CHAPTER THREE
Periodization of Early Cinema j
Ben Brewster
In the introduction to her book on the early history of the film market in Germany, Corinna Müller remarks that "recent research in film history distinguishes between an early 'cinema of attractions' and a subsequent period of 'narrative film'; without saying so, these terms in fact imply the short film on the one hand and the long film on the other."1 By contrast, American sources that make the distinction between a cinema of attractions and a cinema of narrative integration place the transition between them five years or more before the origin of the feature-length film.2 This seems to be more than a misunderstanding of American usage on Müller's part, or a differ- j ence between the United States and Germany. There is here a discrepancy between two kinds of distinction: an essentially stylistic one between a cinema of attractions and a cinema of narrative integration, and an essentially economic or institutional one between a cinema of short films and one of features (for Müller—and I am convinced she is correct—the emergence of the long film is indissolubly linked with changes in the film market). An exploration of this discrepancy is informative for the periodization of the cinema before about 1917, not just in Germany but throughout the film-producing world, including the United States of America.
A closer look at Müller's study reveals, in fact, that she does not simply distinguish between two phases in the early German cinema, one of short films, the other of features. Rather she makes three distinctions: a period when films were screened asjpäff of a mixed bill in a variety theater or in a fairground booth, a period^df permanent cinemas showing programs of short films, and a period of permanent cinemas showing programs built around a feature-length film.3
In the earliest period the main exhibition outlets for German film producers were variety theaters and fairgrounds. Only around 1906-7 did this
66
PERIODIZATION OF EARLY CINEMA 67
change, with the rapidrise of small shop-front moving picture theaters. These permanent sites showed film more or less exclusively, but the variety format persisted in their mixed programs of short films. The first ones were very successful, and, as there were virtually no barriers to entry, their numbers multiplied, giving rise to a highly competitive exhibition market in which theaters, often located more or less next door to one another, competed for customers by showing the latest films at the lowest prices. This resulted in the characteristic problem called by the contemporary trade press "excess competition." Exhibitors could not differentiate between their own offerings and their competitors' other than by price and novelty, and the premium commanded by novelty was easily eroded thanks to the universal availability of films as soon as they went onto the market. Ticket prices thus had to be as low as possible, exhibitors could not pay distributors or manufacturers anything but rock-bottom prices for the films they showed, and manufacturers had no incentive to raise the cost of production of their films. Meanwhile French and American producers, with more lucrative domestic markets (as we will see below), could export films to Germany at prices that undercut anything local producers could afford to offer. German film production was almost driven out of existence.
This situation changed with the emergence of what was called the "Monopol-Film,"the monopoly film or exclusive. Hitherto films had been sold or rented on the open market; that is, anyone could buy any film offered and screen it anytime and anywhere. Given the impossibility of differentiating among films, a more expensive film had no takers in this market. Another system of booking entertainment acts had long existed: the exclusive contract, where the manager of the act—for example, the production of a stage play—agreed with the theater that the act would not appear elsewhere in a given territory for a certain period of time, giving that theater exclusive local rights to that act and thus allowing a monopoly price to be charged. Exclusive contracts became significant in the German film market around the end of 191 o. The exclusive contract allowed the exhibitor to charge whatever price he or she thought the prospective audience would pay without fear of being undercut for that title by nearby theaters; it allowed the exhibitor to play the film for as long as it would draw an audience up to the limit of the clearance allowed in the contract; and, within the same limits, it allowed him or her to delay the opening as long as necessary to conduct an effective publicity campaign for the film. The first monopoly films usually shared two qualities: they were longer than other films, and they were imported. The first characteristic derives from the need to be able to sell the particular exclusive subject as special and a draw for an audience in itself; it thus had to constitute a substantial part of the film program offered. The second follows from the first; for the reasons already adduced, no one in Germany was making "special" films of any type, let alone long films, and
68        BEN BREWSTER
PERIODIZATION OF EARLY CINEMA
the films therefore had to come from abroad (some reasons as to why they could do so will be discussed later in this study). Apart from one early boxing film imported from the United States, the first German monopoly films were Danish and featured Asta Nielsen. Italian subjects soon followed. Once the principle of the monopoly film was well established, indigenous versions could be made, and, of course, Nielsen moved to Germany in 1911 and began the successful series of short features (usually four-reelers) made for Deutsche Bioscop from that year to the outbreak of the World War I.
The situation in Britain was essentially similar to that in Germany.4 Initially films were screened in music halls and fairgrounds. From 1906 there was a rapid growth in the number of permanent small cinemas. Films were sold on an open market, with the same inhibiting effect on domestic production as in Germany; from one of the most advanced and prolific of filmmaking countries, Britain very rapidly saw its production decline and stagnate formally (an important factor here, as well as in the domestic market situation, was the loss of outlets in America with the foundation of the Motion Picture Patents Company there). In Britain the monopoly film was called an "exclusive." Whether the rise of the exclusive film resulted from borrowing from Germany, or vice versa, or the two developments were independent, I do not know; in neither country was there anything stardingly new about the concept. In Britain, too, the first exclusives were from abroad and were usually longer than one reel; and in Britain, too, the rise of the exclusive gave the domestic production industry a new lease on life but nothing like such a vigorous life as the German (in 1914 German domestic producers were effectively protected by the war, whereas their British equivalents remained as vulnerable as before to American imports).
France was somewhat different. Richard Abel divides the prewar years into four periods.5 First was the founding period up to 1902, when film companies were essentially producers of equipment. Second was the period from 1902 to 1907, dominated by exhibition in variety shows and fairgrounds. The third period, from 1907 to 1911, saw a rapid rise of permanent movie houses, with Pathe owning a substantial share of them and instituting its own distribution system to serve them. The large market share and vertical integration Pathe enjoyed and its strong position on the world market allowed it to sustain relatively high production costs and hence compete in quality with its rivals, without the ruinous competition characteristic of England and Germany. The relative control of the film programs in its own theaters also enabled Pathe to extend the length of the films it made fairly freely (in contrast to American producers, as we will see)XNevertheless, the real impetus to much longer films in France came from thVmiport by firms outside the Pathe circle of films from abroad, particularly Italy^-which were distributed as exclusives, thus establishing the basis for the fourth period, beginning in 1911. Pathe and its rivals Gaumont and Eclair were in a position to match
these productions with long films of their own, and by 1914 it was the well-established companies of the third period who also dominated the production and distribution of features in France in the fourth (although from that year American features became serious competitors).
If England and Germany were similar, so, too, were France and the United States. As is well known, before 1906 the predominant exhibition site for films in the United States was the vaudeville house. There were traveling projectionists—Charles Musser and Carol Nelson have chronicled the career of one of them, Lyman Howe6—but in the United States fairgrounds as such do not seem to have had the importance they had all over Europe. After igo6 the crucial site became the nickelodeon, the small permanent theater offering a relatively short program of short films screened continuously a number of times each day. The high competition among these theaters did not drive down production costs, however, because producers grouped themselves, voluntarily or involuntarily, into cartels: first into the Motion Picture Patents Company (MPPC); then into the MPPC and its distribution arm, the General Film Company, on the one hand, and a small number of rival combines of so-called Independent producers and their distribution firms on the other.7 As in France, the strong international position of American film producers, but even more the ability their cartelization assured them to close the domestic short-film market to almost all imports, helped sustain production-cost levels.8 Enjoying the advantages of monopoly through membership in a cartel, however, is not the same as having a preponderant position in the market by oneself, as Pathé had in France. Cartel members agree to abandon price competition, and each attempts to extend its market share by other means—but this can only happen within limits, or the members unable to compete by those means will break ranks and cut their prices. Central to the way the MPPC and the General Film Company handled this problem is the modular program: each company contracted to produce a certain number of films in each week's release schedule, totaling two or three times the footage any one theater would need. Local exchanges, affiliates or branches of the General Film Company, made up programs from these offerings and assigned those programs to theaters according to zoning and clearance agreements and the relative bargaining power of each theater. Mutual and the Motion Picture Sales Company, the distribution arms of the other American film cartels, acted in the same way. The possibility of modular assemblage of films, with all the participating producers thereby guaranteed a reasonable number of screenings of their output, produced the extreme standardization of filmJength characteristic of the American industry and of no other. Hence it proved impossible (despite several attempts) for the established short-film producers to extend the length of films in the way Pathé was able to. However, there was an alternative market for film, the "states' rights" system. This was the American equivalent of the exclusive contract,
70        BEN BREWSTER
PERIODIZATION OF EARLY CINEMA        "JI
the territory within which the contract guaranteed the exhibitor a period of monopoly being a state or a traditionally associated group of states. In the United States, too, states' rights booking was a preexisting system used for theatrical and variety acts. By the beginning of the 1910s, longer films produced abroad were being imported and released through the states' rights system. By 1912 American producers were making feature-length films, mosdy as underfinanced speculative ventures for states' rights distribution. Not until 1914 andW.W. Hodkinson's organization of Paramount, the distributor of a complete program of features produced by an alliance of film manufacturers, and hence an equivalent of the General Film Company for features, did feature production begin to dominate the American market.9 On the other hand, the very high quality (i.e., the very high cost of production made possible by monopoly pricing) of American short films in the early 191 os gave American producers the power to penetrate European markets, with the result that, well before the United States was producing substantial numbers of feature films, American production had come to predominate in almost every European market.
About other countries I do not know enough to say very much. Italy and Denmark are clearly crucial, given that it was films produced in those countries that brought about the shift to feature-length production everywhere else. I have seen no satisfactory explanation for how Italian producers avoided the fate of their English and German counterparts. Corinna Miiller suggests that the Danish exception may be explained by the very strict regulation of movie theaters, and particularly of the numbers of such theaters, by Danish municipalities. Given their monopoly position and their elite catchments, it was possible for Copenhagen theaters to charge high prices and present films in long runs; this encouraged production of identifiably high-quality films targeting well-off patrons and generated the excess profits to finance such production, and the metropolitan Danish cinemas invested in filmmaking in just this way. Kosmorama, the company that made Afgrunden, the film that launched Asta Nielsen and the genre that became known as "erotic melodrama," was the outgrowth of a Copenhagen movie house.10
Despite the variations from country to country that I have indicated so far, there is a remarkable uniformity in these accounts of the development of the cinema. There seem to be three broad phases. The first (although probably in all countries it was preceded by an equivalent of Abel's first phase for France—essentially a period in which film-related activity was still too new and too incoherent to qualify for the name "film industry") can be called the variety-theater/fairground period; the second the permanent-moyie-house/short-film period; the third the feature-film period. And in all coun-
tries the first period gives place to the second around 1906-7 and the second to the third around 1912-14.
Were there corresponding stylistic phases? And if so, do they correspond to such stylistic descriptions as "cinema of attractions" and "cinema of narrative integration"? A first point to be made in relation to these questions is that the phases are not watertight. Fairground cinema did not end in 1907— itinerant booths showing films were still seen in European fairgrounds in the 1930s and later. Although in the United States moving pictures were abandoned by major vaudeville chains with the rise of the nickelodeon, some nickelodeons rapidly became "small-time vaudeville" houses, showing a combined program of second-class live acts and films, thus perpetuating a mixed show in the cinema, and this practice persisted to the end of the silent era (and much later in big U.S. cities and in some European countries more widely).11 And cartoon and newsreel cinemas showing hour-long programs of short films similar in format to the nickelodeon program survived in London railway termini until at least the 1960s. What is at issue is the economically preponderant form. If there were exhibitors who persisted in a form that was no longer preponderant, they were to some extent served by producers and distributors who still provided suitable wares for their enterprises, although some items made for the preponderant institution would also serve their needs; thus the British railway terminus cinemas mentioned above showed cartoons and newsreels manufactured to provide the lower end of the bill in the dominant feature cinema.
Moreover, some characteristics of the format are carried over from phase to phase. Although it was much more exclusively a film-centered entertainment, the nickelodeon program allowed space for some live acts (most notably the song slide), and in its film program it retained the idea of a series of contrasting acts that dominated variety entertainment as a whole. Even the feature cinemas rarely showed just one film, so although one or, at most, two films clearly occupied the top of the bill and were the basis for the advertising of the evening's entertainment, the rest still constituted a variety program and retained the anonymity characteristic of the open-market nickelodeon cinema—audiences paid for the feature (although some patrons seem to have gone more for the live acts) but had expectations about the quality and variety of the rest of the bill, and exhibitors increasingly paid distributors a percentage of the box-office take for the feature and flat-rate rentals for the rest of the program.
A more important point concerns the vagueness of the characterization of the first phase. Demographically, the "variety theater" market is so broad as to encompass the whole urban population, and if "fairground" is coupled with it, towns and villages accessible even to remote rural populations are included. These venues, however, did not each cater to the whole social and
72        BEN BREWSTER
PERIODIZATION OF EARLY CINEMA 73
geographical spectrum. Metropolitan variety theaters were luxurious and ex- ; pensive and catered to elites, suburban ones to the respectable working class, i and fairs to the urban and rural poor. Individuals, who attended variety shows selected the venue that was appropriate to their pockets and their self-esteem. The managers of the venues booked the acts (and the film programs) that they thought would be appreciated by the clientele they hoped to attract. Given this situation it is not surprising that the standard way of selling films in this period is via the classified catalog. There is no reason to assume that someone who regularly attended a particular variety theater that showed Pathe films saw all or even a random cross section of the films in the latest Pathe catalog. As far as I know, little work has been done on the question of which films showed where in this period, although Abel remarks that Melies' long feeries were made for a metropolitan variety theater,12 and it is often claimed that, until the rise of the specialized film theater, Pathe films were especially directed at the fairground. If a variety program is the only common characteristic of all these different venues, then it would seem the only characteristic of the films appropriate to this market would be that they should be of all sorts, the negation of a stylistic category. One is left with Muller's suggestion that they are also all short, but even this is only relatively true: passion films could be assembled that were half an hour or more in length, and the longest films produced by the regular producers as single entities could be a full reel. Nevertheless, there were far more films of three to five minutes than became typical in the nickelodeon phase, despite the survival there of the split reel.
This seems to leave the notion of a "cinema of attractions" rather empty, but one characteristic of fairground cinema is that it was a showman's cinema, one where a presenter, whether actually present as a barker or lecturer or merely implicit in the name on the booth, offered an audience a sensational view or act. The same could be said of variety-theater cinema in those instances when the film program was booked as an "act," where an independent entrepreneur, often the filmmaker or his agent, presented a specially constructed program as a unit in the variety show. This "showmanship" aspect seems to accord well with the idea of explicit narration and direct address to the audience that Gunning and others have seen as central to the "cinema of attractions." This, however, is only one strand of exhibition in a cinematic period that could really be defined as one in which films have no character other than their variousness. '     Much of the showmanship was lost in the succeeding nickelodeon period J because the program was assembled for the theater by the exchange from \ what it got week by week and even day by day from the producers via the \central distribution agency; the theater manager had little knowledge of what he or she was going to obtain and could thus have little control over its presentation. Films had to become much more self-explanatory and indepen-
dent of context to work, and this is, again, a characteristic of the notion of "narrative integration." In the American situation the extremely standardized modular format with the relatively long unit of a quarter of an hour in a sense changed the problem of making a film from one of finding something to show the audience to one of filling a time slot in a way that would be interesting and novel, thus calling forth the institution and then elaboration of modes of narration that is such a strong characteristic of American filmmaking at the beginning of the 1910s. Although the slots were less predetermined by the program format in the rest of the world, it is still fair to say that duration became the central concern of filmmakers everywhere.
Although the feature-length film could be, and was, presented with showmanship, because the exclusive contract allowed extensive local advertising campaigns and lobby presentation,[its increased length added to the problem of how to fill the allotted time and also brought films into parallel with nonvariety forms of entertainment, in particular, the stage playjThis period saw the development of new narrative devices to accommodate the longer form. Some of these were elaborations of short-film devices (but some short-film devices were abandoned, notably narration by flashbacks),13 whereas others represented a wholesale borrowing from other media, particularly from the live theater. The result was a cinema little removed from the one we have now, although the significance of theatrical models in this account runs counter to the currently prevailing conception of the feature film as built around cinematically specific devices.
Thus, the discrepancy noted at the beginning of this article between a stylistic opposition between attraction and narrative, on the one hand, and an institutional opposition between short and long films, on the other, can be resolved, with some reservations, by suggesting three periods: the variety-fairground period, which accommodates if it is not coterminous with the cinema of attractions; the nickelodeon period, characterized by the development of sophisticated film narrative in a short format; and the feature cinema, which adopted forms and subject matter from the established narrative and dramatic arts and developed the battery of devices we know as classical cinema. The reservations concern the dates of the transitions, which are vague. This is not because it took a certain amount of time to make the transition from one phase to another but because, the institutions for each type of cinema being essentially different from one another, they could and did exist side by side for quite long periods. Eventually, one or other form became so economically preponderant as to marginalize the others, and the marginal-ization affected the kinds of films that were made, distributed, and seen in the marginal cinemas.
The main novelty of this account thus centers on the second period, that of the one-reel film. If it has been given serious consideration hitherto, it has been conceived as a decade-long transition between the early cinema,
74        BEN BREWSTER
PERIODIZATION OF EARLY CINEMA 75
however characterized, and the classical cinema.14 But as historians often .-point out, the designation "transitional period" is an oxymoron, simply draining the years it covers of any particular characteristics. A failure to recognize ■ that short-film cinema and feature cinema in the early 191 os were essentially parallel institutions has obscured the specificity of the former. Thus, if you calculate the average length of films released between 1910 and 1915 by j! adding up the total footage released and dividing by the number of titles, you get a steadily growing length. But this does not mean that producers were adding a few feet here and there to their productions so that films gradually ,x got longer (which is a reasonable description of what happened between ^ 1905 and igio);rather, itis because more andmore much longer films were , released into the market in the parallel institution of feature cinema. In the short-film cinema, meanwhile, although there were ways to accommodate ■ | some films longer than a reel, most films adhered to a fairly constant length ;J (and the longer exceptions were executed as multiples of that length).15 I Periodization is a dubious enterprise; everything is always changing into | something else. Its value is that it indicates what in the mass of data can be J aggregated together, what averages and what comparisons are revealing and 1 what misleading. Understanding film production, distribution, and exhibition everywhere in the 191 os is impossible unless it is realized that one phase of the cinema, that of the one-reel film, coexisted with another, that of the ■ feature, at the same time that it was giving way to it. I
NOTES f
1. Corinna Müller, Frühe deutsche Kinematographie: Formale, wirtschaflliche und kul- j turelleEntwicklungen (Stuttgart: J.B. Metzler, 1994), 4-5. Müller cites Heide Schlüp-mann as introducing the opposition into German film historiography but dating the I transition in Germany to 1911. I
2. In the locus classicus, "The Cinema of Attractions: Early Film, Its Spectator, |[ and the Avant-Garde" (Wide Angle 8, no. 3/4 [1986]: 64), Tom Gunning dated die ? end of the cinema he characterizes as a "cinema of attractions" to lgoö-y. I know 1 of American scholars who feel this date is too late but none who suggest it is too early. | In this study I use feature to mean a film of four reels or more. This is anachronistic | because for most of the period I am covering the term meant simply that the film | was "special" and hence could be the featured item in a program. But from around I 1910 length became such a constant characteristic of feature films that the term grad- 1 ually came to imply nothing else. See Michael Quinn, "Early Feature Distribution | and the Development of the Motion Picture Industry: Famous Players and Paramount, | 1912-1921" (Ph.D. diss., University of Wisconsin-Madison, 1998). |
3. In a later study, "Variationen des Kinoprogramms. Filmform und Filmgeschichte," in Die Modellierung des Kinofilms. Zur Geschichte des Kinoprogramms zwischen m Kurzfilm und Langfilm (1905/6—1918), ed. Corinna Müller and Harro Segeberg, vol. I 2 of Mediengeschichte des Films (Munich: Wilhelm Fink Verlag, 1998), 43-75, Müller |
tries to extend "periodization by length" to include a fourth stage, characterized by the "abendfüllende Film," the film that by itself provides an evening's entertainment (i.e., runs about three hours, including intervals). She dates this transition to after the war, so it is beyond the scope of this study, but I would like to indicate that I am skeptical. Films that constituted a full evening's program existed by 1913; Germinal (165 minutes) was the only item on the bill at the Omnia Pathé in Paris in October 1913. On the other hand, they never became standard. There was never an annual program of three-hour movies, and such films were usually road-showed—a form of exhibition by exclusive contract—rather than regularly released, or they were only regularly released in a cut-down form after their initial road-show run. It is not length as such but the distribution system that determines periodization.
4. See Rachael Low, The History of the British Film, vol. 1, 1896—1906, and vol. 2, 1906-1014 (London: Allen and Unwin, 1948-49).
5. Richard Abel, The CinéGoes to Town, French Cinema, 1896-1914 (Berkeley: University of California Press, 1994).
6. See Charles Musser and Carol Nelson, High-Class Moving Pictures: Lyman H. Howe andtheForgottenEraofTravetingExhibition, 1880-1920 (Princeton, N.J.: Princeton University Press, 1991).
7. We are not dealing here with a struggle between brave "Independents" and a sclerotic "Trust," as American film historians from Ramsaye to Hampton liked to picture it, but with a relatively stable market dominated by a small number of competing cartels. The American film industry was already an oligopoly in 1910.
8. See Kristin Thompson, Exporting Entertainment: America in the World Film Market, 190y-34 (London: British Film Institute, 1985), esp. chap. 1.
g. See Quinn, "Early Feature Distribution."
10. See Müller, Frühe deutsche Kinematographie, 124-25, esp. note 74.
11. See Robert Allen, Vaudeville and Film, 1895—191y. A Study in Media Interaction (New York: Arno Press, 1980); and Richard Abel, The Red Rooster Scare: Making Cinema American, 1900-1910 (Berkeley: University of California Press, 1999).
12. Abel, Ciné Goes to Town, 160.
13. See Ben Brewster, "Traffic in Souls: An Experiment in Feature-Length Narrative Construction," CinemaJournal^i, no. 1 (fall 1991): 48-49.
14. See esp. Kristin Thompson, "The Formulation of the Classical Style, 1909-2 8," pt. 3 of David Bordwell, Janet Staiger, and Kristin Thompson, The Classical Hollywood Cinema: Film Style and Mode of Production to i960 (London: Routledge and Kegan Paul, 1985), 155-240. (Admittedly, Thompson is writing a prehistory of classical cinema rather than an account of preclassical cinema as such, so she is legitimately indifferent to the latter's specificities.) See also Charlie Keil, Early American Cinema in Transition: Story, Style, and Filmmaking, 1907-1913 (Madison: University of Wisconsin Press, 2001).
15. See Brewster, "Traffic in Souls," 39.