Source: Hans Beller (Hrsg.): Handbuch der Filmmontage. Praxis und Prinzipien des Filmschnitts. München 1999. p.9-32.

A Kind of Introduction

Every film has to be cut; a film is not a film until it has been edited. That is why both film theory and practice have always concerned themselves with the subject. In Germany the term “film editing”, though it actually comes originally from the practical side, is more often used when analysing or discussing films than in practical film work. Those who do the job refer to it generally as film cutting. As film editing, however, can be understood as a collective term for a great many things regarding both theory and practice it will be employed here. Should the term not be sharply defined in the following pages, then the reason is that film editing stands for a complex procedure which structures the film itself. For example, film editing can mean in the abstract sense the selection, reduction and sequencing of visual and acoustic elements in a film. Eisenstein was convinced that the “main purpose of editing – as with every artistic product – lies in mediating insights: in depicting the task, the subject, the motif, the plot, the action, the dynamics within an episode as well as within a film as a connected, consistent whole.”¹

Editing Units

The smallest unit in film editing is the single frame or picture on an exposed roll of film. Cinematographic (moving pictures) film (which means literally a veil, membrane or thin layer) was made possible when Eastman put the first film roll on the market in 1888 for the Kodak photographic camera. He thereby provided the necessary basic material; for then, as now, cinematographic films (motion pictures) consist of frames that have been exposed singly and which only “move” when they are projected one after the other onto a screen. This impression of movement created by a rapid succession of static images has been investigated by cognitive psychologists and is known as the “phi phenomenon”. It is the principle on which all forms of editing are based: single pictures melt into an overall impression which is more than the sum of its parts. One of the all-time classics, a monument as it were to this phenomenon, is at the end of the famous scene on the steps in Eisenstein’s “BATTLESHIP POTEMKIN” (cf. the following illustrations).

Another phenomenon known to cognitive psychology is the image echo. An image that has reached the retina leaves its impression for a little longer than it is actually present. Eisenstein demonstrated this in “OCTOBER”, a film he made in 1927/28. The face of the soldier at the machine gun, who is shooting at the strikers, “flows into” his weapon, because Einstein has alternately joined two frames showing the man’s face next to two showing the machine gun (cf. illustrations below).

The phi phenomenon and image echoes make it possible to create an overall impression of movement in cinematography by projecting the chronologically sequenced single frames on the film roll one after the other on to a space. In his film “ÜBER DIE TRÄGHEIT DER WAHRNEHMUNG” (“On the indolence of perception”, script writers and directors: Helmuth Herbst and Klaus Feddermann, F. R. of Germany 1981) Werner Neckes developed the following scheme to demonstrate the system of the smallest possible film unit:

A₁ to A_n represent the single frames in a shot or take, represented here by E₁; B₁ to B_n are the frames in take E₂ that has been added on, and so forth. In video editing the time code is quite simply the number given to the respective single frame information which is on the video film cassette.

Hence the frames, as a group, form the next smallest unit which is called a take. A take is a recording of a single shot during which the position of the camera, the focus etc. remain the same and there are no stops and starts. Normally filming time and real time are identical. Fast and slow motion or any other special techniques are usually considered the exception. (In German the term “Einstellung”, which means both adjustment and attitude, originally referred to the former, .i.e. adjusting the camera itself, now the second meaning has also come into play – the ‘attitude’ of the camera and direction to the people or things being filmed).

Eisenstein himself argued strongly against adopting a too technical and mechanical attitude to takes. Russian uses the term “assemble” for editing, which is borrowed from industrial processes and is therefore a further temptation to see “a take” as “A little rectangle with some form or other of a bit of an event. Sticking them together is what’s known as ‘editing’. […]. That’s about what the old film school teaches. Screw for screw, brick for brick …”²

In the course of the controversy, Eisenstein opposes Kuleshov’s static view of takes expressed in the word ‚assemble’, and replaces ‚element’ with the more vigorous term: ‘cell’: “A take is not in any way an ‘element’ in the editing process. A take is a cell within it.”³

The more aggressive English term ‘shot’ came into general use in the fifties. At the beginning, a ‘scene’ was the smallest unit on a strip of film and later this was replaced by the word ‘take’.

Theorists have more difficulty in defining the next size up in film units. It consists of edited takes. There are all sorts of very different terms for this: a scene, a sequence, a syntagma or a segment. They are all used to designate the parts that go to make up an entire film.

In practice, people thought in scenes, which at the beginning consisted of one take, as we shall see below. According to theatrical tradition a scene is action that occurs in one place at one time. In filming the unity of a scene can, however, be broken down further within a sequence by fading out one take and fading in the next. When shooting a film real continuity of time or space does not have to be observed; the individual takes can be shot at different times, they do not have to be shot in chronological order and they can be interrupted at will. Later, when the takes are put together, any scene in a film must have an organic, formal unity which makes sense within the edited sequence. This is the reason why in cinematography a scene – which consists of a series of takes arranged in a certain order – is referred to as a sequence in English (in German Sequenz and in French séquence) which all come from the Latin meaning ‘following’ (i.e. one take follows the other).

In film semiotics, a specialised field of film studies, the term “sequence” has been replaced with “syntagma” (Lotman, Metz etc.). This has been greeted with criticism from other experts in the same field: “Substituting the term “sequence” by “syntagma” suggests that an imprecise, naïve term has been replaced by one that is precise and more sophisticated. This impression is false. The object itself is still a sequence and it remains problematic, no matter what name is given to it. The new term is more likely to cause further confusion, for the term “syntagma”, which is not used as yet either in linguistics or semiotics in the same way, can be employed in every conceivable syntactic construction and is therefore much more abstract and, semantically speaking, ‘woolly’ than the term “sequence” … There is no need, per se, to object to regarding a sequence as a one of a number of possible forms of syntagma; but to replace the descriptive term “sequence” with “syntagma” would be entirely misguided…”⁴

And finally let us not forget to mention the word “segment” which is occasionally used instead of sequence. Nonetheless, from now on we will be using the term “sequence” to designate a number of takes (at least two) which create one unit of thought and/or one formal unit. As can be seen from this definition, a sequence is the largest unit in film editing and one around which the work of film cutting revolves. Film makers had to learn to master the technique, as well as the logical and practical problems the concept brought with it, both on location and at the cutting bench.

The Evolution of Film Editing in Fast Forward

Similarly to natural evolution, in which simple organisms develop into more complex ones, in cinematography editing progressed, sometimes in small steps, sometimes in leaps and bounds, from rudimentary to complicated forms. Although film history only goes back about a hundred years, by taking a closer look at the beginnings we can get some idea of why today’s films look the way they do – the way we are used to. Again, as with biological evolution, claims about the origins of the cinema can only be working assumptions until another missing link is found; a strip of film in a canister which throws new light onto the question of “how it all began”. Until that happens we will content ourselves with describing the stages at which certain developments became established practice and we will not attempt to put an exact date to the very first time that a particular editing form took shape.

In 1893 Edison had his kinetoscope patented; a coin-operated, hand-wound film box. The rolls of film (produced by Eastman) went round in circles and were up to a minute in length (25-50 feet). They were shown from 1894 onward on Broadway and in all large American towns in so-called parlours or peep shows. The rooms, or halls, in which each person peered into a box, winding on the film as he did so, were known as penny arcades. In the beginnings of film-making and editing only the first and last part of the take was cut so that the strip of film could be stuck together more easily or so that it fitted better into the projector. At that time each take constituted an entire uncut scene, so that these films were so-called single shot scenes. The first screening in 1895 - carried out by the Skladanofsky brothers in Berlin, using a bioscope, and in Paris by the Lumière brothers by means of a cinématographe - showed different scenes of this kind, each consisting of one take which was projected on to a screen. The film programme was itself only one part of an entire variety or vaudeville show and therefore reached a larger paying audience than the kinetoscope could. For this reason Edison developed the vitascope, patented in 1896, a projector designed for public screenings in front of a larger audience.

The French film artist and illusionist Georges Méliès “learned” - purely by chance - that a piece of film could contain more than one take. Whilst filming in the Place de l’Opéra in Paris the film got caught in the camera and Méliès was only able to wind it on about a minute later: During that minute, of course, the passers-by, omnibuses and coaches had moved on. “When I came to project the film, which I had joined together again at the place where it had torn, I suddenly saw that the omnibus Madeleine-Bastille had turned into a hearse and the men had become women.”⁵

This was the point when a film principle first became apparent. From then on, takes showing different things happening in identical places were projected on to the screen in smooth transitions. The first of Méliès films showing two different takes in succession did not therefore come about through cutting, but through happenstance, something that was later called a “stop trick” and which Méliès used frequently afterwards. This camera trick – the oldest one there is – can make objects or people disappear, or can replace them with other things, by stopping the camera in the middle of a shot. Méliès’ early films followed in all other respects theatrical traditions. They were, in other words, simple multiple-scene films. They consisted of single shot scenes which were shown one after the other another in their original unedited form. In French and American films of this kind the transitions were often dissolves or “irises” (seen as a circle closing down over or opening up on a shot) in order to avoid “hard” cuts.

Another innovation in the field of editing and cutting was triggered off by cameramen filming real-life events which were not, by virtue of their ‘reality’, as thoroughly predictable as action on the stage. These cameramen “cut”, so to speak, with their cameras. Either the camera would be placed in a different position in order to follow the action logically, or they only began turning the camera handle again when they had found a closer or more interesting viewpoint. In this way the first films consisting of a variety of takes were made - but still without editing. This occurred between 1897 and 1900 at events such as QUEEN VICTORIA’S DIAMOND JUBILEE (GB 1897) and DEWSBURY FIREBRIGADE (GB ca. 1900). Following actors on set or in the studio in this manner, i.e. focussing on a particular point, dividing the audience’s attention between different points etc., did not occur until later⁶.

Film – which is an art taking place in time, like, for example, drama and dance – was thus developing into a new art form which could manipulate its way out of time constraints. As film time did not have to be identical with real time; since two takes edited in direct succession need not have taken the same amount of time to shoot nor have been taken at the same time, all sorts of creative opportunities were there to be grasped. It was possible to release an event from its actual duration and chronology. Cutting and editing began gradually to interrupt takes, to switch between them, put them in contrast or apposition to one another and to redistribute them.

These developments in editing technique did not however come about independently; they were closely connected to the other advances and qualitative improvements made in the aesthetic, technical and economic aspects of cinematography. Overall films progressed from being simple portrayals of movement – living pictures – to recounting entertaining stories and events because audiences were continually demanding new ways of seeing and new stories to watch. Whilst in 1900 87% of films were the kind of documentaries mentioned above, by 1907 this number had halved and in 1908 had fallen to a mere 4% of the total film output: by that time 96% were fictional films.⁷

During the so-called primitive period (1894-1908) films developed from being initially about one-minute-long single takes to film strips lasting from 4 to 6 minutes and consisting of several edited parts by the turn of the century. From 1905 onwards they became one-act ‘plays’ or one-reelers about 13 to 15 minutes in length.

It was also from that year on that films in America were given a home of their own, the Nickelodeon, as the precursors of today’s cinemas were called. From 1906 onwards making films was no longer a one-man job and cameramen, who up to then had also been largely responsible for directing as well, surrendered the task to directors who had often learnt their trade in the theatre. One of these was David Wark Griffith, a man who covered more ground in the field of editing than any other.

The “invention” of the star system occurred in 1910. Thereafter film actors in America were mentioned by name. The star system, because it allowed the audience to identify with the protagonists, resulted in camera moves that suggested to the viewer that s/he was looking through the eyes of the hero at whatever was happening and taking an active part in it. The position it adopted, or the ‘view it took’ was no longer solely that of a bystander or a member of a theatre audience, but also that of the hero. This meant that the focal length as well as the position and angle of the camera changed, so that film editing had to become accordingly more complex.

Between 1909 and 1914 the dream factories were established. Here, the film manufacture of the early days was turned into an industrial process, and films evolved from hand-made to mass-made products. Norms, standards, division of labour, hierarchy and centralisation soon prevailed in the film branch as elsewhere. After 1911 the “one-reelers” with their 15 to 20 takes were replaced by “multi-reelers” because often the stories could no longer be fitted onto only one reel and audiences had been complaining about the occasionally abrupt endings of many films. At that time film editor became a profession in its own right. When feature films lasting over an hour made their appearance as ‘a full evening’s entertainment’ in 1913, it became the norm to produce five and more reels to tell a story.

1914 was the year in which the producer, beside the director, gained an important role. Due to his influence screenplays were broken down into individual parts so as to make the production economically viable. All the indoor shooting or all the outdoor shooting that took place on one set, for example, were bundled together and filmed one after the other, irrespective of their place in the chronology of the story. Place, time and action were divided into single takes on the set by the director and cameraman on the basis of their aesthetic qualities. These fragments then had to be put together again on the cutting bench.

In order to maintain the continuity of action and film narration as well as the orientation in time and space in the face of these fragmentary single takes it became necessary to develop a method of shooting. This was introduced in 1917 and later became known as the 180 degrees principle.

The 180 Degrees Principle

If one traces the development of filming and cutting through to the present day, all sorts of different camera positions and traditional cuts can be shown to originate from this principle. The 180 degree principle - also known as the centre line scheme – works on the basis that the viewer remains on one side of the action, just as the audience does in a theatre with a proscenium arch. The centre line becomes the imaginary vector of the moves (making up part of the action), the positions of the actors and the direction in which the scene as a whole is pointed. The following diagram illustrates some of the possible set-ups.⁸

Everything within the area marked ABC serves as orientation within the action space, whereas everything in the XYZ area creates a complete change of axis and disorientates the viewer. This would not be the case, however, if the camera itself was positioned on the centre line (B₃ or C₃), which would mean that switching from the ABC to the XYZ area makes sense in terms of human vision. Cuts now have to mediate between the various takes of the whole scene that has been shot in such a way that they create a meaningful sequence. This type of sequence consisting of takes from various positions edited together is known as a multiple shot scene and should not be confused with the earlier multiple scene films which were made up of a series of scenes each of which consisted of one shot.

The camera position A₁ produces a scenic tableau type take, this is usually done with a normal lens, at about 4 ½ metres (15 feet) distance from the action with a central focus. This type of take is known as a master shot, or a cover shot if the entire area in which the action takes place is included in it, alternatively it can be a wide or long shot, depending on the camera angle and distance, and, finally, it can be an establishing shot when this kind of shot is used at the beginning of a sequence to establish where the action is about to occur. A₁ was the most common form in the “primitive period” of cinematography, the age of single-shot and multiple scene films.

The camera positioned at A₂ is closer to the action. The cut from camera A₁ to A₂ would be termed a cut-in, for example from an establishing-shot to one in the midst of the action, from a long shot to a medium shot or from a medium shot to a close up or a big close up. Of course, it can be the other way around: instead of cutting in from far off to close to, the editor can cut back from a shot taken by the camera at A₂ to one from position A_1.

This was another technique that had to be mastered. When, in 1916, that film genius Griffith was making his block-buster INTOLERANCE, even he did not shoot the cut-ins straightaway on the set but in some cases only after he had seen the cover or master shots.

The principle of shot/reverse shot (SRS), refers to the exchange between cameras B₁ and C₁. Shot reverse shot cutting is usually used for conversation scenes, the alternating shots showing each character speaking to or looking at the other. The SRS technique began to be implemented in a wider variety of ways between 1911 and 1914. As early as 1915 the cutting method (and hence perception mode) was well established enough to be parodied: in YE GODS! WHAT A CAST! the main actor plays six different parts, he looks at himself and converses with himself by means of the SRS method. Altogether there are 103 cuts which are based on this kind of continuity and ‘eye contact’. This sort of bi-polar awareness in an SRS pattern needs careful organisation. Besides taking into account the centre line, the direction in which the protagonists look also has to fit the position of the person or object s/he is looking at, this is known as eyeline match. This means that there are two kinds of axis switch, one being switching from the centre line to a point behind it, the other switching from one eyeline to another. When switching from one eyeline to another the camera positions in the ABC area are correct, but the protagonists look past each other. Cutters have often solved the problem by inserting a cut with a shot of something else or by turning the incorrect reverse shot around and copying it. Such cheat cuts are still used nowadays when solutions have to be found after the film material has been shot.

Shot reaction shot patterns also belong to the SRS principle. Here the order of the cuts might be, for example, camera B to C then back to B. This kind of to-and-fro is used to show action and reaction. A showdown at the climax of a Western, when the protagonists face each other gun in hand, is usually cut according to this pattern.

If the camera is in exactly the same position as the protagonist (is supposed to be) and it shows us the part of the action that s/he sees, as is the case with camera positions B3 and C3, this is called a point of view shot, or POV for short. These also follow the SRS pattern.

The over-the-shoulder shots B₂ and C₂ became established within the SRS scheme when the talkies arrived in 1930. Reaction shots could, from then on, show the person who was not speaking because the voice of the person over whose shoulder the viewer is looking could be heard at the same time. In the film DEAD MEN DON’T WEAR PLAID (USA 1981), directed by Carl Reiner, there are a lot of SRS sequences in which Steve Martin, as the detective, talks to long-gone stars from old movies in over-the-shoulder shots. The film “quotes” dialogue sequences from around 25 old Hollywood films which have been so cleverly edited into the material shot later that the viewer believes he is watching Martin talking to the stars directly though s/he knows they are long dead. In a similar way to the film mentioned earlier; YE GODS! WHAT A CAST!, dialogues are “simulated” by means of SRS again here.

If the reverse cut is introduced by a view which is directed outside the action radius of the 180 degree plan, as is the case with camera positions B₄ und C₄, then we’re dealing with what is known as a cut away. This does not cross the taboo axis, the centre line, because it is motivated by the direction in which someone involved in the action is looking. If, for example, one of the protagonists hears a sound during the dialogue, then he or she will look past the person they are talking to and we see in the reverse shot whatever it is that has caused the sound. As this object or person is outside the 180 degree area shown so far, the cut away opens up the space by showing something that lies beyond it.

The SRS pattern and its variations make it plain what an analytic process shooting and editing a film needs to be in order to avoid confusing the viewer. This is why editing according to the 180 degree scheme is also termed analytical editing. It is in essence the same as the order of cuts which was developed back in 1910: establishing shot, cut-in and re-establishing shot (cut-back). The 180 degree principle is still used today to ensure continuity of action within a scenic space in a multiple shot scene. Multiple space scenes demand a different sort of continuity that has to be taken into consideration when editing.

The Concept of „Continuity“

The term „continuity“, used in the sense of visually uninterrupted, coherent film narration, made its first appearance in the language of (English speaking) cinematography in 1910. When speaking of continuity the filmmakers meant both intra-sequential continuity,– in a multi-shot scene in which different things take place in the same space, for example one based on the 180 degree principle; and also trans-sequential continuity – when the action is stretched over a number of different spaces and locations. Even when the action was designed on easy-to-follow cause and effect lines, it was still necessary to learn how to direct the viewers’ attention where it needed to be to in order to be able to follow what was happening where and in what chronological order. The question was: how can the action in the take at location A be continued in the take at location B? That was the beginning of the continuity system.⁹

Continuity of movement in one direction through several sequences was achieved by simple means: if someone moved from left to right in one particular take, then that person must continue to move in the same direction in the following take. If s/he moves out of the take on the right, then after the cut in take B s/he has to enter from the left, providing there is no reason why s/he should have changed direction between the two takes.

Continuity was something that had to be observed not only so that the viewer could orientate him/herself but also for practical and economic reasons. After 1910 continuity had to be transformed into a production schedule based on the script. This schedule in turn determined the guidelines for shooting. This meant, for example, as has already been mentioned on the subject of breakdowns, that all the interior and all the external scenes were organised into an efficient and economic order from the point of view of production. The order in which they were shot did not have to tally with the order in which they appeared in the finished film. Using different cameras to shoot one scene, e.g. a scene involving a lot of action, required care in solving the change of axis and continuity problems involved. The most important thing was to tell the story as “fluidly” as possible and without constructing abrupt leaps from one place or from one time to another within the narration. It took another two or three decades however for the cuts to become really smooth and to be a hall-mark of the “classical Hollywood style”. First of all the necessary editing technology (Moviola, cutting benches) had to be developed. The continuity script came into being at the same time as cutting became a separate profession, around 1911. Script girls and continuity clerks, whose job it was to record and check continuity and connecting links, began to appear on the scene between 1917 and 1920. Around this time filmmakers were beginning to shoot scenes from different camera angles and the SRS principle became an established convention.¹⁰

To achieve continuity when cutting there has to be overlap when shooting. A classical example of a cut within a single movement: a man gets up from his armchair; medium shot (E₁) and this is followed by a close up of him in an upright position (E₂). The whole movement is recorded as a medium shot by the camera at E1 and then the movement is repeated and recorded by the camera at E2 as a close up, sometimes from another angle and with a different focal length so that the cutter has sufficient material to produce a well-rounded overall effect.

This was something else that filmmakers and viewers had to learn. Edwin S. Porter, in his film THE LIFE OF AN AMERICAN FIREMAN, made in 1903, showed part of the action twice: a fire engine is called out to save a mother and her child from a burning house. They are rescued at the very last moment and the event is shown once from the inside i.e. from the point of view of the endangered mother and once from the outside; from the firemen’s point of view. “And hence the continuity of simultaneous actions is divided into two discontinuous blocks or sequences, which are to be understood nevertheless as taking place at the same time!”¹¹ The film, interestingly enough, was re-edited in 1910 to suit the change in people’s ways of seeing: i.e. instead of showing first the inside, then the outside view, cross-cutting is used so that the film switches continually from one point of view to the other.

When it comes to a movement or a series of movements which continue through different sequences or locations it is preferable to have the action begin at the end of a sequence and to carry it over into the sequence that follows. The action can be anything from opening and closing a door, to car journeys, chase or combat scenes. The first sequence passes the baton on to the next, or to put it another way, the second sequence echoes the first, so that the joining places match one another logically, like in a game of dominoes. The viewers should be able to immerse themselves in what is happening and not suddenly be flung back on the shores of reality by abrupt cuts. The main aim in American style film narration was a harmonious whole, which they wanted to achieve by gently leading the viewer’s gaze from one cut to the next. This was completely different to the Russian cinema as exemplified by the early works of Eisenstein, who declared himself that he aimed for conflict and collisions (see Oksana Bulgakova: Montagebilder bei Sergej Eisenstein, in her book of the same name, p. 49 ff.).

Kuleshov’s Experiments

Lew Kuleshov was the first person to systematically experiment with cinematographic editing. I should like to mention his work in this field at this point because I shall be building on it in the experimental section of this volume (see Hans Beller: Montage-Experimente … and Han J. Wulff: Der Plan macht's, in this book p. 157 and 178 ff.)

A potted biography: Lew Vladimirovich Kuleshov was born in 1899 in Tambov, a town situated to the south-east of Moscow. He moved to the capital when he was ten and died there aged seventy-one. He began his working life as a painter and then went on to become a film artist and assistant to Yevgeny Bauer. In 1917 he started directing films and between 1918 and 1919 he was a cameraman in the Red Army. In 1920 Kuleshov joined and conducted workshops at the film academy in Moscow, which had been founded the year before and was the first of its kind. Among those who formed the inner circle of his classes was for example Pudovkin, whilst in the outer circle his listeners came from the theatre and literature faculties; Eisenstein was numbered among the latter. Kuleshov believed that artistic creativity in cinematography could be steered by scientific calculation.¹²

Two basic assumptions were the foundation of his experiments. Firstly, Kuleshov saw those who played a role in a film not as actors but as organic “film models” which function as “perfect technical tools” when their emotions and physical movements have been especially trained. The second basic premise was directly related to editing: “The essence of a film need not be sought within the boundaries of a filmed fragment, but in the interconnection between these fragments.”¹³ In 1928 Kuleshov even went so far as to say that “…it was not so important how the takes were filmed but how the takes were edited.”¹⁴

The editing experiments on highly flammable celluloid are no longer accessible to historical researchers. They were probably burnt during World War II. In any case, they have vanished. What has been passed down by word of mouth about the experiments has led people to imagine all kinds of things, to the extent that the master himself became confused: “Film historians and people writing their memoirs have mixed up the facts so much that it is sometimes simply impossible to make any sense of it. Sometimes I am not sure who I really was or who I am, this Lew Kuleshov…” he says in a letter written in 1966.¹⁵

In order to avoid resorting to too much speculation in the face of the difficulties involved in research into the sources on Kuleshov’s programme of experiments, the following information is taken directly from Kuleshov’s own statements.

On the „Creative Geography“ Experiment

“Khoklova and Obolensky worked in it, and this is how they were filmed: Khoklova walks along the Petrovka near the Mostorg department store; Obolensky walks along the Moscow River embankment – three versts away. They each catch sight of the other, smile and hurry to meet each other. Their meeting is filmed on Prechistensky Boulevard – in an entirely different part of Moscow. They shake hands in front of Gogol’s monument (in a fourth place) and look at something off-screen; here we cut in a shot from an American film – the White House in Washington. Next shot: they are on Prechistensky Boulevard; they decide to move on, leave the shot and climb the great staircase of the Cathedral of Christ the Saviour (a fifth place). We film them, edit, and it appears that they are climbing the steps of the White House! […]In doing so we had employed no trick, no double exposure; this effect was achieved solely by organisation of raw materials and cinematographic method […].”¹⁶

On the “Ideal Woman” Experiment

“In the first experiment we had arbitrarily created our own geography, against which a single line of action was played. In the second experiment we maintained a single background as well as a single action, but now by re-combining the people themselves. I photographed a girl sitting before her mirror, making up her eyes and eye-lashes, rouging her lips, lacing her shoes. Solely by means of montage we showed a living girl, but one who actually did not exist, because we had filmed the lips of one woman, the legs of another, the back of a third, the eyes of a fourth. We cemented these shots, fixing a certain relationship among them, and we obtained an entirely new personage, using nothing but completely real material. This example also demonstrated that the whole power of cinematographic effect is in montage. Nowhere else can you achieve with nothing more than raw materials such absolutely unexpected, and seemingly incredible things. This is impossible in any form of spectacle other than cinema, in which the achievement comes not from tricks, but only through the organization of the materials …”¹⁷

On the "Kuleshov Effect "

“This experiment went like this: I took photographs of the actor Mozhukhin from old films and combined them with a variety of other takes from different films. First I made it look as though Mozhukhin were sitting in a prison cell, then he seemed to be enjoying the sunlight, the countryside and the freedom he had regained. In another combination I had Mozhukhin sitting in the same position, with the same expression on his face, this time looking at a half naked woman. In another combination he was looking at a child’s coffin – oh, I really can’t remember – there were so many different combinations … And that is the experiment that has become known as the ‚Kuleshov Effect’.”¹⁸

“We had a difference of opinion on the degree to which the psychological state of the actor is linked to montage. Some said that montage cannot alter that. To an important film-actor who held this view we said: imagine this scene – a man starving in prison is brought a bowl of soup; he is wonderfully happy and devours the soup greedily. Another scene: the prisoner is now fed, but now he longs for freedom, for birds, for the sun, for cottages, for clouds; the door is opened; released, he now sees all he dreamed of. We asked this actor: “The face reacting to the soup and the face reacting to the sun – will these, in cinema, be the same or not?” His answer was indignant: they will certainly be different, for no one would have identical reactions to soup and freedom.

Then we filmed these two shots and – no matter how often I transposed them or how many people examined them – no one could detect any difference in the face of the actor, although his playing of the two reactions had been quite different. An actor’s play reaches the spectator just as the editor requires it to, because the spectator himself completes the connected shots and sees in it what has been suggested to him by the montage.”¹⁹

All kinds of experiments are based on Kuleshov’s exploratory work. They will be discussed in later chapters.

Editing Patterns and Attempts at Classification

Several editing patterns have already been mentioned in the section on the evolution of editing: cut in, cut back, cut away, shot/reverse shot … (cf. p.16 ff.) These are not abstract terms but ones that are taken directly from the professional language used in cutting rooms where English is spoken. Most of them refer to an easily understandable succession of one or two cuts within one sequence. Moving on from this, we come to trans-sequential patterns, such as cross-cutting, which defines a particular structural ordering of sequences.

Cross-Cutting

Cross-cutting satisfies two basic requirements in cinematography at the same time:- variation and repetition. An abstract version of the basic structure of this kind of cutting to and fro could be formulated in the following way: A₁/B₁/A₂/B₂/A₃/B₃ ... A_n/B_n. In the example taken from Eisenstein’s OCTOBER mentioned earlier (cf. p. 10) the basic pattern is followed on the level of single frames: A the soldier, B the machine gun…

Yet cross-cutting in its true sense, as it is used in a ‘last-minute rescue from death’, or in a chase scene for example, was developed and perfected in particular by D. W. Griffith. The best known early Griffith films with this editing pattern that have survived are THE LONELY VILLA (1909) and THE LONDALE OPERATOR (1911). Both of these films are about bandits staging a hold-up, women who fall into their hands, help approaching - and both have happy endings. The film cuts from location A, showing the women in trouble or the bandits, to location B, the place from which men are hurrying to the victims’ aid. The women are freed in the nick of time by their manly rescuers who reach location A and overwhelm the bandits.

This kind of chronological and simultaneous narration, however, is not the only field in which the cross-cutting pattern is effective. Parallel editing is the name given to alternate cuts between events that are related in some way, though they need not occur simultaneously or in chronological order. Technically speaking this is also a form of cross-cutting, however, on account of its individual importance, parallel editing is often not considered as a sub-species of cross-cutting but as a separate editing pattern.²⁰ In Griffith’s film A CORNER IN WHEAT (1909) he juxtaposes, by means of cross-cutting, the poverty of people in a bakery shop with its cause, namely the luxurious living of a rich wheat king. Here the contrasts and symbolism are primarily important, simultaneous or chronological action, on the other hand, are not necessarily interconnected through the narrative plot. Or, to take another example, this time from Eisenstein’s work: in STRIKE, in which he refers directly to Griffith, a workers’ strike is suppressed by the Tsar’s cavalry. These scenes are interspersed with scenes of a cow being slaughtered in an abattoir.²¹ Here, too, the scenes are not linked by chronology and simultaneity; instead the direct connection between the two storylines must be create in the minds of the viewers by means of similarities and symbols, in accordance with Eisenstein’s theory: take A and take B added together will produce supra-image C.

Francis Ford Coppola, who became interested in Eisenstein already as an eighteen-year old, frequently quoted him in his own film work later. In APOCALYPSE NOW (1976-79) he undercuts the scene of Colonel Kurtz’ (Marlon Brando) brutal murder at the hands of Captain Willard (Martin Sheen) with shots of the ritual slaughter of a water buffalo by ‘natives’. Whilst contrasts and comparisons predominate and the simultaneity is secondary and, in fact, could be left out altogether, both of these elements have been integrated into A CORNER OF WHEAT and STRIKE. In Coppola’s film the slaughter of the water buffalo is symbolically comparable to the murder of Colonel Kurtz and the two scenes take place simultaneously. They are interlinked by cross-cutting.

This editing pattern allows narratives to jump across time and space in a way only literary story-telling could. Griffith acknowledged his debt to this literary tradition: “I introduced (…) the idea (of using cross cutting between two scenes in order to increase the tension), but it wasn’t my own invention by any means. I discovered it in the works of Dickens. He had always been my favourite author and through reading his work I became convinced of how effective such methods of switching off could be.”²²

Griffith adopted cross cutting methods not only, as in his early films, to dramatise the action in takes and sequences but also as a principle of integrating films within films. In INTOLERANCE (1916) four episodes are placed one within the other. Each of them contains a complete story. The “modern” version takes place in 1914 and was conceived as a separate film entitled THE MOTHER AND THE LAW; the mediaeval version is a portrayal of the Bartholomew Massacre of 1572, the Judeo-Christian episode takes place 27 years after the birth of Christ and gives an account of Jesus and the Pharisees, and, finally, the colossal Babylonian scenes take place in the year 539 BC. These episodes are not shown en bloc but in small portions which are interspersed. This type of hyper-cross-cutting, in which entire film plots are intermingled, never became a common editing pattern. According to Bordwell, however, almost half of all the films produced in America in 1912 contained cross cutting.²³

In the nineteen-twenties this technique was employed less often in America, whereas in Russia Eisenstein was using it more and more frequently. He too indicates that literary narrative techniques stimulated his work. In his famous essay entitled “Dickens, Griffith and Us” he goes into enthusiastic detail about the “montage techniques” in Oliver Twist.

More recent examples of the omnipresent narrative style of cross cutting are to be found in JACK THE RIPPER – THE MONSTER OF LONDON (GB 1988, directed by David Wickes). Towards the end of the film as the tension leading to the solution of the murders reaches its height, five different suspects are shown driving in five different coaches towards one of the crime scenes, the shot pattern is: A₁/B₁/C₁/D₁/E₁ followed by A₂/B₂/C₂ ... . In the Canadian film THREE APPLES ON THE EDGE OF A DREAM (1989) by Jacques Leduc, the tortured protagonist, struggling through his midlife crisis, thinks back over his chaotic love-life whilst at an indoor swimming pool. The flashbacks are interspersed among shots of him as he undresses, showers, swims and gets dressed again. The cross cutting works on the principle of associating forms, noises and other sense stimuli which thereby link memories with ‚the present’.

It is noticeable that cross cutting is often used in sequences depicting sex or violence. The cutting serves methodically to increase the impact of the acts of violence or sensuality portrayed.

Ellipses

The literary precursors of film editing have already been mentioned. The term “ellipsis” originally comes from the language of rhetoric. It refers to a not quite perfect circular form and means “the omission from speech or writing of a word or words that are superfluous or able to be understood from contextual clues” (OED). In pragmatic, dramatic terms: anything that is boring or redundant can be left out. In practical editing that means that everything that is not “necessary” can be cut out entirely. Cross cutting also allows the editor to connect parts of the action, not only entire scenes. The most informative and exciting moments of a chase can be shown, for instance, by cutting to and fro between the pursuers and their quarry.

The other great Russian film and editing theorists, besides Kuleshov and Eisenstein, was Vsevolod I. Pudovkin. In his opinion the vital task of editing lay in connecting details of events and details of scenes: “Separations everywhere, gaps of all kinds, sometimes measuring only a few metres, and sometimes thousands of kilometres; sometimes lasting a few minutes and sometimes dozens of years. Separations and gaps can go very deep. What is apparently the simplest movement or action of an actor can prove to be divided up into separate parts. ( … ) The art of uniting single, separate shots in such a way that the viewer has the impression, on seeing the result, of a complete, continuous, uninterrupted movement, is what we are accustomed to call ‘editing’.”²⁴ Bob Fosse’s ALL THAT JAZZ (USA, 1979) begins by showing a typical elliptic sequence. In order to characterise the life of the central protagonist (played by Roy Scheider) as that of a workaholic Bob Fosse shows, in the exposition, how his day begins, reducing it to the most important morning activities without paying any attention to smoothness. What prevents the fragments of the takes from falling apart completely is one piece of music which runs through and connects all the disparate shots so that they form a whole. Something similar happens in Michel Deville’s “PERIL EN LA DEMEURE” (France, 1985) when at the beginning of the film the music teacher (the main part) is characterised by the way in which he eats. By concentrating solely on a few gestures enacted in preparing a meal and on the sparseness of the meal itself, the homeless loneliness of the young man is brought across clearly. Once again the disparate elements are linked together by one piece of music.

Whilst when discussing cross cutting and elliptic editing we looked at the structural aspects, it should not be forgotten that the greatest craftsmanship is required when connecting single takes within a sequence (inter-sequential cuts) and when connection one sequence to another (trans-sequential cuts). In a continuity cut the shots or frames are matched as closely as possible, without making a great to-do about it. It is part of the professionalism of the people working in the cutting room to be able to achieve cuts that are so smooth they will not be noticed by any viewer. Cuts that are glaringly obvious do not belong to the cutter’s established and generally recognised repertoire; in fact his/her art lies mainly in being as unobtrusive as possible. Nevertheless, they can get up to sly games with the established practice of matching, namely in what is known as ‘match-cutting’.

Match cuts are a special form of editing continuous movement. In most cases a figure moves continuously through several takes, or, alternatively, through several cuts, whilst the space and time leaps occurring simultaneously are often considerable. Match cuts are special cases – ones in which established practice for inter-sequential continuity cutting is observed as a trans-sequential connection; in other words, the matching refers to the action alone and not to time or space.

The match cut occurring in 2001: SPACE ODYSSEY (USA 1968) is well known. The thigh-bone thrown into the air by a primate is matched, on reaching its highest point and beginning to descend, with a space ship which has the same shape. In this case, the objects have been exchanged and the background altered from a light blue sky to dark night, yet the movement appears as one uninterrupted action. With the help of a single cut the evolutionary leap across thousands and thousands of centuries has been realised and yet, at the same time, the continuity of human evolution is demonstrated.

In the following examples the continuity of movement has been matched by the cutter; the figure remains the same. In Maja Derne’s Film AT LAND (USA 1944) the director appears in her own film, crawling in one continuous movement, divided only by a single cut, from the sea shore directly across a laden banqueting table, whilst the people around it continue their conversations undisturbed. The following example is drawn from the work of Karel Reisz, director and editing theorists: in his film “THE FRENCH LIEUTENANT’S WOMAN” (GB 1981). There are two parallel stories or narrative levels, which are often connected by means of match cuts. One level, which is, in a sense, the meta-level, depicts the shooting of the film “THE FRENCH LIEUTENANT’S WOMAN” itself. The story related in the film takes place in Victorian times. In the film the actress playing the title role (Meryl Streep) rehearses a particular scene with her opposite number (played by Jeremy Irons), script in hand. According to the screenplay the two of them meet on a cliff-edge and she falls (a ‘fallen’ woman!) and he catches her. The two actors rehearse the scene outside the conservatory of the hotel in which they are staying. The film connects the two storylines at this point: the action taking place in Victorian times is linked by means of a cut from one falling movement to another: the movement begins on the terrace during a rehearsal and flows seamlessly into the scene on the edge of the cliff. The protagonists in both scenes are the same, only the costumes have changed. Meryl Streep’s bride-like white dress that she was wearing during the rehearsal has turned into an earth-coloured dark cloak with a hood. The focal lengths and camera angles vary, too: the medium shots (figures taken from the waist up) used in the rehearsal scene turn into a medium close-up, over-the-shoulder shot (figures photographed over the shoulder of the opposite number, i.e. Jeremy Irons, down on to the falling figure of Meryl Streep). The crucial point is the continuity which has been achieved by matching one falling movement in one shot so precisely with another in a different shot, so that the movement itself appears uninterrupted. Hitchcock’s “NORTH BY NORTHWEST” (USA 1959) ends with a similar kind of repetition and double-take. Cary Grant rescues Eva-Maria Saint, - this a real cliff-hanger – holding on for dear life, below her the chasm of Mount Rushmore, and – cut – pulls her to safety, namely onto the upper bunk of a sleeping compartment in the train that is taking them on their honeymoon.

A jump cut is the opposite of a match cut – it is in fact a mismatch. The jerky hopping to and fro which can be seen in early films at the end of inter-sequentially cut takes, and which was the result of insufficient skill and experience, was later used on purpose to make a clear division between two sequences. Godard used this method, which, at the time, confused cinema-goers, in his film “A BOUT DE SOUFFLE” (F 1959) (Breathless), in an inter-sequential dialogue (cf. Joachim Paich, Wiping – Godards Videomontage, in his book p. 242 ff.)

Inserts

A whole range of possibilities are covered by this apparently simple cutting pattern. Some of these, such as cut aways and continuity shots have already been mentioned. Insert editing, as the name suggests, means that frames, shots or sequences are placed between, interrupt or undercut the main action. Newspaper headlines, letters, maps, documents, sometimes titles indicating a sub-section of the plot, all these are typical insert material. In general, however, the function of an insert is not to break the continuity, but to make it appear to flow, in other words, to ‘paste over’ gaps or holes that occurred during shooting: perhaps something was omitted or a change was made. In feature films particularly it is usual to hide these omissions, breaks and alterations from the viewer with the help of inserts. Nevertheless, inserting a close up of the politician’s hands, when he is wringing them in desperation, then cutting back to a shot of his eloquent face as he appears to continue his statement during an interview, will nowadays arouse suspicions in the mind of the viewer. Was this really one and the same statement made at one and the same time; is the interview authentic? Have parts of the conversation been removed and has the insert with the shot of the politician’s hands been used to simulate a complete, unedited flow?

The biggest shock up to now that has occurred in the world of avant-garde cinematography was caused by using an insert. Luis Buñuels UN CHIEN ANDALOU (F 1928) attacks our eyes in a very special way.

Right from the beginning of the film - the ‘once upon a time’ part - war is declared. A man (Buñuel himself) is standing by a window, a cigarette in the corner of his mouth, sharpening a razor. He looks out of the window. A cloud is moving toward the moon. A woman’s face appears; her eyes wide open. The razor cuts into the eye, dividing it in two. The cloud has passed across the moon. – ‘Once upon a time …’ –That scene marks the end of everything that intellectual, sophisticated audiences had appreciated up to then. Through that drastic attack, the eye, which had expected to see something beautiful, or something interesting, has been cut in two. From then on not vision but the central nervous system was called into action.²⁵

The sequence described above is illustrated below (figures 1 - 6) by one frame from each shot. The function of the insert (figures 2 and 5) showing the moon and clouds should thereby become clear.

The idea of correspondence and „matching continuity“ between the take of the scrap of cloud trailing across the moon and the razor blade slicing across the eye was inspired, incidentally, by a dream Buñuel himself had had: “Dali had invited me to spend a few days with him in Figueras and when I arrived I told him that shortly before I had had a dream in which a long thin strip of cloud cut across the moon and a razor blade sliced an eyeball in two.”²⁶

In this example, the shot of the moon/cloud functions as an insert and, at the same time, it is a cut back to a previous motif which was originally introduced in the form of a cut away and which, as such, was placed in a shot-reverse-shot sequence. All these editing elements add up to a sophisticated and shocking sequence with the result that an avant-garde attack on the human eye is undertaken within a harmless, established editing pattern.

This example may serve to make it clear how difficult it is to find precise terms for editing steps and patterns, to differentiate between them, or to classify them in a cut and dried manner. Finally, here is a look at the “negation” of cutting within a “plan sequence”.

Plan Sequences

A ‚plan sequence’ – a term borrowed from French (plan séquence), is a long take that extends for an entire scene or sequence. It is composed of only one shot without editing. Sometimes the camera moves, but there are no breaks in filming.

The term ‘plan sequence’ is used to describe a long take in which the movement which takes place during the uninterrupted shot as well as the dynamics are not produced or constructed by means of editing. This aesthetic principle is closely connected to the technique of deep focus filming and depends upon it to a large extent. Deep focus makes it possible for the camera to record the action (here the French term ‘mise en scène’, or English theatrical term ‘blocking’ are more appropriate) without being irritated by other objects in the fore- or background interfering with the take and appearing on screen as ‘fuzzy’ or out of focus.

The single shot scene, which was originally the basic unit of cinematography, can also be called a plan sequence, if one stretches the meaning a little. Later on, plan sequences are to be found in such films as Lubitsch’s AUSTERNPRINZESSIN (D 1919).

Jean Renoir works with plan sequences in his film LES REGLES DU JEU (F 1939) and Orson Welles likewise in CITIZEN KANE (1941), which he made at a time when the most popular filming techniques showed a preference for blurred backgrounds and soft-focus lenses. These were methods that demanded editing as they rendered a ‘mise en scène’ with spatial depth impossible. Here is André Bazin, a French film critic and historian on the subject: “The entire revolution initiated by Orson Welles is based on the systematic use of deep focus, which was not common at the time. Whilst in classical cinematography the camera focuses one at a time on different points within a scene, Orson Welles had his camera capture the entire stage on which the action was taking place, such that everything was in focus. It was no longer the cut that selected an object for us, deciding which object we had to concentrate on, and that therefore determined for us what was important; no, it was the viewer’s consciousness that he (Welles) forced to decide where the real dramatic centre of a scene is by creating a situation in which reality and representation run parallel to one another and merge.”²⁷

Plan sequences as a methodological editing principle have made appearances in all kinds of different films throughout the history of cinematography: STRANGER THAN PARADISE (USA/GDR 1984), directed by Jim Jarmusch, consists of a series of plan sequences. Each of these sequences, divided one from the other by cuts of unexposed (black) film, depicts an entire, hermetic scene. Andrei Tarkovsky also worked with plan sequences. In NOSTALGHIA (1982/83), for example, when the Russian author Andrei Gortshakov (Oleg Yankovski) attempts to fulfil the promise he has given to the madman Domenico (played by Erland Josephson): holding a lighted candle in his hand he tries to walk from one side of the Bagno Vignoni to the other. It is an empty pool, once fed by a thermal spring and dedicated to Saint Catharine. At the first two attempts the candle is blown out by a breeze, the third time Andrej succeeds in keeping the light burning until he reaches the other side of the stone pool. He is exhausted by this time, falls to the ground and dies. This plan sequence lasts seven minutes. The camera follows the protagonist’s every step, arousing empathy in the viewer who suffers, fears and, at the end, experiences a feeling of relief when the promise has been fulfilled. As the camera runs parallel to the action, so the viewers’ emotions reflect those of the actor.

Tarkovsky considered his work as being antithetical to that of his Russian predecessors: “It is equally hard to agree with the view that editing is the most important form-constructing element in a film; that a film is created so to speak at the cutting bench, as was maintained in the twenties by the followers of Eisentein’s and Kuleshov’s so-called ‘montage’ cinematography.” [ … ] “Editing is, ultimately, only an ideal variant on a number of different length shots stuck together, a variant which has already been determined a priori by the material which has been captured on the strip of film. Editing a film properly means not disturbing the organic connections between individual scenes and shots which have, as it were, already pre-edited themselves because there is a natural law alive within them according to which they are to be joined together. This is what needs to be intuited when cutting a film and sticking it together.”²⁸

We have gone full circle then, when we return to Eisenstein’s image of the take or shot not as a unit but as a single cell. In this respect Tarkovsky is closer to Eisenstein than he thinks.

This brief look at the evolution and development of film editing highlights the following points: editing patterns form an open system within the art of cinematography that is continually evolving and progressing. That it is possible nowadays for editing conventions and patterns of all kinds to exist side by side is proof of their ability to create great cinematographic diversity. Furthermore, knowledge, awareness and analysis of these conventions, patterns and principles empower the cutter to imitate or reject them and hence to edit films freely, according to his/her own style and intentions.

This work, and any part of it, is copyright. Putting any part of this work to any unauthorised use is a punishable offence and liable to prosecution. This applies in particular to reproduction, translation, copying, micro-filming, electronic storage or any other electronic re-working.

6Cf. Stephen Bottomore: Shots in the Dark. The Real Origins of Film Editing. Sight and Sound – International Film Quarterly, British Film Institute London, Summer 1988, Volume 57, No. 3.

8Cf. also on this: David Bordwell, Janet Staiger and Kristin Thompson: The Classical Hollywood Cinema. Film Style & Mode of Production to 1960. London 1988 (1985), p.56.

12cf. Vance Kepley, Jr.: The Kuleshov Workshop. In: Iris, A Journal of Theory on Image and Sound. 'The Kuleshov Effect.' Vol. 4, Nr. 1, Sommaire 1986, see p. 6 ff.

15Vcf. Steven P. Hill: Kuleshov Prophet without honour? In: Filmculture Nr. 44, Spring 67. The letter was addressed to Hill. Hill published the letter in Filmculture together with a „Biographical Interview“ which he had conducted with Kuleshov on 25^th, 26^th and 29^th July 1965 in Moscow.

18Quoted from Steven P. Hill’s interview, see above, p. 8. “Literary legends” (Dadek) have already accumulated around this experiment. Kuleshov’s own statements on the subject are contradictory. In the interview with Hill he puts the experiment as having occurred in 1917/18. But Hill comments on this date as follows: “… this could be too early, because Kuleshov dates it (the experiment) as taking place in the spring of 1920“. Leyda, (see above, p. 165) on the other hand, states that Kuleshov says he carried out the experiment in January 1923. This is contradicted by the fact that Kuleshov comments on the experiment in an article he wrote for the magazine Kino-Photo, issue no.3, which appeared in 1922. This confusing information may perhaps be due to the switch from the Gregorian to the Julian calendar after the Russian Revolution. This makes it difficult for those who lived in Russia between 1918 and 1923 to say retrospectively in which year something occurred.

21cf. also Carl-Dietmar Möller: Auszüge aus einer Geschichte der Parallelmontage. In: Aspekte einer wirkungsbezogenen Filmdramaturgie. Die Oberhausener Filmgespräche 1980-1982. editors: Thomas Kuchenbuch et al., Oberhausen 1982, pp. 90-103.

24Vsevolod I. Pudovkin: Über die Montage. (Beginning of the 1940s) In: Texte zur Theorie des Films. Edited by Franz-Josef Albersmeier. Stuttgart 1979, p. 77ff.