The wholesale conversion to sound of all three sectors of the American film industry took place in less than 15 months between late 1927 and 1929, and the profits of the major companies increased during that period by as much as 600 percent. Although the transition was fast, orderly, and profitable, it was also enormously expensive. The industrial system as it had evolved for the previous three decades needed to be completely overhauled; studios and theatres had to be totally reequipped and creative personnel retrained or fired. In order to fund the conversion, the film companies were forced to borrow in excess of $350 million, which placed them under the indirect control of the two major New York-based financial groups, the Morgan group and the Rockefeller group.
Furthermore, although cooperation between the film companies through such agencies as the MPPDA, the Academy of Motion Picture Arts and Sciences, and the Society of Motion Picture Engineers ensured a smooth transition in corporate terms, inside the newly wired theatres and studio soundstages there was confusion and disruption. The three competing systems—Vitaphone, Movietone, and Photophone—were all initially incompatible, and their technologies were under such constant modification that equipment was sometimes obsolete before it was uncrated. Whatever system producers chose, exhibitors during the early transitional period were forced to maintain both sound-on-disc and sound-on-film reproduction equipment. Even as late as 1931, studios were still releasing films in both formats to accommodate theatres owned by sound-on-disc interests.
It was in the area of production, however, that the greatest problems arose. The statement that “the movies ceased to move when they began to talk” accurately described the films made during the earliest years of the transition, largely because of technical limitations. Early microphones, for example, had a very limited range. In addition, they were large, clumsy, and difficult to move, so they were usually concealed in a single, stationary location on the set. The actors, who had to speak directly into the microphones to register on the sound track, were therefore forced to remain practically motionless while delivering dialogue. The microphones caused further problems because they were omnidirectional within their range and picked up every sound made near them on the set, especially the noisy whir of running cameras (which were motorized in 1929 to run at an even speed of 24 frames per second to ensure undistorted sound synchronization; silent cameras had been mainly hand-cranked at rates averaging 16 to 18 frames per second). To prevent the recording of camera noise, cameras and their operators were initially enclosed in soundproof glass-paneled booths that were only 6 feet (2 metres) long per side. The booths, which were facetiously called “iceboxes” because they were uncomfortably hot and stuffy, literally imprisoned the camera. The filmmakers’ inability to tilt or dolly the camera (although they could pan it by as much as 30 degrees on its tripod), combined with the actors’ immobility, helps to account for the static nature of so many early sound films.
The impact of sound recording on editing was even more regressive, because sound and image had to be recorded simultaneously to be synchronous. In sound-on-disc films, scenes were initially made to play for 10 minutes at a time in order to record dialogue continuously on 16-inch (41-cm) discs; such scenes were impossible to edit until the technology of rerecording was perfected in the early 1930s. Sound-on-film systems also militated against editing at first; optical sound tracks run approximately 20 frames in advance of their corresponding image tracks, making it extremely difficult to cut a composite print without eliminating portions of the relevant sound. As a result, no matter which system of sound recording was used, most of the editing in early sound films was purely functional. In general, cuts could be made—and the camera moved—only when no sound was being recorded on the set.
Most of these technical problems were resolved by 1933, although equilibrium was not fully restored to the production process until after the mid-1930s. Sound-on-disc filming, for example, was abandoned in 1930, and by 1931 all the studios had removed their cameras from the iceboxes and converted to the use of lightweight soundproof camera housings known as “blimps.” Within several years, smaller, quieter, self-insulating cameras were produced, eliminating the need for external soundproofing altogether. It even became possible again to move the camera by using a wide range of boom cranes, camera supports, and steerable dollies. Microphones too became increasingly mobile as a variety of booms were developed for them from 1930 onward. These long radial arms suspended the microphone above the set, allowing it to follow the movements of actors and rendering the stationary microphones of the early years obsolete. Microphones also became more directional throughout the decade, and track noise-suppression techniques came into use as early as 1931. Postsynchronization
The technological development that most liberated the sound film, however, was the practice known variously as postsynchronization, rerecording, or dubbing, in which image and sound are printed on separate pieces of film so that they can be manipulated independently. Postsynchronization enabled filmmakers to edit images freely again. Because the overwhelming emphasis of the period from 1928 to 1931 had been on obtaining high-quality sound in production, however, the idea that the sound track could be modified after it was recorded took a while to catch on. Many motion-picture artists and technicians felt that sound should be reproduced in films exactly as it had originally been produced on the set; they believed that anything less than an absolute pairing of sound and image would confuse audiences.
For several years, both practice and ideology dictated that sound and image be recorded simultaneously, so that everything heard on the sound track would be seen on the screen and vice versa. A vocal minority of film artists nevertheless viewed this practice of synchronous, “naturalistic” sound recording as a threat to the cinema. In their 1928 manifesto “Sound and Image,” the Soviet directors Sergey Eisenstein, Vsevolod Pudovkin, and Grigory Aleksandrov denounced synchronous sound in favour of asynchronous, contrapuntal sound—sound that would counterpoint the images it accompanied to become another dynamic element in the montage process. Like the practical editing problem, the theoretical debate over the appropriate use of sound was eventually resolved by the practice of postsynchronization.
Postsynchronization seems to have first been used by the American director King Vidor for a sequence in which the hero is chased through Arkansas swamplands in the all-black musical Hallelujah (1929). Vidor shot the action on location without sound, using a freely moving camera. Later, in the studio, he added to the film a separately recorded sound track containing both naturalistic and impressionistic effects. In the following year Lewis Milestone’s All Quiet on the Western Front and G.W. Pabst’s Westfront 1918 both used postsynchronization for their battle scenes. Ernst Lubitsch used dubbing in his first American sound films, the dynamic musicals The Love Parade (1929) and Monte Carlo (1930), as did the French director René Clair in Sous les toits de Paris (Under the Roofs of Paris, 1930). In all these early instances, sound was recorded and rerecorded on a single track, although some American directors, including Milestone and the Russian-born Armenian Rouben Mamoulian (Applause, 1929; City Streets, 1931), had experimented with multiple microphone setups and overlapping dialogue as early as 1929. Generally, through 1932, either dialogue or music dominated the sound track unless they had been simultaneously recorded on the set. In 1933, however, technology was introduced that allowed filmmakers to mix separately recorded tracks for background music, sound effects, and synchronized dialogue at the dubbing stage. By the late 1930s, postsynchronization and multiple-channel mixing had become standard industry procedure.