4
The Films: Alternate Scenes

They [the films] are handed out over the counter like so many feet of sausage.

—Moving Picture World, 31 July 1909, p. 151

American films began to undergo a series of changes in 1908-1909 that would be Aas radical as those at any time in film history, marking as they did a major shift in the perception of the nature and function of the motion picture. Rather striking evidence of this shift is provided by Edwin S. Porter's The Life of an American Fireman (Edison, 1903), which was actually reconstructed sometime after its initial release, either to modernize the film or because the original order was thought to be a "mistake." But as modern researchers have now discovered, it is the reconstructed prints that are in error: in them, the fire rescue scenes inside the room and outside the house have been intercut to show their simultaneity, while in the 1903 version, the action of the woman and baby being rescued through the window was shown twice, in consecutive shots, once from the interior and again from the exterior, in the same way that such scenes were shown in slide shows. It is not known when the reconstruction took place, but the same kind of "correction" can be found in several other films made before 1908.¹

As suggested in previous chapters, one important reason for such changing perceptions was the new audience for the motion picture. To be sure, the pre-1908 films were the ones that had attracted that audience to the nickelodeons in the first place. As Charles Musser has demonstrated in a study of the Edison sales records, it was the emergence of the story film, in 1904-1906, that drew people to the nickelodeons. Even though greater numbers of actuality films were still being produced at that time, many more copies of story films were sold. The attraction of the story film for the new audience was finally understood by the producers and sharply reflected in film production after 1907. The movie fan, the one who attended every day or went from show to show, existed before there was a star system or a fully integrated narrative system. It was this audience that presented at the same time a demand for stars, for a more complex narrative, and for more clarity in narrative techniques.²

By 1908 it was evident that the popularity of the story film was presenting new problems for the filmmakers. The most common criticism of specific films concerned the need for clarity. The Moving Picture World reviewer noted, " 'The Devil,' 'Dr. Jekyll and Mr. Hyde', etc., are clever plays but they have been presented in motion pictures in a way that the public do not understand them. The spectators cannot follow the plot and therefore they lose interest." Said a commentator on the task of the filmmaker in 1908, "His plot must be simple, the difficulties into which the characters are led as obvious as the sun, and the solution of them intelligent to the lowest understanding." Another, praising Edison's Saved by Love, remarked, "As the spectators could follow the plot, without the help of a lecturer, they were deeply interested." The help of the lecturer was available only in certain theaters, not in others, or was called upon only for special films. Films that could be understood without the lecturer to explain them could be shown everywhere.³

A greatly expanded audience from diverse cultures no longer had the same frame of reference. This explains some of the complaints about lack of clarity that were heard with great frequency in 1907-1908. Filmmakers could no longer expect the majority of the spectators to recognize the narrative events of a classic tale, a work of literature, a popular play, a familiar myth, unless they were in some way explained. Stereotypes of character and gesture would have to be reestablished for the new audience.

The second reason for change, and the one that the filmmakers all understood, was the uplift movement. Producers were being urged to make films that would be morally improving and educational for the mass audience. How was this to be done? Not the only but perhaps the most effective way would be to enlist the emotions of the spectator in a story. The film would have to carry a lesson or preach a sermon, and to do that it would have to learn to be expressive. The advantage of this method of educating was that it also would attract the audience, and that made good business sense. One could make moral and educational films and still lose the audience if they were dull or if the audience could not understand them.

A third cause of change was the need to standardize film production, to bring out films of the same length and within established genres. The film as standardized product would be one that was clearly understood without the help of the lecturer; one that would be reliably available on a regular schedule, that the exchange and the exhibitor could count on; one that would draw the consumer to brand names, that would be familiar; and one that would be under the control of its producer. The standardized film would make it possible to rationalize production methods and would be more profitable for the producer.

Another contribution to change was the fact that most of the new directors were entering the expanded motion-picture field with backgrounds in the legitimate theater and the road-show melodrama. The previous generation had been chiefly businessmen, technicians, and cameramen, with some variety-show people, including magicians. The new producers and directors, without previous film experience, were much more apt to draw their models from the theater than from the variety shows, lantern-slide shows, comic strips, cycloramas, and all those varied pre-film sources that had inspired earlier filmmakers. It will be noted that changes in the style of filmmaking tended to come from the new generation, a D. W. Griffith, a Ralph or Thomas Ince, a Herbert Brenon, or a Sidney Olcott. Those who had begun in the early days, such as Edwin S. Porter or Alice Guy Blaché, were much likelier to hold onto the earlier styles.⁴

Throughout the period from 1907 to 1915, there were many who believed that the mission of film was to bring theater to the new audiences. They thought it signaled a great cultural advance for the masses. In 1913, Frank Dyer, a man involved in the administrative end of the industry for many years, could only describe the art of the film as photographing great stage plays. But even in 1908, a few people were able to articulate some of the differences between theater and film as they were then understood. One of them was Rollin Summers, writing in the Moving Picture World. In his view, several important characteristics were unique to the motion picture: (1) the silent film cannot explain action that happened before it begins or that happened elsewhere—it must be shown; (2) film cannot show precise mental states, but it can better show elemental emotion; (3) film, using real scenery, gains in atmosphere and heightens illusion; (4) where emotion is to be expressed, the film can be taken at close range: "The moving picture may present figures greater than life size without loss of illusion"; (5) film can have as many scene changes as wanted: "The principal characters, having once been well identified, may be separated and the scene may shift from one to the other and back again. If the sequence of the scene is well contrived there is a decided gain in the quality of the action and a perfected illusion of reality in the method." These were all prescient comments concerning the way in which film was going to develop, but it is the last that I would like to examine in this chapter.⁵

From this time onward, most people assumed that movies aspire to reality, for reasons having to do with the new emphasis on the story film. In earlier days, moving pictures were accepted as a spectacle, a magic show, an amusing entertainment. Verisimilitude was one of film's entertaining characteristics. The first audiences could marvel at how "lifelike" the motion picture was when it documented the world around them. When filmmakers began to relate more complex stories, however, there was a shift in attitude. Reality began to be demanded for the staged fiction film. The screen is not reality, of course, and every spectator knows that very well. What the fiction film came to represent was a dream world, founded on an illusion of the real world. No other medium could give such concrete "evidence" of reality. From this point in history, anything that dragged the spectator out of that dream was subject to criticism on the grounds of breaking the illusion of reality. When the French firm Pathé Frères attempted the popular American genre of the Western with A Western Hero (1909), it was severely criticized by a reviewer because "[The audience] like to believe that the pictures are real photographs of the real scenes, and the public mind seems to resent any jolt which awakens them out of the pleasant dream." The details of the dream are concrete and must be consistent: dream logic must be observed, lest illusion break.⁶

Those who gave away the secrets of how movies were produced were criticized by others, who thought this to be unfair to the audience. Such an idea reminds one of the earlier concept of the cinema as magic act: the magician does not give away the secrets of his profession. In fact, audiences were fascinated with the details of how movies were made, as long as this information came from outside the film. But for the duration of the film, the spectator wanted to be able to suspend disbelief. In the fall of 1909 a spectator at the Bijou Dream on Fourteenth Street in New York described the rapt audience around him: "A tense, well-knit, immobile mass of human faces, with eyes alertly fixed on the screen." The audiences were caught up by the new narrative systems, held in a way they never had been before.⁷

To understand what the fundamental changes taking place in this period were, it may help to begin by examining more closely the films of 1907 and 1908 as the existing models on which the new films would be based.

All films in 1907 were one reel or less in length, from only two or three hundred feet to just over one thousand feet, with the rare exception of passion plays (Biograph's last release of a foreign-made film was Gaumont's two-reel La Vie du Christ in 1907) and the occasional prizefight film. There was room for variation within these lengths, but during 1909 and 1910, the lengths would become standardized. The time came when film after film directed by Griffith at Biograph would measure precisely 990 to 998 feet, or else a reel of just the same length would contain two films known as "split reels." This kind of regularity marked the films made for the Biograph Company, where standardization was perhaps more rigorous than at other companies. Nevertheless the tendency existed with all the producers, and as we shall see in chapter 12, exhibitors would complain if there was too much variance.

The comedies, chases, and actualities of the previous period were still the basic material of the films produced, but as Tom Gunning has suggested, the vein was becoming exhausted. The limits of these forms may have been reached and audiences may have tired of them.⁸

To be sure, many variations enriched the chase formula. For an example, there is Edison's charming Jack the Kisser (1907), in which an overenthusiastic masher snatches kisses from unsuspecting ladies on the street and in the park, in one episode after another, until the actions culminate in the chase sequence typical of the genre. Here too there are fresh variations in the unusual locations, such as the stacked sewer pipes in a factory yard, through which the pursuit continues. The film exemplifies the expansion of narrative in this period by the adding on of episodes with a common motif—in this case, the kiss thief. Since the episodes are variations on a repetitive theme, the film may be lengthened or shortened according to the number of episodes the filmmaker chooses, without altering the narrative conception. Whenever the film is long enough, a final climax may be added to give closure to the narrative.

John Fell has called this kind of structure "the motivated link," while Tom Gunning prefers the term "linked vignettes."⁹ The same loose structure may be found in many comedies of the period. Biograph's Trial Marriages (1907) shows the comic effect of the inspiration a man gets from reading a newspaper article by a woman of advanced ideas recommending trial marriages. He then gives a tryout to a series of potential wives, all of whom have various faults. Repetition of the motif provides a relationship from shot to shot, one that is easily broken. It would not make much difference in which order the various "tryout wives" were shown, or if one were dropped or another added.

The chase comedy of 1907, however, maintained a formal relationship from one shot to the next that was not felt to be necessary for the other types of linked-episode comedy. The shots of the early chase films were linked by the direction of the pursuit in relation to the camera position. More often than not, shots show the participants in the chase, both pursued and pursuer, approaching from the distance and exiting in the foreground, to one side or the other of the camera, with the action and its direction repeated in the next shot but in another location. The shots are linked by this movement through space. Variations included reversing the direction of the chase or entering near the camera and running into the distance, as well as crossing numerous obstacles. This construction, familiar to audiences through endless repetitions, suggests a model for new narrative films. It connotes continuous space-time unrolling from one shot to the next.¹⁰

The type of story that can be told with the motivated link or by the movements in space of a chase comedy is limited, even with imaginative variations. In 1907 comedies were the most common fictional form. However, a handful of melodramas were also made that year, and producers of this genre found it necessary to explore other options. As they attempted to narrate more complex stories, viewers were beginning to notice a lack of connection. Alfred Capus, a French playwright involved in the Film d'Art movement, was quoted as saying, "If we wish to retain the attention of the public we have to maintain unbroken connection with each preceding scene. " By February 1909 a writer in the Moving Picture World could say that "In the past few months there have been put out very many really beautiful, dramatic, tragic and serio-comic films. Films with real acting in them, which tell a beautiful and connected story." Yet the following month, with some fundamental elements of the new construction already in place but used with unequal skill by various filmmakers, another World commentator complained that "many film pieces that are produced fail because of some obvious disconnectedness in action." Even at the end of the year the World carried an editorial on the poor construction of many films, the lack of clarity, the lack of relation to printed synopses, quoting a spectator as remarking at the finish of a film, "Bit disconnected, isn't it?"¹¹

The development of new ways to connect shots, or editing, was probably the most important change in film form to take place during the 1907-1909 period. Creating a spatiotemporal world, a kind of geography made of separate shots related to one another, was crucial in the construction of a complex narrative. The development of new editing methods would also greatly increase the potential for enlisting the spectator's emotions in the film. The new forms would integrate the spectator more deeply into the film experience. Louis Reeves Harrison's articles on the fine art of films were usually incoherent, but he was uncharacteristically acute when he noticed, in late 1910, that when the vaudeville act was over, the lights dimmed, and the pictures went on, the audience became quiet and all attention. "There can no longer be perception without attention," he said; "the loss of one scene means a chapter torn from the story."¹²

In attempting to analyze the emerging forms of editing, David Bordwell, in The Classical Hollywood Cinema (1985), has proposed a distinction between what he calls crosscutting and parallel editing. He defines crosscutting as "the intercalation of two or more different series of images. If temporal simultaneity is not pertinent to the series, the cutting may be called parallel editing; if the series are to be taken as temporally simultaneous, then we have crosscutting." In another part of the same book, Kristin Thompson recapitulates this formulation a bit more simply: "Part One has defined 'crosscutting' as editing which moves between simultaneous events in widely separated locales. 'Parallel editing' differs in that the two events intercut are not simultaneous."¹³ (I would suggest that "widely" should be dropped from this definition. "Separated," yes, but the distance of separation might be only inches: imagine a sequence alternating on the two sides of a wall, for example.)

These definitions have a certain kind of logic that clarifies two functions of alternating shots. It should be noted, though that since the thirties at least, the term "parallel editing" has been used for both functions and especially for the kind of cutting Bordwell and Thompson have called "crosscutting. " To change the definition at this point in time may be too arbitrary and confusing; perhaps it would be more useful to differentiate the types of editing with some new terminology.

In any case, during the period under discussion, when all kinds of editing were new and not yet established as conventions, the distinctions were not always articulated clearly in the films or in the language. Yet it might be suggestive to look at the terminology of the time, for what it reveals of the intentions of those who used it. The terms "parallel editing" and "crosscutting" do not appear at all. In 1908, when such editing was quite new, it was referred to as "alternate scenes." That was the term first used to describe the sequence designed to increase suspense, showing two or more simultaneous streams of action that will come together in the resolution.¹⁴

In 1910 a New York journalist explaining the advantages that moving pictures had over vaudeville and melodrama wrote, "The divided stage, with its broken partition end on to the spectator, may also be sent to the lumber room. [He refers to the divided screen found in many early films.] The motion picture has it in its power by alternating scenes to show us what is going on simultaneously in two different places, inside and outside a house, for example, or in adjoining rooms."¹⁵

The same term is very often found in the advice of the World to musicians who accompanied the films: "Some pictures are shown in which the scenes alternate so rapidly as to make it impractical to change music with every change of scene." Or, suggesting music for A Dixie Mother (Vitagraph): "The next scenes alternate quickly. … This journey is shown in a number of scenes, and after each one the mother is shown. " In 1911, commenting on "a really good picture," Selig's The Mission Worker, the music columnist described how "the scenes alternate between Chinese scenes, a sick-room, and a parlor—many of them rather short." (The advice given to the musician in the case of rapid alternation of scenes was to select the dominant mood and stick with it.) In any case, the term "alternate scenes" was clearly understood by everyone.¹⁶

As for the kind of cutting that Bordwell and Thompson would call "parallel," the interweaving of events that are not specifically related in time, the terms often used at the time were "inserts" or "flashes," reflecting the usual brevity of the shots thus edited. At first, "insert" was applied to extreme close-ups of letters or newspaper articles used to forward the narrative, and also to extreme close-ups of objects, such as photographs, when it was necessary to identify them. As the concept of editing developed, the term was used for any shot that was "inserted" into the chronological sequence of events to show another time or place. In the beginning of this period, inserts were considered to be interruptive of the narrative flow. Although they were seen as a more "realistic" way of giving narrative facts than intertitles, it was nevertheless thought desirable to minimize their use and to keep them short when they were used, so as to interrupt as little as possible. (The terms "alternate" and "insert" will also be familiar to readers of Christian Metz.)

The famous 1913 advertisement in which public-relations men claimed for D. W. Griffith the invention of practically everything in use at the time refers to " 'the switchback,' sustained suspense." The "switchback," here, refers not to the flashback as we know it, but to a cut to the scene, in another location, that immediately preceded it.¹⁷

An English writer in a 1916 manual for scenario writers actually uses the term "flashback" for "the keeping of two apparently distinct stories running at the same time and not allowing them to converge until the time is ripe for them being dovetailed." This technique, the manual indicates, "will allow the writer to switch away from the scene when he wishes, either to show a lapse of time, or … the end of… action." (At the time, what we know as flashbacks were more apt to be called "memory flashes," indicating how they were usually understood, as the memory of a specific character.)¹⁸

The term "cut back," meanwhile, is clearly explained from a 1915 vantage point in Louella O. Parsons' How To Write for the "Movies," although she may have the chronology of its function a little wrong:

In the beginning, the cut back was used … to avoid showing the actual theft or murder or to close up a break in the action. Later, the cut back was used to give added surprise and intensity to the plot. The cut back refers to some particular scene that has gone before. The interest and suspense of the audience is intensified by using the cut back. … Griffith changed [the practice of direct continuity] by flashing from one scene to another, by a series of short scenes … keeping each character before the audience. … It is not used as much as it was formerly. The new school of motion picture acting seems to favor longer scenes and more deliberate action. The flashes or cut backs are still used when necessary, but not to the entire exclusion of longer scenes (How to Write for the "Movies" [Chicago: A. C. McClurg & Co., 1915], pp. 48-49).

In her explanation, Louella Parsons gives an additional motive for using alternate scenes: to avoid images that would be censored. Epes Winthrop Sargent had already suggested this idea in his advice to photoplay authors in 1911. He noted that a fight or struggle with a burglar might not pass the censors, but there would be no objection if it were broken "by contrasting scenes showing the police hurrying to the aid of the victim."¹⁹

The term "simultaneity" in Bordwell and Thompson's definition of crosscutting is not quite accurate because there is nearly always some time dropped out between shots. (In alternate editing, time may be expanded as well, in order to increase suspense.) The character racing to the rescue in an automobile is usually a greater distance farther on when we return to the scene than would be possible in continuous time. The ellipsis of time between the alternate shots permitted a compression of narrative events that was very useful for the one-reel film. As Parsons said, the use of alternation lessened in the early days of the feature film, but not only for the reasons she gave—there was also less need for such compression.

The Mill Girl (Vitagraph, 1907) is a useful reference film because, while it is not typical of the other surviving films of its year, it illustrates the point reached by film narrative on the verge of its great expansion. At a time when most of the films produced were comedies or actualities, this one is a melodrama. Over the next few years, it was the melodrama, particularly the psychological melodrama as articulated by D. W. Griffith, that would become the most popular medium for moral uplift. The Mill Girl has a young working-class man and woman for its hero and heroine. A girl in a textile mill is subjected to the unwanted advances of her employer. The young man she loves, a worker in the mill, comes to her rescue by knocking down the boss. The boss hires two ruffians to beat up the worker, who defeats them on two occasions by his superior strength and wit but is then dismissed by the boss. Free of interference, the boss again forces his attentions on the girl but is interrupted by a fire in the mill, which brings about his death. The heroine is rescued from the fire by the young worker.²⁰

The film has one intertitle and thirty-one shots, a very high number for 1907. It is filmed at a kind of "stage distance," as almost all films were at this time: space appears at the bottom showing the floor in front of the actors' feet, and there is inactive space above their heads at the top of the frame. The distance between camera and actors is narrowed a bit more in the climactic scenes. When the actors enter and exit the frame, however, they often move on the diagonal to the axial plane, thus appearing much closer to the camera, in nearly a "three-quarters shot. " At this point, the actors' knees are near the bottom of the frame and their heads near the top. For 1907 audiences, this composition would be familiar from the structure of the chase-film comedy. However, the significant action of the scene takes place only when the actors reach the center of the frame. Many films of the pre-1907 period presented a screen full of action, with no key, other than the lecturer who might have accompanied the film, as to what the spectator should be seeing. The movement into the center frame in The Mill Girl has the effect of leading the spectator's eye to the significant action.

The diagonal entry and exit contributes more than a directional arrow, however: it also serves as a system for linking the shots and outlining a geography for the action. The first two shots of The Mill Girl may be used to illustrate this. The film opens with a shot of a gate set in a hedge, representing the home of the heroine, where the hero joins her in the morning to accompany her to the mill. The hero enters from the right, stops at center to meet the girl at the gate, and they continue together toward the left, approaching the three-quarters shot as they exit. The following shot, in front of the mill, shows groups of workers entering from the same three-quarters position at left, going into mid-distance, and entering the factory gate in the center. Among them are the young couple. By following this direction principle with a fair degree of consistency, the filmmakers keep the action flowing continuously in a kind of synthetic space. The principle was taken over so thoroughly by other filmmakers that by 1911, spectators and critics alike called attention to any inconsistent direction of action from one shot to the next: "Any one who has watched pictures knows how often his sense of reality has been shocked by this very thing … as if [the actor] has gone halfway around a building during the change of scene and was entering it from the opposite side."²¹

The Mill Girl contains a sequence that exemplifies alternate editing in its basic original form, inside and outside a building. This occurs when the thugs hired by the boss come to the worker's home during the night:

1. Exterior: hero enters scene from left and goes through gate to his home.
2. Interior: bedroom, the hero enters from right foreground, goes around bed, yawns, gets ready for bed, closes window, sits on chair to remove his shoes.
3. Exterior: another view of house, the employer leads his thugs into the scene, then sends them back out while he stands and waits.
4. Interior: A slightly closer view of bedroom, hero is now in bed. Hearing a noise, he puts hand to ear, goes to window, and looks out.
5. Exterior: the thugs come back with a ladder and place it against the house so that the top disappears from sight, while the boss gestures a command of silence.
6. Interior: the hero at the window, he turns and makes up a dummy shape in his bed, gets a stick and crouches below foot of bed, foreground.
7. Exterior: the thugs climb the ladder while the boss holds it.
8. Interior: the hero cups his ear, listening, the thugs raise the window, enter, attack dummy. The hero jumps up, knocks one man down while the other flees, then throws the first one out the window.
9. Exterior: the man who is thrown out falls down on the man at the bottom, the other already having fled the scene.

There are many sequences in pre-1907 cinema showing people jumping or being thrown out of doors and windows. Sometimes there is an overlap in time, or repeated action, from interior to exterior shots, as in the frequently cited The Life of an American Fireman (Edison, 4903) by Edwin S. Porter, in the sequence of the rescue of the woman and child from the burning building. In this film, the same action is shown twice, once from inside and the second time from outside the building. In the interior/exterior editing of The Mill Girl, there is simultaneity and/or an ellipsis of time between shots. In 1907 few films other than Porter's could supply any examples of repetition like that of The Life of an American Fireman, and a few years later, as we have seen, such editing would simply look like a "mistake."

In contrast to the model of alternate editing using adjacent spaces found in The Mill Girl, the use of the insert to link distant spaces turns up in a March 1908 film, Old Isaacs the Pawnbroker (Biograph). Based on a script by D. W. Griffith and directed by Wallace McCutcheon, Sr., Old Isaacs the Pawnbroker is an edifying melodrama about the life of the people who were considered to be the most significant audience for the movies: the new immigrants and urban poor. The slumdwellers depicted in this film included some of high moral character, and they are shown to be more sensitive than their upper-class counterparts. The old Jewish pawnbroker wears a putty nose, but he is stereotypical only in appearance, not in character. In the course of the film he is shown to be very brave and to have a kind and understanding heart.

It should not be concluded from the two examples of melodrama provided here that most films in 1907-1908 supplied working-class heroes for a working-class audience. Only a small number did so, but it does appear that such films were particularly popular, judging from contemporary comments as well as the number of copies sold.²²

By 1907 standards Old Isaacs is for the most part conventionally made. The camera is always at stage distance, the diagonal entrances and exits of the Vitagraph film are lacking, the characters are stereotyped ethnic types without much individuality, and the acting is stylized to ensure the clarity of the action. A small girl goes to a charity institution to seek help for her invalid mother who is about to be evicted, but no action is taken pending investigation. In desperation, the girl goes to a pawnshop and tries to hock her mother's old shoes. They are rejected, and she brings her doll instead, drawing the attention of the kindly pawnbroker to her plight. The pawnbroker arrives with food and help for the little family just in time to stop the eviction. Continuity is achieved by using the little girl as the motivated link, followed from shot to shot. While the girl is at the charity offices, she is sent from one room to another. When she reaches the final office, an image is cut in, showing her mother at home (a "switchback" to a scene we have seen earlier), rising from her bed coughing and falling back down. The next shot returns to the charity office, where the child is refused the immediate help she needs. (The heartless charitable organization is a frequent theme of the era, occurring several times in the work of Griffith.) The shot in the office has been cut to "insert" the scene of the mother.

Elsewhere I have attempted to trace the origins of this shot to the "vision scene" common to stage melodrama.²³ Such scenes are also frequent in earlier films, but in those films the visions would be shown in the same scene with the person or persons having them, as a double exposure, as a superimposition, or with stop motion, as in trick films. The vision scene by double exposure would continue to be widely used for many years to come. However, quite another way of thinking is needed to imagine the vision scene as a separate shot in 1907. The insert represents a leap across space to link two people distant from each other. One doesn't know whether this shot represents the girl's thoughts, since she shows no reaction at all. She simply stands and waits. It may well be that the spectator of the film is the only participant in this drama who is witness to the mother's suffering.

The shot could be considered as the invisible narrator of the story who provides comments on it, a narrator who is neither a character in the story nor a real person standing outside the film, but a role that exists in some sense in the structure of the film. This "narrator" plays an essential role in the system of narration that Tom Gunning believes to have been developed in the work of David Wark Griffith during the next few years.²⁴

About a month after Griffith began to direct, he took up the idea of alternate shots in The Fatal Hour (directed in July 1908). In that film, the purpose of the alternation was to increase suspense. A more direct link with Old Isaacs the Pawnbroker can be found in Behind the Scenes, directed by Griffith in August 1908.

Once again it concerns a mother-daughter relationship, but this time the daughter is at home, seriously ill, while the mother is out earning her living as a cabaret dancer. She knows that her daughter is quite sick, and she is in haste to finish her work and hurry home. During her dance, she trembles a moment, and at this point there is inserted an image of her child, dying. Again, one cannot tell from the film itself whether it is the mother or only the spectator who sees the child. As the mother discovers when she hurries home, the child has died. According to the Biograph Bulletin issued with this film, "A mother's intuition assets itself and in her mind's eye she sees her little one—but only for the moment."²⁵ These words, we assume, express the intentions of the inserted shot, even though the film itself does not quite articulate it.

A couple of weeks later, Griffith used another mental image in After Many Years (released November 1908), the first of his films inspired by Enoch Arden. The faithful wife thinks of her husband lost at sea while he, in another shot, thinks of her. Their positions within their respective shots reflect the emotional relationship, each gazing at the space that the other would occupy if they were in the same scene. Nevertheless, the spectator knows for a fact what the wife does not: that her husband is still alive. Here is a clear example of Gunning's "narrator-system," at the same time able to show us the thoughts of the character and a fact that is not known to her.

Each change in film structure at this period seems to result from the new view-point of employing several shots where, earlier, one shot would have done. The idea of the "mental image" was not unusual at this time in films, but it would traditionally be shown in one shot. In Edison's Fireside Reminiscences (January 1908) and Biograph's The Music Master (May 1908), the characters' mental images are implanted in the fireplace or on the wall, in the same scene with them. The Biograph Bulletin describes the latter as "a phantasmagorial portrayal of his thoughts, which bring him back to days of yore." In 1912, Theodore Wharton took advantage of the ease of the Bell & Howell camera's new frame counter to photograph the story a man is confessing to a priest in dissolving views placed within the same scene as that of his confession and in the same size. This was in an Essanay film called Sunshine, which was much admired at the time. Not everyone wanted to follow Griffith's way. Examples of the mental image shown in the same scene are very common in this period, more so, perhaps, than Griffith's editing system. In the World of 6 May 1911, it was said that keeping the characters on the screen while the story is told is "a much more realistic way of telling such a tale, than to tell it with the story teller hidden." On the basis of so few surviving films, however, it is difficult to decide what was a dominant mode, if any. In a film by a decidedly less talented filmmaker than D. W. Griffith, The Pugilists Child, made at the Powers studio in September 1910, insert editing is used very much as it was in Old Isaacs the Pawnbroker. Here a professional fighter is winning in the ring while his child is dying at home, and editing cuts back and forth several times between the distant scenes.²⁶

Some reluctance to adopt the new editing may be seen in the way scenes showing a telephone call were filmed. Here was a real link between distant spaces, newly familiar to everyone, since the telephone system was spreading across the land nearly simultaneously with the movies. In the 1901 film Are You There?, made in England by James Williamson, a young man telephones his girl, but unknown to him, her disapproving father takes the receiver from her hands and listens. In the next and final shot, the angry father visits the young man and beats him. The participants in the phone conversation are shown in the same image, separated by a divided screen, which is signified only by the hanging of a curtain between the two telephones. In 1904, the Biograph Company's The Story the Biograph Told showed a man at his office talking on the phone to his wife at the same time that he is embracing his secretary. In this case, to show simultaneity, the wife at home on the telephone is inserted as a superimposition on a part of the office scene. When Edwin S. Porter made College Chums for Edison in 1907, he wanted to show both participants in a phone conversation and at the same time indicate the great distance separating them. The distance between is shown by a cityscape of rooftops; the woman telephoning is seen in a circle in the upper corner, the man at the left, and the two sides of the conversation in animated letters float across the screen, first to the left, then to the right. The space and the simultaneous event are shown graphically.²⁷

The idea of showing two telephones with a landscape or city streets between by means of a triple screen became quite popular, especially when used for comic effect. A Max Linder comedy, Max and His Dog Dick, imported to the United States in 1912, used it so that Max's dog could telephone him across Paris to tell Max that his wife was receiving a visit from another man. In the same year, Canned Harmony (Solax), a little comedy directed or supervised by Alice Guy Blaché, contained another triple-screen version of a telephone call, this time with the central panel showing a quiet country road and a small animal wandering in the distance. Suspense

(Rex, 1913), directed by Lois Weber and Phillips Smalley, offers a new variation: the screen is divided into three triangles, with a woman speaking on the telephone at the top right, a tramp at the top left who is outside the woman's house and trying to break in, and the husband at his office, at the other end of the phone, in the center. (Many years later the space separating two phones was made subject of gags in such films as Pillow Talk [1959], where Doris Day and Rock Hudson sit in their baths, talking on the phone, with their naked feet touching the wall that separates the split screen.)

Another approach to the telephone call was the alternate scene, which had been used in this way since 1908 at least. The Pathé film The Narrow Escape, released in the United States in March of that year (although probably made in 1907), employs alternate editing for a sequence in which a family is being attacked by intruders, while the father, a doctor who has been called away from home, talks to his wife by telephone. The telephone call is alternated in three shots: the doctor phoning his wife, the wife speaking to him from the phone in his study at home, the doctor speaking to her; then, in another shot, the doctor hangs up and starts to the rescue.²⁸

This kind of thriller based on alternate editing became one of the most successful genres of the time, with D. W. Griffith as one of its leading practitioners. In The Medicine Bottle (1909), for example, Griffith used the alternating method to build suspense around the story of an innocent child about to give a spoonful of poison to her grandmother instead of medicine. The absent mother discovers she has carried off the wrong bottle and tries desperately to reach her little daughter in time by means of the telephone but is thwarted again and again by switchboard operators gossiping among themselves and neglecting their duty. The film alternates between the mother, the switchboard operators, and the little girl four times, until in the twelfth shot of the sequence the call goes through. The child answers, the mother talks, the child listens, and the position then shifts to a more distant view as the girl hangs up and goes to the bedside of her grandmother.²⁹

In the same year, Griffith expanded on the interrupted telephone call in his famous thriller The Lonely Villa. In this variation on the theme of Pathé's The Narrow Escape (which is one of its possible sources), the phone is seen prominently on the wall at right in the second shot. When the doctor called away from home experiences trouble with his car, he enters an inn to telephone his wife and warn her of the expected delay in his return. At the very moment that he calls, his wife and children are besieged by burglars trying to break into the house. The sequence of the telephone call contains seventeen alternating shots, including various elements of interruption, until the wire is cut and the frustrated couple talking on the telephone find themselves to be "a bit disconnected. " "The entire audience was in a state of intense excitement," reported the World's critic after a showing of The Lonely Villa at the Fourteenth Street Theater in New York, noting that at the end, the woman seated behind him had exclaimed, "Thank God, they're saved!" Indeed, the form of this kind of thriller is so effective with audiences that it remains in use even today.³⁰

The Vitagraph producers, already having used the rudiments of alternate editing for suspense in 1906 with The One Hundred to One Shot and in 1907 with The Mill Girl, easily took up the new style. Their telephone thriller of 1910, The Telephone, shows the same kind of editing as The Lonely Villa. In this version, a woman and her child are at home alone and the husband at his club, when she

discovers that the house is on fire. In a twelve-shot sequence, the film alternates between home, switchboard, firemen, and club. Several years later Griffith made another quite outstanding telephone thriller called Death's Marathon (1913), with Blanche Sweet at one end of the telephone listening to her husband, Henry Walthall, threatening suicide on the other end. In this case the conventions of the wellestablished genre are broken and the husband shoots himself before help arrives.³¹

But by this time Griffith and many others were also using alternate editing as the normal way to show any telephone conversation. In the light comedy Teaching Dad to Like Her (March 1911), Griffith directed a fifteen-shot sequence showing the lively conversation of a young man and his sweetheart, cutting from one to the other alternately, until the boy's disapproving father overhears and puts a stop to it. In this fluidly edited sequence, the overall average length of a shot is about five seconds. The sequence does nothing much to forward the story; it is there only to enrich the characterization. Vitagraph's Playing at Divorce (December 1910), Reliance's Jealousy (March 1912), and Nestor's Mum's the Word (March 1913), among surviving films, all use alternate editing to show telephone conversations, even if in some of the cases the filmmakers did not think of using it in other sequences. Yet, as we have seen, the older method of showing a phone call in one scene by dividing the screen did not disappear.

For some filmmakers, the literal "links between spaces" of the sounds of a telephone call may have been needed to stimulate this kind of alternate editing. In other cases, a sound heard by a character appears to motivate a cut to another scene, although the sound was not in the silent film but was probably provided in many of the nickelodeons by external means. One has only to think of the sequence of alternate scenes in The Mill Girl, for example, in which the hero hears the attackers outside his window. Another example may be found in Edwin S. Porter's The Trainer's Daughter; or, A Race for Love (Edison, November 1907), in which a shot that shows a man blowing the horn to start the race, enclosed in a round matte, is cut into the middle of the shot in the stables where the characters react to hearing the horn. A Vitagraph production of May 1910, Convict No. 796, directed by Van Dyke Brooke, uses the sound of a pure young woman's singing to motivate alternate cutting. An escaped convict in the adjacent room holds a knife ready to murder the judge, her father, who had sent him away; through alternate editing we see the convict's hand stayed by her voice.

The kind of alternate editing that compared images in order to contrast them came into use at the same time. According to Kristin Thompson, it was much more frequent in this early period than it would be when the classical cinema systems were established.³² This is logical because, as we can see from the films themselves, it is a particularly apt narrative system for making moral statements, and the years from 1907 to 1915 were precisely the time when moral uplift was in demand. (Subsequently, it was not classical cinema that would take up this kind of editing, but Soviet cinema of the twenties, where it would be found useful for expressing another ideology.) A Corner in Wheat (December 1909), for example, Griffith's most famous film of social comment, uses alternate scenes to contrast the harsh life of the farmers and the poor, who depend on the price of bread, with the exotic luxury indulged in by the wheat speculator. The characters never meet, although a direct cause-and-effect relationship exists among them. Kalem's A Lad from Old Ireland (November 1910), directed by Sidney Olcott, similarly contrasts the successful life of the Irish boy in America with the poverty of the loved ones he has almost forgotten back in Ireland. Here the use is mixed, as the characters are more directly related, and eventually they do meet, but the moral lesson of the contrast remains the dominant element.

Aside from these special uses of editing, which I would call "ideological editing," alternate scenes became a standard way of narrating a story, of ending a scene, of eliding time between events, and so forth. Audiences must have been accustomed to the technique by the time of Bison's The Empty Water Keg (February 1912), directed by Thomas Ince. Here, two threads of a story are told in alternate scenes with no link between them until very late in the film: one thread follows a discouraged prospector as he searches the desert for gold as a last resort, only to find it too late and die of thirst, while the other presents a romance between a cowboy and his sweetheart. It is only in the last quarter or so of the film that the lovers, who are out riding, look through their binoculars to see, in a circle matted shot, the dying prospector. The two threads are brought together in time for the man to give the lovers the location of his discovery, and they become rich. No particular moral lesson is evident in this use of alternate scenes. By 1912, the construction had become an ordinary way to tell a story.