* STARWORD *         for the best in Film History & Analysis 




During the years of the First World War there was not a great deal of development in film technology, but stylistic  development continued rapidly in the United States of America. It is  often said that the way the war cut  off the European film producers from many of their  markets was responsible for the post-war dominance of the American film  industry, but in fact the American industry was moving into a commanding position even before the  war started at the end of 1914. This can be seen  from the figures for the numbers of films shown in Germany in 1912, and  those put on sale in France from 1911 to 1914, as quoted in Georges  Sadoul's Histoire Générale du Cinéma (Tome III, `Le Cinéma  devient un Art', Premier Volume, p.10). In 1912 as many American films  as  French films were shown in Germany, and in Berlin in particular far more, while in France the French industry's share of the home market fell steadily from 1911 to 1914, with the share taken by  American films rising steadily  to take the leading position in 1914. Eye-witness  accounts of the American takeover of the market in Paris can be read in  Richard Abel's French Cinema: The First Wave 1915-1929 (Princeton, 1984). Italian films held  third place in both the French and German markets  throughout, followed by Danish films, and then the rest. The rapid  expansion of the American industry pre-war must have been aided by the  size of its home market, but  when one compares a sample of American and French  films made in 1913, one can see that what the European cinema-goers were already voting for in the fairly free competition for their money was:- more shots per reel,  more shots in each scene, more close shots, and more naturalistic acting. In other words, a semblance of the more  interesting parts of reality improved and accelerated by leaving out the dull bits, and serving the exciting bits  right up to the audience. The years 1914 to 1919 in  the American cinema were concerned with the further development of these formal features, and also with the appearance of the newer features of  `continuity cinema'.

The most obvious characteristic of the period, which  was the establishment of films several reels long as the major part of  production, had little influence on most aspects of the formal  developments taking place, though it  did contribute to the increasing profitability and  expansion of the American film industry. The many new directors entering the profession, mostly drawn from the ranks of the actors, were  important in establishing the new  developments, since they were not hidebound by the  earlier formal practices, and in fact the decade from 1914 onwards was  the period when film directors had their greatest power in the American  cinema. This is indicated  by the slogan of the Triangle Distributing  Corporation: 'The greatest pictures by the greatest moviemakers', and by the fact that many directors who had made a name for themselves were  able to set up personal production  companies towards the end of the war.

Film Stock

In 1916 the standard Eastman Kodak camera  negative was improved to give what became known inside the company as  Cine Negative Film Type E. A year later this was replaced by Type F, but this had no major visual  effect, for it had the same speed as the previous  Kodak negatives, and like them had an orthochromatic emulsion. It seems  likely that there was some improvement in its granularity and  definition, however. Some years later this  standard Kodak movie negative came to be called  Negative Film Par Speed (Type 1201). There were no other developments in this area during the years 1914-1919.


The studio of the American Film Manufacturing Company in 1916.  Sunlight diffused by thin cotton sheets suspended just above the sets  still forms a large  part of the  lighting, but the figure modelling is sharpened by arc floodlights on  floor stands shining in from the left, with a reflector being held to  the right of the camera to bounce  sunlight as a fill onto the actors. A row of arc floodlights (‘scoops’) suspended on  on a beam across the top of the walls of the set are not switched on.

Lighting Equipment

As is well-known, these years saw the  introduction of spotlights for the lighting of studio interiors in  America, but  the details of the process are not simple. The  lighting units themselves were standard theatrical-type spotlights,  with the carbons producing the arc contained in an  oblong box of black sheet steel, and the light from the source  being concentrated into a beam by a large glass lens several inches in diameter set in the front of the casing. By later  standards they were rather inefficient, since only a small fraction of  the light from the arc made its way  through the lens, and most was scattered around  inside the walls of the housing. These spotlights could be focussed by  moving the arc inside the housing with respect to the lens, and from  1915 they were used in a range  of sizes, from those drawing 60 amperes of  electricity to those drawing 120 amperes. The principal American  manufacturer of such lights for theatrical purposes was Kliegl Brothers, and they, and later others, supplied them  for film purposes as well.

Theatrical spotlights had been used as props within  the scene in a number of films with a backstage story from at least as  early as 1911 (A Stage Romance), but in these cases they were just standing round in the background  unlit. The earliest possible instance of an arc spot effect being used as part of the lighting scheme is in At the Foot of the Stairs, where the principal scene is lit in low-key for suspense purposes. This Universal film was  released in July 1914, well before Wyckoff and  DeMille had done anything with arc spots. But it is very difficult  to make out exactly which type of arc light is doing the lighting unless the instance is in a very low-key situation, so I  may have missed something earlier amongst the films I have seen.


A scene in the short thriller At the Foot of the Stairs (Universal, 1914) lit only by the beam of an arc light.

The use of arc spotlights was very limited at first, and most American  films continued to use no backlighting on  interior scenes for the next few years. In this  matter, as in others to do with film lighting, there was something of a  split between films shot in the East Coast studios  and those shot in California. In the years 1914-1916 films from  the East, though more completely lit by artificial  light than the Californian ones, stayed with either frontal light, or  side, or three-quarters back light done with arc  floodlights in the way that had begun to develop before 1914, while the  films from the West Coast had more of a tendency to use full backlight  on interiors. This backlighting of  the actors was still sometimes done with sunlight in 1915, as in the illustration fromBetween Men, which was  made in the Thomas Ince studios in that year. This  kind of backlighting was done by constructing the set so that the sun  was behind the actors, with its light diffused by the usual  overhead cotton screens, rather than in front of  the actors, as had previously been the case. The  frontal light came from Cooper-Hewitts and arc floodlights as usual.

A studio interior shot in Between Men (Reginald Barker, 1915), with the actors backlit by the slightly diffused light  from the sun behind  them, and with the general diffuse daylight from the front supplemented  by an arc floodlight just to the left of the camera.


On New York films at this date three-quarter backlight from arc floods  on floor stands was sometimes used on close shots, and rather more often than in the previous period. Indeed by 1916 the usage was just starting to  spread to some European films, as in some close  shots in Signorina ciclone, as was the idea of using diffuse sunlight through the studio roof to give a weak backlight.


An interior scene in Their One Love (Thanhouser, 1915), lit by general diffuse light, plus light from an arc spotlight of  the theatrical type hitting the back of the head of the more distant  actress.

By 1917 backlighting with an arc spotlight from overhead was appearing in more American features such as Forbidden Paths, and for the first time cameramen tried using two slightly separated arc spotlights from high  behind so that they hit the side of the head from  glancing angles on either side, producing a bright rim round the whole  of the upper side of the figure. This began to appear in a limited  number of close shots in some Famous  Players-Lasky films in 1917, for example Rebecca of Sunnybrook Farm and A Mormon Maid, lit by Walter Stradling and Charles Rosher respectively, and it seems  the idea spread slowly from there. Walter Stradling took  it with him to the other Mary Pickford films he lit  afterwards, such as Stella Maris (1918), and Rosher also began to use the idea consistently. By that year a single backlight was being used in some, but not all, interior  scenes in other quality American features; e.g. A Modern Musketeer (Allan Dwan, 1918) and The Gun Woman (Frank Borzage, 1918). Walter Stradling had also begun to use arc spotlights for key and fill lighting on  the front of figures as early as 1915, in Mr. Grex of Monte Carlo (1915), and frequently followed the practice  thereafter. However, most cameramen continued to use floodlights for the key and fill in figure lighting, but there were  exceptions in this, as can be seen in The Ghost of Rosie Taylor (1918), which is also lit using spots for  much of the key and fill on the figures.

A Close Up from Daddy-Long-Legs (1919) lit by Charles Rosher  with two backlight spots coming in from an angle a  little to each side of the head, plus key and fill lights of almost  equal brightness to the left and right of the camera at the front. There is no softening of the image by  the use of a diffusing filter in front of the lens.


In 1918 another major technological development in lighting equipment  occurred with the introduction of a new type of arc spotlight, which was based on the military searchlight. This formed the spot beam with a  large  parabolic mirror a couple of feet in diameter, behind the arc, and it had no lens in front of the arc. This type of  spotlight was much more efficient, and could throw a fairly broad beam over a large area from a distance, although this beam was less precisely controllable than that from the earlier type of  spotlight. Over the next few  years such reflector spotlights, referred to as  `Sunlight arcs', came to be principally used for lighting large-scale  night exteriors, but it would seem that in some of the daylight exterior scenes of Daddy Long-Legs (1919), the  cameraman Charles Rosher used them for long-range  fill light on the figures on daylight exteriors, rather than  using the usual reflected sunlight. He also tried  using the `Sunlight arc' as a floodlight on at least one interior scene  in the film.


A large arc spotlight of the type based on the military searchlight  introduced in 1918, and referred to as a ‘Sunlight’ arc.  The beam is focussed by a parabolic mirror behinf the arc. The carbons  between which the arc is struck are just visible in the centre of the  housing behind a protective screen  stretched across the front opening.

An exterior scene in Daddy-Long-Legs lit by high sunlight from the left back with fill light from big arc  reflector spotlight out of  shot at the right front. Note also the soft-edged vignette mask around  the edge of the shot.


Yet another important technical development in lighting during these  years was the introduction of diffusing  screens which were fixed in front of arc  floodlights. These screens, which were made of ripple glass or spun  glass sheets, were fixed to the front of the arc housing, completely  enclosing the carbon arc inside it, rather than  letting the arc shine unimpeded through the square  opening as before. These diffusing screens completely changed the  quality of the light coming from arc floodlights, making it more  diffuse, so that it now cast softer -edged shadows instead of the hard-edged shadows  that had been so characteristic of arc floodlighting. This was  particularly significant in figure lighting, as the shadows cast by the  protuberances of the human face upon  itself were also softened, and if the key floodlight was fairly close to the actor's face, the result was an approach to the sort of `soft lighting' coming from a north-facing window that had long been considered attractive in  portrait photography. (It must be made clear that  this sort of softening of arc floodlights was only tending in the direction of modern `soft lighting', as the  diffusing screen over the arc opening was only about 15 inches across,  which is far smaller than true softlight sources.)  The origin of this use of diffusion on arc floodlights, which came in  fairly suddenly in the work of the better cameramen about 1916, is not  clear. Fanchon the Cricket (1915) is  the earliest film that I have noticed with the light from arc floodlights softened. Since arcs had been used for filming for over a decade in many studios, and since the use of diffused arc light  in still photography goes back  years before that to the beginning of the century,  it is a little difficult to see its sudden appearance as late as 1917  as due to the influence of still photography, though that is not completely impossible. Another possibility is that it  arose accidentally as a side effect of the attempt  to cure the `Klieg eye' condition that began to afflict film actors  a couple of years before, when the use of arc  floodlights first started to become the main source of light on the  East coast, rather than just a supplementary  addition. `Klieg eyes' was an inflammation of the eyes resulting from  their irritation by the fine dust in the studio atmosphere coming from  the burnt carbons of the arcs, perhaps  supplemented by the large amount of ultra-violet  light given off by the arc flame. (Arc lights were generically  referred to as Klieg lights at this time, after the  Kliegl company, the principal American manufacturer of arc lights for theatrical use. In Germany arc floodlights were called `Jupiter lights' for similar reasons.) Certainly the cause and  cure of `Klieg eyes' was a subject of considerable discussion in  1916-1917, and even if this was not the  reason for putting glass diffusion screens on the  front of arc lights, their enclosure in this way certainly stopped  the arc dust getting out into the atmosphere, and  also absorbed the ultra-violet radiation, as ordinary glass is opaque to it. On the other hand, the fact that some 1917-1918 films use arc  floodlights both with and without  diffusion in various scenes might suggest that the  use of diffused arcs was a purely aesthetic decision, as a response to  the dropping of the general frontal diffuse light which had previously  concealed to some extent the  harshness of the light from open arcs. This point  can probably be decided by further research.

Although by this period most of the studios  were using electricity from their own direct current (D.C.)  generators to power their arc lighting, one can  occasionally see one of the cheaper films that has scenes lit by  alternating current (A.C.) arcs. The visible result  is a periodic fluctuation of the light level several times a second  due to the stroboscopic effect between the frequency of the A.C. fluctuations and that of the opening and closing of the  camera shutter. This is quite a different matter from the occasional  flicker of unsteady burning or near  extinction to which arc lights have always been  prone, as a result of irregularities in the automatic feeding mechanism  advancing the carbons as they burn away in the arc. Although the carbon  feed mechanisms were  improved over the decades, all older cinemagoers  will have sometimes experienced the dimming and extinction of an  unattended arc-source film projector resulting from the same cause.

Mention of both shortcomings of arc light sources,  the first eventually eliminated by the use of D.C., and the second still with us, can often be found in the reminiscences of people who were in  films in that period, and both  can be seen in its films, as retakes were not usual  in the early silent period if an arc floodlight lighting a minor part  of the background happened to go out in the middle  of a take. This attitude began to change in the nineteen-twenties.

Throughout the years 1914-1919 the general large-area lighting of sets continued to be supplied by racks of Cooper-Hewitt  mercury vapour tubes, and in some of the smaller companies by arrays of  street-lighting type  arcs hung overhead.

The General Development of Lighting Style

The first generalization to be made about the  development of American film lighting during these years is a fairly  well-known one -- it speaks of a transition from  films being lit with the help of the general diffuse light through the  glass studio roofs, to films being shot entirely  under artificial light in blacked-out studios by 1919. Although the  latter situation is fairly true, as already  indicated there were a number of films shot in East Coast studios even  before 1914 which had scenes lit entirely by  artificial light, and this separation between the lighting practices on  the two sides of the United States persisted to some extent till around 1918. This is best made clear with some examples. To speak of Californian film-making first, the interiors of Birth of a Nation (1915) were lit entirely by  daylight controlled in one way or another, and a  sketch of G.W. Bitzer's lighting procedures can be read in Karl Brown's Adventures with D.W. Griffith. There are one or two pieces of backlighting on interiors in this film  done by letting in a patch of direct sunlight  through a gap in the overhead cotton screens, and the `spotlight' effect on Lincoln's assassin was created with sunlight  reflected from a mirror. An even better example of what could be done  solely with controlled daylight is given by the lighting of Raoul  Walsh's Regeneration (1915), for which  Georges Benoit created remarkably precisely  controlled gradations and localizations of sunlight across some of the  interior scenes. He also used sunlight for backlight in many of the  interiors, with the key light coming from  reflectors put in front, in the same manner as was  now standard for exterior scenes.

Even by 1915 the more common approach, both on the  East Coast and on the West Coast, was to add some arc lighting to  sharpen up the general diffuse lighting of the set; in the New York  studios the diffuse light came  mainly from Cooper-Hewitts, and in California from  diffuse daylight. One of the most elegant demonstrations of this earlier (and about to be superseded) style is given by the lighting of David Harum (Allan Dwan, 1915).  Although the photography of this film has been  recently credited to Harold Rosson, it seems probable to me that he was  only assistant cameraman on this film, as he was only twenty years old  at the time, and his next claimed  solo credit was four years later.

The less common, but more advanced, style to be  observed in 1915, which consisted in lighting a fair number of scenes  principally or entirely with arc lights, seems to have been confined to  some of the New York and New  Jersey studios. A good example of this is given by  the remarkable short film His Phantom Sweetheart, which was made  by Ralph Ince for Vitagraph. In this film the majority of the interior  scenes are lit solely with arc  floodlights; by necessity in the case of those shot  inside a real theatre auditorium and its foyer. As can be seen  from the illustrations, the theatre interior is lit  with two groups of arcs, one on each side of the camera, and both coming in at roughly 45 degrees to the lens axis, on the pattern established  at Vitagraph some years before.  What is new in this film is the scale and complexity of the scenes. His Phantom Sweetheart also contains a climactic scene lit with genuinely low-key lighting done with arcs to contribute to a succession of moods --  sensuality, suspense, and terror. The rather similar, and much better-known example in Cecil B. DeMille's The Cheat (1915) is no more thoroughgoing and extended in its use of low-key arc lighting.


A low-key interior shot in His Phantom Sweetheart lit solely by a small arc concealed under the shade of the table lamp.

By 1916 there were many films coming out of the East Coast studios  that had most of their studio interior scenes lit solely with arc  floodlights. One example is Silks and Satins (J. Searle Dawley),  which had the lights disposed  so as to give a stronger key light from one of the  side-front directions, and weaker fill light from the other, and also  sometimes three-quarter back lighting through a real or implied door or  window opening in the set. (Due to  the way that the light intensity from an arc  floodlight falls off rapidly with distance, it was not possible to use  one  to do backlighting from directly behind and above in Long Shots. This was only possible with spotlights.) In Silks and Satins no diffusing screens were used on the lights to soften them. Relatively unsubtle arc floodlighting like  this, or worse, can still be seen in some films made in 1918, particularly in California, e.g. A Modern Musketeer.

By 1917, as already remarked, the floodlights were  being diffused a good deal of the time on some films, and one of the  better examples of this was The On-The-Square Girl. (Although the cameraman and director of this  film are named on the titles as Morris E. Hair and  Frederick J. Ireland, Kevin Brownlow has suggested to me that these are  pseudonyms for Arthur Miller and George Fitzmaurice. Given the high  quality of the lighting and  direction of this film, and that it comes from  George Fitzmaurice's company Astra, and also that it was listed in a  trade source at the time as a George Fitzmaurice  production, this seems quite likely.) A fully accredited example of what the change to diffusion on arc lights looked like is given by Till I Come Back To You (Cecil B. DeMille,  1918), which was lit by Alvin Wyckoff and Charles  Rosher. In this film there was still some contribution from the  old-fashioned general overall light as well as the  well-managed directional components from diffused arc lights on the  closer shots.

Figure Lighting

From these years onwards it became the practice  in America to treat the lighting of the closer shots of the actors  separately from the general lighting of the set as  it was visible in Long Shot, and indeed to make changes in the positions of the lights when shooting the closer shots which were to be cut into  the main scene. (Naturally there  has to be some sort of very rough correspondence  between the look of the lighting in more distant and in closer shots,  but nevertheless quite substantial changes are not noticed by the  audience, now as well as then.) Although  the essentials had already been independently  developed in a crude way, it was around 1917 that a few of the best  cameramen such as Walter Stradling and Charles Rosher polished up what  were to be the standard patterns  of figure lighting, presumably drawing on still  photographic practice.

The most basic pattern of figure lighting is to have a key (or brightest) light directed at the figure from the front on  one side of the lens axis, a weaker fill light from  the other side of the lens axis, and a backlight shining forwards onto  the back of the actor. What would be considered the ideal angles along  which to direct these lights  depended in the first place on the exact direction  in which the actor was facing with respect to the camera, and in  the second place on the shape of the actor's face.  Besides the skill required in selecting the angles for the lights,  there was also the matter of arranging the relative  levels of brightness of the key and fill lights. It was here that there  began to be a marked improvement over the practice of a few years  before, for earlier it had been quite  common to have the lights from either side of the  camera of equal, or nearly equal, brightness. This produced two shadows  from the nose, one falling on either side of the face.

Charles Rosher's progress in this respect can be illustrated by the difference between his work on The Sowers (1916) and The Secret Game (1917). In the former the lighting on the figure is rather flat, since  the key and fill  lights are of almost equal brightness, but in the  latter the relative intensities and positions of the lights are very  well  judged. The key light was still being placed only  slightly above actor eye-level in most films in 1917, but by 1919 some  cameramen were beginning to place the key light rather higher when  appropriate, as did George Barnes in Dangerous Hours.

As already mentioned, a modification to the initial  use of a single backlight now began to appear, with a few cameramen  using two backlights, one directed from each side at the back of the  figures. This can be seen  intermittently in such films as Fighting Odds photographed by René Guissart in 1917, and Stella Maris and The Whispering Chorus, photographed in 1918 by Walter Stradling and Alvin Wyckoff  respectively. This use of  double backlights could be combined with either a  single key light and no fill from the front to give an alternative  form of three-point lighting to the basic form  described above, or with both key and fill from the front to give four -point lighting, which was less common initially.  Later the use of two backlights and a weaker light straight on from the  front came to be the standard way of treating a `profile two-shot' (two  actors facing each other), but  this did not happen at first.


Early  close double backlights ---- Double backlights spread apart ---- One backlight with key & fill

It is at present impossible to tell who were the principal forces  behind the developments I have outlined above, in part owing to the lack of cameraman credits before 1917, which was the year in which it became usual to name  the cameraman on the better class of production. We  also do not know what were the aesthetic assumptions behind the adoption of the standard techniques, though it might be possible to find out  more about this with  extensive research.

Interior Lighting in Europe

In 1914 the best European lighting of interiors  was being done in a rather similar way to that in America, with  general diffuse lighting through the studio roof  being sharpened up a little with arc floodlights in many scenes, but  also with the rare occasional scene done mostly with arc floodlights when a special lower key effect was wanted. Over the  war years in Scandinavia there was a tendency, just as there was in  America, to move towards heavier  use of arcs, but the Scandinavian studios were never blacked out permanently at this time, and even in the early  'twenties many scenes were still being lit in part  with diffuse daylight, as can be seen in the films of Dreyer, Stiller,  and Sjöström.


Well-placed figure lighting by Johan Ankerstjerne in Haevnens Nat (1916). Side-back lighting from the  left, and side lighting from the right, both produced by diffuse  sources.

The most accomplished Danish cameraman during this period was Johan Ankerstjerne, and his work on Benjamin Christensen's Haevnens Nat (1916) shows the way that a few European cameramen were also  developing the kind of three-point lighting that has already been described in American films. But Ankerstjerne only did  this on closer shots, where the light applied from the three-quarters  back position could come from an  arc floodlight. There was no introduction of  backlighting from directly behind with spotlights in Europe, nor any  use of spotlights at all for that matter, until well after the end of the war. Ankerstjerne also did some notable low-key  work -- such things as a hand-held lamp casting looming shadows in a  flight down subterranean  passageways -- in Verdens Undergang (1915), which was yet another of the speculative and apocalyptic epics like Civilization which were produced during the early stages of World War I.Homunculus (1916), a German  contribution to this genre, shows the retarded state of lighting in that country, in that its low-key effects were done  solely by the control of daylight with blinds and the way the set was  constructed, in the manner of the  Danish films of some years before.

1918 the Germans had fallen badly behind in lighting, as is obvious even in a Lubitsch film such as Die Augen der Mumie Ma, in which the interiors were still done with general overhead diffuse  light, with only a little  sharpening here and there with arc floodlights. And  even this was not well done. Things began to improve a little in 1919,  particularly in the best productions of that year (Lubitsch again), but  in most films the way light was  applied to the figures was still rather crude by  American standards.

The flowering of Swedish cinema during the war also  involved some notable camerawork by Julius Jaenzon and others, and not  surprisingly the styles used owed quite a lot to earlier Danish and  French examples. The most  common approach in Sweden to lighting the general  shot of a scene was to bring the light in from one major direction,  either from the side or from above at the front of the set. The source  of side light was usually a large  opening such as an actual (or implied) door or  window letting in the diffuse daylight coming through the glass wall  of the studio. The result was not an even flood of  light over the whole extent of the set, but moderately localized  light in one area tailing off into the further  corners. Sometimes foreground features of the set or actors were left  relatively dark. A good example of this is given by the lighting of  Mauritz Stiller's Balletprimadonnan (1915).  An alternative form of this localized lighting with  natural light which was very popular with the cameramen at Svenska  Biografteatern was to allow direct sunlight to fall frontally from high  above onto a central area of the  foreground of the set where most of the action took  place, with more general diffuse lighting working its way into the  farther parts of the set to light them more dimly. This second method of using mostly natural light gave  somewhat the same effect of separation of lighter  figures from darker background that had earlier been achieved by  American cameramen using artificial light sources. When using both these methods of lighting the placement of  the actors at the various stages of the evolution of the scene becomes quite important if their faces are to be clearly  visible, and not heavily shadowed when they move into some parts of the  set. It also precludes the use of  a lot of cutting around to different angles within  the scene. As the Swedish directors only used a limited amount  of cutting within scenes at this point in history,  they experienced no difficulties because of this. Nor did the major  exponent of a somewhat similar approach in America.

Maurice Tourneur and Cinematography

There were some American film-makers who used  lighting styles other than those already described, and the most  important of these was Maurice Tourneur. His cameramen used the most  precise and subtle form of lighting  from a single direction, and had probably evolved it from the rather more primitive French forms current before 1914.  Although Tourneur used more cuts within a scene than the Swedish  directors, he overcame the problems  in staging associated with the single direction  lighting style by having a certain amount of localized fill light on the actors when they were in some positions, and also by unobtrusive  relighting for the closer shots.

The most obvious visual feature of Tourneur's films,  which was the creation of foreground silhouette effects, fitted  in well with the handling of the main light source  as described above, but this particular feature does not appear  continuously in every scene in his films. On the  contrary, during the years 1914 to 1916 many of the scenes in his  films are lit in the more conventional manner usual  at that time in American films. In fact, some of Tourneur's most  striking images were obtained purely through the  compositional arrangements in shots which were lit in a high key, i.e.  with moderately even illumination over the whole frame.

Although hardly any of their films from this period  have survived, it seems probable that the other emigré French directors  at the Fort Lee studios -- Perret, Capellani, and Chautard -- worked in  similar, though less exalted,  versions of this `pictorialist' style. (The sense of the description `pictorialist' is that the compositional style is  closely based on that used in the painting of past  periods -- say the Salon painting of the end of the nineteenth  century -- as opposed to the relatively nondescript  compositions of the already existing film tradition, where the  connection with fine art is at second or third  hand.) The principal continuers of this tradition in America during the  'twenties were Rex Ingram and Josef von Sternberg,  though in the case of the latter it evolved into a style with quite new  qualities. Traces of the Tourneur influence can occasionally be seen  elsewhere, for instance in von  Stroheim's Foolish Wives, and also in the films of Clarence Brown, Tourneur's former assistant.

The Italian Picture

One aspect of the decline of the Italian cinema  during the First World War was the pursuit of pictorialism, particularly in exterior scenes, regardless of its relevance to the narrative. In  short, the story stopped while the  leading characters stood around in a beautiful  landscape picture. This can be readily seen in the Eleanora Duse  vehicle, Cenere (1916), but it touched even the best Italian films of the period, such as Assunta Spina (1915).

The Photography of Night Exterior Scenes

In 1914 exterior scenes purporting to be taking  place at night were still exclusively shot under full daylight, and  the impression of night was conveyed by the standard blue tinting, usually with the help of a previous descriptive  title. But in 1915 the first night scenes actually  shot at night with the help of artificial light appeared in a few  American films. Notable early examples include a street scene in Cecil  B. DeMille's Kindling which was lit by  Alvin Wyckoff solely with a few strategically placed arc floodlights, and a night battle scene done the same way in a short  Thanhouser Company production, Their One Love. G.W. Bitzer's cruder solution to the same  problem in Birth of a Nation was to use  pyrotechnic flares to light the scene of the farewell ball before the  battle . By 1916 the use of arc floodlights on moderate  scale night exteriors was becoming more common in the better American  films, and by 1918 even an ordinary Western such as Henry King's Six Feet Four has a large-scale  night street scene lit in this way. But very distant landscape scenes still had to be given normal daytime photography, and  then integrated with closer shots which had been photographed `night for night' with artificial  light by applying the same overall blue tinting to  the whole sequence. Some cameramen and directors show the first signs of trying to improve on this method however, and there are a few very rare cases where distant night  scenes have been photographed in the half-light of  dusk (Female of the Species, 1916), or where the upper part of the scene has been darkened by putting a partial filter over it in front of the lens, as in Less Than the  Dust (1916). In such cases the cameramen avoided getting any appreciable amount of sky into the shot, but in general  such an approach was neither possible nor striven for.


A battle scene shot at night, and lit solely by arc floodlights in Their One Love (Thanhouser, 1915)

In Europe the idea of lighting night exteriors with arc lights was also just beginning to appear, as in Bauer's Zhizn za zhizn (1916) and Sjöström's Berg-Ejvind och hans Hustru (1917). Elsewhere I have seen no sign of this  technique before the 'twenties, but it must be  emphasized that I am discussing location or `back-lot' exterior scenes,  and not sets representing exteriors constructed inside a studio, such as those in Lubitsch's Die Puppe (1919), or the subsequent Das Cabinet des Dr. Caligari (1920).

Shadowplay and Other Lighting Effects

It was not until this period that the use of cast shadows for expressive purposes began its true development, though  occurrences still remain fairly rare. Cecil B. DeMille began using the  shadows of objects outside the  frame cast into the frame area as early as 1915. The well-known example here was the shadow of the prison bars falling on  the husband in The Cheat, but the first instance of this in DeMille films was earlier in 1915 in The  Girl of the Golden West, where a rope casually  dangling on the set casts the threatening shadow of a hangman's noose at an appropriate moment. Variants of the idea also occur in Maria Rosa, where the villain casts his  shadow on the wall before he enters the scene, and  it had spread to other countries by 1916, e.g. Protazanov's Pikovaya Dama and Abel Gance's Barberousse, where the shadow of the clutching hand of a criminal slides  onto the white pillow of the sleeping heroine. After a scattering of other similar examples we find a natural part of the  set creating the shadow of a cross on the heroine at a suitable moment  in Until they Get Me, by which time  such devices were available to any really  enterprising director. Looming shadows had begun to spread to other  directors by 1917, e.g. The Whip by Maurice Tourneur, and Kidnapped. And DeMille's The Whispering  Chorus of 1918 uses looming shadows cast on the  walls from lights placed low in a scene in which the hero begins to  stray into wrong-doing. There was no apparent light source in this scene motivating these upcast  shadows as there had been in the earlier Italian  examples of low placed lights that I have mentioned, and this is also the case in Sidney and Chester Franklin's Going Straight (1917),  in which a low placed light shining up into  a face in Close Up was used in a nightmare sequence  without any apparent or reasonable source, purely to convey a sinister  atmosphere as the hero's fears and worries were played out.

The shadow of a threatening hand slides over a sleeping woman’s face in Barberousse (Abel Gance, 1916)



It was during these years that the Bell &  Howell camera, described in a previous chapter, began to displace the  Pathé studio camera as the major tool for American cameramen. Another  new camera, the Akeley, was first  produced in 1917, but since it was a few years  before it had any significant use, I will defer a description of it till the next chapter.

Angle Shots

In this period shots taken from really high- or  low-angles continued to be rare, and mostly they were used in a  situation where they could be understood as representing the Point of  View of one of the characters in the scene  in question. However, there are examples of extreme  high-angle shots which are objective, and definitely not POV shots, in  films from most countries, from America to Russia. When they do occur,  the allowance is one per  film. Ignoring distant shots from the ground towards a first floor window, or something similar, real low-angle shots are  even rarer. By far the most striking instance in this class is a  low-angle Close Up in Abel Gance's Barberousse. This is of the titular protagonist, at the point where he declares that he is `the King of the Forest',  and this must be intended to be expressive.


A Close Up shot from a very low angle in Abel Gance’s Barberousse (1916). The character has just boasted that he is ‘The King of  the Forest’.

Camera Movements

During the years 1914-1919, just as in previous  years, there was little change in the way the vast majority of shots  were taken with fixed framing, particularly in interior scenes. Very  rarely one finds panning shots being  used to follow actors across a set, as in some  Reginald Barker films such as Typhoon (1914) and Between Men (1915), but there was some increase in the use of  framing movements -- i.e. small pans and tilts to keep the actors  well-framed -- as some directors started to take even more of their  shots closer to the actors. (When  shooting close in it is possible to avoid the use of framing movements if the movements of the actors are carefully  controlled, but if the cameraman had the ability to  turn the panning or tilting crank while also cranking the film  drive, it was easier on the actors to let the camera conform to them.) Early examples of this slight trend towards the  greater use of framing movements can be seen in David Harum (Allan Dwan, 1915) and The Right Girl (Ralph Ince, 1915), but by 1919 it is much easier to find examples amongst the increasing numbers of films that were now  being shot from closer to the actors; e.g. Jubilo (Clarence Badger, 1919). As before, exterior action  scenes were the likeliest place to find camera movements.

Tracking Shots

Parallel tracking shots, in which the camera  moves at a fixed distance from actors moving on a parallel course,  continued to occur on rare occasions such as car and train chases, but  tracking towards and away from groups  of actors who were not moving a great deal (which I  call `tracking on a quasi-static scene') had a world-wide vogue in the  wake of the Italian film Cabiria (1914). Such tracking shots were referred to at the time as `Cabiria movements', for it seems that no-one had taken much notice of the earlier tracking shots on quasi-static  scenes in American and English films, except perhaps Giovanni Pastrone, the director of Cabiria. At the time  Pastrone stated that his intention was to create a  `three-dimensional' effect in the photography to show off the vast solid sets of his film, and for this reason his tracking shots were made  moving inwards on a diagonal to his  sets. These tracks are also of a fairly limited  extent, slow, and do not end too close to the actors. In 1915 and 1916  every bright young director had to have one or two `Cabiria movements' in one of his films, but they used  them slightly differently to Pastrone.

To pick just a few examples of this fashion from well-known directors, I will mention David Harum (Allan Dwan, 1915), Ditya bolshogo goroda (Yevgeni Bauer, 1914), Evangelimandens Liv (Holger Madsen, 1915) , and The Vagabond (Chaplin, 1916), all of which move in much closer to the actors rather faster than the originals in Cabiria, and also have trajectories fairly straight in or out from the scene. And all of these tracking  shots incorporate a certain amount of panning as well, which those in Cabiria did not. The example in The Vagabond is the most elegant application: a track out from a close shot of a  painting reveals the people standing  around looking at it. Everyone seems to have been  satisfied with at most two tracking shots on quasi-static scenes in  their films, with one well-known exception. This was The Second-in-Command (William J. Bowman,  1915), which though of no great interest otherwise,  contains about two dozen tracking shots. These go closer in to the  actors than those in any of the other films, even as close as a Big  Close Up at one point, and one of them  is of greater complexity than any in other films as  well. The tracking shot in question follows a couple round a dance floor amongst other couples, panning the while to keep them in frame, and the general effect is exactly the  same as it would be twenty years later in any  tracking shot following a dancing couple. The Parson's Horse Race (Edison, 1915) has a track back from the final group of characters at the end of the film, and can be seen  as a development of D.W. Griffith's idea for the conclusion of A Girl and Her Trust (1912).

By 1917 the tracking shot craze in America was  declining as fast as it had arisen, and by 1918 and 1919 tracking shots  on quasi-static scenes had again become rare, the only examples I have  come across being in The Blue Bird (Tourneur, 1918), and Stella Maris (Marshall Neilan, 1918), though there probably some more  amongst the large number of lost films. The example inStella Maris is a further development of a usage that was to become popular much,  much later: as the hero and heroine embrace in the final shot of the  film the camera  pulls back from them, and there is a slow fade-out.  There are also still a few examples in European films, such as Herr Arnes Pengar (Stiller, 1919), Jacques Landauze, and Malombra.

Camera Movement and Expression

Cases where a camera movement could reasonably be considered to produce meanings through its conjunction with the action  in the filmed scene are hard to find in this period, apart from the  marginal case in Stella Maris mentioned above. The only other instance that springs to mind is in von Stroheim's Blind Husbands, in which  what was to be a characteristic effect in his films  first occurs: a Point of View shot tilting up from the feet to the face  of a potential prey as the villain sized her up.

Depth of Field and Other Photographic Variables Influencing the Film Image

Depth of field(often erroneously  called depth of focus) is one of the central factors controlling the  appearance  of the film image, and it is really necessary to get a clear understanding of the way it is related to other variable  factors if one is to appreciate the interconnections between the visual qualities of films and film technology. The four  central quantities whose variations are strictly connected one with  another are Depth of Field, Lens Aperture, Focal Length of Lens, and Lens Focus.

Depth of field is the range of distance in front of the camera lens inside which objects produce sharp images of themselves as seen on the cinema screen when the film is finally projected. The boundaries of this range of  sharp focus are approximate, as objects just outside it appear only slightly unsharp, or may even perhaps appear in focus to the casual glance at the cinema screen. The range of sharp focus as it  appears on the ground- glass  screen of any camera view-finding system is not  necessarily the same as that on the cinema screen, though usually close  to it.

Lens Aperture is the size of the  variable opening in the diaphragm built into the middle of the lens. Its size is  measured in `f-numbers' or `stops', and these  f-numbers are inversely related to the actual diameter of the lens  diaphragm opening. the basic series of f-numbers  runs f1, f1.4, f2, f2.8, f4, f5.6, f8, f11, f16, f22, f32, f45, f64,  though other numbers may appear on actual lenses. Each of these  f-numbers is said to differ from the next by  `one stop', and each change of a stop proceeding  from left to right along the series halves the amount of light  passing through the lens to the film, and conversely in the other direction the amount of light passing is double for  each change of a stop. The smallest aperture on a  film camera lens is now usually f22 or f32, but in the early days it  could be f45, and the largest or maximum aperture was usually between f2 or f4.5. Determining the  correct exposure means determining the amount of  light that has fallen on the scene and is then reflected from it  into the lens, and then determining the lens  aperture that will permit just the right proportion of this light to  fall  onto the film to give the right amount of activation of the silver halides contained in it. It is colloquially said by  cameramen that when there is twice as much light on  the scene, then the light has `increased by one stop', and  that a photographic film that needs only half as  much light as another is `faster by one stop'. Likewise, a film that  needs four times as much light as another is `two stops slower', and so  on.

Focal Length of a Lens is the  distance behind its `optical centre' of the plane in which an image of  an infinitely  distant object is formed. The `angle of view' of a  camera lens is inversely proportional to its focal length for the  same size of film frame, so short focal length  lenses have a wide angle of view, and are colloquially referred to as  wide-angle lenses, and long focal length lenses have a narrow angle of  view. This brings me to the awkward  question of what constitutes a standard lens. the  opinions of film cameramen on this point have changed during this  century, and as already remarked, some cameramen before 1914 considered a 3 inch (75 mm.) lens to be  standard, though most considered a 2 inch (50 mm.)  lens to be standard, which was exclusively the case in the 'twenties.  Later on, there was some move towards considering even shorter focal  lengths as standard, as I shall  detail later. There has been another approach to  this problem through experimental investigation of which camera lens focal length gives audiences the best impression of correct  perspective in projected images of real  scenes, and this work suggest that in this sense a  standard lens has a focal length of around 35 mm. to 40 mm., with the  uncertainty corresponding to a real experimental variation.

Lens Focus is of course the  distance at which lens focus is set so that objects at that distance  will produce the very sharpest images on the film and on the screen.

Now the value of any one of these four quantities is  determined by the values of the other three, but it is usual to  consider the effect of the depth of field of holding any two of the other three fixed, and varying the third. The results of this are nowadays set down in depth of field tables, but these were not used in the period we are  considering, and cameramen relied on experience to  determine what would be in focus or not. Given that the other two  factors are kept constant, the depth of field increases with (1)  reduction of lens aperture, (2) decrease  of focal length of the lens, (3) increase in  distance at which the lens focus is set (up to a certain distance called the hyperfocal distance).

As has already been indicated, the aperture cannot be freely chosen in any particular case, for it depends in its turn on the light level on the scene to be photographed, and also on the  sensitivity to light (the `speed') of the  particular type of film in the camera. And on this  point there was no real choice till the end of the silent period.

Lens Apertures Used In 1914-1919

Towards 1919, for the first time since the use of diffused sunlight was established for the filming of studio interior  scenes, there began to be signs of a change in the  lens aperture used, and hence in the depth of field. In a few films such as Stella Maris (1918) and Jubilo (1919), there is quite  clearly a visible reduction in the depth of field  when the actors are in Medium Shot, when compared  with the situation at that closeness previously. I estimate that in  these cases, and one or two similar ones that I have seen, the depth of  field corresponds to an aperture of  about f4 with a 50 mm. lens. Although these examples presaged the trend of the next few years, they were not typical in  1919, but restricted to the work of a limited number of leading  film-makers. This phenomenon may  have had something to do with the move towards  shooting in totally blacked-out studios which was taking place around  this time, for although in general the background of diffused daylight  that was lost in this move was  replaced with greater use of Cooper-Hewitts and  diffused arcs, it seems likely that this replacement was not complete,  and hence the overall light level dropped slightly. However the majority of the studios were probably  still working at an aperture of about f5.6 most of  the time, just as before the war. Certainly the now minor and declining  studios of Vitagraph and Edison were, according to an article in The Transactions of the Society of  Motion Picture Engineers (No.8, 1919). However  this is a suitable point to warn against taking such reports of  particular cases as applying in general, for it is  clear from the detailed description in this article of the kind of  lighting set-ups being used at Vitagraph and Edison  in 1919 that the cameramen there had not advanced from the standard  procedures of several years before. Whereas in the major studios there  had been the considerable  changes in lighting style that I have described  earlier. Similarly, a reminiscence by a cameraman that he once took an  exterior shot at f45 around this time does not mean that this was  standard practice. It wasn't.


There was no change in the variety of camera  lenses available during the years 1914-1919, but the first signs of  the use of long focal-length lenses appeared in  entertainment films. There are isolated shots in a crowd scene in Civilization (1916) and the battle on the pyramid in The Woman God Forgot (1917) which are taken with  lenses of focal length in the region of 4 to 6  inches, both scenes clearly having been shot with multiple cameras. This kind of usage remained very rare for decades, even in similar  mass-action scenes, as most film-makers  preferred either to arrange the scene so that they  could get one of the cameras in closer with a standard lens, or  alternatively to restage parts of the action for a separate shot.

Another harbinger was Hendrik Sartov's use of a long lens for shooting Close Ups in Broken Blossoms (1919), though when this practice became common in the next decade most cameramen were satisfied with something  like a 4 inch focal length, rather than the 6 inches plus used by Sartov.

The Use of the Iris Mask

The use of the iris mask came to a peak during  the years 1914-1919, both as a way of beginning and ending a scene, and  also to create a static mask or circular vignette around some shots.  Whether or not Griffith and Bitzer  originated irising and the use of the iris vignette, it seems highly probable that the well-deserved prestige of D.W.  Griffith and the success of Birth of a Nation were responsible  for the popularization of this device. By 1914  Griffith had settled on the standard procedure of  beginning every shot with an iris-out (i.e. opening the iris diaphragm  in front of the lens), and concluding it in the reverse way, though some of these irisings were removed  later in the editing process. Nevertheless, in  Griffith's films a sufficiently large number of shots, even within  scenes  and sequences, remain with the irising still present to create a very discontinuous impression. Very few film-makers in  America went as far as Griffith in this direction, and those few who did soon abandoned the extreme  of the practice, but Griffith himself persisted with it into the 'twenties. This may be because Billy Bitzer kept using a  Pathé camera, which did not have a fading shutter, throughout this  period, whereas other cameramen were  switching to the Bell & Howell as soon as they  could afford it.

The films made at the Ince studios contain relatively few iris-ins and -outs, and those few are confined to the beginning and end of sequences. At the Ince studios, as elsewhere, fades also  continued to be used for the  purpose of beginning and ending sequences, without  any consistent relation to the temporal connection between the sequences they separated. By 1918 the use of the iris to begin and end sequences  was starting to decrease in  the United States, though in Europe it was just  starting to become fashionable. At that date it is quite easy to find  American films such as Stella Maris in which only fades are used.

A variant of the simple iris opening out from the centre of the frame appears at the beginning of 1915 The Girl of the Golden West and Birth of a Nation. In this procedure the opening and closing centre of the iris started  from whichever point in the frame contained the  subject of principal interest in the scene, and it had an effect  somewhat analogous to a modern zoom shot. There are very few other  examples until 1917, when the device  became slightly fashionable. However, the effect was always used very sparingly, and in most films that have ordinary  irising it does not even appear. To produce `directional' irising of  this kind required a special sliding  mount for the iris diaphragm that enabled it to be  centred in front of the appropriate point in the frame.

Yet other variants of the simple iris appeared at  this time, and in these the mask opening or closing in front of the  lens had shapes other than circular. One of the more frequent of these shapes could be called the opening slit; a  vertical central split appears in the totally black  frame, and widens till the whole frame is clear, revealing the scene  that is about to start (The Cossack Whip, 1916). Eventually the  diagonally opening slit appeared as well.  Another form was the single mask that pulled up from the bottom like a theatre curtain, or down from the top, or back from  one side, and yet another was the diamond-shaped opening iris, as in Poor Little Peppina and Alsace (1916), rather than the usual circle. Again, all of these variant forms were very infrequently used, and  when they did occur in American films it was usually in the introductory stages. Before leaving the subject of irising, I  should also mention that by 1918 the edges of ordinary circular irises  were becoming very fuzzy in  American films, sometimes to the point where it is  difficult to distinguish an iris-out from a fade. This is a reflection  of the move that was beginning towards photography at larger apertures,  and hence reduced depth of  field, which put the iris mask in front of the lens  further out of focus than it had been some years previously. The  edge of the iris mask in European films stayed  rather sharper and more distinct into the 'twenties, because the trend  to filming at larger apertures had not yet developed there.

The Return of the Wipe

The true wipe -- i.e. a boundary line of some  shape moving across the frame and erasing the image as it passes over it to leave a new image behind it -- which seems to have dropped out of  use after being invented by Robert  Paul at the beginning of the century, now made its  return around 1917. The Angel Factory (1917) includes several  wipes as transitions to and from scenes representing a character's  thoughts. These wipes have a curved  edge rather than the original straight edge of those used by Paul and Smith, and they proceed from side to side rather than  up and down. A wipe of the same kind gets half-way across the screen to  reveal a mental image  before stopping in Old Wives For New, and there is an instance similar to that in The Angel Factory in Twin Pawns (1919), so there were probably at least a few other films that used wipes at the time. There were also  various approximations to the wipe as a form of transition between sequences, as in The Ghost of Rosie Taylor (1918), where an iris-out is overlapped with an iris-in, and there were quite probably other examples of these  kinds of procedures in the vast numbers of films  which are now lost, so the simultaneous iris-in and iris-out from  opposite corners of the frame that is used a couple of times in Das Cabinet des Dr. Caligari (1920) is not as  unprecedented as has been suggested.

`Soft Focus' and Lens Diffusion

The earliest use of a form of `soft focus' of which I know occurs in His Phantom Sweetheart (1915), and in this  case it is done by putting the lens very slightly  out of focus. That the effect is intentional is shown by the fact that  it occurs twice in successive shots; first as a  mysteriously seductive woman is introduced in Medium Shot behind  foreground actors who are sharply in focus while she is slightly out of  focus, and then in a Medium Close Shot of  her alone which is again slightly out of focus. A  more fully developed example of this technique occurs a few months later in Mary Pickford's Fanchon the Cricket, in which there is  repeated series of Medium Close Shots  of Mary Pickford in an exterior scene with her face  well out of focus, and with strong backlighting as well. The only other  example of soft focus that I have come across from before 1918 is in Ablaze on the Rails, No.96 in  the `Hazards of Helen' series of films. This film,  which was made in 1916, opens with a close shot of the actress playing  the heroine of the film in a glamorous gown introducing herself in the  working clothes of the films. The  central area of the frame covering her face is  softened by some means, presumably the use of a special lens on the  camera, and then this softening vanishes on a dissolve to the next shot, which has an identical set-up.  Although I know of no other examples of this  technique from the next couple of years, this does not mean that they  did not once exist, and indeed there has been a claim made for the use  of `soft focus' in another film made in  1916, but which is now lost. What one does find in the next few years is the use of extremely out-of-focus circular vignette masks, which are so out of focus that the blurred edge of the  mask extends its effect to the  centre of the frame, slightly reducing the  definition of the image there. Then in 1918 there was a completely new  development in D.W. Griffith's Broken Blossoms. In this film all  the Close Ups of Lillian Gish are heavily  diffused by the use of layers of fine black cotton  mesh placed in front of the lens, and also by the intrinsically poor  definition of the special long focal length lens used by Hendrik Sartov  to photograph these shots. Heavy  lens diffusion was also used on all the other shots  carrying forward the romantic and sentimental parts of the story, though whether these were done by Sartov or Bitzer is not known. Heavy lens  diffusion was also used in a  similar way in France by Marcel L'Herbier in his  film Rose-France. This could well have been a case of direct  influence, since that film was first shown at the  very end of 1919. After that date lens diffusion occasionally appears in a more limited way in the works of the so-called `French avant-garde',  but not elsewhere in Europe  for a few more years.


A Close Up in Fanchon the Cricket (1915), with the focus sharp on the frame of leaves in the foreground, but with the actress  behind appreciably out of focus. She is backlit by the sun, and there is strong reflector fill from the front as well.

Masking of Other Kinds

Masks of shapes other than circular also began to appear in American films during the years 1914-1918: first such simple  shapes as the `cinemascope'-shaped narrow rectangle formed by a black  band masking the top and  bottom of the frame in Intolerance, then moving on to more complicated shapes such as a mask with a cruciform cut-out in Stella Maris (1918). The Girl Without a Soul (Wm. Bertram, 1917) also has shaped  vignettes, while A Little Patriot (Pathé,  1917) has elaborately shaped vignettes used on a scene of a children's  mock battle, and also a white vignette to  concentrate attention on a detail. In 1918 Maurice Elvey in Britain took up the idea, and, along with a number of other new tricks, introduced  it into his Nelson; The Story of England's  Immortal Naval Hero. This has a couple of scenes framed in a heart-shaped mask, as does his subsequent The Rocks of Valpré (1919). The most elegant variants occur in some films Ernst Lubitsch made in 1919 and later.  In Die Austernprinzessin a triple layer of  horizontal rectangles with rounded ends enclose sets of dancing feet at  the frenzied peak of a foxtrot, and in Die Puppe a dozen  gossiping mouths are each enclosed in individual small  circular vignettes arranged in a matrix. Unlike most of the vignettes used in American films, the vignettes used by  Lubitsch were `hard' or sharp-edged, as was  necessary for clarity in his particular application. In France again,  unusually shaped masks play a large part in Rose-France (1919), and later continue to appear in a small way in  subsequent films.

Elaborately shaped vignette mask used for a shot in a children’s battle scene in The Little Patriot (1917)


For the sake of completeness I should also mention another celebrated use of hard masks in these years, and this was the characteristic  arch-shaped mask used by Maurice Tourneur in his films to denote fantasy or hallucination . As far as I can remember he used it consistently  for this purpose, and not merely for decoration. Certainly in Poor Little Rich Girl the arch-shaped mask is used solely on the shots of the heroine's hallucinations.


The use of anamorphic (distorted shape) images first appears in these years with Abel Gance's la Folie du Docteur Tube. In this film the effect of a drug administered to a group of people was suggested by shooting the  scenes reflected in a distorting mirror of the fair- ground type. Although this film still exists, it was not shown at the  time of production, which Gance claims was 1915. It would be nice to  have some independent confirmation  of this date. There may well have been other uses of anamorphosis during the war years, but in any case the next use I know  of was in Till the Clouds Roll By (Victor Fleming, 1919). Here it was used to depict the nightmare  effects of indigestion in a comic manner. In fact,  like so many film effects that distort the representation of reality,  anamorphosis was first used exclusively in comic contexts.

Other Subjective Effects

In fact, it was during this period that camera  effects intended to convey the subjective feelings of characters in the  film really began to be established. These could now be done as Point of View shots, as in Sidney Drew's The  Story of the Glove (1915), where a wobbly hand-held shot of a door and its keyhole represents the POV of a drunken man. In Poor Little Rich Girl a rocking camera shot is intended to convey delirium, and by 1918 the  idea had got to Russia, in Baryshnya i khuligan, where the Hooligan's infatuation with the Lady is conveyed, in a less  than ideal way, by his Point of View of her splitting into a multiple  superimposed image.

`Poetic Cinema' and Symbolism

Symbolic effects taken over from conventional  literary and artistic tradition continued to make some appearances in  films during these years, and it is possible that there were yet more  examples among the vast number of films  from the war years that are now lost. In D.W.  Griffith's The Avenging Conscience (1914), the title `The birth  of  the evil thought' precedes a series of three shots  of the protagonist looking at a spider, and ants eating an insect,  though at a later point in the film when he prepares to kill someone  these shots are cut straight in without  explanation. The inspiration for this may well have  come from the widely distributed Italian film Gli ultimi giorni di Pompei, which I mentioned in this connection in the previous chapter.

Possibly as a result of Griffith's influence, 1915  was a big year for `symbolism', allegories, and parables in the American cinema. Films following this route invariably included female figures  in light, skimpy draperies, and  indeed sometimes wearing nothing at all, doing  `expressive' dances or striking plastic poses in sylvan settings. Titles include Lois Weber's Hypocrites, Vitagraph'sYouth, someone else 's Purity, and so on. All of it was  thumpingly obvious, and usually done at considerable length, as in The Primrose Path, which starts with a large  painting illustrating the concept, which dissolves  into a replica of the same scene with actors posed, and then they  come to life. This is amplified by closer detailed  live action representations of stations on `The Primrose Path' before  the film proper gets under way.

Giovanni Pastrone's Il fuoco (1916) represents an advance to some extent, in that the symbolic effects, though  admittedly fairly obvious, were not explained as they occurred. Il fuoco was an entry in the already established  `vampire' genre, of which the best-known example is Frank Powell's A Fool There Was (1915), but in fact these tales of a man enticed and destroyed by an evilly seductive woman had been developing in European  cinema for years before that. The central figures of Il fuoco are introduced as `He - The Unknown Painter' and  `She - The Famous Poetess', and the three stages of  the affair are introduced by illustrated titles showing The Lightning  Flash, The Flame, and The Ashes. Throughout the early stages of the film her dress and poses are  arranged so as to suggest a bird of prey, and at a  key moment a shot of one is cut in without explanation. An interesting  German example from a few years later is Robert Reinert's Opium (1919), which has some notable  innovations in the use of Insert shots to help  convey the sensation of the drug reveries. These are travelling  landscape shots taken from a boat going down a river, and they are  intentionally shot out of focus, or  underexposed, or cut into the film upside down. The  last of these devices in particular seems to me very striking, and also  quite successful in conveying a feeling of disorientation.

Symbolist art and literature from the turn of the  century also had a more general effect on a small number of films made  in Italy and Russia. The supine acceptance of death resulting from  passion and forbidden longings was a  major feature of this art, and states of delirium  dwelt on at length were important as well. Although such features  were mostly in what I would call the content of  these films, there was an interaction of this content with their formal  features, so I will mention some of them. The first Russian examples  were all made by Yevgeni Bauer for  Khanzhonkov during the First World War, and include Grezy, Schastye vechnoi nochi, and Posle smerti, all  from 1915. These to some extent live up to the  promise of the `decadent' aesthetic suggested by their titles; Daydreams, Happiness of Eternal Night, and After Death. Schastye vechnoi nochi includes a visually very  striking vision of a medusa- like monster superimposed on a night-time snow scene, and Posle smerti has a somewhat subtler dream vision of a dead girl, picked out by extra arc lighting, walking through a wind-blown  cornfield in the dusk. Later examples from the rival Ermoliev company such as Protazanov's Pikovaya dama and Satana likuyushchi lacked the true Symbolist feel. In Italy, another country somewhat isolated filmically by  the war, the same kind of realization of the fin de siecle decadent symbolist aesthetic can be found, mostly in films associated with the diva phenomenon. I have already mentioned Il fuoco, but there were others afterwards  developing the theme further, such as Malombra, and the most complete example, which also has decor to match, is Charles Kraus' Il gatto nero. This last is one of the few films of this kind to use atmospheric  insert  shots to heighten the mood. Films from other  countries did not show this tendency to any significant extent, either  because Symbolism had never had much of a grip on their major arts, or  in the case of France and Germany,  because newer artistic movements had made Symbolism  thoroughly old-fashioned.

The first film explicitly intended by its maker to be a visual analogue of poetry, Marcel L'Herbier's Rose-France (1919), continues further along these same paths.

Art Direction and Design

The general style of design for film interiors  remained a tidied-up naturalism, and it is during this period that it  became established that room sets in American films  be built about 50% bigger than they would be in actuality.  The other generally notable characteristic of  interior sets in American films is that the walls are always of a rather dark tone. It is largely this convention, which  lasted till the end of the nineteen-twenties, that gives the films of  these years their `old-fashioned' look. As is  well-known, it was during the war years that greater attention came to  be paid to art direction, and as well as care being given to visual  co-ordination in films with contemporary  subjects, the first efforts at stylized design were  made in a few films. Most of these have often been discussed and  illustrated, but a brief survey should mention The Female of the Species (1916), in which the art director  Robert Brunton did not go much beyond what might  have been the very latest ideas of refinement in actual interior  decoration. Though the abstract designs round the intertitles in this  film are a little more advanced. The  same concern for putting into a film the latest kind of `modern' elegance that a wealthy contemporary with the most advanced taste might hypothetically use in his home can be glimpsed in some of  the sets in Benjamin  Christensen'sHaevnens Nat (1916) and Ernst Lubitsch's Schuhpalast Pinkus (1916).

In Fighting Odds (1918) Hugo Ballin went  beyond this to a real degree of stylization; the furniture is sparse to a point well beyond the simplifications of the  stylized naturalism in ordinary films, and such solid features of the  decor as fireplaces are simplified to the barest  possible geometrical shapes, and integrated into the walls by being  covered with the same coating of uniform dark grey  paint. This rather peculiar approach was not copied in other films of  the period.

The films made in Russia during the war by Yevgeni  Bauer are quite interesting from a design point of view, and some of  them closely reproduce what was the most advanced work there in the  interior design of real houses,  mostly that being done by Fedor Shekhtel'. Most of  this does not appear particularly forward-looking today, with one  exception. In Yuri Nagorni (1916) the sets are done by Bauer  himself in a slightly simplified, rectilinear  way that resembles the mature style of Shekhtel', as in his Yaroslavl' Railway Station interior of 1902, and his 1903  project for the new Moscow Arts Theatre, and the furniture in Yuri Nagorni is clearly influenced by the  work of Ivan Fomin from the same period. This is  perhaps not so surprising, as advanced stylized set designs had appeared in the Russian theatre before the war, and Bauer had been a set  designer in the theatre before he  turned to film-making. Also, Ian Christie tells me  that Bauer knew Shekhtel' quite well. It is also worth mentioning that  some of the exterior scenes of Bauer's films have a definite flavour of  the paintings of Konstantin  Somov done in the early years of the century, with  their peculiarly Russian blend of Symbolism, Art Nouveau and  Impressionism.

The film Thais (1916) made by the Italian  Futurist Bragaglia is usually mentioned as the first instance of the use of fully stylized decor, and this does seem to be  correct, though it only applies to one set used in the last few scenes.  The greater part of Thais seems to be a very conventional and  inept entry in the `diva' genre that gripped  the Italian cinema at the time - those films in  which a female star anguished for love in the midst of rich and  glamorous suitors and surroundings, struck Art Nouveau poses, and then  died desperately. (It is just possible  that Thais was intended as parody, but if  that was the case it is still inept.) However, the decor of the final  fatal  room is highly stylized, with the walls covered with sets of alternating black and white rectangles and triangles  nesting inside each other, but, contrary to some  suggestions, the geometrical regularity of these designs sets them apart from true Expressionist art.

Maurice Tourneur's The Blue Bird and Prunella, both made in 1918, were rather more in the mainstream of  cinema. In the first of these films some of the sets were partially done as simplified and stylized scenery painted on  backdrops behind the action area. The style used for this was rather  like some of the most advanced  commercial art of the time, but certainly not in any of the manners used in the most advanced easel painting such as Cubism  or Expressionism, or one of the abstract styles. Other parts of the  design of The Blue Bird went  straight back to nineteenth century Salon painting. In Prunella the stylization of houses, trees, etc. in the decor into simplified  flat patterns was carried much further, with much more consistency. Prunella was also unusual in  that these stylized sets were part of the framing  action, which was set in a fantasy world, whereas the central section of the story was set in the real world, and had realistic sets, so  reversing the usual large-scale  construction of such films. Both films were designed by Ben Carré.


One of the many sets with ‘Toy Town’ stylization in Ernst Lubitsch’s Die Puppe (1919)

Then in 1919 Ernst Lubitsch moved in the same direction with the decor of Die Puppe, though in this case the very definite `Toy Town' stylization of the  sets was justified by the framing presentation of the narrative as  representing the doings of dolls from a toy-box.  Lubitsch's Die Austernprinzessin made earlier in 1919, and  likewise designed by Ernst Stern, also used slightly stylized sets, but  this did not go much further than the  enlargement and geometricalization of the kind of  decorative features to be found on the walls of real houses, etc..  Incidentally, all this happened before Das Cabinet des Dr. Caligari was made at the very end of 1919. (Its  premiere was on February 20, 1920.) The use of stylized decor in Die Puppe may well have suggested a similar approach in Caligari, though Ernst Stern's work had nothing to do with the characteristic forms of Expressionist  painting. The choice of a truly Expressionist style for the design ofCaligari was presumably due to the impact of the stagings of Expressionist plays in the Berlin theatre that year. For instance, Toller's Die Wandlung, which  was premiered on September 30, 1919 had decors by Robert Neppach in a genuine Expressionist style.

Glass Shots and Glass Matte Shots

Although the earliest examples date from the  previous period, extensive use of glass shots did not occur till after  1914, in part because of the poor registration of  cameras prior to the introduction of the Bell & Howell. Norman O.  Dawn made the first glass shots in 1907 by painting additions to the  scene being photographed -- which were  roofs for roofless buildings -- on a sheet of glass  fixed several feet in front of the camera. The progress of the  painting had to be continually checked by examining  the image focussed on the film to ensure that additions to the image  exactly obscured the unwanted parts of the scene, and also exactly  matched the other parts of the scene in  tone and shadow disposition. In this initial form of the technique the camera and glass had to be shielded from  direct sunlight by canvas to prevent reflections in  the glass, and the painting had to be specially illuminated, either  by reflected sunlight or by artificial light. There  were many obvious disadvantages to this process, not least the time  required to make the painting, so in 1911 Dawn introduced a modified  form of the process called the glass  matte shot.

In glass matte painting a sheet of glass is set up in front of the camera as before, but it is not specially shielded or  lit. A matte or mask of opaque black paint is  applied to the glass so as to obscure the unwanted areas of the scene in front, and this can be done rather quickly, checking the image on the  film the while to see that just the  unwanted parts of the scene are covered. Next the  scene is filmed with the action taking place in the areas still  visible through the parts of the glass which are not blacked out, and further lengths of test footage are exposed in the  same way. Back at the studios one of the test sections, but not the main negative, is developed, and then  threaded in the gate of a camera which is set up in  front of an art board on an easel. Light is shone through the  back of one frame of the test film to project an  image of the test film onto the white art board. Then the artist is  free to slowly build up painted additions to the  scene, checking all the while for matching, and he finally blacks out  the parts of the board where the filmed parts of the scene fall. The  resulting painting is then filmed as a  second exposure on the undeveloped negative after a  series of test exposures and developments have been made using the other undeveloped test sections. In this way a correctly combined scene can  be obtained on one  negative after it has been developed

A glass matte shot in Daddy-Long-Legs (1919) combining a real exterior scene in the bottom right corner of the frame with a  painting occupying the rest of the frame.


The successful application of this technique can be seen in Civilization (1917), and the result of trying to make matte combinations in a camera with poor registration can be seen in Birth of a Nation, in the `burning of  Atlanta' scene.


During the war years the trend towards carrying  most of the narrative through dialogue titles used in combination  with the action solidified into standard practice in the American cinema, though all films still continued to use a  small proportion of narrative titles. However, as  with other aspects of film form, there were a few directors who clung to older practices to a greater or lesser extent, and here D.W. Griffith  was one of the extreme cases. He  continued to use large numbers of narrative titles  into the 'twenties when such a practice was quite obsolete. In Europe,  as usual, these developments lagged some years behind American practice, with most directors using  few dialogue titles even in 1919.

It must not be understood from what I have just said that all the lines of dialogue which were visibly spoken by  the actors came to be given in intertitles; what is  at issue is the proportion of dialogue to narrative titles. All films  continued to leave some visibly spoken lines of  dialogue untitled, and as early as 1915 there were films such as David Harum and The Cheat which mainly used dialogue titles, but still left a large number of  spoken lines  untitled. In these films and many subsequent ones  quite active co-operation from the audience was needed to deduce what  might be being said. (I am not talking about lip-reading here, but about purely intellectual deduction , given what had happened in the film up to the  point in question.) In The Cheat one of the many untitled lines  quite clearly contained a proposition which would have been unacceptable to the various censorship boards of  the time, though not because it was obscene in the  strict sense. From this point onwards the pleasure of guessing what was  being said came to be an occasional and intentional feature supplied to  audiences by the brighter film -makers. One of the masters of this device was Ernst Lubitsch, though he did not use it in all of his silent films. His  earliest really distinctive use of untitled dialogue occurs near the  beginning of Carmen (1918).

Art Titles

Even in the early years the development of most  formal and stylistic features of film was gradual, with one or two  isolated instances appearing first, and then over  the next few years an increasing frequency of examples. But the  use of `art titles', which were title cards with  illustrations on them, occurred rather suddenly, without preparation, in 1915. At least two Lasky films of that year, The Girl of the Golden West and Mr. Grex of Monte Carlo have illustrated titles, as does Maurice Tourneur's Trilby. Then in 1916 quite a large number of American films use the device.  Usually the art work, which sometimes covered the whole frame area with  the text superimposed , was an illustration of, or comment upon, the  intertitle, but sometimes it was just a neutral decorative background or border. The style of the illustrations and decorations was almost  always that used in middle-brow  book and magazine illustration of the period, but  the abstract backgrounds to the titles in The Female of the Species (1916) were in the manner of the embryonic Art Deco style, which was the very latest thing at that date.  Art titles never caught on in Europe to any great extent.

A unique way to treat dialogue titles also turned up in this period, but it was not generally adopted for technical reasons. In Dolly's Scoop (J. De Grasse, 1916), the lines of dialogue at the climax of the film  are superimposed  directly over the image of the person speaking them, rather like the sub-titling used in modern films. However, in this case the lines of dialogue were superimposed across the top of the  frame rather than the bottom. Obviously  this would create problems with the production of  foreign versions of a silent film, not to mention the extra difficulty  of carrying out the superimposition in the camera at exactly the point  at which the lines were spoken, so  it is no surprise that there were no more examples  of this technique.

A clever variation on the illustrated title idea  which obviously had no future was the use of live action vignetted into  decorative cut outs around the title in the title card in Twin Pawns (1919).


During the years 1914 to 1919 the range of acting styles used in American films narrowed, basically by the elimination of the last traces of the more exaggerated kind of miming. Acting towards  the camera had been fairly  well eliminated in American dramatic films by 1914,  but this was not altogether the case in European films, and it  is still easy to find examples of this in 1919. In  the previous period Asta Nielsen had established the occasional look  into the camera lens as an element of personal acting style, and in  Russia the famous Mozukhin pushed this  further, with the aid of the more frequent close  shots that were now appearing even in the films of that distant country. The other aspect of European film acting, which had already begun to  appear years before this period,  was its slowness. There were theories about this  sort of acting when it appeared on the stage, and these apparently were  still in vogue in Russia where it had its most extreme manifestations in the films of Yevgeni Bauer . It was just possible to do work in this style that still seems striking, as does Aleksandra Rebikova in Yuri Nagorni, but frequently it just seems like very protracted ham acting to the modern sensibility, as in the case of  Emma Bauer's acting in the same film.

The most naturalistic extreme of American acting  after 1914 occurs in some of Maurice Tourneur's films -- some of the  performances in A Girl's Folly come close to being not acting at  all -- and it is difficult to think of anything  going further in this direction until recent  decades. By 1919 American acting style had developed to a point that  left the acting in D.W. Griffith's films at the more emphatic end of the spectrum. What had earlier been  outstanding invention of acting detail in the  context of the general production of the time was now beginning to show  its contrivance -- the hand of the puppet-master was becoming visible.  In any extended piece of acting by  the young actors in Birth of a Nation one can clearly see that they make the moves and expressions registering  one thought or emotion, then there is a brief pause  before they register the next thought and emotion, and so on. This is  presumably the result of Griffith talking them through the scene, and  although the general and detailed  dramatic construction of his films was still  sufficiently strong to over-ride this flaw, this was not to be the case  in the next decade, when exaggeration in acting was a thing of the past. After 1919 dramatic acting became so  standardized in American films that there are no  more general trends to be discussed, though the fine detail and  differences in individual performances can be profitably considered in  other contexts.

The Rise of Continuity Cinema

The years 1914-1919 in America also saw the  consolidation of the forms of what was to become the dominant mode of  commercial cinema <196> that mode which I shall call for sharpness and brevity `continuity cinema'.  During this period there were other styles that were still important, and these can be considered to lie along a spectrum  between the best examples of `continuity cinema' at one extreme, and at  the other extreme the  `discontinuity cinema' of D.W. Griffith.

There are a number of factors involved in the strong  and apparent visual discontinuities between successive shots in  Griffith's films, and the use of cross-cutting between parallel actions  is only the most obvious of these. Cuts  within the duration of a scene are still relatively  infrequent in his films, and when they do occur they are frequently  from Long Shot or Medium Long Shot (which were the  shots he most used) to a Big Close Up of an insert detail which only  occupied a small part of the frame in the previous shot. This in itself  introduces a fairly strong visual  discontinuity across the cut, but as well as that,  the cut-in shot might often have a circular vignette mask if it were  a Close Up of a person, so reinforcing the effect.  And sometimes the now-standard Griffith iris-out and iris-in  might also be left on the inserted shot, even though it had action continuity with the shots on either side of it. As  well as all this there was Griffith's habit of  moving the action into another shot in an adjoining space, and then back again if it was at all possible, which produced a marked change in  background which also made its small  contribution to the discontinuity between shots.  This discontinuity between shots in Griffith's films can be demonstrated in a particularly striking way by taking a reel from towards the end ofBirth of a Nation or Intolerance and showing it out of context alongside any other climactic reel from a film made by anybody else at  that time or later.

Because of the custom of attributing all technical  developments to D.W. Griffith, the first masters of continuity cinema  are largely unsung, and sometimes even unknown, but it is possible to  mention films that show particular  continuity techniques making some of their early  appearances. One of these techniques involves the exact way the movement of actors from a shot in one location to another in a neighbouring  location is handled. At best this  kind of transition had previously been dealt with by having the directions of travel of the actor in the two shots  correspond on the screen, though there were still some directors in 1914 who could not manage that much. But  in a film such as The Bank Burglar's Fate (Jack Adolfi, 1914), one can see shot transitions in which a cut is  made from an actor just leaving the frame, to a shot of him well inside the frame in an adjoining location, which have the  positions and directions so well chosen that to the casual eye his  movement appears quite continuous,  and the real space and time ellipsis between the  shots is concealed. So thorough-going is the demonstration of barely  noticeable shot transitions (in my terminology, `soft' cuts) in this  film that I am tempted to take it as a  consciously virtuoso performance by the director.  Strangely, this film, so exceptionally advanced for 1914 in this  respect, and also in other respects, entirely lacks  dialogue titles, as the story is entirely supported by narrative titles. Anomalies between the sophistication of the handling of the different  dimensions of the medium are not  uncommon during this period; for instance crude  acting sometimes occurs in films with good scene dissection, but this is the most singular example of this kind I have noted. Other good  examples of this technique for eliminating  several yards of waste space and a few seconds of  waste time can be seen in Ralph Ince's films, particularly The Right Girl (1915), and by 1919 it was widely diffused in American films, but not in those made in Europe.

Exactly the same approach came to be applied to  breaking interior scenes down into a number of shots -- a character  could leave one shot and be picked up immediately several feet away on  the other side of the room in  the next shot, again with apparent continuity. This  became important as more and more of the shots in a scene came to be  taken from close in during the war years in America, but for the  technique to work really well, it was  necessary that there also be a substantial angle  change between the two shots. This is because if both shots were taken  directly from the front, the omission of several feet of the actor's  path across the room would be more  apparent from the obvious sudden background change.  All this connects with the rise of the use of cutting to different  angles within a scene during the years 1914-1919, and in particular to  the development of reverse-angle  cutting.

Reverse-Angle Cutting

It was only in 1915 that cutting to different angles within a scene became well-established as a technique for  dissecting a scene into shots. As already described, this approach had appeared a few times in earlier years, but in general cuts to or from a closer shot within a scene were still being made more or less down the lens axis as  established in the Long Shot of the scene in  question. There were a few instances in which the disposition of objects within the filmed scene were such as to prevent the camera being moved  absolutely straight forward to  take the closer shot, but the deviations were never  so great as to have the camera shooting in the opposite direction. This  applies to D.W. Griffith as well as nearly everyone else, but I must  make one more exception to  this generalization, and this is in connection with  scenes taking place in a theatre. In such cases cuts with a change  of direction of approximately 180 degrees between  shots of the audience, and of the show they were looking at, were used  even in Europe before 1914.

The leading figure in the full development of  reverse-angle cutting was Ralph Ince, who has already been mentioned in  this connection in the previous chapter. Films that he made at Vitagraph in 1915 such as The Right Girl and His Phantom Sweetheart show him putting the final polish on the technique of using a large  number of  reverse-angle cuts in interior, as well as exterior, scenes. Other directors were also just starting to take up this style  in 1915, for instance Reginald Barker in Bad Buck of Santa Ynez,  but none matched Ralph Ince's  command. It must be emphasized again that this  development has nothing to do with Thomas Ince, for the films he most  closely supervised, such as Civilization (1916), lack the features I am discussing, and indeed it is quite  possible that Thomas Ince was responsible for the other positively retarded features of Civilization. As for Griffith, in Birth of a Nation there are just eight cuts to reverse-angle shots in the scene in Ford's Theatre, while  elsewhere throughout the two-and-a-half hour length  of this film there are only four more true reverse-angle cuts. (I define a reverse-angle cut as one in which the camera direction is changed by  more than 90 degrees, which  corresponds closely to the way film-makers use the  term.) None of these cuts occur at any of the major climaxes in Birth of a Nation where they would be most effective, such as the pursuit of Flora Cameron and her leap  from the cliff, whereas there are more than a dozen such cuts within the ten minute length of Ralph Ince's His Phantom Sweetheart.

WARNING Since Birth of a Nation is such a frequently seen film I must point out that to the uninstructed glance there might appear to be more reverse-angles in it  than I have stated, but careful consideration of the relative  positions of the actors will show that in what might at first appear to be possible instances of reverse-angle cuts  the camera is in fact shooting from almost exactly  the same direction in the adjoining shots; i.e. from the `front'.

Nevertheless, the Griffith style of film-making was  still followed in its full idiosyncrasy, with extensive use of side by  side spaces and a definite `front' for the camera, in most slapstick  comedy, and this was because of the  success and influence of the Keystone company, which was already rigidly using this style before 1914. Directors of dramatic films such as James Kirkwood, Lloyd Ingraham, and W. Christy Cabanne,  who had all  previously worked for Griffith, also followed his  style fairly closely, though by 1916 Ingraham could sometimes manage to  use the occasional reverse-angle cut when the two shots concerned also  formed a watcher-POV pair.  In fact the Griffith style, with only a slight  weakening of his relentless frontality of scene dissection, was the  standard for films made by his Fine Arts section of the Triangle  company, and was followed by all who worked  there. D.W. Griffith's prestige ensured that many  American film-makers elsewhere were very slow to adopt true  reverse-angle cutting during this period, and on into the years after  the First World War.

By 1916 there are a number of films in which there  are around 15 true reverse-angle cuts per hundred shot transitions --  which I shall refer to as 15% reverse-angles -- and two such are The Deserter (Scott Sidney) and Going Straight. By the end of the war such films form an appreciable but minor part of production: e.g. The  Gun Woman (F. Borzage, 1918) with 18% reverse-angles, and Jubilo (Clarence Badger, 1919) with 16%, and  by that date most directors of quality films were  making more use of reverse-angle cutting than D.W. Griffith did, though  they tended to restrict the device to one or two major climaxes in their films. Anyone who did not move  with this trend when it became dominant in the next  decade was in danger of having their films look old-fashioned, and such  was the fate of D.W. Griffith himself. Other qualities in a film could  surmount this handicap,  but not if it was combined with yet other retarded  stylistic features, and old- fashioned subject matter as well. All this  hardly concerned European cinema, where those few reverse-angle cuts  used were mostly between a  watcher and what he sees from his Point of View,  both being filmed in a fairly distant shot. However, after the end of  the war some of the brighter young directors such as Lubitsch started  using a few reverse- angle cuts,  mostly in association with Point of View cutting.

Cutting On Action

A major feature of `continuity cinema' was the  establishment of cutting on action as a standard way of smoothing the  transitions between cuts within a scene. This meant making the cut to or from a closer shot, not when the  actor concerned was more or less stationary, as had  usually been the case, but when he was in the middle of a definite  movement, and as well as that, making sure that the movement across the  cut had reached exactly the  same point, to the very frame, in the shots on each  side of the point where the cut was made. As in other aspects of the  development of continuity cinema, a leading figure was Ralph Ince, and  his 1915 films contain a number of  demonstrations, for the first time, of how to do  this in a number of standard situations. His Phantom Sweetheart, The Right Girl, and The Juggernaut use perfect cuts on action in such places as the middle of the movement of a person sitting down in a chair, or when they  were making some other sort of broad body movement, and increasing  numbers of other American film-makers took this up over the next few  years.

Editing Equipment

Those film-makers concerned with the development  of continuity cutting seem to have felt the need for some mechanical  assistance with the editing task under these new stylistic conditions,  for around 1916 the first editing  viewers appeared in the United States. Initially  these machines were no more than a projector film-gate through which the film was pulled by the usual intermittently moving sprocket wheel,  which was driven by the Maltese  Cross gear mechanism which was now becoming standard in projectors. The gear train was driven in its turn by a small  crank-handle at the side of the device, and the frames passing through  the aperture were viewed through  a magnifying lens supported a few inches in front of the film by a tube attached to the front of the gate. The whole  device was only several inches high and was mounted  on a little stand which could be put on the top of an editing  bench. As the film was cranked through, it had to be fed into the bottom of the gate from a small roll held in the hand, and illumination of the frame of film in the gate was from behind in some  sort of ad hoc manner. There  were no loops of film formed in the machine to  smooth out the intermittent motion through the gate, so the editor had  to keep unrolling the film from the feed roll so that there was no  tension between it and the machine. This  was not too difficult to do for small rolls of film. No doubt this machine was only used to deal with the most tricky points of action-matching across a cut when the figures were small in the  frame, since it is actually quite  possible to do good continuity cutting `in the hand' most of the time, with no aid other than a simple magnifying glass, as  had been done before, and as still continued to be done.

The Use of the Insert Shot

As already described, the use of Insert Shots --  Close Ups of objects other than faces - was established very early, but  apart from the special case of Inserts of a letter that was being read  by one of the characters, they  were infrequently used in American films of the  previous period, and hardly at all in European films. It was also before 1914 that D.W. Griffith had begun to bend the use of the Insert towards truly dramatically expressive  ends, but he had not done this often, and it is  really only with his The Avenging Conscience of 1914 that a new phase in the use of the Insert Shot starts. As well as the symbolic inserts I have already mentioned, The  Avenging Conscience also made extensive use of  large numbers of Big Close Up shots of clutching hands and tapping feet  as a means of emphasizing those parts of the body as indicators of  psychological tension. Griffith  never went so far in this direction again, but his  use of the Insert made its real impression on other American film-makers during the years 1914-1919.

Cecil B. DeMille was a leading figure in the further  development of the use of the Insert, and by 1918 he had reached the  point of including about 9 Inserts in every 100 shots in The Whispering Chorus. He also pushed the  insert into areas of visual sensuality inaccessible  to D.W. Griffith, with such images as a Close Up of a silver-plated  revolver nestling in a pile of silken ribbons in a drawer in Old Wives for New (1918).

The impact that the increased use of the Insert Shot  had at the time is difficult to recapture now, for at that date there  had never before been accurate images of relatively small objects  presented with such definition and  enlargement in any medium, be it painting,  photography, or whatever. Things like pistols when shown in Big Close Up could be several times the size of a real pistol when held at arms  length, and for instance in Her Code  of Honour (John Stahl, 1918), the scratches on  the metal and the movement of the internal parts as the trigger is  squeezed can be quite clearly seen in an Insert Shot of an automatic. Since the evolution of the use of the Insert had been  quite gradual in the United States, there was no comment upon it there,  but in France a number of  young aesthetes felt its full force in 1917, when  the American films that had been withheld by the war during the previous three years were suddenly released to the public. Louis Delluc and  others then explicitly formulated the idea of the Point of View  shot and the Insert in their critical articles, and this had a  significant influence on the  development of the so-called `French avant-garde' of the early nineteen-twenties. (Detailed information on this subject can  be found in French Film Theory and Criticism by Richard Abel  (Princeton University Press, 1988 ). When Louis Delluc and others of like mind came to make films after the war, the fact that they had conceived of these  sorts of shots as a separate idea tended to promote their use in a more  isolated and discontinuous way  than in their original source. Combining this with  the influence of Griffith's cross-cutting in its most extreme form in Intolerance helped to promote a European avant-garde cinema of discontinuity which was some distance  apart from the mainstream of continuity cinema that had already formed in the United States.

The Atmospheric Insert

Like many other devices that were more fully  developed in Europe during the next decade, what could be called the  `atmospheric Insert Shot' made its first appearance in American films  during the years before 1919. This kind  of shot is one of a scene which neither contains any of the characters in the story, nor is a Point of View shot seen by one of them. It first appears to my knowledge in Maurice Tourneur's The Pride of the Clan (1917), in  which there is a series of shots of waves beating on a rocky shore which are shown when the locale of the story, which is  about the harsh lives of fisher folk, is being introduced. Simpler and  cruder examples from the same year  occurs in William S. Hart's The Narrow Trail, in which a single shot of the mouth of San Francisco Bay taken  against the light -- the Golden Gate -- is preceded  by a narrative title explaining its symbolic function in the story.  This film also contains a shot of wild hills and  valleys cut in as one character comments that the country far from  the city is so clean and pure. By 1918 we can find a shot of the sky being used to reflect the mood of one of the characters without specific explanation in The Gun Woman (Frank Borzage),  but it must be emphasized that  these examples are very rare, and did not either  then, or within the next several years, constitute regular practice in  the American cinema. The Tourneur example just mentioned also could  stand as part of the beginning of the  `montage sequence', which probably had its true  origin in American films during this period. Another case that has  crossed my attention is in The Woman in 47, which includes a  chain of shots joined by fades discovering the  heroine in the middle of typical New York scenes, as she discovers the city for the first time. Maurice Elvey's Nelson - England's Immortal Naval Hero (1919) has a symbolic sequence dissolving from a picture of Kaiser  Wilhelm II to a peacock, to a battleship, which is  probably more startling now than then, given our awareness of  Eisenstein's subsequent films.

The atmospheric Insert began its notable career in European art cinema in Marcel L'Herbier's Rose-France. Here amongst the intentionally `poetic' uses of  vignettes and filters and literary intertitles, a shot of the empty path once trod by the lovers is used to evoke the past.

The Flash-Back

The fashionable interest in the flash-back  continued into this period, and it could now be entered with very little preparation, as in Between Men (Reginald Barker, 1915). In this  film the hero reads a letter which refers to a  past incident in his life -- we see the letter in an Insert Shot -- then after a cut back to him sitting thinking, there is  a dissolve which goes straight into a representation of the past scenes referred to in the letter, without any explanatory  titles occurring at any point. During these years the usual way of  entering and leaving a flash-back  was through a dissolve, and this was in fact the  principal use at this time for this device.

(The subsidiary use for a dissolve was to bridge a  suspected mis-match in actor position on a transition from a Long Shot  to a Close Up, and although the technique of American directors and  actors was already sufficiently  good to render this unnecessary most of the time,  there are enough occurrences of this usage in 1915 and 1916 for it to be described as standard.)

On the other hand the dissolve was still not  being used to denote a time-lapse, though there are one or two films in  1914 where it does happen to correspond to a time-lapse as well as to  other things. In that year the  enthusiasm for the new possibilities of the medium  led to considerable complexity being crammed into one reel of film, as  in The Family Record (Selig, 1914), in which an aged man and  woman separated for most of their life  have his flashback and then hers shown in succession within the framing story. In fact fully developed flashbacks occur in  more Selig films during this period than in those from any other company contained in my sample. The  Vitagraph company's The Man That Might Have Been (William Humphrey, 1914), is even more complex, with a series of  reveries and flash-backs that contrast the protagonist's real passage  through life with what might have  been, if his son had not died. In this film  dissolves are used both to enter and leave the flash-backs, and also the wish-dreams, and also for a time-lapse inside a  reverie at one point. But fades are also used for these purposes  in this and other films of the period, and flashback transitions are also done with irising in other films, and even  straight cuts in Bauer's Grezy and Posle smerti, so that  all that one can say on the basis of these examples is that  the understanding of a particular transitional  device depended totally on the context. To reinforce this point, I will  mention what seems to have been a unique occurrence of a novel way of  getting into a flashback during this  period. In The On-The-Square Girl (F. J.  Ireland, 1917), a flashback is shown as a succession of scenes inset  into the centre of a letter which one of the  characters is reading. Since this is a fairly standard sort of film, it  would seem that this device was expected to be as understandable to an  audience then, on its first occurrence,  as it is now. This kind of lack of regularity in the significance of style features, which was to become even more marked  with the emergence of the avant-garde in the 'twenties, is one of the  main reasons for the failure of  attempts to create a science of film considered as a language system. This is not to say that aspects of film cannot  be studied by scientific methods, or that there are  no regularities in the forms of films at all, but just that these  regularities are insufficient, and also change too fast, to be  considered as a language system.


The beginning of the flash-back scene in The On-the-Square Girl done as a series of shots inset into the middle of a letter  recalling the past events in question.

The fashion for flash-backs at the beginning of this period was such  that one gets some instances where the use of flash-back construction  was completely pointless, but on the other hand there are instances  where an  extensive series of flash-back scenes serves a  contrasting function essential to the plot, as in Silks and Satins. During the war the use of flashbacks occurred in  films from all the major European film-making countries as well, from  Italy (Tigre reale) to Denmark (Evangeliemandens Liv) to Russia (Grezy and Posle smerti), where it  arrived in 1915. As the years moved on a sudden  decline in the use of long flash-back sequences set in around 1917, but  on the other hand the use of a transition to and from a brief single  shot memory scene remained quite  common in American films. However, I have come  across one more final example of complex flash-back construction in  American films in the case of W.S. Van Dyke's The Lady of the Dugout (1918). This film has a  story that happened long before narrated by one  character in the framing scene, and initially accompanied by his  narrating dialogue in intertitles, though after a while this stops, and  the intertitles then convey the dialogue  occurring within the flashback. Inside this main  flashback there develops cross-cutting to another story, happening at  the same time, and at first apparently unconnected with it, though the  connection eventually appears . Next, inside this first flashback, the Lady of the title narrates another story, presented in flashback form, but  with cut aways inside it back to events occurring in the time frame in which she is doing her narrating. Actually,  all this is fairly easy to follow while watching the film, in part because what happens in all these strings of action is  relatively simple.

Cross-Cutting Between Parallel Actions

After 1914 cross-cutting between parallel actions came to be used whenever appropriate in American films, though this was not the case in European films. It should be noted that a good deal of  the American use of cross -cutting was not the rapid alternation between  parallel chains of action developed by D.W. Griffith, but a limited  number of alternations to make it possible to leave  out uninteresting bits of action with no real plot function. In Europe,  some of the most enterprising directors did use cross-cutting sometimes, but they never attained the  speed of many American examples, and their lack of  ease with it is indicated by the fact that some of them felt it  necessary to make the initial transition to the  first shot of the alternate strand of action with a fade, as in Benjamin Christensen's Haevnens Nat (1916) and the Cines company's Il sogno patriottico di Cinessino. And in 1918  the quite experienced Russian director Protazanov  still found it necessary to cover important simultaneous action inside  and outside Father Sergius' cell in the film of the same name by having  the wall of the set split apart to  show these actions at the same time, rather than by  cutting between them.

In the United States some directors became so  enraptured with the idea of cross-cutting that they sometimes used it  when it was not really necessary, and contributed nothing to the film;  in other words, when nothing of any  significance was shown happening in the alternate  action, and no acceleration of the main action was accomplished either.  One example of this is contained in the Selig company film, The Lost Messenger (1916).  On the other hand, cross-cutting was used to get new effects of contrast, such as the cross-cut sequence in Cecil B.  DeMille's The Whispering Chorus, in which a supposedly dead husband is having a liaison with a Chinese  prostitute in an opium den, while his unknowing wife is being remarried in church. Or the sequence in The Female of the Species (Raymond B. West, 1918), in which a man is crawling into a woman's  sleeping-berth on  a train while in the cross-cut scene another train  is speeding towards them in the opposite direction on the same track.  The crash comes as they embrace.

Of course all this was simple compared to The Master's Intolerance, in which four parallel stories are intercut  throughout the whole length of the film, though in  this case the stories are more similar than contrasting in their  nature. The use of cross-cutting within these  parallel stories as well as between them produced a complexity that was  beyond the comprehension of the average audience of the time, and  effectively though unintentionally turned Intolerance into the first avant-garde film masterpiece. (Only loosely speaking, since Intolerance was intended  to be commercially successful, whereas real avant-garde films are not.) The influence of Intolerance produced a few other films that combined a number of similar stories having similar themes, such as Maurice Tourneur's Woman (1918), but the box-office failure of Intolerance ensured that these later films had simpler structures.  The true line of descent from Intolerance curves away from the mainstream through Abel Gance's la Roue (1921), and some of Eisenstein's films, to the real avant-garde.

Scene Dissection

Another new fashion of 1915 was the practice of  beginning scenes with a close shot of some detail in them, and only then tracking or cutting back to show the whole scene, rather than following the usual practice of starting  with a general shot, and only then cutting in  closer. The first example I have come across is in the Thanhouser  company's The Center of the Web, released at the very end of  1914, though this may not be where the idea  started. This film begins with an insert shot, and  then the camera tracks back to reveal the whole scene. Other instances  of this new idea can be seen in David Harum (Allan Dwan) and Elsa's Brother (Van Dyke Brooke),  released in 1915. A couple of years later, the Franklin brothers' Going Straight includes a scene which starts  with a series of Close Ups of actors interacting  with each other, before where they are doing this is revealed. Although  not common, once it had been established, this new variant in the way of dissecting scenes never  completely vanished after this initial burst of  enthusiasm, but has been returned to from time to time ever since by  imaginative directors.

The possibility of breaking a scene down into shots  in markedly different ways that now existed in the American cinema was  intimately connected with a number of developments, one of which has  already been mentioned,  namely the use of reverse-angle shots. Also involved was the general tendency to cut scenes up into more and more shots, and along with this the tendency to use a greater proportion of close  shots. All these developments  are obviously interconnected to some extent, but  perhaps surprisingly they could also be relatively independent. And the  films in which each of these different tendencies was most prominent may also be found a little surprising . For instance, by 1918 there were more shots per  hour in a Kay-Bee Triangle film such as The Hired Man (Victor Schertzinger) than in Griffith's Broken Blossoms, while Jubilo (Clarence Badger, 1919) and Until They  Get Me (Frank Borzage, 1917) are shot from much  closer in throughout their length than contemporary films by the  best-known names of the period. And Badger and Borzage used far more  reverse-angle cuts than Cecil B.  DeMille, while in his turn the latter used more  Medium Long Shots than D.W. Griffith, who tended to avoid this range of  camera closeness by 1918.

When I add that other films by other directors were  now using various other combinations of these variables of film style,  the response might well be to ask for a better, briefer, and clearer way of handling and describing all  these matters than the imprecise words I have used  up to this point. I shall now begin to provide this new approach in the  following chapter.



This is a typical chapter from the book, but with some of the illustrations left out. If  you are interested in early cinema, the previous chapter (1907-1913)  contains even  more important stuff that is not available  elsewhere.