5
Technological Change and Classical Film Style

David Bordwell

Kristin Thompson

The 1930s was an era of enormous technological development in the Hollywood cinema. This chapter considers major changes in the look and sound of the studios' product. That is, its concern is with the systematic use of film techniques that constitutes a film's style. It seeks to show how certain technological developments during the decade affected Hollywood's canonized style.

To explain such developments, we must take account of two broad factors. First, to a considerable extent, the aesthetic norms of Hollywood studio filmmaking as a whole constitute a group style. This style uses particular devices, such as three-point lighting and match-on-action editing. As might be expected, technology often creates new devices or reinforces or revises existing ones. The Hollywood style also embodies assumptions about how a film is constructed and the sorts of effects it should have. For example, the Hollywood style can be said to facilitate story continuity or to construct a unified space for a scene. Such assumptions can in turn guide technological changes. Filmmakers' beliefs about proper film technique make certain kinds of change more acceptable than others.

A second explanatory factor is nonstylistic. We must take account of the social processes that translate filmmakers' goals and standards into new materials, equipment, and procedures. In an industry as complex and wide-ranging as American movie production, these processes inevitably involve institutions. Just as the Hollywood style is a group phenomenon, so is technological change. In sum, the maintenance and development of this film style owes something to technological change. At the same time, technology is itself guided by both the characteristic devices and broader goals of that style. And both are subject to social practices within particular institutions.

A Specimen Scene

Consider a scene from The Charge of the Light Brigade (Warners, 1936), directed by Michael Curtiz. Late in the film, near the climactic charge, the scene shows Geoffrey (Errol Flynn), who has been made an officer in the Crimea, forcing his brother Perry (Patrick Knowles) to convey a dispatch to general headquarters. One aspect of the scene involves the historical action: Geoffrey has forged an order for the regiment to charge, and he needs to convey his decision to a sympathetic adviser. But there is a personal side to the scene as well. Since Perry is in love with Geoffrey's fiancée, Elsa, Geoffrey makes a noble sacrifice by sending Perry away from a suicidal charge. Perry, however, thinks that out of jealousy Geoffrey aims to shame him by sparing him from the fighting. Perry angrily accepts Geoffrey's order, not knowing that Geoffrey has already given Elsa permission to marry Perry. Geoffrey sacrifices his brother's honor and his own reputation as a judicious officer to make Elsa happy.

The scene, laid out in Figures 1-11 (pp. 111-112), provides a condensed example of some central stylistic features of the period.

The Charge of the Light Brigade Scene 34

Dissolve to:

1. (Fig. 1) Long shot: Geoffrey at his desk in the foreground. Perry enters from rear and walks to desk; he starts to salute.
2. (Fig. 2) Medium close-up: Perry salutes, then starts. Oh, I thought Sir Benjamin sent for me.
3. (Fig. 3) Medium shot: Geoffrey at desk. No, I sent for you. I want you to take this dispatch to general headquarters …
4. Long shot, as 1: Geoffrey rises (Fig. 4) as he continues, … and wait there for further orders. As Geoffrey approaches and the camera tracks in (Fig. 5), Perry protests: Wait at general headquarters! But I've learned the regiment's moving forward! Geoffrey now faces him (Fig. 6). If I have to wait at general headquarters, I'll be out of the fight altogether. Geoffrey replies, You'll deliver it personally to Sir Charles Macefield, do you understand?
5. (Fig. 7) Medium close-up, Geoffrey's shoulder in the foreground: Perry glowers: I believe you're trying to keep me out of this deliberately because of Elsa.
6. (Fig. 8) Medium close-up, Perry's shoulder in the foreground: Geoffrey answers, No; no, Elsa's got nothing to do with this. You're not in the diplomatic service now, you know. You're with your regiment at the front. You'll obey my orders immediately. Perry replies, If I refuse? Geoffrey: You'll be courtmartialed.
7. (Fig. g) Medium close-up, as 5: Perry lowers his head.
8. (Fig. 10) Medium shot, as end of 4: Perry, looking down at the dispatch, takes it from Geoffrey. I see. Disgraced either way. He turns away toward the door. Very well. As he walks out the rear, track in to medium close-up of Geoffrey, who raises his head as if in pain and closes his eyes (Fig. 11). Dissolve.

Most fundamentally, the scene exemplifies what has come to be called the "continuity style" characteristic of Hollywood. In general, this style aims to present a coherent, stable, clearly defined space in which film technique is used to call the audience's attention to the most salient narrative information at each moment. In this style, lighting, costume, and figure placement will not change noticeably from shot to shot. Spatial continuity will be further assured by staging and shooting the action so that the camera is always on one side of the action.

Sound will also aid temporal continuity by blending speech, music, and noise into an intelligible, smoothly modulated pattern. Music may lead into the action or underscore strong moments, while sound effects provide a background ambience. Nonetheless, dialogue is the principal carrier of story information, and it will dominate the other factors and stand out crisply.

This Charge of the Light Brigade sequence shows that in many respects the continuity style of the 1930s continues traditions developed during the silent era. For instance, the staging of the action defines the interpersonal situation. Geoffrey is seated at his desk when Perry enters and salutes. Geoffrey rises, strides around the desk to him, and offers him the orders. When he threatens court-martial if Perry refuses his mission, Perry stalks out, leaving Geoffrey to agonize over what he has been forced to do.

Lighting serves to emphasize salient aspects of the staging. Throughout the Hollywood tradition, the human figure tends to be the center of the spectacle, and film technique serves to guide audience response by making aspects of that figure salient. Thus, lighting picks out the faces and the edges of the bodies (Figs. 7-8). The lighting also gives a soft sheen to the figures, eliminating blemishes from the faces and picking out highlights in the men's uniforms.

Similarly, when the human figures are framed in closer views, it is usually in order to emphasize what they say. For example, a cut-in to a medium close-up of Perry underlines his recognition of who his commanding officer is (Shot 2). Similarly, Geoffrey's reply is rendered during a closer view of him at his desk (Shot 3, Fig. 3). Sometimes the closer framing simply facilitates the audience's view of an unspoken reaction, as when Perry realizes the threat of court-martial and starts to look down at the order (Shot 7, Fig. 9).

The scene's editing is organized in terms of classical continuity's "axis of action," or "center line." This refers to the vector of movement, such as the direction of motion or the line of interaction between characters. In this scene, the line develops rather simply. At first, the line puts Geoffrey, sitting at the desk, in the foreground (Shot 1 to beginning of Shot 4). When Geoffrey rises and goes to Perry, he stops so as to place Perry slightly in the foreground (end of Shot 4). At one point (Shot 8), the line connecting the brothers is slightly diagonal to the camera.

Given a shot establishing an axis of action, the director assumes that no shot of either character can "break" it; that is, new camera positions will occupy any spot on one side of the center line. All the shots in this sequence respect such constraints. Shot 2, a medium close-up of Perry (Fig. 2), is taken from the same side of the line as was Shot 1. Shot 3, a reverse shot of Geoffrey (Fig. 3), is of the opposite end of the center line, but the camera position remains on the same side of the axis as it was in Shot 2. Such adherence to axis-of-action continuity ensures that when two characters face each other, their bodily orientations, facial orientations, and eyelines "match." Geoffrey is offscreen in Shot 2, but adherence to continuity editing assures that we understand that Perry is looking at him. Imagine, by contrast, if Curtiz had filmed Perry from the other side of the axis: he would now be facing and looking right, and the viewer might wonder if he had turned around.

Editing governs temporal factors no less than spatial ones. The sequence under discussion also reveals that unless there is an indication to the contrary, each cut is presumed to be durationally continuous. No moments of story action are left out between shots. Two cuts display explicit temporal continuity by means of a match on action. At the end of Shot 1, Perry salutes, and the next shot continues his movement. A similar match connects Shots 7 and 8 when Perry lowers his head. When some time must be elided, the Hollywood editing system handles the matter in one of two ways: either by cross-cutting two actions in different locales (so that when we leave one action to see another, time is skipped over in the line of action we do not see), or by use of conventional "punctuations." Instances of the latter frame the sequence under consideration: the scene begins and ends with dissolves to indicate that some time has elapsed.

The bookended symmetry of the dissolves is echoed by repetitions within the scene. The overall angle of view, along the diagonal running from the desk to the door, is reiterated in three shots showing the two men facing each other (Shots 1, 4, and 8). In Shot 1, they face each other unmoving; in Shot 4, Geoffrey moves from foreground to background; in Shot 8, Perry moves from foreground to background. Interspersed with these three two-shots are three medium closeups of Perry (Shots 2, 5, and 7). The last two of these are from the same camera set-up. Two shots give us closer views of Geoffrey, one at the desk (Shot 3), and one when he forces Perry to take the dispatch (Shot 6). In a sense, there is a third close view of Geoffrey—at the end of the scene when the camera tracks in on him (Shot 8, Fig. 11). Shot 8, then, at first functions as a two-shot of the pair and then becomes a close-up of Geoffrey. This creates a pattern:

This sort of balanced repetition has been pointed out by Raymond Bellour as a basic principle of the classical style.¹

In many cases, the repetition makes for greater efficiency in production. The repeated setups of Perry (Shots 5 and 7) permit ease of filming: no adjustment of lights or camera would be needed. More strikingly, the two-shots as edited (into Shots 1, 4, and 8) are phases of what was probably during production a single take, the "master shot." In the silent era, it was common practice to film all shots with the same framing at the same time, but no continuous take of the entire scene would typically have been taken. By the early 1930s, however, directors usually made a master shot of the entire scene, with all dialogue and camera movements played out. Closer framings of particular lines would then be shot, to be inserted during the editing. The master-shot technique yields the sort of repetition we see here. The setup at the end of Shot 1 is that seen in the beginning of Shot 4; that at the end of Shot 4 supplies the framing for the beginning of Shot 8. Here classical style pays double dividends: the repetitions yield visual redundancy but also provide a financially economical filming procedure.

Such tactics of staging, lighting, framing, and cutting emerged during the first two decades of American film history and were by and large consolidated in a systematic way by 1917. Thus, the Light Brigade sequence relies upon stylistic principles already firmly in place in the Hollywood silent cinema.

Yet the sequence also stands as a prototype of how the sound cinema of the 1930s could build upon these principles. Perhaps most notable is the rapid development of the action made possible by sound. The fairly complicated intrigue is developed in a scene employing merely 123 words and eight shots, and lasting for under a minute of screen time. In a silent film, the same lines would have required at least nine dialogue titles, each minimally five seconds long. Moreover, with sound the filmmakers can supply several pieces of story information at the same time by using different tracks. In Shot 4, while Geoffrey rises and approaches Perry (visual information), he continues issuing his order (auditory information). The sequence thus serves as a fair instance of the concision normally associated with studio filmmaking in the 1930s.

By the 1930s, "continuity" had come to connote a perceptually smooth conveyance of story information. Sound could enhance this sort of continuity in several ways. In the specimen sequence, beginning Geoffrey's line in Shot 3 and ending it in Shot 4 makes the cut itself less noticeable. Moreover, scriptwriters strove for a verbal continuity to match that on the visual track. The dialogue's flow from line to line proceeds from question and answer ("Do you understand?" "If I refuse?") and from echoing repetitions from one line to another:

"…Sir Benjamin sent for me."/"No, I sent for you."
"Take this dispatch to general headquarters and wait
there…"/"Wait at general headquarters!… If I
have to wait at general headquarters…"
"If I have to wait…"/"If I refuse?"
"You'll deliver it…"/"You'll obey my orders…"/
"You'll be court-martialed…"

While some of these techniques can be found in the intertitles of the silent cinema, 1930s dialogue became a dense tissue of such repetitions.

Music yields a comparable auditory continuity. Here Max Steiner's score continues from the previous scene before fading out as the dialogue begins. Ten seconds later, when Geoffrey announces Perry's order, the line is followed by a two-chord "sting" in the woodwinds. The second chord is prolonged and provides a background for the next phase of the scene. After Perry's "Wait at general headquarters!" the two-chord motto is repeated in the brass. It recurs during pauses in the dialogue, before rising to a climax in the timpani under Geoffrey's "You'll obey my orders immediately." This line is further emphasized by the dead silence after it. With Perry's reply ("If I refuse?") the music resumes, underlining his resistance to the order. After his "Disgraced either way" comes a two-second pause. The music reappears, and as he walks out, a lyrical string theme (associated with Geoffrey and Elsa's romance) emerges to end the scene, underlining Geoffrey's closing his eyes in agony.

The sequence neatly illustrates the wide range of ways in which music can organize and point up patterns in the dialogue. Sometimes a line is delivered and music emerges in the pause after it. Sometimes the music continues underneath the line and continues during the pause. And sometimes the music runs under the line and abruptly halts to yield a few seconds of emphatic silence. In any event, few spectators will notice that about 80 percent of the scene has musical accompaniment. Steiner would write some years later that the "great problem" of film composing is "to give the scene continuity, to keep the audience unconscious of any break, yet to make the music perform its function of sustaining each mood and scene."²

The smooth flow of dramatic action is also aided by virtually unnoticeable camera movement. All the basic types of moving camera had been used during the silent era, and at several points camera movement had been a prominent stylistic device. Again, however, the 1930s gave certain stylistic options greater salience. When Geoffrey rises from his desk in Shot 4, the camera pans slightly right to reframe him. Such reframing movements, while not unknown in the silent era, were far more frequent in the sound period. As Geoffrey strides to Perry, the camera tracks (or "dollies") forward to frame them in medium-shot. The camera movement is quite unobtrusive because it follows the character's movement. At the end of the last shot, as Perry leaves, the camera tracks forward once more, stressing Geoffrey's pained reaction. Observe that again this movement is almost unnoticeable because its timing and pace coincide with the action of Perry stalking off in the background.

Our prototype is a very simple scene; a more complex sequence would require many more pages of analysis. But the Charge of the Light Brigade passage does suggest that we can usefully consider the style of the 1930s Hollywood film to be a modification of certain devices and principles already established in the Hollywood tradition. Stylistic novelty during this period emerged within a frame of reference that valorized certain devices, such as matches on action and the 180-degree system, while also promoting certain broad goals: centrality of the human figure; spatiotemporal continuity; and a clarity and economy in emphasizing the ongoing dialogue, physical action, and psychological reaction.

The Interplay of Style and Technology

In any industry, one goal of technological change is efficiency—to make a process consume less time, energy, or money. The film industry is no exception. Many tools and procedures were introduced to cut the costs of film production, especially after the advent of talkies had raised expenses considerably. The 1930s was the era in which sound production came to be streamlined in a form that is recognizable today.

Technological change also has an aesthetic aspect. By canonizing a particular style of filmmaking, the Hollywood industry set engineers an agenda. Sound had to be integrated into an existing set of stylistic priorities. More specifically, the fluency and economy found in our specimen scene were goals consciously pursued during the early 1930s. Hollywood filmmakers strove to recover the stylistic flexibility that they had enjoyed during the late silent era and that had often been curtailed during the early days of sound. Technological change in the Hollywood studios thus had two goals: maximal efficiency and maximal integration with, or extension of, the classical stylistic norms.

These two goals came into sharp focus in the earliest days of talkies. Between 1929 and 1931 most filmmakers staged the action for the sake of sound recording. Getting a clear, complete soundtrack had the highest priority, and since microphones were heavy and hard to move, the picture track often became static and rather flatly lit. In addition, most scenes were shot with several cameras running simultaneously, all filming the scene in toto. By placing the cameras at various angles and by using lenses of different focal lengths, the filmmakers could preserve the changes of framing essential to Hollywood continuity editing.

A characteristic example is furnished by a sequence from Millie (RKO, 1931). Millie has just eloped with Jack, and they have checked into a hotel. The scene depicting her reluctance to go to bed is staged quite "theatrically" in a wide, fairly shallow set. The hotel room's parlor gives onto a curtained bedroom in the rear. In the course of the sequence, five cameras follow the characters around the room. After the couple check in, Millie sits down on stage right and Jack joins her; they are filmed in medium long shot (Fig. 12). After she goes out frame left (Fig. 13), a long-shot camera follows her walk to the door, panning right to reframe Jack as he approaches (Fig. 14). At the door, the couple are framed in medium shot as he tries to persuade her to come to bed (Fig. 15). In an extreme long shot he draws her to the bedroom in the rear (Fig. 16). As she resists, a medium shot enlarges them again (Fig. 17). The two medium shots are filmed with long lenses, while the other views are filmed with lenses of shorter focal lengths.

Multiple-camera filming guaranteed synchronization in an era when matching picture and soundtrack posed great difficulties. One cinematographer recalled, "In those days we didn't know how to cut sound, so we'd shoot the sound in one solid unit, and then cut the film from our twelve [!] cameras to fit the track."³

Yet the multiple-camera procedure had many drawbacks. It was inefficient, wasting time and film and requiring a large crew. Since most sound shots could not be made separately, any error in performance or recording necessitated starting the entire scene over. There were also aesthetic drawbacks. The cameras were frequently housed in soundproof booths, and the thick glass reduced photographic quality. Long lenses could not achieve the tight, precise framings of the silent era. In the cinema of the 1920s, the camera was frequently placed close to the players, and actors' glances and movements in a sense flowed around the spectator. In early talkies, however, the placement of the cameras outside the zone of action made the action seem more distant and uninvolving. In the Millie scene, in contrast with the sequence from The Charge of the Light Brigade, the viewer does not really enter the hotel room, either through camera movements or cut-in close-ups. Space becomes less voluminous, actors more distant. The analysis of the action is accomplished by enlarging portions of the scene, as if we were looking at a sporting event through binoculars.

Given the economic and aesthetic drawbacks of multiple-camera shooting, it is not surprising that technical personnel devoted a great deal of energy to returning to single-camera filmmaking. Filmmakers quickly learned that shots not requiring synchronized dialogue could be filmed silent and spliced into multi-camera sequences, perhaps with some sound effects added. Soon directional microphones and silenced cameras allowed closer framings and more nuanced lighting. New sound-editing equipment permitted picture and sound to be recorded and cut separately. Mobile camera carriages enabled filmmakers to recover and even enhance the fluidity of silent cinema. By the end of 1931, most studios were able to film with one camera again.

More generally, contemporary observers insisted that there was no going back to the plodding pace of most early talkies. Some writers foresaw a new cinematic rhythm born of ablend of talk, music, gesture, and silence.⁴ One writer proposed

building smaller sets so that characters could pass through them more swiftly, eliminating "dead footage."⁵ Another commentator asserted that film technique, instead of merely recording theatrical dialogue, was an intrinsic part of cinema's appeal. "And the tortuous way back to the old silent days' technique of many angles and rapid cuts, plus sound-camera equipment, was begun."⁶ Yet another critic advocated increased tempo: more and shorter scenes, swift lap-dissolves and montage sequences, rapid-fire dialogue, and "an almost constantly moving camera."⁷

These writers were not calling for that "creative use of sound" by Mamoulian or Lubitsch canonized in film history textbooks. These directors' innovations often minimized speech, letting the audience grasp the situation by means of sound effects or purely visual cues. Rather, the commentators mentioned above foresaw a cinema grounded in dialogue. Nonetheless, the actors' performances would be enhanced by all the visual resources revealed in the silent era, as well as others yet to be developed.

Technological change thus fulfilled several roles. Not only was it aimed at greater economy or efficiency, but new technology could also help filmmakers maintain established standards of visual intelligibility, and it could enhance the appeal of what was essentially a dialogue-based cinema.

Sources of Innovation

While technological and artistic innovations can usually be attributed to individuals, those individuals operate within a broader context. The norms of Hollywood film style emerged and held sway within the routines of film production as fostered by the studios. Similarly, technological change and stability are processes that occur within particular institutions. The individual inventor works within a context of goals, constraints, and opportunities defined not only by the hardware's "state of the art" but also by the unspoken assumptions about what is needed for this particular mode of filmmaking.

During the 1930s, three types of institutions supplied filmmaking technology: the studios themselves, the large and small service firms, and professional organizations attached to the industry. A survey of these sources of technological innovation will illustrate how particular changes emerged.

The film studios were sometimes a source of technological innovation. At MGM, John Arnold oversaw a Camera Department whose job was not only to maintain equipment but also to solve technical problems. Douglas Shearer, head of the same studio's Sound Department, developed several pieces of recording equipment, as well as pressing for more uniform reproduction in theaters. At 20th Century-Fox, Daniel Clark, the Executive Director of Photography, devised a new camera and developed a useful system of standardizing f-stops. Less well endowed studios also put some resources into technology, most notably in the design of special-effects equipment and of camera carriages and cranes.

On the whole, however, the production companies were not well equipped for large-scale technical innovation. Most studio "research laboratories" seem to have been glorified machine shops, and the "experiments" sometimes referred to in the trade literature were seldom more than skillful tinkering. For basic and standardized innovation, two other sorts of institutions played more central roles.

Most visible are the firms, large and small, which served the industry's technological needs. Eastman Kodak, for example, had since late in the nineteenth century committed itself to an ambitious program of research and development. Recruiting outstanding graduates from the nation's engineering schools, Eastman undertook basic research into the optics and chemistry of photographic processes. Eastman supplied the movie industry with raw film, both positive and negative, and equipment, such as filters for cameras and sensitometers for laboratory work. Similarly, Eastman's Rochester neighbor, Bausch and Lomb, was the principal source of lenses for motion-picture cameras, projectors, printers, and other equipment.

With the coming of sound, other corporations entered the picture. Both the Radio Corporation of America and Bell Telephone had invested heavily in basic research, and, so, large and sophisticated laboratories stood ready to assist Hollywood in standardizing and improving sound processes.

Less committed to basic research were the more specialized companies serving the industry. Bell & Howell had been a mainstay since 1908, when its continuous printer helped standardize the industry's printing practices. During the 1920s, the Bell & Howell camera was considered the most sophisticated and precise available. The firm continued its triumphs with the introduction of new film gauges (16-mm silent in 1923, 16-mm sound in 1932, and 8-mm in 1936), of automated printers for sound (1930, 1935), and of perforators and splicers.⁸

Bell & Howell's chief competitor was the Mitchell company, whose camera also became widely used in the 1920s. The coming of sound gave the Mitchell NC (for "New Camera") a strong advantage, since the Bell & Howell camera's noise was picked up by microphones. Bell & Howell introduced a quieter version of its product in 1933, but it did not win out over the BNC (for "Blimped New Camera") that Mitchell introduced a year later. The BNC had a compact soundproof housing and other desirable features, such as a shutter that could be adjusted during a shot. By the end of the decade, Bell & Howell cameras were being consigned to newsreel and special-effects work, while the Mitchell was becoming the standard studio camera.⁹

Other specialized companies offered tailor-made engineering. The Moviola company had since the 1920s provided a variety of viewing machines, all of which obliged the cutter to view the film through a magnifying lens; in 1938 the firm created the Preview Moviola, which projected the film on a small screen. The firm of Mole-Richardson concentrated on supplying lighting equipment and mobile carriages for cameras and microphones. In 1928 the company won an Academy Award for its research into illumination, and in 1935, another for developing special spotlights. Mole-Richardson often built prototypes on demand and then manufactured the equipment that studios deemed most useful.

Along with such service companies—and there were many others—we can distinguish a third sort of institution that shaped technological change during the 1930s. During the teens and twenties there emerged several professional organizations that played important roles in the process.

A key instance is the American Society of Cinematographers (ASC), which grew from an informal club into the professional association of studio cinematographers. Chartered in 1919, it sought "to establish and maintain standards of professional and technical skill and integrity, and to bring into closer personal and professional fellowship all of the leaders of the motion picture camera profession."¹⁰ During the thirties monthly ASC meetings became occasions for cinematographers to share discoveries and display new equipment. The results were often also disseminated in American Cinematographer, a monthly magazine.

The Society of Motion Picture Engineers (SMPE) took a more stringent approach to innovation. Founded in 1916 and composed of technicians from the supply firms, the SMPE was primarily devoted to establishing engineering standards for the motion-picture industry. The society served its members' companies by pressing for uniformity and efficiency in such matters as film dimensions, apertures, perforations, sound characteristics, screen materials and sizes, and so on. Research in the companies' laboratories was announced at twice-yearly conventions and in the Journal of the Society of Motion Picture Engineers. Relatively few employees of Hollywood studios belonged to the SMPE in the 1920s, but the arrival of sound filming gave engineering a new pride of place. In 1929 the SMPE opened its Pacific Coast section under the aegis of such major suppliers as Mitchell and Mole and with membership including major studio personnel.

The increased technological development of the decade required broad coordination, which could not be achieved by the rather specialized ASC and SMPE. The Academy of Motion Picture Arts and Sciences, founded in 1927, had many functions, but by the early 1930s it had taken on the role of coordinating technological innovation and its diffusion.

The Academy established its Technical Bureau in 1928, but it was not particularly effective until 1930, when the Motion Picture Producers Association transferred its technical office to the Academy and donated $15,000 to help coordinate the work of existing research facilities. In 1932 the Academy board of directors revamped the bureau and created the Academy Research Council, which announced that it intended "to coordinate motion picture research and standardization efforts and to function as a central bureau for the exchange of technical and artistic information."¹¹ The council's quarterly meetings brought together representatives of all branches of the membership, technical staff from each studio, and advisory engineers from supply companies. Across the years the council helped standardize apertures in cameras and projectors, process projection equipment, costume colors, and other factors. By representing the studios' interests, it provided a forum wherein producers could communicate their needs directly to the major suppliers.

During the early 1930s the supply companies developed close ties to these organizations. Not only did company employees become leaders of the SMPE, but in 1931 the society created the category of "sustaining member," which allowed firms to contribute money to collective efforts. Two years later, the ASC allowed representatives of Bell & Howell, Eastman, Bausch & Lomb, and other firms to become associate members, while some outstanding figures such as Howell and Mitchell became honorary members. Supply firms' innovations were often premiered at ASC meetings, and cinematographers were invited to try out new materials and equipment on the set. Similarly, when the Academy was reorganized in 1933, it created the category of "corporate member" to allow the suppliers to join and to help fund the council's activities. The Academy Research Council became heavily dependent upon the manufacturers; by 1937, its thirtysix committees were said to involve 180 different company representatives.¹²

Studios, supply firms, and professional associations were, then, the major innovators of technology during the 1930s. They cooperated closely with each other. When in 1931 new faster film stocks needed testing, the ASC cooperated with the Academy Technicians' Branch on the project. In 1936 the head of the SMPE's Sound Committee reported working with the studios' sound departments and the Academy to compare the recording frequencies of different studios.¹³ As the following case studies show, most technological changes involved a degree of standardization that required a consensus among the product's users.

That consensus in turn rested on assumptions about what technology should be used for. Materials and equipment had to satisfy the criteria of efficiency characteristic of any industry: durability, low cost, portability, flexibility, uniformity, and so on. But more intangible were those requirements that bore on style and artifice. Studios proved startlingly uninterested in standardizing the most basic practices of their industry, such as perforation size and aperture dimensions; such issues would drag through the organizations' technical committees for years. But filmmakers had their own priorities. Peter Mole confessed that when he arrived in Hollywood, he discoverd that studios did not wholly share the engineer's gospel of efficiency. The most elegant technical solution would not always suit the studios' goals.¹⁴

In addition, the studios faced divisions within their own industry. Often the production sector's pursuit of novelty and spectacle would exceed exhibitors' willingness to adapt. Such innovations as the widescreen, 3-D, and stereophonic sound, all introduced during the 1930s, could not be developed on a large scale. The changeover to sound had required a huge investment in theaters, and exhibitors resisted attempts to revamp their facilities during the Depression years. Declining attendance in the late 1940s would make exhibitors more enthusiastic about 3-D, Cinerama, CinemaScope, and similar novelties in the following decade.

The dynamics of innovation are thus fairly complicated. Studios could make broad demands, such as for a greater fidelity of sound recording, or propose specific problems, such as the removal of unwanted equipment noise on the set. The Academy, the professional organizations, and the supply firms could investigate the matter. Possible solutions could be tested in various venues, often with studio cooperation. Reports from different agencies could sharpen the issues and suggest new prospects. Eventually a solution could be proposed, promulgated, and accepted, perhaps even becoming rigidly standardized. At every stage, contending factions within the motion-picture industry might have to be consulted.

In the course of the investigation, other technical issues might arise. Or an innovation in one area might have unexpected implications in another, such as improvements in rear projection. At certain points, the supply firms took the initiative and, anticipating filmmakers' needs, innovated in ways that offered filmmakers hitherto unexpected possibilities for improving production routines or exploring fresh options within Hollywood's characteristic style.

Here we can only suggest how this process worked in certain key instances. There is much that researchers still do not know about the dynamics of technology and style in Hollywood. It seems clear, however, that during the 1930s, Hollywood stabilized its sound-filmmaking practices by defining goals fairly clearly, pursuing them rationally, and integrating the efforts of various technical institutions. The case studies that follow show these processes at work in more detail.

Case Studies

What follows are case studies of five major technological developments over the decade. They have been chosen somewhat arbitrarily, since they are connected to each other and have links to still other developments that cannot be pursued here. For example, the increase in camera movement necessitated new instruments for racking focus, which drew in turn upon research undertaken for entirely different ends, the problems of focusing the bulky Technicolor camera. All this chapter can do is indicate how some important developments met the two central criteria: gaining greater production efficiency and fulfilling or extending norms of the classical storytelling style.

Innovations in Sound. Early sound-filming technology employed either sound on disk (the Warners Vitaphone system) or sound on film (the Fox Movietone system). Very quickly the problems with sound on disk—primarily related to synchronization—assured the dominance of sound on film. By 1930, most studios were using so-called double-system sound; that is, a separate strip of film ran through a recording device that registered sound impulses as optical patterns.

Since the Hollywood sound cinema was to be centered upon a narrative carried by dialogue, the crucial problem was that of recording the voice while retaining the flexibility of silent visual style. Here the purely acoustic difficulties were severe. Personnel, lights, and cameras created considerable noise on the set. Most microphones were "omnidirectional" and picked up such noise indiscriminately. They were also bulky and difficult to move around the set. In addition, the signal they captured was narrow in frequency range, muddy in texture, and often too low in volume.

Some of these difficulties were overcome by quieting the set. The camera was put in a booth or a soundproof metal casing ("blimp"). Noisy arc lighting was replaced by incandescents. Mole-Richardson designed lightweight microphone booms that replaced the crude rope-and-pulley system of the early days.

More complicated problems of recording required the help of RCA and Western Electric. For instance, in 1931, RCA introduced the "ribbon" micro-phone, which had a flat response, high sensitivity, and a bidirectional range of pickup. The ribbon microphone became widely used because it could be placed between conversing characters, and if it was perpendicular to the camera, little equipment noise would register. RCA's 1936 "cardoid" microphone yielded a unidirectional pickup over a heart-shaped field. This device was coming into wider use by the end of the decade, when Western Electric introduced a comparable series of directional microphones.

RCA and Western Electric engineers also strove to increase the range of sound that could be recorded. The most significant step here was the introduction in 1934 of "push-pull" sound recording. This procedure split the sound into two signals and recorded them 180-degrees out of phase, either by the variable-area method (the RCA design) or by the variable-density method (the Western Electric solution). This innovation increased the volume range of the recorded material by "pushing" the fundamental frequency and "pulling" overtones and equipment noise.¹⁵ In addition, in 1936, RCA developed ultraviolet recording, which captured higher frequencies. Thanks to such innovations, by the end of the decade a Hollywood film could record a range of 50-8,000 cycles per second and could achieve up to sixty decibels in dynamic range.¹⁶

In the early days of talkies, nearly all sound recording took place during the shooting phase. Music had to be performed on the set, even if the source was never to be seen. If an actor had to be "dubbed," the dubber recited the lines off-camera. For the sake of efficient division of labor and greater control over the soundtrack, filmmakers strove to reduce the amount of sound recording occurring on the set. For this to be done, noise, music, and dialogue would have to be recorded and manipulated on separate tracks. The 1930s is thus the decade that established principles of multitrack sound that still govern film production.

The separation of tracks altered the handling of noise effects and music. By the end of 1932, most such sound was recorded independently of the image. At about the same time, prerecorded or "playback," sound came into use for musical numbers. A musical piece would be recorded on film or acetate disc and then played back during shooting so that songs and dances could be performed in synchronization. For the finished film, the sound staff would synchronize the piece with the footage. For One Hundred Men and a Girl (1937), Leopold Stokowski experimented with recording on eight channels: six for different parts of the orchestra, one for the whole orchestra, and one for the soloist, Deanna Durbin. Although this was exceptional, it foreshadowed the multitrack recording of later decades.¹⁷

Rerecording dialogue lagged behind rerecording music and sound effects because vocal perspective was initially difficult to duplicate off the set. Soon, however, the improved microphones and recording equipment made dubbing voices feasible. By 1935, actors could rerecord voices guided by a track made during filming. Each stretch of dialogue was kept on a separate roll of film (a procedure still called "looping"). During the second half of the decade, the sound mixer might be handling as many as fifteen tracks for a scene. A 1938 observer commented that a finished eleven-minute reel could have a hundred sound cues.¹⁸ For this reason, the sound department received the script before filming and, from it, built a "sound script" that would govern the search for music or effects.

These and other innovations in sound recording were coordinated by the Academy. Throughout the decade it held conferences and courses to train technicians in new equipment. In 1937 the Academy persuaded the major companies to standardize the frequency range of their release prints so that theaters could in turn establish a uniform sound characteristic.¹⁹

Developments in sound and picture editing proceeded alongside innovations in recording and reproduction. Double-system sound posed the problem of synchronizing picture and sound. This was solved through two procedures. At the start of every take, picture track and sound track were punched simultaneously so that they could easily be lined up. In addition, since 1918, Eastman negative stock had borne stenciled footage numbers on its edge; this enabled editors to distinguish various takes of the same shot. After 1930, the sound track was recorded on film with edge numbers identical to those on the picture track. Thus, the editor could immediately match any frame with its corresponding sound. The mechanics of the job were eased by the gang synchronizer, a mounted set of sprocketed clamps that allowed several reels of sound and image to be aligned in parallel. Precise editing was also assisted by the 1931 double-system Moviola, by a 1933 model with heads for three sound tracks, and by the 1937 Preview Moviola, whose 5.5-by-6.5-inch viewing screen replaced the peephole view of the film provided on earlier models.

Technology such as this, when integrated into the editing routines inherited from the silent era, enabled the filmmaker to time cuts quite exactly. What Barry Salt calls the conventional "dialogue cutting point" in sound cinema—cutting away from the speaker just before the last word or syllable has been spoken—became feasible with frame-by-frame matching of picture and sound.²⁰ Although the Light Brigade example does not exemplify this tactic, the sequence does bleed one line over a cut, and it cuts on pauses in a way that shows how sound and picture editing cooperate in creating a scene's rhythm.

The new recording, mixing, and editing technology also permitted intricate manipulation of music. Silent films had had continuous musical accompaniment in the theater, but early talkies largely relinquished the use of nondiegetic (i.e., non-story-source) music. Once rerecording techniques could assure reasonable sound quality, however, Max Steiner, Alfred Newman, Erich Wolfgang Korngold, and others began to write long stretches of music. King Kong (1933) contains an hour of original music, with some passages running for many minutes. Korngold's score for Captain Blood (1935) lasts over half of the film's running time. During the 115 minutes of The Charge of the Light Brigade, there are only about twenty-three minutes when Steiner's score is not heard!

Though these have come to be known as "symphonic" scores, they more closely resemble operatic accompaniment to vocal recitative.²¹ As noted in the Light Brigade analysis, the music ebbs and flows underneath the dialogue, emphasizing certain phrases, filling the pauses, linking scenes, or cutting off dramatically. The studio composer might write eleven- or twelve-minute passages, or bits only ten seconds long; the music might have to "sneak in" under dialogue or rise to mezzo forte very quickly.

The sound-recording and mixing techniques reviewed above made such split-second timing possible, as did some more specialized devices. The "click track" was originally a black leader punched with holes at intervals that matched the pacing of the action; these clicks gave the composer the rhythmic pulse of the music needed. Later, in recording the score, the conductor would lead the orchestra in time to the click track. The mixer could blend the score with other tracks with the help of Western Electric's "automatic balance regulator," which controlled the mixing balance between dialogue and music. Now the music could comment, moment by moment, upon the dramatic action without distracting from it.

In general, improvements in sound technology in the thirties reinforced the basic assumptions of Hollywood's stylistic tradition. Storytelling was enhanced by a variety of unobtrusive technical devices. The image track's emphasis on the human face was paralleled by an emphasis on the voice, and various technological innovations enabled clearer recording of speech. Certain canonized devices, such as the continuity cut, were strengthened by the judicious use of sound. Overall, the sound film gained a controlled economy and a brisk pace governed by dialogue.

Camera Movement. Camera movements had been part of Hollywood's technical repertory during the silent era, and the coming of sound did not eradicate them. Many early sound films display rather complex camera movements, either shot silent or filmed from wheeled booths. Some observers believed that The Last Laugh (1924), Variety (1925), and Applause (1929) had started a vogue for flamboyant traveling shots.²² In the course of the decade, more unobtrusive camera movements became quite common, partly because of technological factors that made them reasonably efficient and partly because they created a fluidity that was considered valuable in the talking cinema.

In the silent era and in the early days of sound filmmaking, most shots were made with the camera on a tripod. But the weight of a heavily blimped sound camera strained ordinary tripods and made the camera difficult to move around the set. It soon became evident that greater production efficiency and stylistic flexibility could be achieved by retooling the sort of camera carts used in the silent era.

The initial prototype was the Bell & Howell Rotambulator, introduced in 1932. This was a seven-hundred-pound carriage about five feet long and four feet wide. Using a central column on a turntable, the Rotambulator could set the camera at any height between eighteen inches and seven feet. The seated cinematographer could pan, tilt, track, rotate, or raise the camera on the vertical shaft. On the most sophisticated model, hydraulics enabled the cameraman to control panning with foot pedals while tilting by means of a hand lever.²³

The Rotambulator, introduced in 1932 and disseminated in the following year, was quickly revised. The Fox Velocilator (introduced in 1933 and manufactured by the Fearless Camera Company) had many of the same features as the Rotambulator, but it was only thirty inches wide and weighed only three hundred pounds. Moreover, instead of a central column on a turntable it made use of an angled boom arm that raised or lowered the camera. Cinematographer John Seitz explained of the Velocilator, "While its primary purpose is, of course, facilitating the making of traveling shots, it has proved equally valuable as a means of saving time and effort in changing camera set-ups."²⁴

Advantageous features of the Rotambulator and the Velocilator were combined in the Fearless Company's Panoram Dolly, which debuted in 1936. The camera was anchored to a boom arm, which in turn rested on a turntable on a four-wheeled chassis. It could pass through a thirty-six-inch door, and two seats were provided for camera personnel. The Fearless dolly, or copies of it, quickly became standard studio equipment.²⁵

Such standardized camera carriages naturally increased filmmakers' ability to make lengthy tracking shots. During the mid 1930s some films utilized virtuosic camera movements. The Velocilator appears to have been the basis of Erik Charell's Caravan (FOX, 1934); the film's average shot lasts thirty-seven seconds, and Charell uses camera movement to shift in and out of flashbacks within a single shot. More orthodox films also made increased use of camera movements. By 1938, one SMPE committee could report that the growing emphasis on camera movement had made "a very definite contribution to the continuity and smoothness of action of the motion picture story."²⁶

But many cinematographers complained about the "orgy of indiscriminate dollying": moving shots were complicated and expensive.²⁷ Most cameramen and directors favored the unnoticeable tracking movements that followed the movements of the players, as in the Charge of the Light Brigade sequence. One observer spoke for many when he explained that the best talkies' style offered "a skillful blend of cuts and mobile camera, achieving the fullest cinematic effectiveness by an entirely unself-conscious 'participation' in the 'central emotional strain' threading rhythmically through the story."²⁸

Highly noticeable camera movements did have their uses, however. Most often, they could be used at the beginning of a scene, as John Cromwell noted:

New York: Norton, 1937], p. 60">

Fundamentally [the director] has a choice of two kinds of technique: starting on a fairly close shot of some action that is either revelation of the characters or story, then proceeding back to a revelation of time and place, or establishing the locale first and then progressing up to the characters. Using the first method, the picture might open on a close shot of a newspaper lying on the pavement with the camera tilting up to the little boy who is standing over the newspaper reading it, and from there moving horizontally or panning over to a man standing on the street and watching the little boy, when the story begins. Using the second method, you might open on a long shot of the street, then move the camera towards the little boy in what is known as a trucking shot, and finally pan over to a shot of the man as the story begins. ("The Voice Behind the Megaphone," in Nancy Naumberg, ed., We Make the Movies [New York: Norton, 1937], p. 60)

While Michael Curtiz uses neither tactic at the start of the specimen scene, he does obey Cromwell's suggestion in other sequences. We can thus get a sense of the most common functions of 1930s camera movement: following figure movement or opening a scene by stressing a detail.

The new camera devices had other advantages as well. Hydraulic control of panning and tilting reduced the sort of jerky reframing that can be seen in some early talkies. Moreover, allowing the camera operator to sit comfortably and to manipulate reframing with feet and one hand represented an improvement over tripod heads that required the use of both hands. The new dollies made feasible the sort of tight, minute adjustments of framing seen in the Light Brigade sequence.

The growth of the moving-camera style affected other equipment and procedures. Lighting units on the set floor hampered the dolly, so production became more reliant on hanging units. Since these put the light source farther from the action, spotlighting units became preferable to the broader units used on the floor. The new demand for spotlights in turn triggered a series of refinements in them, most notably the use of Fresnel lenses to focus the illumination.²⁹ Through such ripple effects, the preferred stylistic alternatives shaped the direction of technological innovation.

While various dollies were being designed and adopted, some practitioners experimented with larger camera carriages. The enormous Universal camera crane used on Broadway (1929), with its twenty-five-foot-long arm, had represented the summit of such efforts at the end of the silent era. The small sets in early talkies did not favor such equipment, but after 1931, cranes were proving feasible again. The Broadway behemoth was trundled out for All Quiet on the Western Front (1930), and MGM and UA also began using cranes. In 1932, at an Academy equipment demonstration, Paramount displayed its "Baby" camera boom, which could rise from floor level to a height of about eight feet.

Often such cranes were used for transporting the camera from one setup to another and for taking static high-angle shots (thus obviating the need to build a special camera platform). But a 1931 article pointed out that filming from a moving crane yielded fluidity and spectacle. Citing an unspecified film, the writer noted that a powerful effect was achieved by tracking with the main characters as they entered a ballroom and then craning up to show the entire dance floor. "Such photography, interspersed throughout the more commonplace scenes, does much to determine the success or failure of any motion picture. The problem of keeping the plot moving along and at the same time impressing the beauty and color of setting upon the spectator is solved in many instances by the camera crane."³⁰

By 1936, most production companies had flexible dollies and either owned or rented large cranes that could rise twenty or thirty feet. In addition, the Velocilator had shown that a small unit could utilize a cranelike angled arm, so many firms sought to build middle-range cranes that could work between eight and twenty feet off the studio floor. One of the most noteworthy of these was John Arnold's MGM boom, introduced in 1939, which had a maximum height of sixteen feet. The crane's light weight, underslung camera mounting, and highly maneuverable wheels were designed to make it the preferred camera carriage for virtually any shot.³¹ The boom won Arnold an Academy Technical Award.

Flamboyant crane shots are visible in several films of the early and mid 1930s, often being used to establish the setting in an arresting way (e.g., Street Scene, 1931). Crane shots are a major attraction of the musical numbers in such Busby Berkeley films as Footlight Parade (1933) and Gold Diggers of 1933. Stately crane movements are particularly in evidence in MGM's The Wizard of oz (1939) and Gone With the Wind (1939). There was usually no attempt to make such shots as unobtrusive as the mild track-ins and reframings already noted. Instead, crane shots were used to impart a flow to ensemble musical numbers and to indulge in a momentary display of spectacle, as in Gone with the Wind's legendary crane back through the railroad yard full of wounded soldiers, which utilized a sixty-foot construction crane.³² On the whole, the crane shot remained a valuable but secondary tool in a cinema devoted to storytelling rather than camera pyrotechnics.

Technicolor. Although it would be some decades before most Hollywood films would be made in color, the 1930s saw a significant interest in color filmmaking, and this revolved chiefly around the firm of Technicolor.

In outline, the firm's history is fairly clear. The years 1929-1930 saw a boom in Technicolor, chiefly in such musicals as On with the Show (1929), Gold Diggers of Broadway (1929), and Whoopee! (1930). These films used the two-color Technicolor process, which favored reddish and greenish hues. After the box-office decline of the early 1930s and complaints about color quality, interest in the process waned. Technicolor developed its imbibition-printing method in 1930 and its three-color process in 1932. These greatly improved color techniques were tested in Disney's cartoon Flowers and Trees (1932), in the short subject La Cucaracha (1934), and in the feature Becky Sharp (1935). A new color vogue gradually emerged, signaled by the success of The Trail of the Lonesome Pine (1936). This second wave of success gathered momentum in 1939, when a faster film stock was used on Drums Along the Mohawk, The Private Lives of Elizabeth and Essex, Gone with the Wind, and other major productions. Throughout the 1940s between twenty and fifty Technicolor features were produced every year.

The Technicolor camera used prisms to split the light beam and record different wavelengths on three strips of film.³³ As a result, the cameras were cumbersome and difficult to thread. Similarly, Technicolor's imbibition developing of color relied on a process akin to lithography and could not be accommodated by standard studio laboratories. Throughout Technicolor's history, it therefore had to convince the production companies that its technology was worth the trouble. Technicolor financed independent films to demonstrate various processes: The Gulf Between (1917) displayed the red-green additive process, and The Toll of the Sea (1922) exhibited the two-color subtractive method. After the decline in studio interest in the early 1930s, Technicolor used the Disney cartoons as a showcase for the three-component and imbibition processes. John Hay Whitney, a Technicolor stockholder, produced La Cucaracha and Becky Sharp under the auspices of his own firm, Pioneer Pictures. Only then did the studios feel that Technicolor's increased cost (almost 30 percent over black-and-white film) might be justified for certain productions.

The Technicolor company exemplifies how a specialized service firm could be governed by engineering principles of standardization and efficiency. Founded in 1915 by two engineers from the Massachusetts Institute of Technology (MIT), Herbert Kalmus and Daniel Comstock, Technicolor was named after their alma mater. Kalmus and Comstock, along with W. Burton Westcott, had already served as scientific consultants for the chemical industry. The three men decided to ground their color-film enterprise in practical industrial research. They secured investment capital from New York lawyers, advertising agencies, and manufacturing companies. They also hired MIT students, one of whom, J. A. Ball, was eventually to devise the three-color system. In 1925, Comstock left the firm, making Kalmus the executive head of Technicolor. By 1930, he had a Hollywood laboratory said to employ six hundred people. Knowledge was kept compartmentalized for the sake of secrecy; only the executives had an overall view of the enterprise.³⁴

Kalmus maintained that the quality problems in the two-color process had been due largely to inexperienced studio cameramen, so he insisted that the company had to supervise every phase of the process. For most 1930s productions Technicolor supplied not only the film stock but the cameras, the processing, the release prints, a company-approved cinematographer, the camera crew, and several advisers, notably the "color consultant" (usually Kalmus's ex-wife, Natalie). At the start of every shooting day, the crew would bring loaded camera magazines and the cleaned and inspected cameras from the company. Since threading a Technicolor camera did not go quickly, the production unit typically rented two cameras so that one could be shooting while the other was being reloaded.

For all his autonomy, Kalmus was obliged to work with related service firms. While his engineers knew physics and chemistry, their camera blueprints had to be turned over to other companies for realization; Mitchell produced three-strip cameras to Ball's design. The firm also depended upon Eastman, not only for supplying raw film stock but also for such innovations as the 1939 emulsion that cut required lighting levels in half.³⁵

Also notable was the change in lighting equipment that followed from the mid-1930s color boom. Three-strip Technicolor was balanced for daylight, which was best matched by arc illumination. On the studio floor, however, arcs had largely been replaced by the quieter incandescents, which could be used with Technicolor in limited ways but could not deliver the intensity of light desired. Spurred by this new demand, Mole-Richardson devised silent, flickerless are spotlights and broadsides for color production. These were in turn quickly adapted to black-and-white filming.

Technicolor created no radical changes in classical studio filmmaking. Instead the process accommodated itself to the reigning aesthetic norms. For example color risked calling attention to background details that would have gone gray in black-and-white filming. Natalie Kalmus therefore urged that backgrounds be darkened so that the warmer flesh tones would come forward.³⁶ This recommendation assumed that the human figure was the basis of the action. "The one key color present in every scene," wrote a Technicolor consultant, "is the face of the actor."³⁷ Technicolor aesthetics in the 1930s were governed by producer Edgar Selwyn's advice to Daniel Comstock: "The human being is the center of the drama. The face is the center of the human being. And the eyes are the center of the face. If a process is not sharp enough to show clearly the whites of a person's eyes at a reasonable distance, it isn't any good no matter what it is."³⁸

At the same time, Technicolor provided an unrivaled degree of spectacle. Used for maximum intensity, as in The Wizard of Oz (1939), it could yield bright saturated hues suitable for fantasy. Cinematographers of the 1930s also learned to exploit Technicolor's rich shadows and subtle pastels, as in Trail of the Lonesome Pine (1936), The Garden of Allah and Goldwyn Follies (1938). Most circulating copies of Nothing Sacred (1937), A Star Is Born (1937), and The Adventures of Robin Hood (1938) have lost the dee] blacks and vibrant purples and greens that the process could render.³⁹

Technicolor also had straightforward narrative uses. Like music and camera movement, it could underscore a dramatic moment: in Drums Along the Mohawk (1939) a gloomy medium shot bursts into red-orange when a rifleman is hit by a flaming arrow. And, like music, color could create motifs associated with a certain character or situation—a tactic developed in the green-faced Wicked Witch of the West. When Technicolor succeeded in making the splendors of color flatter the actor and intensify the ongoing drama, it proved itself suitable for adoption into the mainstream Hollywood style.

Special Effects. Although the term "special effects" conjures up thoughts of films of fantasy like King Kong and The Wizard of Oz, it is worth considering this characterization of special effects, offered in 1943 by William Stull, president of the American Society of Cinematographers:

Ninety per cent of the trick and special effects shots in Hollywood movies are made with no thought of fooling or mystifying the audience. The great majority of camera trickery is used simply because filming the same action by conventional means would be too difficult, too expensive, or too dangerous. Virtually every picture that leaves Hollywood's major studios includes a greater or lesser proportion of scenes made by one or more of the trick processes. Only in the rare instances of a film like Dr. Cyclops, The Invisible Man, or Topper is the process work done as a deliberate trick. ("Process Cinematography," in The Complete Photographer, vol. 8, edited by Willard D. Morgan [New York: New York Educational Alliance, 1943], p. 2994)

Because special effects offered economy, ease, and safety, as well as spectacle and fantasy, they were used extensively, and they affected the style of many films made during the 1930s.

During the silent era, the cinematographer and camera assistants created most special effects in the camera itself, without the aid of additional work in the laboratory. To make a dissolve between shots, the camera operator faded out at the end of one shot, wound the film back, and faded in at the beginning of the exposure of the second shot. Similar techniques were used for superimpositions, split-screen exposures, and other effects. If the two shots involved were not to be filmed in succession, the cinematographer often had to remove the film from the camera, carefully mark the footage already filmed, and replace it in the camera when the second shot was to be made. Such procedures were time-consuming and required enormous skill on the part of the camera staff, but they often yielded results that are still impressive. In Buster Keaton's 1921 comedy The Playhouse, for example, multiple images of Keaton perform onstage in the same shot, as a result of great precision in winding the film back repeatedly to expose one tiny portion of the frame after another.

In general, during the silent era such effects were called "tricks." By 1928, however, the term special photographic effects came into common usage. By then, effects work was increasingly carried out by specialists working in separate special-effects departments in the main studios.

There are a number of reasons for this change. Most important, the introduction of sound took much control over the creation of the film away from the cinematographic team. In multiple-camera shooting, successive shots that might be linked by a dissolve would not necessarily be shot by the same camera; hence, the camera could not make the transitions. Moreover, shooting the whole scene straight through precluded stopping in the middle to rewind the camera and make a superimposition or a dissolve. (Even with the return to single-camera filming, the method of making a single master shot would inhibit in-camera effects.) Sound filming also called for less location shooting. Recording usually had to be done in soundproof studios; going on location to a busy street would make it difficult to exclude random noises. Instead, it was more practical to shoot the street scene separately and project it on a screen behind the actors in the studio as they were filmed speaking their lines. Such a procedure would also save money. Thus, for a variety of reasons, the blending together of images in the camera was minimized in the 1930s. Instead, a film became more of an assemblage of separate images (and sound tracks) shot and combined by specialists. This trend intensified over the decade.

Most types of special effects involved combining separate images in one of two ways. One familiar technique introduced during this era is rear projection (also called "process projection"). Here the background of a shot is filmed separately and projected onto a screen behind the actors. Alternatively, an optical printer might be used to expose a negative twice or more in order to create a composite image out of different components: painted settings, miniature landscapes, partially exposed positive prints, and the like. Such techniques cut costs and allowed filmmakers to redo the effect until they were satisfied—a choice usually not available to the silent-film cameraman.

Given these advantages, why were such techniques as back projection and optical printing not introduced earlier? Only in the late 1920s did a combination of technological factors make them possible. Farciot Edouart, head of special effects for Paramount, described how cinematographers had dreamed of using back projection: "Three key factors were lacking to make the dream into reality. We needed a simple, non-mechanical method of synchronizing the background projector and the foreground camera. We needed negative emulsions of sufficient sensitivity to enable us to record the back-projected picture. And we needed optics and light sources of increased power to enable us to get a brighter image through our background screens."⁴⁰ All three conditions were met within a short span of time. The introduction of sound brought motors to synchronizing camera and recording equipment, and these were easily adapted to link camera and back projector (so that the camera and projector shutters were open at the same instant). Highly sensitive panchromatic negative film, introduced in 1928, made it possible for the camera to pick up the relatively faint image thrown by the back projector. Rear-projection technology also made use of the powerful projector lamps manufactured for the big screens of the movie palaces and the wide-gauge formats that emerged in the late 1920s (e.g., the 70-mm processes Fox Grandeur and MGM Realife).

Since no commercial back-projection systems were available, studio personnel built their own equipment to suit their individual needs. Fox tried out the new technology in Liliom and then on Just Imagine (both 1930), a futuristic fantasy; the company received a special 1930-1931 Academy Award for this work.⁴¹ By the middle of the decade, rear projection had become a common way to save money. In one shot from Fury (1936; Fig. 18), Spencer Tracy sits in a three-dimensional car, but the landscape and road that appear to recede behind him are projected on a flat screen. More elaborate is one scene from Young Mr. Lincoln (1939; Fig. 19) that features a larger studio set, but the sky and river behind are a projection. A big battle scene in The Plainsman (1936) was staged primarily on a sound stage at Paramount, where footage of six thousand Indians and a cavalry regiment were added on two large screens at the rear, the gap between the two screens being hidden by a dead tree in the set.⁴² The 1937 film Captains Courageous was set primarily on the deck of a fishing schooner at sea, yet again the main actors stayed in the studio, with real water in a large tank below them and seascapes and sky projected on a screen above. A film could even use miniature settings or animated figures as part of rear-projection footage, as in the complicated special effects in King Kong (1933).

The industry made efforts to standardize the rear-projection process. In 1936 the studios agreed to cross-license almost fifty patents they had acquired on process projection. Out of its work for Technicolor, Mitchell created a rear projector possessing increased steadiness, and Eastman's Super-X panchromatic film was designed partly to improve rear-projection filming. In 1938 the Academy Research Council sponsored a committee of fifty experts to devise a set of standard specifications that all rear-projection systems should meet. After a year

of deliberations, the committee released a report detailing uniform rear-projection specifications.⁴³ Mole-Richardson, Bausch & Lomb, and Mitchell soon set to work furnishing the requisite equipment.

The optical printer was a more complex device than rear-projection equipment, and it offered several possibilities for rephotographing material. The main obstacle to the invention of the optical printer had been the lack of an adequate film stock for duplicate photography. Before 1927, all such stock yielded too contrasty an image; in that year, Kodak introduced the first successful negative duplicating emulsion. That made it possible for a machine to rephotograph portions of existing images onto one strip of film. As with rear-projection equipment, there was no standardized optical printer available commercially. Nevertheless, by the early 1930s most major studios were establishing effects departments centering on optical printers built by their employees.

Essentially an optical printer consisted of a camera, a lens mount, and a projection device facing the camera lens, all mounted to slide along a single track in a lathe bed. The projector displays an image, portions of which are photographed by the camera, usually with only part of the negative being exposed at one time. A second run through the camera exposes another portion of the negative by photographing a different image in the projector, and so on. The optical printer may be used to add a "matte painting" to a shot. A matte is simply a mask that blocks part of the negative while it is being exposed, either in the original camera or in the optical printer. When the matte is removed and a second matte covers the exposed portion, the unexposed portion is run through the camera to register a second image. In Figure 20, from Queen Christina (1933), a harbor scene has been created by exposing the original film of the actors and a three-dimensional portion of the set at the lower left and then by covering that part of the frame and photographing a painting of roofs, ships, sky, and water at the right and above.

Such an effect, in which the line separating the two portions of the image does not move, is called a "static matte." A "traveling matte" is far more difficult to execute, since the effects expert must make a whole series of pairs of mattes, each with a slightly different join line. Then the film must be exposed one frame at a time with only one of the mattes and then exposed again with only the other matte. The result is a moving joint between two separate images.

During the 1930s, traveling mattes were commonly used to make "wipes," in which a line moves across a shot, bringing a new image gradually onto the screen without superimposing it over the previous one. Optical-printer virtuoso Linwood Dunn experimented with wipes of various shapes and directions in such RKO films as Flying Down to Rio (1933); sawtooth lines, starbursts, and spirals create elaborate transitions between shots. Dunn devised more subtle and virtually undetectable traveling mattes so that Cary Grant and Katharine Hepburn could appear to be in the same shots with a leopard in Bringing Up Baby (1938).

Optical printers took over most of the functions originally performed in the camera in the silent era. Such printers easily provided fades and dissolves, made superimpositions and split-screen effects, and so on. For example, in Love Me Tonight (1932), a split-screen effect used mattes to join separate images of the hero and heroine in their respective beds; thus, the scene suggests that they are really together in the same bed (Fig. 21). A series of images could be linked through superimpositions, wipes, and split-screen effects into the familiar "montage sequence." Such a sequence offers a compressed account of passing time or large-scale processes, as in the newspaper montages in Mr. Deeds Goes to Town (1936; Fig. 22).

As these examples suggest, optical-printer effects were extremely useful and quite common, and their impact is evident in the stylistic devices of many 1930s

films. Even the introduction of color caused only brief problems for experts in matte painting and other optical-printer effects. Experimentation soon revealed how the colors in the paintings could be matched to those in the photographed live scene. Matte work was first used extensively for a color production in The Garden of Allah (1936), and techniques devised at that point were ready to be expanded for Gone with the Wind (1939), which depended heavily on matte paintings and other special effects.⁴⁴

Although optical printers were generally built by the studios rather than the supply companies, America's entry into World War II created a need for them in military cinematography units, and Linwood Dunn, still at RKO, designed one that was manufactured by a small firm as the Acme-Dunn printer. Dunn's invention was significant enough to win his colleagues one Oscar in 1944 and another, for a later version of their device, in 1981.⁴⁵

Cinematography Styles. Throughout the sound era, multiple-camera shooting never completely vanished. It remained a staple for coverage of unrepeatable action, such as burning buildings or cars careening off cliffs. It was sometimes considered an efficient way to shoot spectacular musical numbers.⁴⁶ Certain performers also required multiple-camera procedures. "Oh God, the Marx Brothers were difficult!" recalled Joseph Ruttenberg. "For one thing, you never knew what they were going to do, so you had to have a camera on each one of them. With the long-shot camera, that added up to four cameras, and lighting for four cameras is pretty dull."⁴⁷ The results can be seen in Horse Feathers (1932): Groucho's cavorting with the college widow is framed from two slightly different angles (Figs. 23-24).

On the whole, though, the return to single-camera shooting allowed filmmakers to revive and refine tendencies that had emerged in the silent era. One of the most prominent of these was the "soft" style of cinematography, which had become particularly prominent in the late 1920s.⁴⁸ Cinematographers habitually diffused the lights or the lens in order to blur contours and lower contrast, yielding an image with muted highlights and more detailed shadow areas. By 1927, orthochromatic film had largely been replaced by panchromatic film, and the use of panchromatic stock and incandescent lighting in early talkies continued to produce rather soft images. After single-camera shooting was reestablished, many cinematographers sought to return to the soft style.

Eastman cooperated with the trend by introducing in 1931 its Super Sensitive Panchromatic, designed for use with incandescent light. Two to three times as sensitive as earlier stocks, the new panchromatic could also soften light values. The ASC formally recommended the new film because it enabled the cinematographer to lower the lighting levels on the set. The film was used on several major 1931 and 1932 productions, notably Five Star Final, Dr. Jekyll and Mr. Hyde, and Shanghai Express. Soon Du Pont introduced its own panchromatic negative, and Bausch and Lomb designed its Raytar lenses for use with incandescent light and the new emulsions. By 1932, the more sensitive panchromatic became standard in most circumstances.

The new film stocks offered various possibilities. Most cinematographers used them to guarantee a somewhat soft image with more latitude in developing. As a rule, the soft style survived in only attenuated form in the 1930s; few films were as heavily diffused as those in the silent era. The sparkling image might be used to convey a romantic atmosphere, as in A Farewell to Arms (1932; Fig. 25); a more pervasive softness was reserved for fantasy, as in A Midsummer Night's Dream (1935; Fig. 26).

Cameramen also discovered that Super Sensitive Panchromatic yielded unprecedented definition. On exteriors, pointed out one cinematographer, the film was almost too crisp, piercing haze and revealing contours not visible to the naked eye.⁴⁹ The new emulsion's power was signaled when the ASC recommended using Eastman Super Sensitive: "With the present lightings and smaller lens openings, improved definition can be obtained without sacrifice of those qualities of softness which have always been the artistic aim of cinematographers."⁵⁰ The range of possibilities is spectacularly visible in Shanghai Express. For some shots, the Eastman stock enabled Lee Garmes and Josef von Sternberg to soften the highlights with relatively little diffusion (Fig. 27). In Figure 28, however, the new emulsion renders edges and planes razor-sharp. (Compare the much blurrier contours, partly a result of filming through the windows of camera booths, in Millie; Figs. 12-17 earlier.)

Some filmmakers were alert to the new possibilities of filming in depth, because this represented one option in the stylistic paradigm of the late silent era. In multiple-camera filming, however, staging and filming tended to create a flatter space of performance. During the 1930s, "deep-space" and "deep-focus" techniques reemerged.

Several stylistic choices are relevant here. A filmmaker can stage a scene in a relatively "shallow" fashion, putting all the relevant action on a single plane standing out against a background. The profiled two-shots from Millie exemplify shallow staging. Alternatively, the filmmaker can stage the action in greater depth by putting salient elements of the scene in two or more distinct planes. By choosing to stage in depth, the filmmaker faces a new set of alternatives. The filmmaker may adjust camera and lights so that only one of the planes will be in sharp focus, as in Figure 29, from William Wyler's Jezebel (1938). Or the filmmaker may opt for what was called "pan focus" or "deep focus"—that is, rendering all of the planes in sharp focus, as in The Public Enemy (1931; Fig. 30).

Most often, when deep focus depicted deep-space staging, the foreground plane was framed in a long shot or a medium long shot, as in the Public Enemy example. This option simplified photographic control, since sharp focus was usually easy to carry from several feet to infinity. But in the cinema of the 1910s and 1920s one can find occasional examples of filmmakers setting one plane fairly close to the camera and seeking to carry focus all the way to the rear plane. A famous instance is Figure 31, from Greed (1924). If such a shot looks "modern," it is because Citizen Kane's close-up foregrounds helped launch a vogue for extreme deep focus in the 1940s and 1950s. In the context of 1920s cinema, however, the deep-focus style, with its hard-edged planar separation, was a functional alternative to the soft style.

During the early sound era, several filmmakers sought to present images with significantly deep focus. The art director William Cameron Menzies strove for closer foregrounds in Bulldog Drummond (1929) and Chandu the Magician (1932; Fig. 32). James Wong Howe's cinematography for Transatlantic (1931) also experimented in this direction. At Goldwyn studios during the early 1930s, there was something of a deep-focus vogue, typified by John Ford's Arrowsmith (1931; Fig. 33).

Technological developments facilitated the trend toward depth. The film stocks already mentioned were the beginning of a series of faster and more sensitive emulsions from Eastman, Du Pont, and Agfa. These stocks facilitated the lowering of lighting levels on the set, and producers welcomed this as an efficiency measure. But inquiries by the Academy and the SMPE revealed that some cinematographers were keeping the lighting levels high and stopping down the camera diaphragm.⁵¹ This raised the likelihood that the shot would render more planes in focus. In addition, during the mid 1930s, Mole-Richardson's arc lights developed for Technicolor began to be used with black-and-white, and these gave a harder, more directional light that favored the deep-focus approach.

Over the same period, cinematographers began gravitating toward lenses that could render several planes in focus. These "wide-angle" lenses were of short focal length (usually 35 mm or 25 mm). In the silent era they had been used principally for long shots, but Bausch & Lomb's improved Raytar lenses (introduced in 1931) enabled cinematographers to use them for closer views as well. In 1932 one cinematographer claimed that he was shooting 85 percent of a film with a 25-mm lens.⁵² Late in the decade a new chemical coating increased the light-gathering capacities of the lens, and this, too, allowed cinematographers to stop down the apertures for greater depth of field.

Some cinematographers took advantage of these developments to cultivate a deep-focus look. James Wong Howe, for instance, designed Viva Villa! (1934) to have strongly foreshortened compositions, and he filmed all close-ups with wide-angle lenses.⁵³ Tony Gaudio's work in Anthony Adverse (1936) yielded a marked depth in certain images, with some exaggerated foregrounds (Fig. 34). Most famously, Gregg Toland cultivated a reputation as a cinematographer identified with deep-space staging and deep-focus filming. In his early 1930s work for Goldwyn, such as Nana (1934), there are strong foregrounds but without deep-focus treatment. In Toland's work with William Wyler later in the decade, however, we sometimes find a dense deep-focus style. Dead End (1937), for example, is dominated by stifling foregrounds (Fig. 35). Such tactics would become central to Toland's work with Ford, Welles, and Wyler during the 1940s.⁵⁴

Most deep-focus efforts were more subdued than the Wyler-Toland extravagance; the Light Brigade sequence, with its mild diagonal entrances and exits in depth, is more typical of the period. Many directors sought to avoid obtrusive

pictorial effects by staging in significant depth but framing the foreground plane only in medium shot. Examples are Ford's Young Mr. Lincoln (1939; Fig. 36) and Howard Hawks's Only Angels Have Wings (1939; Fig. 37). The technique could also be used to intensify reverse-angle framings, as in the wide-angle close-ups of King Vidor's Stella Dallas (1937; Fig. 38) and Raoul Walsh's The Roaring Twenties (1939; Fig. 39). By the end of the decade, the depth approach had established itself as a significant stylistic alternative. It remained for Welles, Wyler, Hawks, and Toland to push deep-space staging and deep-focus filming to still greater extremes in Citizen Kane, The Little Foxes, and Ball of Fire (all 1941).

Conclusion

The truly new technology of the 1930s, talking pictures, set the pace for much of the technological change of the decade. Sound technology had to be made more faithful and flexible, but it also had to be integrated into the stylistic priorities of classical storytelling. Much of the research into microphones, recording media, reproduction equipment, and editing instruments was therefore devoted as much to problems of narrative coherence and spectacle as to strictly engineering matters.

At the same time, filmmakers and researchers bent their efforts to recovering the visual and narrative dynamism that the early sound cinema was believed to have lost. Abandoning multiple-camera shooting, filmmakers sought to increase flexibility in editing, camera movement, special effects, and photographic qualities of the image. The industry and its professional organizations defined, often with considerable precision, the goals that technology had to meet, and the manufacturing and supply firms set out to achieve them. The ASC, the SMPE, and the Academy created channels of communication and promoted favored solutions. In sum, it was during the 1930s that the film industry fully realized its ability to standardize the very process of technological innovation.

At the same time, the pursuit of efficiency and stylistic standardization led to the exploration of new possibilities, such as the reintroduction of arc lighting and the devising of sophisticated back-projection equipment. Such developments could not have been predicted, but they were seized upon as ways to enhance the classical style.

During the 1930s, Hollywood recalibrated its filmmaking techniques, creating in the process many stylistic qualities that still endure. It also consolidated a systematic approach to technological innovation and assimilation that remains in force today.