Bess in Breaking the Waves.


to the top of the page
Anderson, Joseph. The Reality of Illusion: An Ecological Approach to the Cognitive Film Theory. Southern Illinois University Press, 1996.

Arnheim, Rudolph. Art and Visual Perception. Berkeley: University of California Press, 1954.

Aumont, Jacques. The Image. London: BFI, 1997.

Bordwell, David. Film Art: An introduction. New York: Addison-Wesley, 1979.

Bordwell, David. On the history af Film style. Cambridge: Harvard University Press, 1997.

Bottomore, Stephen. Shots in the Dark in: Elsaesser, Thomas (ed.). Space, frame, narrative. London: BFI, 1990.

Carroll, Noël. "Toward a Theory of Film Editing," Millennium Film Journal no. 3, 1979.

Eisenstein, Sergei. Film Form and Film Sense. London: Faber and Faber, 1970.

Højbjerg, Lennard. Visuel stil som repræsentation; in: Sekvens 1997.

Kimergård, Lars Bo. Fresh Cuts; in: DOX #12, august 1997.

Kimergård, Lars Bo. Tiden og det magiske mellemrum, in: Dansk Film #3, November 1996.

Murch, Walter. In the Blink of an Eye. Australian Film, Television & Radio School, 1992.

Nilsen, Vladimir. The Cinema as a Graphic Art. New York: Hill and Wang, 1959.

Reisz, Karel and Gavin Millar. The Technique of Film Editing. London: Focal Press, 1982.

to the top of the page

[1 ]Definitions and terms on traditional continuity editing are used as in Bordwell and Thompson: Film Art.

[2] As I have described in the article Fresh Cuts; DOX no. 12, 1997.

[3] In Bordwell and Thompson, Film Art.

[4 ]A bottom up-process is a fast, involuntary operation, whereas a top down-process is guided by expectations and assumptions, and therefore tends to be more conscious.

[5] In "Farvefilm og farvet film", in Om Filmen (Gyldendal, 1964).

to the top of the page