GBRAD 808.29 Al4lfr I zLi brary -"v French Cinema  The First Wave, 1915- 1929 Richard Abel PRINCETON UNIVERSITY PRESS .II Copyright C 1984 by Princeton University Press, Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, Guildford, Surrey All Rights Reserved Library of Congress Cataloging in Publication Data will be found on the last printed page of this book ISBN 0-691-05408-8 ISBN 0-691-008132 (pbk.) First Princeton Paperback printing, 1987 Publication of this book has been aided by a grant from the Paul Mellon Fund of Princeton University Press This book has been composed in Linotron Garamond Clothbound editions of Princeton University Press books are printed on acid-free paper, and binding materials are chosen for strength and durability. Printed in the United States of America by Princeton University Press, Princeton, New Jersey "When you wake up in the morning, Pooh, said Piglet at last, "what's the first thing you say to yourself?" "What's for breakfast?" said Pooh. "What do you say, Piglet?" "I say, I wonder what's going to happen exciting today?" said Piglet. Pooh nodded thoughtfully. "It's the same thing," he said. -A. A. Milne, Winnie the Pooh (1926) For Barbara, with love and gratitude  Introduction xi Acknowledgments xvii Note on Terms xxi I. The French Film Industry Introduction 5 The War: Collapse and Reconstruction 7 Production: From Independent Artisan to International Consortium 15 Distribution: The Divided Country 38 Exhibition: "We're in the Money" 49 Silence in the Face of Sound 59 II. The Commercial Narrative Film Introduction 69 Serials 71 Bourgeois Melodramas 85 Realist Films 94 Fairy Tales, Fables, and Fantasies 138 Arabian Nights and Colonial Dreams 151 Historical Reconstructions 160 Modern Studio Spectaculars 205 Comics and Comedies 220 III. The Alternate Cinema Network Introduction 241 The Beginnings of a Film Criticism 241 Film Journals 245 Cin6-Clubs 251 Specialized Cinemas 257 Proliferation and Crisis 260 Preservation 272 IV. The Narrative Avant-Garde Introduction 279 The Alternate Cinema Network, Auteurs, and Genres 282 Stylistics and Subjectivity 286 The Fields of Discourse and Narrative 290 J'Accuse (1919) and Rose-France (1919) 295 Contents vii CONTENTS L'Homme du large (1920) and El Dorado (1921) 305 Fièvre (1921) and La Femme de nulle part (1922) 313 La Roue (1922-1923) 326 La Souriante Madame Beudet (1923) and L'Inondation (1924) 340 L'Auberge rouge (1923) and Coeur fidèle (1923) 351 Le Brasier ardent (1923) and La Fille de l'eau (1925) 367 Paris qui dort (1924) and Entr'acte (1924) 377 L'Inhumaine (1924) 383 Menilmontant (1926) and Rien que les heures (1926) 395 En rade (1927) and L'Invitation au voyage (1927) 408 Feu Mathias Pascal (1925) and Maldone (1928) 415 Napoléon, vu par Abel Gance (1927) 428 L'Image (1926), 6½ X 11 (1927), and La Glace à trois faces (1927) 446 La Chute de la maison Usher (1928) and La Petite Marchande d'allumettes (1928) 463 La Coquille et le clergyman (1928) and Un Chien andalou (1929) 475 La Passion de Jeanne d'Arc (1928) 486 Finis Terrae (1929) and Gardiens de Phare (1929) 500 L'Argent (1929) 513 Afterword 527 Notes 531 Bibliography 601 Filmographies 631 Index 641 viii The discovery of the cinema is as important as the discovery of printing. -Adolphe Brisson (1914) Ca, c'est du cinema!-Louis Delluc (1921) It is from the cinema that our era borrows its color, its picturesqueness, and the moral atmosphere in which it breathes; one lives as a function of the other. . . -Albert Valentin (1927)  To paraphrase Jean Epstein, the book that one writes is never the book that Introduction one initially conceives. A book that presumes to be a history is especially both less and more. Less than one had hoped for in all sorts of ways: in the consis- tency and comprehensiveness of its design; in the thoroughness of its investi- gation of sources; in the articulation of its findings; in the balance of its mediation between past, present, and future. But also more: for the miscon- ceptions corrected (one's own as well as others), the new questions posed, the conceptual models considered, the discoveries made in progress, and the ways found to speak them. The process of writing a history, in fact, can be so engrossing, stimulating, and open-ended that to close it off seems little more than arbitrary and expedient. Any project that aims to reconstruct the past, of course, is fraught with difficulties. If we tend to iron out ambiguities in the light of more recent developments and the questions raised by them, as John Berger has noted,' it is imperative that we recognize the interests and patterns of thinking that guide our work. Here, Shoshana Felman's pedagogical questions may be help- ful: "What is the 'navel' of my own theoretical dream of understanding? What is the specificity of my own incomprehension? What is the riddle which I in effect here pose under the guise of knowledge?"2 Such questions may keep us from repeating the fable of a history already agreed upon or from accepting the illusory idealism of an absolute knowledge (and the mastery it bestows) as well as allow us to counter the conclusion that all things deeply searched merely become confusing. The project for this book had its origin in the summer of 1976. A sabbatical leave from Drake University permitted me to travel to Paris where I could continue a study, begun three years earlier, of French films and critical writ- ings from the 1920s. At the Cinematheque fran aise, Marie Epstein (with the approval of the late Henri Langlois and Mary Meerson) generously agreed to let me examine closely and repeatedly a large number of French films, espe- cially those of her brother, Jean Epstein, on an old, hand-cranked editing machine in one of the dark, cluttered offices on rue Courcelles. Within a month, I discovered that my stay chanced to coincide with an extensive ret- rospective of early French films-"Eighty years of French cinema (part 2)"- mounted, with the then usual lack of publicity, by the Cinematheque fran- §aise.3 From late July to the middle of September, for three to five hours, five afternoons a week (with the legendary delays and substitutions of film prints), I perched uncomfortably on what passed for cushions in the black-walled Pe- tite Salle at Chaillot and madly took notes as film after film unreeled in the darkness and silence. The idea of writing a history of the French cinema of the 1920s already tempted me, but I preferred first to do a number of studies of individual filmmakers and film texts before tackling such an immense proj- ect. Gradually, after more and more editing-machine sessions and screenings, conversations with Stuart Liebman (then a graduate student at New York University), Dugald Williamson (then a graduate student at Griffith Univer- sity in Australia), and Peter Cowie (Tantivy Press) made me realize the unique opportunity at hand. By the time I returned to the United States in December, 1976, I was committed to an historical project that would occupy over five years of research and writing. xi INTRODUCTION During the course of that writing, I came to see that the project actually covered more than a decade and was marked off by two major disruptions- the halt in French film production (from 1914 to 1915) that resulted from the outbreak of World War I and the slowdown that came from the industry's belated acceptance of the "sound film revolution" in 1929. I also realized that, to more clearly differentiate between the commercial and avant-garde cinema during this fifteen-year period, I needed to distribute the work into two major sections. The first would focus on the dominant film industry and the com- mercial feature films that were the staple of its production. The second would focus on the alternate cinema network which developed in parallel to the industry and on the narrative and non-narrative avant-garde films that were determined, in part, by that network. In effect, this project gradually has grown into a major reassessment and, if I may presume, a much-needed "revisionist" history of the French cinema during the period, 1915 to 1929. The absence of prior studies-with the exception of Georges Sadoul's mammoth yet uneven Histoire gindrale du cinma (six volumes) as well as Jean Mitry's slightly less imposing Histoire du cinma (now five volumes)-demanded that I give more space than originally planned to the changing material conditions and policies of the French film industry and to the (previously unexamined) generic nature of its output of commercial feature films. Futhermore, the general lack of sustained close analysis of the narrative avant-garde films (both in French and English) led me to sacrifice not only the short non-narrative avant-garde works (many of which have re- ceived a good deal of attention in the United States) but also the usual auteur- based study of those films. Instead, I decided to position the theory and prac- tice of the narrative avant-garde within the context of the conventions of narrative film discourse then operating in the French as well as the American cinema. This somewhat original working definition then provided the frame- work for a series of individual textual studies of the narrative avant-garde's exploration of the cinema's systems of signification, the means to what they saw as a new aesthetic practice. Some years ago, Jean Mitry defined a proper history of the cinema as, simultaneously, a history of its industry, its technologies, its systems of expression (or, more precisely, signification), and its aesthetic structures, all bound to- gether by the forces of the economic, psychosocial, and cultural order.4 Writ- ing such an inclusive history is perhaps an impossible task. As one French film historian confessed recently, "It is already too late to write the history of the silent cinema. "5 In my case, the received level of knowledge about the early French cinema (in English) as well as the constraints of time, access to sources, intellectual acumen, and personal interest have led me to emphasize certain subjects and lines of inquiry at the expense of others. Still, the purpose of this history is multiple. First of all, for English-speaking viewers, it pro- vides a good deal of "new information" about this neglected period of French film history. That information includes not only data on specific films and filmmakers, on industry policies and practices, on institutional as well as in- dividual relations, on ideological and aesthetic constructs, but also resource references-as a means of stimulating further inquiry. It also singles out par- ticular areas of historical development for special attention, either because of a lack in prior histories and critical studies or because of serious misconcep- xli tions and misrepresentations. That is why so much space is given over to the INTRODUCTION industry and its more commercial products and why even more is given to the narrative avant-garde practice. Finally, it continually raises questions of his- torical accuracy, conceptual formulation, and textual reading and interpreta- tion in an attempt to offer directions for further research on the French cinema of the silent period. May at least some of this work meet the challenge thrown down, in another context, by Walter Benjamin: A writer who does not teach other writers teaches nobody. The crucial point, therefore, is that a writer's production must have the character of a model: it must be able to instruct other writers in their production and, secondly, it must be able to place an improved apparatus at their disposal. This apparatus will be the better, the more consumers it brings in contact with the production process-in short, the more readers or spectators it turns into collaborators.6 The American researcher interested in the French cinema of the 1910s and 1920s faces a number of obstacles. Access to existing films and written sources is severely limited because the best archives are in Europe. The largest repos- itory of French silent films is the product of Henri Langlois's extraordinary, lifelong passion for film collecting, the Cinematheque franaise in Paris. Until recently, however, the Cin6matheque had shown little interest in (nor could it probably afford) providing viewing facilities for historians and critics, except on an irregular, rather personal basis. Specific films had to be "caught" in their infrequent public screenings at Chaillot or Beaubourg (and previously at rue d'Ulm). Now that the Cinematheque's relationships with the French gov- ernment and with the state Archives du Film at Bois d'Arcy have stabilized and become more clear, that attitude has begun to change. Smaller collections of French silent films are housed at Bois d'Arcy, at the Royal Film Archive of Belgium in Brussels, at the National Film Archive in London, and at the Cinematheque de Toulouse. Except for the latter, all provide (at some cost) excellent viewing facilities for visiting researchers. The most important librar- ies for written sources pertaining to the period are also located in Paris: the Bibliotheque de l'Arsenal, the Bibliotheque d'IDHEC, the Bibliotheque na- tionale, and the Bibliotheque de la Cinematheque franqaise (just recently opened for the first time to the public). Another valuable collection is housed at the Cinematheque de Toulouse. For a list of sources consulted for this history, see the appended bibliography, especially the section on film journals. In the United States, the most important collection of French silent films is at the Museum of Modern Art in New York. During the past few years, the museum has been adding excellent prints of 1920s French films to its archives; and, beginning in January, 1983, it included these as well as prints from other archives in an extensive retrospective of the French cinema.7 Smaller collections of French silent films are housed at the George Eastman House in Rochester, at Anthology Film Archives in New York, at the Library of Con- gress in Washington, at the UCLA Film Archives in Los Angeles, and at the State Historical Society of Wisconsin in Madison. Along with the Museum of Modern Art, all provide superb viewing facilities for researchers. Written sources pertaining to the period are less easy to come by. The best collection of ma- terial related to the narrative and non-narrative avant-garde films is housed at xiii INTRODUCTION the Museum of Modern Art, while those related to the commercial film in- dustry can be found at the New York Public Library, the Library of Congress, and the University of Southern California. For researchers seriously interested in the French cinema of the 1910s and 1920s, consequently, it is almost imperative to arrange to study in Europe. If access to sources presents a problem, the number and condition of exist- ing film prints presents one even more severe. No one seems to know exactly how many of the several thousand films (excluding newsreels, educational films, and other short works) produced in France from 1915 to 1929 still exist and in what condition.8 Raymond Borde recently reported that his queries to some thirty film archives around the world had turned up only 222 titles of the approximately 1,400 French films produced just between 1919 and 1929.' Although most of the films of the narrative avant-garde were diverted from the industry's economic cycle of production, exhibition, and destruction- through the action of cine-clubs, specialized cinemas, and private collectors- the more commercial products usually had less chance of survival. A good number of important films seem irretrievably lost-for example, Baroncelli's Ramuntcho (1919), Hervil's Paris (1924), Grmillon's Tour au large (1927), Renoir's Marquitta (1927), Feyder's Thrdse Raquin (1928), Bernard's Taraka- nova (1930). Others, though listed as surviving, actually exist in incomplete or condensed versions-e.g., several of Antoine's films for Pathe-Consortium; several of Poirier's films for Gaumont; Dulac's La Mort du soled (1922); Rous- sel's Les Opprimis (1923), Violettes impriales (1924), and La Terre promise (1925); Luitz-Morat's La Citefoudroyee (1924); Perret's Madame Sans-Gene (1925); Vol- koff's Michel Strogoff (1926); Bernard's Joueur d'checs (1927). And this list does not even consider the popular serials, comic shorts, and other lesser films.10 Despite the special attention paid to them, several of the narrative avant-garde films are also lost or survive in less than complete versions-e.g., Dulac's La Fete espagnole (1920), Delluc's Le Silence (1920), Feyder's L'Image (1926) and Epstein's Sa Tate (1930). Added to this destruction and dismemberment is the loss of two features considered essential to the films' exhibition. The first to be lost were the musical arrangements and special scores (for large orchestras, organs, or small chamber groups) which accompanied the film screenings. These were partic- ularly important to the films of Marcel L'Herbier, to Bernard's Le Miracle des loups (1924), to Marodon's Salammb6 (1925), to Epstein's La Chute de la maison Usher (1928), and to others as well. Only recently has any attention been given to this specialized music, and very few scores have yet been found." Perhaps the most important "reconstruction" to come out of this effort has been Clair's Entr'acte (1924), meticulously timed to synchronize with Erik Satie's music. The second feature to disappear was color. Most French films of the period were tinted in a half-dozen different colors according to a conventional set of codes-blue for night scenes and seascapes, mauve for early evening scenes, light green for daylight exteriors, amber for interiors, red for passionate scenes or those-lit by firelight. 12 Some were even toned (in the dark areas of the frame) as well as tinted (in the light areas), producing a relatively refined two- color process. In scenes of horror, for instance, the light areas were tinted red while the dark areas were toned green; in Gance's Napoleon (1927), the gen- eral's reaction to the burning of the French fleet off Toulon was described in xiv a close shot of his face (tinted orange-red) against a background (toned deep blue). '3 Certain Pathe films-e.g., from La Sultane de l'amour (1919) to Ca- sanova (1927-were even printed in a complicated stencil process that could accommodate three or four different colors within a single frame. 14 Tragically, very few film prints survive with their original color intact, and it is quite expensive to reproduce it. I myself have seen only a half-dozen or so original prints-Gance's J'Accuse (1919), Pouctal's Travail (1919-1920), L'Herbier's El Dorado (1921), Poirier's Jocelyn (1922), Le Somptier's La Dame de Monsoreau (1923), Epstein's Pasteur (1922) and Mauprat (1926), and a brief fragment of Volkoffs Casanova (1927). The loss of these features changes the works dras- tically-the parallel to Greek and Roman statues is both presumptuous and apt. Futhermore, it lessens the impact of those few films that were exhibited in sepia-e.g., Clair's Un Chapeau de paille d'Italie (1928)-and in stark black- and-white-e.g., Feyder's Therise Raquin (1928), Dreyer's La Passion deJeanne d'Arc (1928), and Epstein's La Chute de la maison Usher (1928) and Finis terrae (1929)-as well as those that may have included scored moments of silence. Consequently, writing a history of the French cinema of the 1910s and 1920s is a treacherous operation, somewhat like constructing a work of land- scape art over partially visible terrain. Not only is one sometimes cut off from the "primary evidence"-the individual film texts-and forced to rely on writ- ten documents for description and analysis, but one can rarely view even those films that survive under conditions of their original projection and as fre- quently as one would wish. Nearly all of the films I have selected for extended analysis are those I have been able to study closely on a viewing machine or view several times projected on a screen. The section on the narrative avant- garde is predicated exclusively on the shot-by-shot descriptions that this kind of study allows. As scrupulous as it may seem, such a method inevitably, produces gaps and misrepresentations. I have tried to keep these to a mini- mum; and, whenever discrepancies between archive prints have cropped up, I have taken note of them. January, 1983 In order to reduce the cost of this edition, only a few corrections in spelling, dating, translated words, and captions have been made in the text. Other, more extensive corrections which could be made include the following: 1. page xiv: Many of the French film titles listed as lost or incomplete have been restored now and projected either by the Cin6matheque francaise or by other archives. 2. page 23 ff: The spelling and date of Fescourt's film should be Mandrin (1924). 3. page 287 ff: David Bordwell's dissertation, French Impressionist Cinema, should have been referenced as published by Arno Press (1980). 4. page 307: L'Herbier's El Dorado (1921) has been projected along with its original musical score (recently rediscovered by Theodore Van Houton) in the. Netherlands several times and at the Salle Gaveau in Paris, on 13 November 1986. 5. pages 269 and 480: The rightist attack at Studio 28 occurred during a screening of L'Age D'Or (1930) rather than of Un Chien Andalou (1929). INTRODUCTION xv INTRODUCTION Any serious rewriting of the text-stimulated by the Cin6matheque fran- caise's recent screenings of restored film prints, the publication of Raymond Chirat's Catalogue desfilms franfais de long metrage: films defiction, 1919-1929 (Cin- 6matheque de Toulouse, 1984), and my own further research in various libraries and archives-will have to wait until later. October, 1986 xvi Any work of history is collective in its dependence on the generosity, re- Acknowledgments source material, ideas, and constructive criticism of others. This book is no exception. Of all those who made this work possible, the most important was Marie Epstein, the former director of technical services at the Cinematheque fran- gaise. Without her trust and her devotion to her brother's films and to the work of the narrative avant-garde generally (characteristically, she downplays her own considerable work as a filmmaker), I would not have been able to make the shot-by-shot descriptions of so many rare films that were essential to my analysis. She was like a fairy godmother to me during the months I sat hunched over that ancient "visionneuse" and during our countless conversa- tions and more formal interviews. I will forever be indebted to her. At the Cinematheque fran aise, I must also thank the late Henri Langlois, Mary Meerson, and especially Lucie and Renee Lichtig (who have taken over Marie Epstein's position now that she has finally retired) for their generous assistance in arranging special screenings of particular films. Other archivists and archives were no less helpful-especially Charles Silver of the Film Study Center at the Museum of Modern Art; Jeremy Boulton and Elaine Burrows of the National Film Archive in London; James Card, George Pratt, and Marshall Deutelbaum, all formerly of the George Eastman House in Rochester; Jacques Ledoux of the Royal Film Archive of Belgium in Brus- sels; Maxine Fleckner of the Film Archive at the State Historical Society of Wisconsin in Madison; Patrick Sheehan of the Motion Picture Section of the Library of Congress in Washington, D.C.; and Edouard Chamard of the Centre national du cinema in Paris. For written sources, I am indebted to the staffs of the Bibliotheque de l'Arsenal, the Bibliotheque d'IDHEC, and the Bibliotheque nationale in Paris; the library of the Museum of Modern Art in New York; the library of the University of Iowa in Iowa City; the library of the University of Southern California in Los Angeles; and the Inter-Library Loan Services of Cowles Li- brary at Drake University. For frame stills and production photographs, I would like to thank Sylvie de Luze of the Cin6matheque franqaise, Mary Corliss of the Museum of Modern Art, Donald Crafton of Yale University, the National Film Archive, the State Historical Society of Wisconsin, and David Phipps (who prepared many of the actual stills) of the Educational Media Services Center at Drake University. Whenever possible, I have tried to use frame stills rather than publicity pho- tographs, especially for the narrative avant-garde films, even if that meant sacrificing image quality. Any gaps and omissions are due either to space limitations or to the absence or inaccessibility of sources. I am indebted also to several people who graciously opened their personal collections of films, photographs, and written materials to my use-Jean Dr6- ville, Gerard Troussier, Kevin Brownlow, Stuart Liebman, Armand Panigel, Bernard Eisenschitz, and Mme. Ruta Sadoul (who arranged for me to examine part of the late Georges Sadoul's private library). Particularly helpful were the transcripts of interviews that Armand Panigel conducted in 1973 with some dozen French film celebrities from the silent period.1 Those who read all or part of the manuscript at various stages and made numerous emendations and provocative suggestions include Kevin Brown- xvii ACKNOWLEDGMENTS low, Barry Salt, Kristin Thompson, Rick Altman, Alan Williams, Douglas Gomery, Donald Crafton, Stuart Liebman, David Bordwell, Dudley Andrew, and Janet Altman. Others who contributed information and ideas were Ber- nard Eisenschitz, Dugald Williamson, Sandy Flitterman, Ernest Callenbach, Glenn Myrent, Wendy Dozoretz, Phil E. Brown, William Lafferty, Jon Gar- tenberg, Robert Hammond, Charles Ford, Claude Autant-Lara, and Lotte Eis- ner. Whether accepted in toto or not, their responses were valuable in shaping this book. At Princeton University Press, I am grateful for Joanna Hitchcock's enthu- siastic and unstinting support for the manuscript as well as her helpful sug- gestions and patience during its final revision. I was also delighted to have Marilyn Campbell's highly efficient and consistently well-judged assistance in preparing the manuscript for production. Support for my research and writing came from several quarters at Drake University-from Carol J. Guardo (the former Dean of Liberal Arts) and the English Department, for a partial leave of absence one semester, and from the Graduate Research Council for several summer grants as well as typing serv- ices. In Paris, my efforts to contact various people, conduct interviews, and do some writing was facilitated by the Bureau d'Accueil des Professeurs d'Universites Etrangeres (formerly the Centre Universitaire International). Finally, my deepest appreciation goes to the person who has encouraged me most throughout this period of writing, who has read and reread, corrected and commented upon, every version of the manuscript-my best reader and collaborator, Barbara Hodgdon. Permission has been granted to reprint portions of this book, which origi- nally appeared in different formats, in Quarterly Review of Film Studies 1 (1976) and 2 (1977); Wide Angle 3 (1979); Film Quarterly 25 (1982); and Cinema journal 22 (1983). xviii Personalcollection of Richard Abel: 1, 2, 3, 4, 6, 7, 8, 9, 10, 13, 14, 15, 17, 18, 19, 23, 29, 32, 35, 45, 47, 48, 55, 56, 62, 63, 64, 68, 69, 70, 72, 78, 86a, 89, 90, 96, 97, 98, 99, 100, 102, 103, 104, 105, 107, 109, 111, 113, 117, 118, 119, 120, 121, 127, 128, 129, 130, 131, 132, 134, 135, 136, 138, 140, 147, 150, 151, 155, 156, 157, 158, 159, 160, 162, 173, 175, 177, 184, 194, 196, 198, 199a-k, 200, 201, 202a-c, 209, 210, 211, 214, 216, 217, 222, 225a, 225c, 233, 234, 239a-f, 240, 241, 242, 243, 244, 245a-d, 246, 247, 248, 249, 250a-c, 251, 252a-b, 253a-b, 254, 255a- b, 256a-b, 257a-b, 258a-b, 259a-b, 260, 261a-i, 262a-b, 263a-b, 264a-g, 265a-h, 266, 267a-c, 272, 275 Cinematheque franfaise: 20, 21, 22, 25, 26, 28, 30, 33, 37, 39, 42, 43, 44, 46, 51, 52, 61, 65, 66, 67, 71, 73, 74, 75, 77, 81, 83, 86b, 87, 88, 92, 108, 110, 112, 116, 123, 139, 142, 143, 144, 161, 174, 178, 179, 180, 186, 204, 205, 228, 229, 277, 278, 279, 280, 281 State Historical Society of Wisconsin: 53, 54, 93, 124, 125, 126, 163, 164, 165a-c, 166, 167a-d, 168a-h, 169, 170, 185a-e, 235, 236, 237a-c, 238 National Film Archive: 24, 27, 38, 41, 79, 80, 82, 84, 106, 114, 115, 145, 153, 171, 172, 176, 187, 188, 189, 190, 191, 206, 207, 208, 212, 219, 268, 269, 271 Museum of Modern Art: 11, 12, 16, 31, 36, 40, 76, 91, 94, 95, 122, 141, 148, 181, 182, 183, 192, 203, 220, 253, 270, 273, 274 Personal collection ofJean Dreville: 5, 137, 146, 213, 215, 221, 223, 225, 226, 230, 231, 232, 282 Personal collection of Stuart Liebman: 34, 133, 154, 193, 195, 197, 276 Yale University Film Stills Collection: 49, 50, 57, 85, 149 Images Film Archives: 101, 218, 224, 227a-c Personal collection of Barry Salt: 58, 59, 60, 152 Personal collection of Donald Crafton: 10 Illustration Acknowledgments xix  In the interests of reading ease, I have translated nearly all of the French quotations into English; the translations are my own unless otherwise indi- cated. In the interests of accuracy, however, I have kept the French titles of books and essays as well as the original titles of the French films under dis- cussion, adding the English translations only when I thought it required or when reference is made to the American release print of the film. Otherwise, I have used the original titles of American films and the English titles of German, Italian, Swedish, and Soviet films, along with their French release titles when it seemed appropriate. The date given for each French film is the date of its public release in Paris. At the end of the war, serials and feature films were presented to the cinema owners and critics about five weeks prior to their planned release.I By the middle twenties, the period between private presentation or preview and pub- lic release had been extended to three to six months. And actual production generally began anywhere from six months to a year before the presentation. Whenever there was an exception to this pattern, usually a delay of a year or more, the dates of both production and exhibition are noted. Occasionally, I call attention to the length of a film-in reels, meters, minutes, or hours. The standard reel of a French film was approximately 300- 400 meters in length and ran close to fifteen minutes (that is, at 18 frames per second or fps).2 However, the cranking speeds of silent films increased gradually during the 1920s so that the screening time of a reel of film became shorter and shorter. According to Barry Salt, American films were being shot and projected at the speed of 22 fps by as early as 1924, whereas French films did not reach that speed until 1929.3 Most French films made at the end of the decade, consequently, were projected at close to sound film speed and should be projected at that speed today. Finally, in the sections of the book devoted to close textual analysis, my descriptions of specific shots resort to notational acronyms designating camera distance and angle that have become familiar and necessary in film criticism. They include the following: ECU extreme close-up (the shot of an eye, a pair of eyes, or part of a body or small object) CU close-up (the shot of a face or object) MCU middle close-up (the shot of a person from the chest up) MS middle shot (the shot of a person from the waist up) FS full shot (the shot of a person from the knees up [American shot or from the feet up, or the shot of a large object) LS long shot (the shot of a full interior space or of a large exterior space) ELS extreme long shot (the shot of a mammoth interior space or of an extensive landscape) HA high angle (a shot taken from above chest level, looking down) LA low angle (a shot taken from below chest level, looking up) When a film's mise-en-scene combines two or more planes of interest within the frame, I resort to a double acronym-e.g., MS/LS. Only one other acro- nym appears frequently-POV-for point-of-view shot. Note on Terms xxi  French Cinema  I METhe French Film Industry I mm / The cinema: two drawers-one marked expenses, the other receipts-and between them this mysterious little box in which films are filed as index cards. . . -Louis Aubert (1923) The facts do not speak for themselves. . . . [The} historian speaks for them, speaks on their behalf, and fashions the fragments of the past into a whole whose integrity is- in its representation--a purely discursive one. -Hayden White (1976) At the outset a question: How was the French film industry structured Introduction during the period from 1915 to 1929, and how did it function? Like the cinemas of other advanced industrial societies, by then the French cinema had become a mass art constituted from a new technological apparatus of percep- tion and representation and thus had acquired some degree of cultural impor- tance. That importance derived, in part, from its emergence within a number of already existing social structures, each of which helped shape its organiza- tion and operation as an industry. But the shaping of the French film industry was unique. A good way of grasping that shaping process, Gerard Talon sug- gests, "is to put in play once more the series of economic and ideological mutations that made the French film industry produce the films that it pro- duced."' To begin reconstructing this period of the French cinema's history, therefore, let me sketch a framework of those mutations.2 The years between the beginning of the Great War and the end of the twenties can be divided politically into four more or less distinct periods. The war years were dominated nominally by the Radicals, who had controlled the French legislature and its cabinet ministries since 1899. The real government leaders, however, were the Moderate Raymond Poincar6 (in. the hitherto cer- emonial post of president), who had become the symbol of the "nationalist revival" just prior to the war, and the aging Tiger, Prime Minister Georges Clemenceau, who for the last two years of the war turned France into a virtual civilian dictatorship. In 1919-1920, partly as a result of internal divisions that wracked the Radicals and the Socialists, a coalition of rightist parties and Moderates, for the first time since the 1870s, achieved a brief, clear-cut ma- jority in the National Assembly. This Bloc national or horizon bleu coalition soon gravitated into a Moderate-Radical alliance, led most prominently by Poincar6, from 1922 to 1924. It adopted a rather harsh foreign policy, espe- cially toward German war reparations, and held strictly conservative domestic positions. The elections of 1924, however, gave power to a Cartel des gauches organized by left-leaning Radicals and Socialists. This government reversed its predecessor's policies by adopting Aristide Briand's strategy of rapprochement with Germany and by half-heartedly advancing Prime Minister Edouard Her- riot's program of economic reforms. In 1926, in the face of escalating inflation and a "capital strike," partly occasioned by overextended government loans for reconstruction, a Radical-Moderate coalition, the Union nationale, assumed po- litical control of the cabinet ministries. Leading this "restoration of confi- dence" once again was Prime Minister Poincar6, who quickly reestablished more conservative positions on internal matters but who also allowed Briand to continue his foreign policy of rapprochement. This return to an "era of stability" was formalized in the elections of 1928. Despite these political changes, which, as we will see, did have a considerable effect on the French cinema throughout the period, the government's attitude toward the film industry was consistently one of benign neglect. As a new industry, like the more important synthetic textiles industry, for example, it was expected to make its own way. That attitude was particularly telling in the areas of film pro- duction financing and the regulation, or lack of regulation, of film imports. France's economic development during this period coincided, in part, with these political changes. Inflation and unemployment, for instance, were high- est in the years immediately following the war, from 1919 to 1921, and in ) FRENCH FILM INDUSTRY the middle of the decade, from 1924 to 1926. At no time, however, did either reach the disastrous levels experienced in Germany. Furthermore, the revolutionary dreams of the working-class movement were shattered by the failure of the 1920 general strike and by the bitter ideological debates that split the Confederation Gnrale du Travail and the Socialists and gave rise to a fledgling Communist Party (and its trade-union arm, the Confed6ration Generale du Travail Unitaire). But the seeming stability of the period could not mask a persistent sense of stagnancy and a fear that the country was on the verge of bankruptcy. In 1919, for instance, industrial production was no more than 60 percent of what it had been in 1914, and much of the country's extensive foreign investments had been lost, particularly in Russia. The fol- lowing decade was marked, especially compared to Germany, by the slow development of key industries-for example, that of hydroelectric power and the concomitant commercial and domestic uses of electricity. The cautious reinvestment strategies and the resistance to new production and marketing strategies that too often characterized this development also showed up in the French film industry. Although the French tended to be chauvinistic in re- sisting the intervention of foreign enterprises-despite the lack of economic regeneration from within-the French film industry actually encouraged for- eign capital and business organizations, almost out of necessity. In the overall economy, however, the industry was small, capital-deficient, and played a distinctly minor role (in contrast to its American counterpart, which was, for some time, among the top forty industries in the United States).3 Also, in comparison to the Americans, the French film industry's structure exhibited much less concentration and vertical integration, and the industry invested much less of its capital in film production and more in distribution and ex- hibition. Finally, the Great War severely strained the ideological bases of French society. Instead of celebrating its victory after the Armistice, France evi- denced, paradoxically, a national mood of suffering and defeat. "The war played the role of catalyst," writes .Jacques Petat. "It burst the nationalist gut strings. The uniforms and flamboyant rhetoric could no longer hide the shit of the trenches. We were disappointed, disgusted. We had been swindled. "' Its eco- nomic effects, adds Jose Baldizzone, completely undermined "bourgeois mo- rality. "5 Stripped of the ideological trappings that decked out the prewar Belle Epoque revival, the French nation seemed in search of its own collective iden- tity. As a popular spectacle, the French cinema gave representation to this search or reexamination in an operation that was doubly motivated by the industry's own sense of defeat or loss. Before the war, according to then ac- cepted, though unreliable statistics, "90 percent of the films exhibited throughout the world were French films." By 1919, only 10 to 15 percent of the films projected in Paris alone were French.6 "If the French cinema is stripped of its glories, it will perish," wrote Mon-Cin6, "and we will have to resign ourselves to being a country that no longer makes good films."' Thus, the industry had its own related question of identity to confront: What was or should be French about the French cinema? Can there be any wonder, then, that, from the beginnings of the Great War to the end of the 1920s, the French cinema found itself in a state of 6 erisis. Year after y'ear, the epithet echoed through the French press devoted ta TF WVAR the cinema. Te risis preceded the war, wvhich precipitated it täthet than being its prineipal cau1se. It wtas inevitahle. . . .Henri Diamant-Berget, 191M Cag ni zant ofi the pir i Lo1 state at our cinema, in hoth men and material, to day we must demol ish these badily lit, enelased htuldings, scrap this obsalete eqoipment, $i«s banish from the screen all that which admirs at the theater or pramenade, expel the, "1 cads and the unfit. -Pierre Henry, 1919' , The first question posed . .. : what remnedies da youi recommend fotr the real enisiso the cinema? René Jeanne, 1921 1, The present si tUa tian is c lear; we are faceed wxith tu in....Jcan Sapéne, 1925'' It has became banal and even a hit ridiculoos to speak ut the (risis af the cinem. Hubert Reval, 1927>' Despire rather steadily grawing bax aflice reeeipts and intermittent reportso a renaissance af the French cinema, the erisis was perceived as ehranic 1n persistent, infecting all levels af the film industry fr om productian tt) exhib1 t1n. I.Iharles Parhe What exactly hadelhappened 1' And was the fl lm industry really as mnaribund, (lisarganized, and secand- at ex-en third-rate as was imnagined: Or cauld the erisis be read as a sign af healthy ferment' Ta discaver mare prcs answers, to better undferstand the French cinema af the period, orne can begin by look- ing mare clasely at the dooble-edgedl effeer of the war. On the eve af Warld War 1, the French bilm i ndustry seemed almast as consal idateel and powerful as was its coLunterpart in the United States. The earl½- artisan-based production-distribution companies of the Lum ifre brothers and af Georges Méliés bard been superseded by a different kind oif campany, modeled on the torporations that were beginning to monopal ize other more important industries. At first, these new co)mpanies employed a strategy 1)1 horizontal integration tt) ensure their success. They soght to control the ma- terial conditions at the yaung i ndustry by manofaeturing camera and projector ecquipment, producing and developing film stack, and canstruering stu1ios.- Soon, however. as tbc cammaelity nature of the cinema, with its dependence on rapidl change in cansumer interest - was recognized, this strategy gave way ra ane at verrical integration. Control was now sought over Bach stage af film praeduetian, distribution, and exhibition. The company that led this transtor- mation and thus come to dom inare the indLustry, both in France and abroadi, was Pathé-Freres. Uinder the directian af Charles Pathé, Pathé Freres had grownr out of sell1ing and exhibiting films (lidisan Kinetoscope films, in fat) at traxveling fairs in the Frencb prax-inces. Back-ed by loans fram the Nevret-G rivo las group of the Créd it Lvannais. Pathe stoon expanded the company 5 operations to se-Il cin- ernatograpbic equiptnent (in alliance with Continsouza, edevelopers oo the Maltese: taoss dex-ice to advance the film in the camera), to pradoce its ont film stock The War: Collapse and Reconstruction FRENCH FILM INDUSTRY and films (at Vincennes and Montreuil), to open foreign exchanges for distri- bution (in Germany and Russia, it dominated the market until the war), and to exhibit its product through its own network of cinemas (which were organ- ized into six regional circuits whose principal manager was Edmond Benoit- Levy). In 1907, as part of its vertical integration strategy, Pathe made the crucial decision to begin renting film prints rather than selling them outright, as had been the practice-a decision that quickly forced the industry to adopt international standards of distribution and exhibition. Within a year, through its exchange in New York and its studio in Jersey City (which became the center of a French emigre film colony), the company was allegedly distributing more films in the United States than any of the American film companies. Even Louis B. Mayer, when he was managing a nickelodeon in Boston around 1912, confessed that he could double his receipts simply by showing Path6 films. By 1914, in France alone, Path6-Freres employed 5,000 people. The company trademark (a crowing red cock) was seen by more audiences through- out the world than any other.' Etablissements Gaumont, founded by Leon Gaumont, quickly followed Path's example by shifting its business from photography to all the different stages of cinematography. From 1905 to 1907, the company transformed itself from a family firm into a limited liability corporation, dictated in part by Pierre Azaria and the Banque Suisse et Fran aise (later the Credit Commercial de France). Previously, Gaumont's films had been written and directed almost exclusively by Alice Guy (perhaps the first woman filmmaker); now film pro- duction was turned over to Louis Feuillade-in the world's largest studio at Buttes-Chaumont (La Cit6-Elg6, after Gaumont's initials). Besides producing films and cinematographic equipment as well as experimenting with color processing and sound synchronization, the company expanded into distribu- tion and exhibition. It, too, put together a chain of cinemas that included the grand Gaumont-Palace (the former Hippodrome Theater, remodeled in 1911 to seat close to 6,000 people). By 1914, under the sign of the marguerite (the tiny white French daisy), Gaumont had 52 agencies and 2,100 employees around the world.2 Despite their success, Pathe and Gaumont were model French companies in that they refused to monopolize the film industry by destroying their smaller competitors.3 Consequently, although other film companies had to limit them- selves to one or two areas of operation within the industry, they could compete almost equally with Path6 and Gaumont within those limitations. Eclipse (headed by Louis Mercanton), a former British company, and Eclair (headed by Marcel Vandal) concentrated chiefly on film production, including news- reels. Eclair had even set up a second studio in Fort Lee, New Jersey, as well as a distribution exchange in Germany.4 Another important production com- pany was the prestigious Film d'Art, originally founded by the Lafitte brothers (one of whom was a successful magazine publisher) and then taken over by Charles Delac [Ben-Kaled. Film d'Art had an explicit "cultural mission"-to elicit original scenarios from Comdie Fran aise dramatists for Com6die Fran- qaise actors and directors.' Soon after its debut, the Societe cinematographique des auteurs et gens des lettres (S.C.A.G.L.) was established by Pierre Decour- celle and Eugene Guggenheim, from an idea developed by Edmond Benoit- Lvy.6 As a subsidiary company to Pathe-Freres, S.C.A.G.L. produced adap- 8 tations of literary classics, most of them nineteenth-century novels and short Ti-e WAR stories, directed by Andre Capellani. One of the two major film distribution rivals to Gaumont and Path6-Freres was the Agence g6nerale cinematogra- phique (A.G.C.), which generally handled the films of Eclipse, Eclair, and Film d'Art. The other was Etablissements Aubert, headed by Louis Aubert, whose operation depended chiefly on film imports. In 1913, Aubert was for- tunate enough to distribute the record-breaking Italian film, Quo Vadis?, which incited him to embark on a major program of cinema construction." The French film industry seemed to be in a very lucrative position. It could produce and market films and equipment almost anywhere in the world through its own foreign exchange offices, and it could saturate its own markets in France, thus inhibiting foreign companies from directly distributing their products. While Path6 and Eclair made and, along with Gaumont, distributed films profitably in the American market, most American films sold to France (except those of Vitagraph) initially came through Aubert or Path6, whose American exchange, in 1908, had helped set up the Motion Pictures Patents Company (or simply the Trust). Even before 1914, however, this position had begun to suffer erosion. Pathe left the Trust and soon lost some of its influence in a bitter rivalry with the Eastman Company over the production and marketing of film stock in the United States. Then, as the financial risks increased, French banks began to curtail their support for any film production company.'" The film industries of the United States, Denmark, Sweden, and 2. Le'OH Ga1mont Italy began to flourish, and French film production dropped to 30 to 35 percent of the world's total production.' American film companies such as Vitagraph (especially) and Biograph (briefly) established distribution offices in Paris, and many cinemas (e.g., the Tivoli, Colisde, and Cosmograph) began to exhibit American, Italian, and other foreign films on a regular basis. The crucial fact was that, although more French films were being made, relatively fewer were being shown, even in France. In early June, 1914, for instance, of the 20,000 meters of film being exhibited in Paris, 17,000 meters were for- eign films." The French position collapsed completely in August, 1914, when the declaration of war paralyzed the film industry almost overnight. All branches of the industry immediately closed down. The general mobi- lization emptied the studios of directors, actors, and technicians. Even the French film star, Max Linder, although rejected by the army, left for the front in his own limousine to deliver military dispatches before eventually going off to make films in the United States. The deserted spaces of the studios were requisitioned for military stores and horse barns, and Path6's film-stock factory at Vincennes was transformed into a war plant. The cinemas, along with all other shows, closed their doors in the national interest. But everyone expected that the war would be brief and that life would soon resume as before. Soldiers going off to the front shouted, ''To Berlin!' and 'Home for Christmas!' It was not to be. As the war settled down to a grim, protracted struggle late in the year, the film industry sought to reestablish itself. In November and December, after the "race to the sea" had ended in a stalemate and trench warfare began in earnest, the cinemas (along with some theaters) began to reopen, for the offi- cial purpose of maintaining the morale of the people.'' Early in 1915, in order to supply the cinema owners' demands for film programs, each of the major 9 FRENCH FILM INDUSTRY companies resumed production, but well below prewar levels. For nearly two years, most of their work was geared to the war effort. Pathe and Film d'Art especially felt compelled to propagandize their audiences by producing histor- ical and fictional films on patriotic themes connected with the war. 15 Two of the most publicized were Film d'Art's Alsace (by Henri Pouctal) and Meres franfaises (by Rene Hervil and Louis Mercanton), the latter of which posed Sarah Bernhardt at the foot of Jeanne d'Arc's statue before the ruined cathedral at Rheims. When the Service photographique et cinmatographique de l'armee was created in February, 1915, each of the four major companies assigned a cameraman to it-Path6: Alfred Machin; Gaumont: Edgar Costil; Eclair: Georges Maurice; Eclipse: Emile Pierre-and together they turned out a weekly news- reel on the war, Annales de la Guerre (though they were confined to activity behind the lines until the summer of 1916).16 The French public soon tired of all this attention to the fighting and demanded what they had begun to enjoy just prior to the war. Given the obvious opportunities for cinema exhi- bition-compared to some of the other spectacles (particularly the opera and theater) that remained shut down-the way was prepared for the American invasion. Philippe Soupault has written one of the best descriptions of the French reaction to the new American films and their publicity posters. We walked the cold and deserted streets seeking an accidental, a sudden, meeting with life. To distract ourselves we found it necessary to yoke the imagination to sensational dreams. For a time we found distraction in lurid periodicals-those papers which are more highly-colored than picture postcards. We scoured the world for them, and by means of them we participated in marvelous and bloody dramas which illu- minated for an instant various parts of the earth. Then one day we saw hanging on the walls great posters as long as serpents. At every street-corner a man, his face covered with a red handkerchief, leveled a revolver at the peaceful passersby. We imagined that we heard galloping hoofs, the roar of motors, explosions, and cries of death. We rushed into the cinemas, and realized immediately that everything had changed. On the screen appeared the smile of Pearl White-that almost ferocious smile which announced the revolution, the beginning of a new world.17 It all began during the spring of 1915 when Aubert released the first Mack Sennett Keystone comedies starring Mabel Normand, Fatty Arbuckle, and Charlie Chaplin. By the summer and fall, Chaplin or Charlot (his French nickname) was the rage throughout France. In December, Path6 began dis- tributing the first episodes of Les Mysteres de New York, starring Pearl White, in conjunction with Pierre Decourcelle's daily serialization of the film's story in the Paris newspaper, Le Matin. The posters advertising the film publicized the image of Pearl White to such an extent that they created a fashion out of the heroine's costume-a simple skirt, black vest, narrow-brimmed hat, and white gloves. 18 The following year, in addition to the continuing series of Sennett/Chaplin comedies and Pearl White serials, came the Triangle films produced by Thomas Ince (especially the westerns of William S. Hart) and Famous Players' adaptations directed by Cecil B. De Mille (e.g., The Cheat, starring Sessue Hayakawa). Most of the genres specifically developed by the French prior to the war-- the burlesque (Andre Deed, Prince Rigadin, Max Linder), the serial (from 10 Eclair's Nick Carter series to Gaumont's Fantomas), the adventure film (Jean THE WAR Durand and Joe Hamman's westerns), and the dramatic film or literary adap- tation (Film d'Art and S.C.A.G.L. productions)-had been taken over by the Americans.19 Moreover, technically (in set design, lighting, and editing), the imported films were consistently superior to the French films, and their acting style was strikingly more natural and spontaneous.20 "Forfaiture (The Cheat), Pour sauver sa race {The Aryan}, Molly, then a hundred other films," wrote Jean-Louis Bouquet, "provided the evidence for French technicians. We could no longer explain away such success as isolated incidents. We were surely in the presence of superior methods. "21 Given the limited quantity and variety of French film production, the exhibitors turned increasingly to the highly popular American films. By 1917, American films comprised over 50 percent of the cinema programs in Paris.22 The rapid influx of American films and the resumption of French film pro- duction wrought immediate changes within the French film industry. Al- though established distributors such as Aubert and A.G.C. handled many of the Triangle and Mutual films, they simply could not accommodate the de- mand.23 Several American companies, Jesse Lasky and Adolph Zukor's Famous Players, for instance, opened offices in Paris to join Vitagraph in marketing their films more directly. More importantly, a multitude of European import companies (Monat-Film, Mundus-Film, Harry, Georges Petit, and especially Western Imports or Jacques Haik), sprang up to profit from the tide. At the production end of the industry, the personnel who had been conscripted into the armed forces or who had gone to the United States-e.g., Maurice Tourneur and Andre Capellani transformed the Eclair studio at Fort Lee; Leonce Perret became a director at Paramount-were replaced by a new generation of directors, actors, and technicians.24 From the theater came Andre Antoine (to direct for Path6), Leon Poirier, Jacques Feyder, and Raymond Bernard (all to write and direct for Gaumont), Marcel L'Herbier (to write scenarios for Eclipse), and Abel Gance (to direct for the new production head of Film d'Art, Louis Nalpas). Several stars and young directors formed their own independent pro- duction companies, perhaps in imitation of Camille de Morlhon, whose Va- letta Films had been a prewar subsidiary of Path6. Rene Navarre (the star of Fantomas), Musidora (the star of Les Vampires), and Renee Carl all broke away from Gaumont to head short-lived production units. Several journalists turned from writing to directing. Jacques de Baroncelli set up Lumina Films, in alliance with Film d'Art and A.G.C., and Germaine Albert-Duluc (an inter- viewer and critic for the first French feminist magazines, La Fronde and La Franfaise) founded Film D.-H., in conjunction with her husband and the poet- scenarist, Irene Hillel-Erlanger. The practice of independent or semi-inde- pendent production would become standard during the next decade. The full impact of the American films did not come until the crucial years of 1918-1919. By the end of the long, costly, but victorious war, the industry was faced with a situation aptly summed up by the posters advertising Mun- dus-Film (the distributors in France for Selig, Goldwyn, First National, etc.): a cannon manned by American infantrymen fired film title after film title into the center of a French target.25 According to the Cinimatographie franfaise, for every 5,000 meters of French films presented weekly in France there were 25,000 meters of imported films, mostly American.26 One of the reasons was 11 3. An early Path6 camer obviously economic. Whereas one meter of film exported to the United States or England cost the French the equivalent of .18 to .35 centimes in customs duties, one meter of film imported into France cost only .02 centimes.', It was simply better business to risk a little money on American imports than to risk a lot more on producing French films. Consequently, by 1918, French film production had fallen so low that it made up only 20 percent of the Paris cinema programs. In the words of Henri Diamant-Berger, the publisher of Le Film, France was in danger of becoming "an American cinematographic col- ony."8 How the industry would respond to that threat was determined in large part by Charles Pathe. a Early in 1918, perhaps spurred by the recent foundation of UFA in Ger- many, Pathe proposed "the organization of a cartel of manufacturers, which while allowing competition to continue, would enjoin all the French cinemat- ographic companies to assure the normal existence of some film producers in our country."" In order to do this, he recommended "the institution of a a percentage on cinema receipts as well as a quota system for the importation of foreign film negative into France." Although producers such as Louis Nalpas and Henri Diamant-Berger supported Path , most of his colleagues in the film industry, especially the distributors and exhibitors, opposed the idea, out of short-sighted self-interest and probably also out of fear that Pathe would turn such a cartel to his own advantage. After all, ten years earlier he had opposed a cartel of European production in the United States, only to sign up Path6- Exchange as part of the Trust.'( Moreover, early in the war Path6 had realized that the world banking center had shifted from Paris to New York and that the New York branch office was now the real source of financial power and stability for his company." It was Path6-Freres, after all, which, unable to get much film stock for its own productions, had begun importing Path -Ex- change films (particularly serials such as Les Mysteres de Newr York) and had ca thus opened the way for the flood of American films into France.'' Louis Delluc wondered sarcastically if Charles Pathe ever really went to the cinema." Path6's proposals marked the initial step of a retreat and a reversion to more conservative business methods.'' Soon after their possibly inevitable rejection, he embarked on a reorganization of Path6-Freres to ensure its financial security (the accumulation of capital) and his own semi-retirement, once the war ended. In November, 1918, Path6-Cinema was established according to the tradi- tional French principles of limited production, high profit margins, and mar- ket-sharing arrangements. Because the French money and material available for film production seemed paltry compared to the American resources-and the German as well, since its film industry had expanded greatly due to the war-and because production had become a financial adventure with unusually high risks, Path6-Cinema would concentrate on distribution and exhibition." Also, because the French exhibition market was relatively small (2,400 cine- mas to 18,000 in the United States, 3,730 in Germany, and 3,000 in Eng- land), any French film project the company might finance would have to have "a sure commercial appeal beyond France itself'-which meant that it should cater to the interests and tastes of the American or English-speaking public." For Path6, disingenuously, the only crisis facing the French film industry was a "scenario crisis." Consequently, Path6-Cinema cut back production (by 1920, S.C.A.G.L. had dwindled to financing Andre Antoine's last films), sold off 4. A Debrie camera (cir 1920) 12 R t f ', , 'IJ 4 mile -AL nr . , ?i4- y I. _ V" _", y 1 f i] Ve e. 4E _411 wlk4 T. N"E 5. (left) Germaine Dulac with a Debrie Parvo camera (1925) 6. (right) Gaumont studio at its foreign exchanges, and expanded its network of cinemas in France and Buttes Chaumont (circa 1916) nearby lands. To Georges Sadoul, Path6's new company was a twentieth- century version of Count Ugolin, prospering by devouring its own children. Most of the other major film companies accepted Path6's analysis and con- clusions. "Everything is beginning again, thanks to the Americans," Gaumont told Leon Poirier. "While our factories produced material for the war, theirs made films; and they have conquered the market."(' Recognizing the superi- ority of American film technique ("technique . . . is the key to success") and the French public taste for American films ("th Gaumont-Palace never made more money than when it played a Chaplin comedy or a William Hart west- ern"), Gaumont also turned increasingly to distributing American film im- ports and closed most of the company's foreign exchange offices."' Louis Feuil- lade's highly profitable serials would be continued, but Gaumont's prescription for the rest of his company's product was simply, "American technique and French subtitles, that is what must be done now." For its part, Eclair ceased commercial film production and became the So- ciet6 fran aise du cinema and then the Societ6 industrielle cinematographique. Although it continued to produce the newsreel, Elair-Journal, and to distrib- ute a few films through Union-Eclair, the company sold its moribund foreign exchange in Germany (which became Erich Pommer's Decla-Bioskop, soon part of the UFA system), made its two studios at Epinay-,ur-Seine available on a rental basis, and concentrated on camera equipment and film processing. Eclair's four-lens-turret camera, developed in 1921, together with Debrie's "Parvo" camera (with its automatic dissolve facility) competed on a par with 13 "V .w~. V / I 7. Path6 studio at Vincennes (circa 1916): notice the waist- level two-camera setup, with Path6 cameras 8. (facing page) Path6 studio at Monteuil-sous-bois (circa 1916): notice the waist-level one-camera setup Bell and Howell's cameras on the world market." Eclipse halted production entirely and gambled on distributing the films of several stars-Ren6 Navarre (briefly), Ren6 Crestr (the star of Judex), and Suzanne Grandais (perhaps the most popular French film actress of her day). When Suzanne Grandais was killed in an auto accident in September, 1920, the French cinema lost its third top film star in less than a year--the music hall star of Mercanton's Boudlette (1918) and Pouctal's Dieu du hasard (1919), Gaby Deslys, had died the year before; and Max Linder had returned a second time to the United States." In 1919, Eclipse built a new studio at Boulogne-sur-Seine (short- sightedly following the prewar glass window design that primarily used sun- light for illumination) and rented it out to the other production companies. Thus, a wave of pessimism, coinciding with the defeatist mood of the coun- try, swept through the French film industry at the crucial moment of its reconstruction. The lack of capital for film production, the lag in advanced technological resources (Cini-pour-tous's review of French studio facilities, early in 1920, was dismal)," the loss of exhibition markets and of control over even those in France-all these contributed to the perceived crisis in 1919. And they would remain more or less unchanged throughout the next decade. Per- haps the best way to consider the various ways the industry responded to these economic and political conditions is to examine separately-since they now functioned separately-the different phases of the industry's operation: pro- duction, distribution, and exhibition. For one of the chief characteristics of the French film industry during this period was the fragmentation, decentral- ization, and lack of coordination in policy and practice among its constituent parts. Yet these presumed disadvantages, in part, proved advantageous. 14 5 4 t ! } N . -I In the production sector, the French film industry underwent a paradoxical series of metamorphoses during the 1920s. It found itself reacting to encroach- ments by foreign companies and financing and, in the process, saw at least one faction-the Russian emigre film colony-turn into a crucial source of production. It became dependent, in equal measure, on a cottage industry of small companies and individual producer-directors, on several big studios, and on an increasing number of co-production deals with other European countries. And it created fertile conditions for a range of experimental production strat- egies as well as gambling and profit-taking by opportunistic speculators. When the French fully realized the power and influence of the American cinema in France at the end of the Great War, one of the first things they did was attempt to imitate or associate themselves with it. Much like Louis Re- nault's engineers who before the war had studied F. W. Taylor's management methods in the American automobile industry,' a stream of producers and filmmakers crossed the Atlantic to study the American film industry, partic- ularly the methods and techniques of its factory system of production. They included Charles Path , of course, Henri Diamant-Berger, Germaine Dulac, Abel Gance, Jacques de Baroncelli, Marcel Vandal, and Charles Delac (the last two having now taken joint control of Film d'Art). In 1918, with Charles Path6's approval, Diamant-Berger launched the first effort to link up with the Americans. Familiar with Path6-Exchange's failure to sign a contract with Chaplin after he had left Mutual, Diamant-Berger took as his model instead Louis Mercanton's production of Queen Elizabeth (1912), which Adolph Zukor had made such a success in the United States. Zukor was now president of Paramount Pictures, the new leader (in quantity, at least) Production: From Independent Artisan to International Consortium 15 FRENCH FILM INDUSTRY of American film production. After extensive talks with Zukor, Otto Kahn (Paramount's chief financier), and Adolphe Osso (affiliated with Path-Ex- change), Diamant-Berger proposed an alliance-"an exchange of stars, direc- tors, technicians, and a limited common market of exhibition."2 Zukor agreed, and they drew up a plan: a French-American association based on absolute equality: each will invest one million dollars and hold 49 percent of the shares of the new company; the remaining 2 percent will be put at the disposal of the French partners for the first two years, as long as there are profits; otherwise, the 2 percent will pass to the American partners. . .3 Although personally uninterested in the plan, Pathe arranged for Diamant- Berger to secure financial backing from the bankers supporting Edmond Be- noit-Levy's chain of cinemas. Initially enthusiastic, the bankers cooled to the idea as the postwar inflation began to drive down the value of the French franc. At his end, Zukor agreed to wait, but the French bankers remained adamant-the risk was too great.4 The collapse of this project was a sign of things to come. Several subsequent French initiatives got much further along, but they, too, came to naught. In 1919, after returning from the United States, Vandal and Delac embarked on a series of film projects supposedly made according to American production methods and calculated to be marketable in the United States. They hired the American actress, Fanny Ward (who had starred in The Cheat), and assigned their new "artistic director," Jacques de Baroncelli, to direct her in two bourgeois melodrama adaptations, La Rafale (1920) and Le Secret du Lone Star (1920).5 Unfortunately, it turned out, Fanny Ward was no longer a star in the United States, and not one American distributoragreed to buy the films. Trying another tack, the two producers attempted to rees- tablish a vertically integrated corporation in France. Their plan was to merge Film d'Art with Eclair, A.G.C., and the circuit of cinemas controlled by Benoit-Levy, particularly the newly opened Salle Marivaux. This merger, la Compagnie Gnrale Franqaise de Cinematographie, lasted but a couple of months and produced only one film, Jacques de Baroncelli's Le Rve (1921).6 Another ill-fated venture was the Franco-American Cinematographic Cor- poration, launched in the summer of 1920. Initially, this was announced as a trust, with 300 million francs in capital, that would operate much like the newly formed UFA in Germany and UCI (Union Cinematografica Italiana) in Italy.7 Its supporters, financial and otherwise, gave the corporation an aura of unusual prestige: the banker Henri de Rothschild, the new auto magnate Andre Citroen, the Minister of Education, and the director of the Comedie Fran aise. And its scenario department, including the likes of Tristan Bernard, Albert Carr6, and Andre Antoine, showed a good deal of promise. Interest- ingly, just three months after an elaborate banquet inauguration, the corpo- ration's young secretary-general, Andr6 Himmel, was arrested for embezzle- ment; and the whole project collapsed, without a single film to its credit.8 Finally, Diamant-Berger made a second effort to exploit his American con- nections. On his second trip to the United States, late in 1919, he convinced Adolphe Osso (who had spent seven years with the Path6-Exchange) to return to France and become an independent producer. The two men quickly agreed on a joint film project, Le Secret de Rosette Lambert (1920). Diamant-Berger 16 would provide the scenario (by Tristan Bernard), the director (Raymond Ber- nard), and half the financing; Osso would provide the other half as well as the distribution and the American star. Unfortunately, the star turned out to be an unknown actress, Lois Meredith, otherwise qualified as Osso's mistress, and Osso's money ran out before the film could be distributed. When Diamant- Berger had to take over the film's distribution at the last minute, he resorted to an American-style publicity campaign, including special poster ads in the film magazines and widely distributed postcards. Although Le Secret de Rosette Lambert played at several major Paris cinemas-e.g., the Lutetia, Marivaux, Tivoli, Colisee, Select-it apparently did not regain the filmmaker's invest- ment.9 With the failure of these joint French-American ventures and the cutbacks at Path6-Cinema, Gaumont, Eclair, and Eclipse, the burden of French film production fell on a diverse group of smaller production companies and inde- pendent producers. The most important opposition to what was seen as a full-scale retreat by the French film industry came from a loose band of newly independent pro- ducers and filmmakers. At least two of these men dreamed of becoming "American-style producers" within the French film industry structure. "' They and their colleagues were convinced that, since the American "superproduc- tions" now determined the course of world film production, distribution, and exhibition, the French would have to engage in large-scale productions beyond the budgets of their serials and conventional feature films in order to survive even in their home markets.'' Since the median budget for a French film in 1920 was about 100,000 to 200,000 francs (or 20,000 to 40,000 dollars, a mere 10 to 20 percent of a comparable American film), the French superpro- ductions would demand budgets of a half million to one million francs. Paradoxically, both of these dreamers received encouragement and financial support from Charles Path6. Perhaps the most important of the two was Louis Nalpas, the former pro- duction head of Film d'Art, who, with the financial backing of Path6 and Serge Sandberg, an important cinema manager for Aubert, set up his own production company in January, 1919. Films Louis Nalpas was launched with a superproduction of an exotic Arabian Nights story, La Sultane de /'amour, which was filmed at the villa Liserb in Nice. Directed by Ren6 Le Somptier and Charles Burguet, La Sultane de l'anour had a rousing preview in May, 1919, and was a tremendous commercial success when it opened in November, playing for over two months in several major Paris cinemas. Encouraged, Nalpas quickly initiated a series of high- and low-budgeted projects, including Maurice Mariaud's Tristan et Yseut (1921) ["six songs" of 650 meters each) and Germaine Dulac's La Fete espagnole (1920), from a brief Louis Delluc sce- nario. He and Sandberg also purchased an area west of Nice called Victorine and constructed an immense studio there for Henri Fescourt's serial super- production of Alathias Sandoif (1921). Nalpas's dream was to create a French version of Hollywood around Victorine. Although Gaumont, Path6, and the Societ6 des Cineromans (Rene Navarre and Jean Durand's company) had small studios around Nice, most of the French film production was still done in and around Paris. But a combination of technological short-sightedness and polit- ical resistance (Victorine was only minimally electrified, and the local author- PRODiCTION 9. Louis Nalpas 1? FRENCH FILM INDUSTRY ities reneged on several promised commercial concessions) dashed Nalpas's hopes. In the summer of 1920, while Serge Sandberg assumed control of Victorine (for Rene Navarre's Societ6 des Cineromans), Nalpas took his films off to the United States in order to sell them there in person. 13 The other American-style producer was Henri Diamant-Berger, the former editor of Le Film and a prot6g6 of Charles Pathe. After his first plan for joint French-American film production fell through, Diamant-Berger set himself up as an independent "executive producer" in Films Diamant. With a budget of 165,000 francs, personally underwritten by Path6, he engaged the famous French comic, Max Linder (just back from the United States), to make a film of Tristan Bernard's popular boulevard comedy, Le Petit Cafe, at one of Eclair's studios in Epinay. From Gaumont, he hired Raymond Bernard (the play- wright's son) to direct the production. Against great odds (only four Paris cinemas contracted to exhibit it in December, 1919), the film ended up nearly as successful as La Sultane de l'amour-eventually bringing in grosses of over a million francs. Undermined by Adolphe Osso, Diamant-Berger's second French- American project, Le Secret de Rosette Lambert, immediately depleted most of those profits. 14 But his credit was still good, as we shall see, particularly with Path6. Of the established filmmakers who also formed their own production units at this time (e.g., Luitz-Morat from Gaumont, Louis Mercanton from Eclipse, Henry Roussel from Film d'Art), perhaps the most promising were Abel Gance and Jacques Feyder. Gance, it turns out, was also a proteg6 of Path6. After making several very successful films during the war for Film d'Art-Mater Dolorosa (cost: 48,000 francs/receipts: 181,000), La Dixieme Symphonie (cost: 63,000 francs/receipts: 343,000)-Gance began the first of a trilogy offilms, Ecce Homo. 15 Film d'Art became alarmed at the film's costs and closed the production down, leaving the stranded director with a 50,000-franc debt. Path6 personally settled that debt and then arranged for Path6-Cinema to finance Gance's war epic superproduction, J'Accuse (1919).16 When that film proved extraordinarily profitable (3,510,000 francs by 1923),17 the young filmmaker founded Films Abel Gance (again with Pathe's backing) in the summer of 1919. He, too, acted as executive producer for someone else's film, Robert Boudrioz's L'Atre (1920-1923). But his energy went primarily into writing and directing (with the poet Blaise Cendrars's assistance) a scenario originally entitled La Rose du rail. Late in 1919, Gance constructed an on- location studio in the Gare Saint-Roch railyard outside Nice and began the unexpected year-long shooting schedule on his film, which would finally end in the snowy wastes below Mont Blanc. Three years and 3 million francs later-after the deaths of his leading actor, Severin-Mars, and of his young wife, Ida Danis-the film opened as La Roue (1922-1923) and soon surpassed even J'Accuse in popularity. About the same time in 1919, Jacques Feyder found himself fired at Gau- mont, either because of the company's cutback in production or because of a dispute over his direction of La Faute d'orthographie (1919) as a comedy. Aware that Gaumont had turned down an option to produce a film from Pierre Benoit's best-seller, L'Atlantide, Feyder bought the rights to the novel and was fortunate enough to borrow 600,000 francs from a rich cousin, Alphonse Fred&rix, director of the Banque Thalman. Deciding to shoot the film entirely 18 on location, he and his cast and crew filmed all of the interiors and exteriors in various parts of Algeria from March, 1920, to January, 1921. By the time L'Atlantide, premiered at the Gaumont-Palace on 4 June 1921, its cost had soared to nearly 2 million francs. Everywhere in Paris, the hlm's posters pro- claimed Feyder as "One man twho] dared." The risk proved worth it, for L'A tlantide was a smashing success. 8 Another group of independent producers and filmmakers opted for a strat- egy of production quite different from that of the superproductions. This strategy found its most radical advocate in Pierre Henry, the editor of Cin- pour-tous. " Since the French could not produce enough quality films to com- pete with the large-scale American films, Henry argued, they should instead produce top quality short films of two to three reels each (600 to 900 meters). Given the cinema program format of the time, these short films, even made in quantity, would be guaranteed exhibition as a complement to the serials and the American or French superproductions. Henry pointed to several short films then ready for release as worthy examples-Dulac's La Fete espagnole (for Louis Nalpas), Baroncelli's La Rose and Delluc's Le Silence (for Film d'Art), and E. E. Violet's Le Main (for Aubert). Although none of the smaller film com- panies adopted Henry's idea exclusively, several integrated it into their strat- egy of producing a varied series of low- and medium-budget films. As the only major established company to buck the trend toward fewer films, Film d'Art had the potential in the spring of 1920 to act as the leader of this group. The failure of the Fanny Ward films, together with the collapse of the ambitious Compagnie Generale Franqaise de Cinfmatograhie project, however, seems to have curtailed whatever influence Vandal and Delac may have exerted. Still, their work, along with that of Louis Nalpas, encouraged several others to try their hand at film production. In May, 1920, for instance, Franz Toussaint split away from Films Louis Nalpas to form a company called Jupiter (Societ6 Fran aise des Films Artistiques). Besides distributing Italian film imports and collaborating on several projects with Stoll Films in England, Jupiter set up a schedule of French film production that resulted in about ten completed films in little more than a year. Included in its production were Toussaint's own Le Destin rouge (1921), Roussel's Visages voiles . . . ames closes (1921), and Delluc's short feature, Fievre (1921).20 About the same time, Legrand and Tarieux, a pair of producers who took Vandal and Delac as their model, formed Films Andr6 Legrand. For several years, they financed and sometimes wrote scenarios for a number of medium-budget films-Rene Her- vil's Blanchette (1921) and Le Crime de Lord Arthur Saville (1922), Severin- Mars's Le Coeur magnifique (1921), Dulac's La Mort du soleil (1922), and Fey- der's Crainquebille (1923).2 Almost unnoticed at the time (unlike Diaghilev's Ballets russes a decade earlier) a small band of Russian exiles led by Joseph Ermolieff took over an abandoned studio, once built by Melies and later used by Path6, at Montreuil- sous-bois. Ermolieff had controlled one of the three major Russian film pro- duction companies prior to the 1917 Revolution, and his films had been dis- tributed in Russia by Path6's exchange office. Initially based in Moscow, Er- molieff had moved to Yalta and then fled to France when the Civil War broke out in 1918. Now reorganized as Films Ermolieff, his company began turning out a series of French films for distribution through Pathe-Consortium. The PRODUCTION 10. Abel Gance (circa 1917) 19 FRENCH FILM INDUSTRY 11. Louis Aubert company's prestige grew quickly as its output increased-Protozanoffs L'An- goissante Aventure (1920) and L'Ombre du peche (1922), Mosjoukine's LEnfant d carnaval (1921), Tourjansky's L'Ordonnance (1921) and Les Milles et Une Nuits (1922), Boudrioz's Tenp'tes (1922), and Volkoffs La Maison dui mystere (1922)." While Film d'Art was encountering difficulties in its reorganization, Louis Aubert was expanding his company's operations to include film production. His strategy was to draw a number of independent film producers into his orbit of influence by offering them partial financial backing as well as a guar- anteed outlet for distribution. By June, 1920, Aubert was acting as a consor- tium for at least five independent producers: Ren6 Hervil, Le Somptier's Ci- negraphie d'Art, Delluc's Parisia-Film, Violet's Films Lucifer, and Pierre Ma- rodon's Monte-Carlo Films. Although Hervil soon left the fold to work for Films Andre Legrand, the rest each had at least one film ready for the fall season. These included Delluc's Funee noire (1920), Violet's Li-Hang le cruel (1920), Dulac's Malencontre (1920), Charles Maudru's Le Lys rouge (1920), and Le Somptier's La Montee hers l'Acropole (1920). What Aubert called his 'Artistic Effort" nearly suffered the same fate as Jupiter's. But just as his production money was beginning to dry up, in the summer of 1921, he had the foresight to take over the distribution rights to Feyder's L'At/antide (1921 )." This upsurge of independent film production, especially in the face of the continuing flood of American films, even had an effect, momentarily, on Gau- mont and Path -Cinema. Despite cutting back on film production (especially its series of short comedies), Gaumont was receptive to several strategies. For instance, he put up 40,000 francs for Marcel L'Herbier's first film, Rose-France (1919), a "cantilene" on the war which played briefly at just two Paris cinemas and bombed ignominously.'' More important, he used some of the profits from Louis Feuillade's popular series -udex (1917), La Nouvelle Mission de Judex (1918), Tih-Minh (1919), Barrabas (1920), and Les Deux Gamines (1921 to initiate a series of medium-budget quality films called "Series Pax."5 Edgar Costil was named the series' executive producer, and Leon Poirier was elevated to the position of "artistic director. "2" Besides himself, Poirier was able to engage L'Herbier and H. Desfontaines as directors for the series, but Gaumont refused his request to add Louis Delluc as well. Over the next three years, Series Pax produced some of the most imaginative and critically successful French films of the period: Poirier's Le Penseur (1920) and Jocelyn (1922), L'Herbier's L'Homnle du large (1920) and El Dorado (1921). The effect of the independent producers on Path -Cinema is more complex. In September, 1920, Charles Path6 announced a further reorganization of his enterprises by dividing Path -Cinema into two separate companies. Under the direction of Ferdinand Zecca, Path -Cinema would henceforth concern itself with amateur camera equipment (the 9.5mm Path -Baby), film stock, and film processing. A new company, Path -Consortium, was cut loose to manage the commercial film-stock factory at Vincennes, to rent out the old studios at Vincennes and Nice, and to concern itself only with the distribution and exhibition of films. The reason for this division was (and still is) less than clear. There is some evidence to indicate that, earlier in 1920, S.C.A.G.L. was replaced by a production unit called the Societ6 d'Editions Cinematogra- phiques. This unit seems to have adopted a systematic strategy of producing long films divided into several episodes or 'chapters." That summer, it initi- 20 ated a series of expensive superproductions, including Henri Pouctal's Gigolette PRODUCTION (1921), Bernard-Deschamps's L'Agonie des aigles (1921), and Rene Leprince's L'Empereur des pauvres (1922). Perhaps Pathe wanted to nip this speculative venture in the bud. Or rather, to protect his investors with the assurance of growing assets and steady dividends (he was fond of quoting these), he would limit any high-risk speculation in film production to his own personal re- sources (through S.E.C.). Yet, apparently, it was Pathe who immediately involved Pathe-Consortium in financing an even larger superproduction.27 That superproduction was Les Trois Mousquetaires (1921-1922), to be di- rected by Path6's young friend, Henri Diamant-Berger. Diamant-Berger adapted Dumas's novel as a twelve-hour film divided into hour-long chapters, designed to be released in consecutive weeks over a three-month period. For this gigan- tic serial, Diamant-Berger was allotted an enormous budget of 2 million francs.28 From December, 1920, to September, 1921, the film's shooting ranged widely across France and ended in the studios at Vincennes. There Diamant- Berger installed a battery of mobile ceiling arc lights mounted on rails and transformed it into one of the earliest electrified studios in France.29 Given an unprecedented gala premiere that lasted three evenings at the Trocadero, Les Trois Mousquetaires went on to become one of the most profitable films of the decade-quickly accumulating an astounding 17 million francs. 30 Its success was aided, to be sure, by an arrangement with United Artists that kept Doug- las Fairbanks's The Three Musketeers (1921) from being distributed in France and much of Europe.31 Still, the strategy of balancing the high risk of super- productions with the low risk of serials (consistently the most profitable films in France) looked like a sure bonanza. Before the film was even finished, Pathe- Consortium underwent an important internal transformation that seemed to put it in a position to exploit Diamant-Berger's success. According to Georges Sadoul, Path6-Consortium was controlled initially by three major interests: Path , the Lyon Banque Bauer et Marchal, and the Gounouilhou-Bourrageas families of Marseille and Bordeaux. In a surprise move, shortly after the company's inauguration, the Gounouilhou-Bourrageas faction seized power. Besides increasing the cinema circuits under its control, the new administration, headed by Denis Richaud, decided to invest heavily in super- productions similar to Diamant-Berger's film. In anger over what he took to be foolhardiness and insubordination (although, ironically, he himself in part was its cause), Path6 resigned from the board of the company he had created. Unflinchingly, Richaud and his "artistic director," M. Fourel, drew up plans to continue the expensive superproductions announced the previous summer. Bernard-Deschamps's L'Agonie des aigles (1921) was budgeted at two million francs; Leprince's L'Empereur des pauvres (1922) at one million; and Diamant- Berger's second adaptation of Dumas, Vingt Ans apres (1922-1923), at another two million. Furthermore, they reaffirmed the company's support of Abel Gance's La Roue, which had then reached the editing stage. The sum of this capital investment was staggering. Path6-Consortium's strategy had all the appearances of a strong challenge to the American cinema, but it was depend- ent on relatively unmodernized technical facilities, dubious scenarios, and di- rectors (excepting Gance) whose styles were distinctly old-fashioned. The stage was set for another fall.32 By 1922, French film production had risen to 130 feature-length films per 21 FRENCH FILM INDUSTRY year (up from 80 and 100 films during the two years before). Although this figure was still quite small compared to the production levels of the United States (706 films) and Germany (474 films), it seemed, given the industry's severe limitations, a clear sign of health. Below the surface, however, several of the major production companies were experiencing serious problems and undergoing drastic reorganizations. By 1924, partly because of these changes, the annual production figure had fallen to only 68 films. This drop may have represented a setback for the industry, but it was also a bit deceptive, for the French films were now "larger" (longer and more highly budgeted) and there were more production companies making major films." All three of the production companies that had survived the war suffered relapses in 1922. When Jacques de Baroncelli left his position as "artistic director" at Film d'Art, Vandal and Delac did not replace him. Instead, they themselves assumed the role of independent producers (to make three or four films per year) and opened negotiations to link Film d'Art with Louis Aubert for distribution purposes.34 At La Cit6 Elge, Leon Gaumont quietly folded his tent in capitulation to the American films. Decision-making power seems to have passed to Edgar Costil and the company's financiers, the Compagnie Generale d'Electricit6 and the Credit Commercial de France; Gaumont himself became little more than a board member.35 Production was reduced to Feuil- lade's serials and a few low-budget films as the company turned increasingly to the distribution of American films. With Series Pax halted, Poirier and L'Herbier were let go as directors. In fact, L'Herbier's last film for the series, Don Juan et Faust, was shut down near the end of shooting (after its costs had soared to 800,000 francs) and was subsequently released in an oddly truncated form early in 1923.36 At Pathe-Consortium, the superproduction strategy was already in doubt by the end of 1922. Receipts on the three big films that followed Les Trois Mousquetaires did not measure up to expectations, especially in the case of L'Empereur des pauvres (1922). Although the Gounouilhou-Bourrageas faction retained control, Richaud and his associates were sacked and replaced by Andre Gounouilhou and Henri Mege (of the Banque Bauer et Marchal).37 The new administration quickly adopted a restrictive policy on film production: The industry is going to its ruin if it continues on the path it has followed. We do not want to leave anything to chance, to the unexpected, to accident, delay, negligence, or waste. No film will be undertaken without a precise plan. A French film costs so much because we squander time and money.38 The policy change hit Diamant-Berger particularly hard. Pathe-Consortium had earlier agreed to produce a third and final Dumas adaptation, Le Vicomte de Bragelonne, and Diamant-Berger was busy transforming an abandoned Niepce airplane hanger in Billancourt into a studio, he says, "the like of which did not exist in France."3 Besides baths, a restaurant, and a special foyer for the actors, "a veritable electric keyboard-a centralized command post for all the floodlights-would make it the most modern and best-equipped studio in Europe." When Path6-Consortium abruptly canceled production on Le Vicomte, Diamant-Berger was left with a nearly completed studio and little to do. Unable to get financing anywhere else, late in 1923, he had to put Studio Billancourt up for sale to pay his mounting debts.40 Within six months he 22 would escape the disarray at Path6-Consortium and seek out a new career in the United States. Much like Gaumont, for all intents and purposes, Pathe- Consortium was on the verge of becoming a film distribution company. The only major films it helped finance now were the serials produced by its new affiliate, the Societ6 des Cineromans. In that affiliation lay the company's rejuvenation. Early in 1922, Jean Sapene, publicity editor of the major Paris daily news- paper, Le Matin, had taken over the Societ6 des Cineromans from Ren6 Na- varre and drawn the company into an alliance with Path -Consortium and the Lutetia circuit of cinemas in Paris. Cineromans would produce serials to appear simultaneously in Path6-Consortium/Lutetia cinemas and' in the cartel of Paris newspapers headed by Le Matin. Sapene immediately hired Louis Nalpas (freshly returned from the United States) as his executive producer and put Arthur Bernede (Feuillade's former scriptwriter) in charge of his scenario department. He also began to set up a factory line system of film production and distri- bution. As the Cineromans serials, beginning with Henri Fescourt's Rouleta- bille chez les boheiniens (1922), steadily increased in number and popularity, Sapene's influence within Path6-Consortium rose. By the end of 1923, with the active support of the Banque Bauer et Marchal (through Henri Mege) and Path6-Cinema, Sapene was in a position to oust the Gounouilhou-Bourrageas faction, bring back Charles Path6, and take control of Path6-Consortium. Be- hind this takeover, suggests Georges Sadoul, was a power struggle between the cartel of Rightist Paris newspapers (represented by Sapene) and the Right- ist provincial newspapers financed by Gounouilhou and Bourrageas. One of Sapene's first measures was to expand the company's production base by buy- ing and renovating (electrifying) the Levinsky studio at Joinville-le-pont (built by Gaumont's old set designer), turning its five stages into the most modern in France. By the end of 1924, Sapene had reorganized Path6-Consortium into a vertically integrated corporation again, thus making it the most powerful company in the industry.' The Cineromans serials-such as Jean Kemm's L'Enfant-Roi (1922) and Vi- docq (1923), Gaston Ravel's Tao (1923), Fescourt's Mandarin (1923), Le- prince's L'Enfant des halles (1924), and Luitz-Morat's Surcouf (1925 provided a solid production base for Path6-Consortium. Secured by their profits, in 1924, Sapene (much like Gaumont after the war) launched a new series under the aegis of "Films de France." These films would be shown in a single seance, or performance, and would have preferred treatment in the Path6-Consortium and Lutetia circuit of cinemas. For the 1924-1925 season, Louis Nalpas an- nounced no less than ten Films de France, including Luitz-Morat's La CitI foudroyee, Fescourt's Les Grands, and Dulac's Le Diable dans la Vylle.2 Yet Cineromans was only the most important of several French film pro- duction companies that, from 1922 to 1924, attempted to compensate for the production cutbacks at Path -Consortium, Gaumont, and Film d'Art. If "pro- duction in abundance is the sign of industrial power," wrote Rene Clair, then Louis Aubert was certainly next in line.'i In an interview with Andre Lang in 1922, Aubert had offered this famous definition of the cinema: "Two draw- ers-one for receipts, the other for expenses, and this little mysterious box in which films are filed on index cards...." As one of those drawers turned into a cornucopia of profits, from an expanded circuit of cinemas and from the PRODUCTION 12. Jean Sap'ne 23 FRENCH FILM INDUSTRY 13. Clncromlans studio at C1 0o Jominille-le-pont (late 1 20s) sensational distribution of Feyder's L'Atlantide, Aubert decided, in his words, to boldly play "the French film card."'5 He would do his part to revive the French film industry by investing his fortune in an expanded film production. After hiring C. F. Tavano as his executive producer, Aubert gathered Ren6 Hervil, Louis Mercanton, Ren6 Le Somptier, E. E. Violet, Donatien, Charles Maudru, as well as Vandal and Delac of Film d'Art into a loose consortium to co-produce a series of films.' This new "French Effort" opened with Mer- canton's Phroso (1922), followed by a half dozen other films, including Hervil and Mercanton's Sarati-le-Terrible (1923) and Le Somptier's La Bete traquiee (1923) and La Dame de Monsoreau (1923). By the summer of 1924, Aubert had more than a dozen films in various stages of production, and Cinea-Cini- pour-tous devoted a full issue (1 September 1924) to the company's upcoming season. After Cineromans and Aubert, the most important French production com- panies were two that developed out of the Russian emigre colony in Paris. In 1922, Joseph Ermolieff left Paris for Berlin and sold Films Ermolieff to his associates, Alexandre Kamenka and Noe Bloch." Kamenka reorganized the company into Films Albatros and quickly initiated a half-dozen film projects. Under his direction, Albatros's production soon shifted to more expensive, more prestigious films-e.g., Tourjansky's Le Chant de l'amnour triomphante (1923), Volkoff's Kean (1924) and Les Ojnbres qui passent (1924), Nadejdine's Le ChiJ- fonnier de Paris (1924), and Epstein's Le Lion des Mogols (1924). Recently, the scriptwriter, Charles Spaak, described Kamenka, for the consistency of his work, as perhaps the greatest French producer of the 1920s." The second company was financed by a young Russian emigre in the steel industry by the 24 name of Grinieff. It was founded in 1923 by the popular novelist, Henry Dupuy-Mazuel, and by Jean-Jos6 Frapp6, who christened it the Societ6 des Films Historiques (S.F.H.). " The name was apt, for their grandiose scheme, according to Andre-Paul Antoine, was "to render visually the whole history of France."' The company's first project dwarfed, in scope and budget, any- thing previously attempted by the French film industry. It was an immense reconstruction of the period of Louis XI (late fifteenth century) and of the famous legend of Jean Hachette. Directed by Raymond Bernard, Le Miracle des loups became the most popular film of 1924 and the first film ever to be premiered at the Paris Opera. Another sign of the French film industry's regeneration during this period was the continued proliferation of independent production units, most of them small and geared to a particular director. Yet even here there were losses. Germaine Dulac had managed to sustain Films D.-H. for five years through a varied selection of film projects and financial arrangements with other produc- ers and distributors (Harry, Eclipse, Film d'Art, Louis Nalpas, Aubert, Andre Legrand).i In 1921, she dissolved her company and, after a trip to the United States, tried her hand at free-lance work. When that failed (projects in Ger- many and Italy), Vandal and Delac arranged for Film d'Art to finance her personal production of La Souriante Madame Beudet (1923). Then, through the intercession of Louis Nalpas and Henri Fescourt, Dulac signed on with Cin6- romans to direct a serial and one of the first Films de France. 5' The young critic-cineaste, Louis Delluc, worked independently for three years, from 1920( to 1922, through two separate production companies, Parisia Films and then Alhambra Films-both of which he supported with an inheritance; box office receipts; money from Elena Sagrary, who was allowed to star in Fievre (192 1); and financial or administrative assistance from his friends, Henri Diamant- Berger, Louis Nalpas, Jacques de Baroncelli, and Leon Poirier." The budgets for his films were ridiculously low, but the receipts were even lower, so that La Femme de nulle part (1922) brought him close to bankruptcy. He had to sell Cinea, the important film journal which he had founded, and abandon all film projects for over a year. Delluc literally worked himself to exhaustion and died prematurely of pneumonia in April, 1924. Nearly a dozen other film producers and directors, however, set up produc- tion companies in Dulac's and Delluc's place. The more successful of these included several of Delluc's mentors. Jacques de Baroncelli, for instance, left Film d'Art to work independently when he was assured production capital from a Belgian financier, Arthur Mathonet. His consistent success with literary adaptations La Ligende de Soeur Beatrix (1923), the Prix Goncourt NIne (1924), and Pecheur dIslande (1924 made Films de Baroncelli an unexpectedly re- spectable enterprise.5 After leaving Gaumont, Leon Poirier set up Films L6on Poirier, apparently modeled on Baroncelli's company, to produce and direct another Prix Goncourt adaptation, La Briere (1925). Even the popular dram- atist, Tristan Bernard, established his own film production company so that his son Raymond Bernard, from 1921 to 1923, could direct a series of psy- chological and sentimental comedy scenarios.51' While these filmmakers were successful as their own producers, others ex- perienced difficulties or unusual risks in deciding to work independently. After L 'Atlantide, Jacques Feyder deliberately turned away from superproductions PRODUCTION I Alexandre Kamenka 25 FRENCH FILM INDUSTRY 15. Marcel L'Herbier and approached Andre Legrand to produce a small, realist film adaptation of Crainquebille (1923) at a fraction of the cost (only 300,000 francs) of his desert epic.7 In September, 1923, together with the Max Linder, who had returned a second time from the United States, Feyder announced the formation of Grands Films Independents, a production company financed by two Lausanne businessmen, Dimitri de Zoubaloff and Franqois Porchet." Linder's stay with the company was unusually brief, and the venture quickly collapsed when Feyder and his financiers quarreled over Visages d'enfants (1925), whose editing and distribution were held up for a year>' For Julien Duvivier, the means to film production was Celor Films, a small company he founded in 1922. On shoestring budgets from diverse sources, Duvivier turned out a variety of films over a three-year period, including the earliest compilation documentary on cinema history, La Machine a refaire la vie (1924).6( Although Ren6 Clair came to filmmaking through the intercession of others, he, too, got by on paltry means.'" Early in 1923, Baroncelli read a script by Clair, who was then his assistant, and sent it on to Henri Diamant-Berger. Still flushed with success from his two superproduction serials, Diamant-Berger arranged a miniscule budget to finance Clair's first film, Paris qui dort (1924). In order to continue his independent status through another three films, Clair accepted the patron- age of Ren6 Fernand and then Rolf de Mar6, publisher of Theatre et Conioedia Illustri (for whom Clair wrote as a film critic) and owner-manager of the The- atre de Champs-Flysees. Two other young filmmakers were even more lucky than Duvivier or Clair. On the basis of a surprisingly successful fictionalized documentary film-Pas- teur (1922 and the double recommendation of Louis Nalpas and Abel Gance, Jean Epstein won an extraordinary ten-year contract with Path -Consortium late in 1922.62 The contract called for Epstein to make four films a year for Path , but he would have complete freedom in his choice of scenarios and in his manner of filming, within a limited budget. Although this arrangement lasted for little more than a year, it produced three important small films- L'Auberge rouge (1923), Coeur fidele (1923), and La Belle Nivernaise (1924). The most reckless of all the neophytes, Jean Renoir, was also the most wealthy. In 1923, after seeing Mosjoukine's Le Brasier ardent, he dropped his work in ceramics and decided to pursue a career in the cinema. Drawing on an im- mense fortune from his father's paintings, Renoir set up his own production company and began paying for his apprenticeship as a producer and director with Catherine {Une Vie sans joie} (1927) and La Fille de l'eau (1925)." Finally, there was one independent enterprise whose ambitions loomed larger than any other of the period. This was Cinegraphic, which Marcel L'Herbier founded (with the help of Jean-Pierre Weller, Latigny, and others) when he left Gaumont in 1922. L'Herbier's intention was to create "a kind of workshop [or atelier} of creativity" as an alternative artisanal or cooperative practice to what he saw as an exhausted, timid, unimaginative capitalist production sys- tem in France. One of L'Herbier's strategies was to encourage young filmmak- ers with their films Jaque Catelain's Le Marchand de plaisir (1923) and La Galerie des inonstres (1924) as well as Claude Autant-Lara's experimental short, Faits-Divers (1924). Although unsuccessful in convincing his distributor to back Ren6 Clair's script for Le Fantome du AMloulin Rouge in 1923, he had Cinegraphic produce Louis Delluc's last film, L'Inondation (1924). However, 26 L'Herbier's own projects were at the center of Cingraphic's operation, and they were designed to bring together artists and artisans from various disci- plines to create special "synthetic" films. Cingraphic's offices at 9, rue Boissy d'Anglais (in a building owned by the literary patron, le Vicomte de Noailles) became "a school without precedent" where L'Herbier and Philippe H riat presided over the study and execution of scenarios, decors, costumes, camera setups, and editing. The list of films in preparation included adaptations of Racine's Phedre, Maurice Barres's Le Jardin sur l'Oriente, Oscar Wilde's The Picture of Dorian Gray, Pierre Mac Orlan's La Cavaliere Elsa, and Leo Tolstoy's Resurrection. With the decision to go ahead on Resurrection and then on a Mac Orlan scenario for La Femme de glace, Cinegraphic helped 'initiate the first era of international co-productions.64 The outcome of the French superproduction strategy of 1919-1922 was decidedly mixed. With the exception of L'Atlantide, as we will see, none of the big French films made a dent in the large American market. And at home, they failed to ease the near stranglehold of American films in the cinemas. Consequently, at the same time that Jean Sapene set out to strengthen the industry (and his position in it) by revitalizing the serial format, several other major companies and independent producers embarked on a grander strategy to challenge the world cinema markets. The strategy was simple: to produce films that would have international appeal, one must make the basis of their production international. Writing in Cinmagazine, Paul de la Borie put the economic basis of this strategy quite bluntly: the film industry is ruled by economic conditions which do not allow one to treat it nationally, since no nation-except perhaps America-is in a position-cinemato- graphically-to live on its own resources. And because this is so, because all the European markets (whether it please us or not) are constrained to strict interdepend- ence, the conditions of existence for the film industry can be boiled down to this inflexible formula: it will be international or it will not be at all.65 So although the foreign policies of the Bloc national were still in force and although the Poincare government had taken the extreme measure of occupy- ing the Ruhr in January, 1923 (to force the Germans to make some kind of war reparations), the French film industry began to seek out alliances with the Germans and Austrians. From American capital and film techniques, the French now turned to German capital and/or studio facilities for the co-production of so-called international films. Although untried by the film industry, this strat- egy was not without precedent-before the war, one faction of the French government had sought a reconciliation with Germany through various fi- nancial collaborations.66 This shift in alliances was partly defensive, the result of Paramount's shock- ing-to the French-invasion of the French film industry. Apparently, the way was prepared by L6once Perret who, in 1920, had returned from the United States to France and, with Paramount's assistance, had made several independent films under the rubric of Perret Pictures. The most successful of these, Le Demon de la haine (1922)shot in Texas, New York, Paris, London, the French Alps, and along the French-Spanish border-was heralded by Cina as the "first international film." Perret went on (with or without the help of Paramount is unclear) to set up a new production company for the sole purpose PRODUCTION 27 FRENCH FILM INDUSTRY of directing a major superproduction adapted from Pierre Benoit's best-seller, Koenigsmark (1923). Showcasing spectacular sets, Bavarian landscapes and cas- tles, and Huguette Duflos in magnificent costumes, Koenigsmark had a success that rivaled Feyder's L'Atlantide, breaking all the box office records set previ- ously by American films in France. According to Marcel L'Herbier and Leon Moussinac, it became a new model for French superproductions.67 On the strength of Perret's success, Paramount opened a new branch office in Paris in 1922 and named Adolphe Osso its director, with orders to imple- ment a policy of film production in France. Osso soon engineered a major coup by releasing one of the most popular French films of the 1922-1923 season, Henry Roussel's Les Opprims (1923). Although this story of a six- teenth-century Flanders peasant girl was shot in France (and Belgium) by a major French director, employing a French crew and cast (excepting the star, Spanish singer Raquel Meller), it was produced and distributed entirely by Paramount.68 That an epic reconstruction of French history should be made with American money piqued the French no end. While Roussel went on to form an important production company, Lutece Films, in order to exploit his partnership with Raquel Meller in another epic historical reconstruction, Vi- olettes impiriales (1924), Paramount initiated a second more contemporary co- production, Robert Boudrioz's L'Epervier (1924). When First National an- nounced plans, shortly thereafter, to film a co-production of Le Collier de la reine (with Norma Talmadge) on location at Versailles, the French film press exploded in rage.69 No French film company had ever been accorded that privilege. Pride as much as money was at stake. Paramount's involvement with a fourth French filmmaker brought grief of a different sort. In the summer of 1922, Adolphe Osso had asked Marcel L'Herbier to prepare for Paramount a script on Notre-Dame-de-Paris. When that project failed to develop properly, Osso agreed to a contract with Cin- graphic by which Paramount would handle the international distribution of L'Herbier's adaptation of Resurrection. When one-fourth of this ambitious film had been shot at the Epinay studio and on location in Lithuania (at a cost of 300,000 francs), L'Herbier fell ill with typhoid, and the project closed down. After his recovery, Marcel Lapierre writes, L'Herbier worked tirelessly to put together 2 million francs and was about to resume shooting on the film. At that moment, however, Universal reported that it was preparing an American version of Resurrection on a budget the equivalent of 20 million francs! Para- mount promptly reneged on the distribution of L'Herbier's film, and the Cine- graphic project collapsed for good. Undaunted by this demonstration of Amer- ican power and self-interest, L'Herbier was not to be denied. The following summer, he signed a contract with the singer-actress, Georgette Leblanc, and Otto Kahn (the prominent American patron of the arts and Paramount's principal banker) to produce a film starring Leblanc that would be destined for American as well as European audiences. Director and star clashed repeat- edly over Pierre Mac Orlan's scenario-the one wanting to "present a synthesis of the modern decorative arts," the other demanding more emphasis on her character and the story. The result was a "modernist fantasy," L'Inhumaine (1924), whose mixed critical reception and soft returns ended Cin6graphic's status as a fully independent production company.70 28 The conjunction of these film projects with the political tide that was build- ing to the Cartel des gauches victory in the 1924 elections spurred Aubert and Pathe-Consortium to seek a rapprochement with the German film industry. Aubert's own experience with Franco-American productions gave him added incentive. In 1923, he repeated Vandal and Delac's strategy of hiring Amer- ican stars for French films. Aubert hired another eclipsed American actor, Sessue Hayakawa, to star in two major superproductions, Violet's La Bataille (1923) and Roger Lion's J'ai tul (1924). Although the first had considerable success in Europe, the second flopped badly. The owner of the Fordys cinema chain made the same mistake by engaging the old serial queen, Pearl White, for a film entitled Terreur (1924)-with disastrous consequences. In April and May, 1924, while marketing his films in Germany, Aubert developed a fresh tack in negotiations with UFA and Erich Pommer. A new German production company by the name of Vita-Film had constructed a studio in Vienna and was anxious to collaborate with the French film industry on co-productions. Aubert and Pommer arranged for the big spectacle film, Salammbo (1925), to be shot there, with a French director and cast. The path to this co-production was probably laid down by Jacques Feyder who, after the debacle of Visages d'enfants, had accepted the position of "artistic director," as well as a three- film contract, at Vita-Film. Certainly it was Feyder who persuaded Vita-Film to produce Max Linder's Le Roi du cirque (1925), which Violet directed and Aubert distributed. Of the three films in production in Vienna during the summer of 1924, however, Salammb6 and Feyder's own L'Image (1926) did poorly at the box office. Their failure coincided with the fall of the Austrian currency, and Vita-Film was thrown into bankruptcy. Although the experi- ence was anything but fortuitous, Aubert's alliance with UFA and Pommer remained intact, momentarily.71 The most spectacular example of this international strategy, however, in- volved Pathe-Consortium in an alliance with two new firms in France and Germany. In 1923, the important German financier, Hugo Stinnes ("the coal merchant of Mulheim"), and a wealthy Russian emigre, Vladimir Wengeroff, conceived the idea of a grand European film consortium. Together, the two speculators formed the Westi Corporation in Berlin and began setting up affiliated companies throughout Europe, the most important being Cin6-France in Paris, directed by Noe Bloch, the former associate of Kamenka and Ermo- lieff. Late in 1923, Path6-Consortium joined the group and gave its authority to the strategy. The first contract was between Westi, Path6-Consortium, and Films Abel Gance. This was for Gance's mammoth project, a six-part Napo- lon, for which Westi created a separate administrative firm, Wengeroff Films. Gance's immediate response was to buy and complete Diamant-Berger's studio at Billancourt, where Napolon's shooting commenced in January, 1925. The next contract was between Westi and Cine-France for a project announced in March, 1924, Michel Strogoff (1926), to star Ivan Mosjoukine under the direc- tion of Leonce Perret. Another Westi-Cine-France project starring Mosjoukine was signed and begun the next summer, Le Prince charmant (1925), directed by Tourjansky. When difficulties cropped up over the production of Michel Strogoff (Perret signed a contract with Paramount), Tourjansky also agreed to direct the second film after completing Le Prince charmant at the Billancourt PRODUCTION 29 FRENCH FILM INDUSTRY studio of Abel Gance. Finally, late in 1924, a third Westi-Cine-France project put Germaine Dulac under contract to direct what would become Ame d'artiste (1925). Path&-Consortium agreed to distribute all of these co-productions in France; for its part, Westi contracted to distribute the Grinieff-financed Le Mirade des loups in Germany and Eastern Europe. The magnitude of this Franco- German consortium, much of it financed by Russian emigr6 money and em- ploying some of the best French directors and technicians, dwarfed even Au- bert's alliance with Erich Pommer and UFA. So much so that the axis of Pathe-Consortium, Cine-France, and Westi seemed to offer the possibility of a genuine European challenge to the American film industry's dominance.72 Yet strong political and economic forces were already in motion to under- mine the strategy of these alliances. The rapprochement foreign policy of the Cartel des gauches government certainly gave sanction to the cooperative enter- prises of the French and German film industries. Its economic policies, how- ever, had a double-edged effect. The bank capital strike and the high rate of inflation produced by these policies forced the industry to look beyond its national borders for production financing and thus indirectly fostered co-pro- ductions. But not many French producers were in a position to attract foreign capital-the major corporations, yes; the smaller firms and independent pro- ducers (with an exception or two), no. More and more money was going to fewer and fewer films. The consequence was that only fifty-five films were released in 1926, the lowest number until the transition year of 1929.7i Furthermore, the co-production strategy of the major corporations was thwarted by what was happening in Germany. In the summer of 1924, the Allies set up what was called the Dawes Plan to stabilize the Deutschmark (after Germany's rampant inflation threatened the international financial con- munity) and to give the Weimar Republic economic control over all of its territory (ending French occupation of the Ruhr), both of which were designed to provide Germany with the means to pay off its reparations debt.74 The Dawes Plan, in effect, encouraged the investment of American capital in Ger- many to shore up most sectors of its economy, including the film industry.75 One by one, the German film companies found themselves indebted to the Americans, partially absorbed by them, and then systematically robbed of many of their best personnel.76 By early 1927, according to Leon Moussinac, 75 per cent of German film production was financed by American money.77 Although the Dawes Plan may have helped deflect American film investment from France to Germany, it also disrupted the French-German efforts to create a truly European consortium. Consequently, the period from 1925 to 1926 saw a good deal of change in the production sector of the French film industry. The American influence in French film production grew slowly but steadily. Under Adolphe Osso's (sometimes token) stewardship, Paramount's Paris office presented the most lavish spectacle film of the 1925-1926 season, L6once Perret's Madame Sans-Gene (1925), starring Gloria Swanson as the legendary eighteenth-century rags-to-riches duchess. Its 14-million-franc budget clearly separated the fat from the lean, for a similar sum could have produced ten French films. To the consternation of the French film industry, Madame Sans- Gene had a tumultuous reception and was awarded the Jury Grand Prix at the famous Paris Exposition des Arts D6coratifs. Released six months earlier, in a much shortened version, in the United States, however, it drew faint praise 30 and less than expected grosses. Although some in the industry began to ques- tion the whole strategy of international films, Paramount itself went ahead with two smaller film projects for the following year.78 Metro-Goldwyn's alliance with Gaumont countered French film production in a different way. Its distribution contract of April, 1924, led to even further reductions in Gaumont's own filmmaking. When Louis Feuillade died in March, 1925, the company had no other directors or projects and La Cite Elg6 was turned into a rental studio. Within six months, Gaumont had become Gau- mont-Metro-Goldwyn, controlled by the American corporation as an outlet for MGM films.79 Aubert found itself being co-opted by the Paramount and MGM interests on another front, in Germany. Although German film production declined steadily during the first half of the decade (from 646 to 228 films per year) according to Georges Sadoul, the cost and the quality of its films just as steadily mounted. What specifically affected Aubert was the unsuccessful at- tempt by UFA, Germany's largest production company, to challenge the growing American hegemony with a series of monumental films for export--e.g., Lang's Nibelungen (1924) and Metropolis (1927), Murnau's The Last Laugh (1924), Tartuffe (1925), and Faust (1926). This strategy soon had UFA impossibly overextended-with vast studio holdings, scarcely any liquid assets, and an increasing indebtedness to American interests. In a joint agreement, signed in December, 1925, Paramount and MGM became controlling partners in a re- organization of the German firm. Each of the three partners would produce twenty films annually for distribution through a subsidiary, Parufamet. No longer its own master, UFA began using its contract with Aubert, as well as others, merely as one way to fill its production quota of films.80 The alliance between Pathe-Consortium, Cine-France, and Westi suffered an even greater shock. Early in 1924, Hugo Stinnes died suddenly, and Wen- geroff was left as sole director of Westi Corporation and its far-flung affiliates. It was he who had initiated all the projects in France: Gance's Napoleon and the three films under Noe Bloch of Cin6-France. He had even persuaded Jean Sapene to let Westi help finance Henri Fescourt's four-part adaptation of Les Miserables (1925-1926). Sapene needed the Westi money because Les Misrables was the biggest film Cinromans had attempted, with a budget of nearly five million francs and a cast that included fifteen major and fifty minor roles. In the spring of 1925, however, the Stinnes empire was disclosed to be deeply in debt. After a quick examination, the German banks decided to liquidate all the Stinnes companies, among them the Westi Corporation, which was dissolved in August, 1925. Suddenly bereft of financing, Gance's Napoleon was halted in mid-shooting, a shutdown that would last six months. Cineromans' Les Misirables and Michel Strogoff were also severely cut back near the end of production. Unable to operate on its own, Cin6-France disappeared. The dream of a European consortium was in ashes.81 For most of the smaller French producers, ironically, the domestic and foreign policies of the Cartel des gauches also spelled the end of independent film production. Considered on the commercial level, [films] must inevitably become a kind of factory product. More and more, film directors are condemned to bend to the exigencies of PRODUCTION 31 FRENCH FILM INDUSTRY companies-there is only one option: to submit or to be sent packing-and it is because they have not yet been able to replace this system that they support it.82 As their sources of financing dried up, the more commercial independent filmmakers, such as Julien Duvivier, Henry Roussel, and Jacques de Baron- celli, moved within the spheres of influence of Aubert/Film d'Art, the Societe des Films Historiques, or Cinromans. After the Westi collapse, Abel Gance kept his own company alive in name only. Bled white by his ill-fated Franco- American productions, Marcel L'Herbier allied Cinegraphic with Films Alba- tros on a project starring Mosjoukine, Feu Mathias Pascal (1925). Successful though the film was, Cinegraphic remained in debt, and L'Herbier was forced to accept a six-film contract with Cineromans. After his Swiss and Austrian debacles, Jacques Feyder reluctantly signed on with Albatros. He was joined there by Ren6 Clair, whose independent productions with Ren6 Fernand and Rolf de Mar6 were commercially unsuccessful. Finally, Jean Renoir's fortune began to run out. With Pierre Braunberger, the former director of Para- mount's publicity department in Paris, Renoir arranged to produce a film adaptation of Nana (1926), in conjunction with a German studio. When the film failed through a quirk in distribution, much of Renoir's one-million-franc investment was lost. Desperate to direct a more commercial project, he ac- cepted financing from actress Marie-Louise Iribe (the wife of Pierre Renoir), who set up Les Artistes R6unis so she could star in Marquitta (1927).83 Against this tide of retreat and consolidation, only a few producers and filmmakers could pursue their own way. Of the old guard, only Leon Poirier was fortunate enough to find independent financing. From 28 October 1924 to 26 June 1925, he and Georges Specht documented the Citroen-sponsored automobile race across Africa (from Morocco to Madagascar) in La Croisiere noire (1926). After that, however, Poirier was able to complete just one major fiction film during the last half of the decade, Verdun, vision d'histoire (1928). In 1925 and again in 1927, Germaine Dulac interrupted her contractual as- signments at Cin6-France and Cineromans to produce three small films on her own, none of which received more than scant circulation-Folie des vaillants (1925), L'Invitation au voyage (1927) and La Coquille et le clergyman (1928). Intrigued by his experience of starring in Raymond Bernard's Le Miracle des loups (1924) and Joueur d'echecs (1927), the famous Theatre de l'Atelier director and actor, Charles Dullin, formed his own film production company, Societe Charles Dullin. The company functioned much like L'Herbier's Cinegraphic, gathering together a small group of artists, including the documentary film- maker, Jean Grmillon, to make its one and only superproduction, Maldone (1928).84 Perhaps the most important of the independent production companies were those of Jean Epstein and Pierre Braunberger. Late in 1925, Epstein left Films Albatros, where he had directed four films, after breaking his contract at Path6-Consortium nearly two years before. Several months later, with the backing of Marguerite Viel (another Russian emigre) and the indirect support of Abel Gance, he established Films Jean Epstein and gathered around him a group of young film enthusiasts as assistants, one of whom was Luis Bunuel.85 His company produced four major films over the next four years, none of which had much success except in the limited circuit of specialized cinemas through- 32 out Europe. After the disappointment of Nana, Pierre Braunberger founded N6o-Film to both produce and distribute French films. Neo-Film was con- ceived even more along the lines of L'Herbier's earlier Cinegraphic. While financing major commercial projects by Jean Renoir and Alberto Cavalcanti (L'Herbier's former assistant and set designer), Braunberger also sponsored documentaries and experimental short films.86 Much like Epstein's work, his productions also (but not exclusively) often appeared in the specialized cine- mas. What Braunberger did to support several avant-garde filmmakers of this period provided a model for his later support of the young New Wave film- makers, especially from 1954 to 1958. The political and economic changes that occurred in France in 1926 reversed the downslide or slump that was spreading gloom throughout the French film industry. The Union nationale foreign policy of conciliation continued to en- courage international co-productions. But, more important, conservative fiscal policies finally ended the high rate of inflation and stabilized the value of the franc. Even the rate of unemployment fell drastically. As more capital became available again in France, the level of film production rose accordingly-from fifty-five films in 1926 to seventy-four in 1927 and ninety-four in 1928. Al- though the forty-three producers and twenty-five distributors responsible for those films evidenced much the same fragmentation and disorganization as before, the future looked bright.87 France seemed embarked on an era of pros- perity and the easy life. The four major production companies all went through changes in manage- ment and orientation. Albatros lost most of its Russian emigre base-Mosjou- kine signed a contract with Universal to work in the United States and then Germany; Tourjansky and Volkoff both transferred to Films Abel Gance and then to Noe Bloch's Cin6-France. To continue his policy of quality films at Albatros, Kamenka engaged Jacques Feyder, Ren Clair, and Henri Chomette (Clair's older brother and Feyder's assistant) to direct films that were more specifically French. His dividend was a series of brilliant comedies, the best of the decade. By contrast, the Societe des Films Historiques shifted to a non- French subject by putting a second epic historical reconstruction into produc- tion, Raymond Bernard's Joueur d'6checs (1927). Grinieff held up its shooting to divert funds to Gance's stalled Napolon; and after that risky adventure, Films Historiques resumed its own orientation. Its last projects included Rous- sel's life of Chopin, Valse de /'adieu (1928), and two historical films by Jean Renoir, one of which, Le Bled (1929), marked the centennial celebration of French colonization in Algeria. Louis Aubert did both more and less. Al- though he tried valiantly to sustain the high level of production he had ini- tiated in 1924, his 1925-1926 season of films contained some dreadful dis- appointments, such as Salammb&. In February, 1926, he ceded administration of the Aubert cinema chain to a M. Car and, like Pathe and Gaumont before him, began to take a less active role in his company's operations.88 Production was cut back slightly, but without disturbing the alliance with Vandal and Delac of Film d'Art. Although the company lost its major director in Ren6 Hervil, Aubert himself signed up the new film production team of Jean Be- noit-Levy (an educational filmmaker) and Marie Epstein (Jean Epstein's sister), and Film d'Art took Julien Duvivier under contract.89 After the collapse of Westi and Cin6-France, Jean Sapene's Cin romans was PRODUCTION 33 FRENCH FILM INDUSTRY left as the central house of production within Pathe-Consortium. The com- pany's announcement for the 1926-1927 season was down a bit from the previous two years: four serials, of course, but only four new Films de France, plus a special co-production with Cinegraphic.90 Was Sapene concerned about this slippage? Did he now believe himself better qualified than Louis Nalpas to run most phases of Cinromans' operation? For whatever reasons, "around 1926-1927," according to Henri Fescourt, "Sapene personally took charge of the preparation of all the films his company was going to produce."91 What particularly obsessed him was reading and revising all the dcoupages, especially rewriting the intertitles. This intervention even led him to the extreme of launching a publicity campaign around a series of films to make a star of his wife, Claudia Victrix (was he thinking of William Randoph Hearst and Mar- ion Davies?). In effect, Sapene took over Nalpas's position as executive pro- ducer. Jean-Louis Bouquet concludes that, though audacious in his business enterprises, Sapene was unusually cautious and even retrograde in his artistic tastes and that he failed to encourage and promote the best of his young staff within the company.92 Despite such innovations as a far-ranging campaign to advertise his film production schedule nearly a year in advance of its release date, the Cineromans films began to suffer from his lack of taste and judg- ment.93 As he himself admitted later, "Having entered the film industry by error, I remained there out of pride."94 Coupled with this leveling off or incipient stagnation of the major French film producers (which, some have argued, provided a climate of security), the stabilization of the French economy stimulated the development of several new production companies.95 Besides Pierre Braunberger's relatively minor No- Film, the two most important firms were Production Natan and Franco-Film, both of which seem to have developed out of associations with Paramount. Early in 1926, Bernard Natan, the director of a film processing company and publicity agency (Rapid-Film and Rapid-Publicit), turned his speculative sights on the opportunities in film production. First, he arranged distribution rights through Paramount for a film adaptation of Pierre Benoit's novel, La Chtelaine du Liban (1926). Then, after meeting Henri Diamant-Berger, just back from two years in the United States, Natan agreed to produce two of his films, one of which, Education de Prince (1927), was lucky enough to star Edna Purviance. Based on the success of these films as well as his association with Paramount, Natan purchased one of the old Eclair studios at Epinay and constructed an- other completely modernized one, the Studio Runis, on rue Francoeur in Montmartre. From Rene Fernand, he secured the rights to produce Marco de Gastyne's Mon coeur au ralenti (1928) and Maurice Gleize's La Madone des sleep- ings (1928). Then came his most ambitious project, the popular historical reconstruction that took two years to complete, Marco de Gastyne's La Vie merveilleuse de Jeanne d'Arc (1929). By this time, the Cinmatographie frangaise was calling Natan one of the most important French film producers.96 Early in 1927, Robert Hurel, another French producer at Paramount, formed a consortium called Franco-Film out of a number of small firms-Films Leonce Perret, Jacques Haik (Gaston Ravel), and Paris-International (actor-director Leon Mathot and Italian actress Soava Gallone). Within as many months, Hurel announced at least six films for the upcoming fall season, headlined by Perret's Morgane la sirene (1927). What helped the consortium get under way 34 was Perret's reputation plus the profits from La Fenlne nue (1926), which had been financed by Natan, perhaps with Paramount's assistance. By the summer of 1928, Franco-Film was so prosperous that it took over the former Louis Nalpas studios at Victorine (which Rex Ingram had been using for several years). Even though two of its affiliates had close ties to England and Italy, Franco-Film professed its commitment to "the development of French cine- matography" with "programs composed of French films. " The commitment was far from empty in the context of international co-productions. Despite the collapse of the Westi consortium, the costliness of Aubert's agreement with UFA, and a number of outcries from within the industry itself-Ren6 Clair, for one"-the strategy of international co-productions con- tinued in vogue. Georges Sadoul sums it up well: In the last years of prosperity, the French productions evolved toward huge set pieces with sumptuous costumes and decors, in hopes of rivaling Hollywood. For these "prestige" films, the last big French producers resorted, especially after 1925, to formulas which drew them into associations with various foreign countries, Germany mainly, but also England, Sweden, sometimes Italy, and more rarely the United States. For the powerful German cartel UFA, the co-production strategy was a means of enlarging its position in the French market. Collaborations between Berlin and Paris developed considerably." In fact, international co-productions threatened to monopolize the production capital and energy of the French film industry. Much of this investment went into one corporation that emerged from the ashes of Westi and Cine-France. Late in 1925, Grinieff, the Russian emigr6 financier behind the Societe des Films Historiques, along with Henry de Ca- zotte, founded the Societ6 Generale des Films (S.G.F.), on whose board sat none other than Charles Path6. The occasion of S.G.F.'s formation was Gri- nieff's decision to finance the remaining production schedule for Abel Gance's Napoleon (1927). Although only Part I of the six-part epic was completed, S.G.F.'s largesse seemed limitless. By the time Napoleon premiered at the Opera in April, 1927, it had become the most expensive French film of the decade--at a cost estimated somewhere between 15 and 19 million francs. Soon after Napoleon resumed shooting at the Billamcourt studio, S.G.F. an- nounced a second project, Casanova (1927), which drew most of its personnel from Albatros and Cine-France: Bloch (producer), Volkoff (director), Lochakoff (set designer), Bilinsky (costumes), and Mosjoukine (star). Finally, during the later stages of Napoleon's shooting, S.G.F. offered the Danish filmmaker, Carl Dreyer (whose Master of the House (1925] had just had a big success in Paris), the chance to make a film on a famous woman who might complement Napo- leon in French history. Dreyer accepted and began work on La Passion de Jeanne d'Arc (1928), importing his cinematographer and set designers from Ger- many. 100 The fascinating thing about S.G.F. was, as Harry Alan Potamkin put it, that, among the major French corporations, it alone allowed "the director, and not the fiscal policy, to set the pace" and character of its film produc- tion.11 Although restricted to historical reconstruction projects, Grinieff of- fered the filmmaker unbelievable conditions: total control over all phases of production and an almost unlimited budget. Overwhelmed by a screening of PRODUCTION 6. Becrnard Natan 35 FRENCH FILM INDUSTRY La Passion de Jeanne d'Arc, G. W. Pabst confessed that the German film in- dustry could never be that experimental with so much money. 102 This was atelier filmmaking raised to a level that even Marcel L'Herbier had not dreamed possible at Cingraphic. It was an incredible risk, especially on the eve of the sound film revolution. Perhaps inevitably, the result of this concentrated be- neficence was financial disaster for the company. International companies and cross-cultural contracts now proliferated in the place of independent French production units. Henry Roussel's Lutece Films, apparently aided by S.G.F., invited Maurice Tourneur back from the United States for a major French superproduction on the war, L'Equipage (1928).103 For his part Jacques Feyder signed a contract with DEFU-Deutsche First Na- tional to direct a French-German cast in a Berlin studio production of Threse Raquin (1928).104 The Societe des Films Artistiques, SOFAR, which had been based in Italy and Germany, initiated a series of modern studio spectaculars directed in France by the Italian director, Augusto Genina.105 By now, how- ever, the Russian emigre talent and money were shifting substantially to Ger- many. Late in 1927, Cinimatographie franfaise announced a six-film contract between UFA and Cin6-Alliance, a new company headed again by No Bloch. The first film was a lavish spectacle film, Volkoffs Schhrazade (1928), shot in UFA's Berlin studio. 106 Besides Volkoff, this project tied up Lochakoff, Bilinsky, and Nicolas Koline, all formerly of Albatros and Cine-France. Either out of necessity or speculative desire, the older French majors were drawn into the international strategy again. In 1927, no less than three Cin6- romans films were co-productions: Robert Wiene's La Duchesse des Folies-Bergere (filmed in Vienna, Berlin, and Paris), Gaston Ravel's Le Roman d'un jeune homme pauvre (shot in Berlin), and Marcel L'Herbier's Le Diable au coeur (made in conjunction with UFA and Gaumont-British).107 Aubert tried to recover from Salammb with another film directed by Wiene, Der Rosenkavalier (1927), which was shot in Vienna under the baton of Richard Strauss himself. 108 Late in 1927, Vandal and Delac, Aubert, and Wengeroff-Films signed an accord to produce a series of films jointly. 109 The first of their projects was Julien Duvivier's Le Tourbillon de Paris (1928). About the same time, Albatros and Sequana-Films (a production company newly established at the Billancourt Studio) announced "a French-German-Spanish-Swedish collaboration," which, after several comedies, culminated in Cagliostro (1929), with a French-German cast under the direction of Richard Oswald in Paris.110 And, despite its com- mitment to French films, even Franco-Film hired the German director, E. A. Dupont, to shoot Moulin Rouge (1928) in Paris and London as an Anglo-French production." As older production companies such as Cineromans and Aubert seemed to mark time, newer companies such as Franco-Film and Natan's Studio R6unis grew apace. In February, 1928, Gaumont severed its alliance with MGM, by common consent, and began developing its optical sound-on-film process for feature film production. 112 The stage was set for the transition to sound films. At least three conclusions can be drawn at this point about French film production during this period. First of all, the production sector of the in- dustry underwent a kind of compression-explosion that created a multitude of large-, medium-, and small-scale enterprises. Although Pathe-Consortium, 36 Gaumont, Cinromans, and Aubert (at certain moments) could claim that they made more films than any other company, clearly the majority of French films were produced by small firms and independent producers. In one sense, this dispersion of film production was a regression from the consolidated system that had marked the French film industry before World War I. It even looked archaic compared either to the corporate capitalist structure of the American film industry (with its model of division of labor and scientific management, instituted by Thomas Ince, as well as its half-dozen vertically integrated mo- nopolies, spurred by Zukor's Paramount Pictures and perhaps modeled on the old Path6-Freres company) or to the state-supported (through favorable legis- lation) structure of the German film industry.113 Heterogeneous and internally competitive, the French companies were hardly in a position to challenge the larger, more homogenous, more financially endowed corporations in the United States and Germany-though they doggedly persisted in doing so. In another sense, however, this decentralized system of small French production compa- nies depended on an unusual degree of entrepreneurial independence and ar- tisan- or -atelier-based praxis.114 That independence and praxis fostered an at- titude of risk-taking, experimentation, and concern for Frenchness, not only in certain individual filmmakers, but even in such large companies as Alba- tros, Cinegraphic, Cine-France, S.F.H., and S.G.F. On that basis alone, the French film industry could be said to rival its competitors in the United States, Germany, and the Soviet Union. Second, a good portion of the production sector's investment came from outside France. Less came from the United States than one might have ex- pected, at least until the last couple of years of the decade. But then, the British and German film industries and markets offered greater opportunities for the Americans.115 More came from Germany, after 1924, but gradually that investment amounted to the use of German capital and studio facilities for co-productions in Germany. The extraordinary thing was the high inci- dence of Russian emigre money in French film production. The heavy French investment in prewar Russia had not been lost entirely after all. The Russian emigre money supported a variety of production strategies-from Albatros and S.F.H.'s commercial quality films to Cine-France's international co-produc- tions, from Abel Gance's epic Napoleon to Jean Epstein's narrative experiments. Without this lavish financing (through Grinieff, Wengeroff, Bloch, and Ka- menka), the French production efforts of Sapene and Aubert might not have been sufficient to counter the American and German product. Nearly all of this foreign investment, however, went directly into individual films. For modernizing their production base in studios and equipment, the French had to rely on themselves. Technologically, they tended to lag behind the Amer- icans and Germans, from a lack of capital, certainly, but perhaps also from a lack of industrial research, research that linked investor and entrepreneur. 116 Consequently, the work of Debrie and Eclair (on camera equipment) and of Diamant-Berger and Sapene (on studio facilities), for instance, was all the more important for the industry's survival. Finally, the heady champagne of internationalism, which dominated the latter half of the decade, failed to strengthen and broaden the industry as had been hoped. The Americans could not have been more pleased with the results than if they had concocted the strategy themselves and unashamedly sold the PRODUCTION 37 FRENCH FILM INDUSTRY Distribution: The Divided Country French on it. As modern studio spectaculars began to rival in number the historical reconstruction films, the French co-productions increasingly mir- rored American films, reflecting a modern style of life whose characteristics of material well-being and conspicuous consumption were basically American. As early as 1926, an American analyst reported: The peoples of many countries now consider America as the arbiter of manners, fash- ions, sports, customs and standards of living. If it were not for the barrier we have established, there is no doubt that the American movies would be bringing us a flood of the immigrants. As it is, in a vast number of instances, the desire to come to this country is thwarted, and the longing to emigrate is changed into a desire to imitate. 117 As Marcel Braunschvig conceded, "film is in the process of Americanizing the world."118 Increasingly also, the American and German film companies were encroaching on French terrain (in 1928, United Artists and First National joined Paramount as French film producers) or siphoning off its talent (e.g., Feyder, Volkoff, Mosjoukine). More and more French resources seemed to be involved in projects whose profits went to the Germans or Americans (and sometimes both together). This encroachment and exploitation were even stronger in the distribution and exhibition sectors of the French film industry. If the French government's attitude toward French film production was strictly laissez-faire, its attitude toward film distribution and exhibition was a bit more ambiguous. That ambiguity is apparent in the government's fitful attention to film censorship and to film imports regulation. Before the war, according to Paul Leglise, the control of film distribution and exhibition, like that of theatrical performances, was in the hands of local authorities.' In 1906, the Chamber of Deputies abandoned the system of a national commission to control spectacles in France and delegated the power of censorship to local mayors and police chiefs. This system led to a crazy quilt of standards and wild fluctuations in censorship practices. The southeast provinces, including Lyon, were especially prone to ban films, and even the Minister of the Interior would step in occasionally to recommend the censor- ship of films "susceptible of provoking demonstrations that might disturb the public order and tranquility." The advent of the war brought further crack- downs, particularly on the crime serials. Finally, after continued protests from film distributors and cinema owners, the Minister of the Interior established a temporary national Commission du Contr6le des Films in June, 1916.2 The principal purpose of the cormission was to review the weekly Annales de la guerre, which was produced by the army's section on photography and cinematography. The head of that section, J.-L. Croze, remembers two examples of the commission's censorship practice. It banned shots of General Petain grimacing at the taste of the common sol- diers' wine, and, in 1918, it eliminated all references to the approaching armistice.3 The commission's jurisdiction also extended to the rest of the French film production and even to film imports, thus paralleling the power of local mayors and police chiefs, which remained in force. Marcel Lapierre has argued that another reason for the commission's creation was the popularity of the crime serials and their supposedly harmful effect on the young who, under 38 lessening parental control, were flocking to the cinemas.4 Yet the commission DISTRIBUTION allowed most of the serials to pass uncut in Paris, while some provincial authorities banned the Pearl White serials and even Gaumont's Judex (which, ironically, was quite properly moralistic compared to the previous criminal celebrations of Les Vampires). What the commission did do, however, was halt the distribution of such American films as Griffith's Birth of a Nation and Intolerance (which was screened privately within the industry in 1917, perhaps when Griffith himself visited France in May and October), forbid the produc- tion of "bloody" dramas such as Othello, and cut from Thomas Ince's Civili- zation Christ's words to his disciples, "Let peace be with you!"5 In 1917, the Minister of the Interior appointed a committee to recommend a national regulation system for film distribution and exhibition that could be instituted after the war. As both Leglise and Diamant-Berger have pointed out, the primary motivation of the committee was to protect the film industry from the vicissitudes of local censorship in the provinces. After two years of wrangling, on 25 July 1919, the government established a central commission of thirty members, headed by Charles Deloncle and including Charles Pathe, Leon Gaumont, and even Abel Gance. Henceforth, all films except newsreels would have to obtain visas from the commission to permit their distribution in France. The visa system applied equally to French films and imports, and several films soon came under fire. In 1920, and again in 1921, the Interior Minister personally banned one of Aubert's early film productions, Li-Hang le cruel. In 1920, the exhibition run of Marcel L'Herbier's L'Homme du large was interrupted to excise shots of violence and eroticism from several sequences in the village caf6. Louis Delluc's Fievre (1921) had to be recut for similar reasons. Despite the central commission's status, film bannings continued in the prov- inces, again especially in the southeast; and, surprisingly, several court rulings upheld their legality. Film journals (L'Ecran, Cin6-Journal), newspapers (Co- moedia), as well as filmmakers such as Andr6 Antoine (who had fought censorship in the theaters two decades earlier), protested both the national and local forms of film censorship. However, the industry generally downplayed the issue, and these protests fell on deaf ears in the National Assembly. Sanctions continued against such films as L'Herbier's Don Juan et Faust (1923), Epstein's Coeurfidele (1923), Feyder's L'Image (1926), and Cavalcanti's Rien que les heures (1926).6 Finally, in 1926, various segments of the film industry, particularly the distributors and exhibitors, began to agitate for more equitable controls. Their pressure eventually led to Edouard Herriot's "Decree of 18 February 1928," which strengthened the central commission and limited the power of local authorities. Individual French filmmakers, however, did not escape further reprimands-e.g., L'Herbier's L'Argent (1929), Dreyer's La Passion de Jeanne d'Arc (1928), Benoit-L6vy/Epstein's Peau de pche (1929). Jacques Feyder had the most difficult time with Les Nouveaux Messieurs (1929), which was prohib- ited because it allegedly insulted the French National Assembly (at one point, the satirized hero was made up to look like the president). For the commis- sion's second review, the film was presented by Henry Roussel, who had gained the support of influential friends in the Assembly. In order to save face, the censors asked Albatros, the film's producer, to cut just twenty meters of in- tertitles during the love affair scenes, and Les Nouveaux Messieurs was ap- proved.7 39 FRENCH FILM INDUSTRY However, the primary targets of control were the foreign imports. German films, beginning as early as 1922, with Lubitsch's Madame Du Barry, routinely were re-edited or banned outright-e.g., Pabst's The Joyless Street and Pandora's Box, Lang's Spies, and May's Asphalt. Even some American films came in for excisions. Rene Jeanne recalled that Griffith's Intolerance, for instance, was shorn of its St. Bartholomew's Day massacre story (because it depicted an unpleasant period in French history); and Louis Delluc saw prints of the same film that had been reduced in length from three hours to a mere one and a half hours. But the real brunt of French censorship was borne by the Soviet films. First Eisenstein's Battleship Potemkin and then Pudovkin's Mother were banned, only to circulate widely in special screenings legally organized by the Cin6-Club de France and Les Amis de Spartacus. After the rightist victory in the 1928 elections, the ban on Soviet films became total. In 1929, Cinia-Cin6- pour-tous listed six Soviet films which had been forbidden distribution: Kule- shov's Dura Lux, Eisenstein's Battleship Potemkin, Wiskousky's Black Sunday, Pudovkin's Mother, Eisenstein's October, and Pudovkin's End of St. Petersburg. As it turned out, the central commission on film censorship operated indirectly as France's chief means of limiting film imports.8 During the 1920s, the distribution sector of the French film industry was determined, to a large extent, less by its own production than by the quantity and quality of foreign imports. Although imports, primarily from the United States, comprised 75 percent or more of the films marketed in France by 1918- 1919, most of them were still distributed by French companies (with the exception of Vitagraph). As the American film industry consolidated into a half-dozen vertically integrated corporations, it began to intervene directly in the French film economy. One after another, the major American firms set up offices in Paris or strengthened their alliances with French distributors. In 1920 came Paramount and Fox-Film, not only with their feature films but with newsreels as well.9 Jean Mitry recalls the sensation caused by the huge posters that suddenly materialized everywhere, depicting a tripod and camera straddling the globe: "It's a Paramount film!"'0 A year later, they were joined by Associated Artists (United Artists) and Erka Films (First National).1"In 1922, it was the turn of Universal, which offered a special premiere of Metro's The Four Horsemen of the Apocalypse at the Theatre de Vaudeville."12 Meanwhile Goldwyn and Metro signed exclusive distribution contracts with Gaumont and Aubert, respectively. By 1923, Cinimatographie franfaise was often listing the weekly releases of Paramount, Universal, and Fox-Film ahead of those of the French distributors. Even the two largest companies, Pathe-Consortium and Gaumont, were distributing fewer French films than imports. France had be- come a veritable dumping ground for the American film industry. After turn- ing a profit in the United States, films could be sold cheaply and easily on the French market for an added bonus. At the same time, German films began to appear in the Paris cinemas, for the first time since before the war. Breaking the barrier of prejudice and hostility was Robert Wiene's Cabinet of Doctor Caligari (1919), which Louis Delluc arranged to have screened at the Colis6e cinema in October, 1921. Others soon followed in 1922, imported chiefly by Cosmograph and screened at the Cine-Opera: Lang's Destiny (1921), Wiene's Genuine (1920), Pick's Shat- tered (1921) and Murnau's Nosferatu (1922). To oppose this invasion, the Co- 40 mite de defense du film fransais was formed by director Gerard Bourgeois and DISTRIBUTION actor Jean Toulout, with the support of Germaine Dulac, Ren6 Le Somptier, Ren6 Hervil, Armand Tallier, and others. The National Assembly even tried to ban Lubitsch's Madame Du Barry, claiming that it was an insult to French history. Neither measure was successful. The walls were down, and the pos- sibilities of exchange were once more open. 13 For their part, the French were attempting to rebuild the foreign markets they had lost in the war, especially the more lucrative ones in Germany and the United States. The meager range of their common export market can be gleaned from Cin6-pour-tous's publication of the balance sheet on a moderately expensive (310,000 francs) Path-Consortium film, Pierre Caron's L'Homme qui vendait son dme au diable (1921).14 Twenty-five prints had been sold in France, about twenty in England and its colonies, ten in South America, four in Eastern Europe, three in the Scandinavian countries, two in Belgium, and one each in Switzerland, Holland, Spain, Portugal, Greece, Turkey, Egypt, and Japan. None in the United States, Germany, or the Soviet Union. With those sales, the film barely turned a profit; had it been released just a year or two later, when the American industry had taken control of the British market, it would not have done so. As their first goal, the French distributors blithely took on the herculean task of breaking back into the American market. Several ambitious efforts quickly ended in failure: Diamant-Berger's film with Adolphe Osso, Vandal and Delac's films with Fanny Ward, and the embezzled Franco-American Cin- ematographic Corporation. Yet some French films seem to have made their way into the United States just after the war-e.g., Mercanton's Bouclette (Eclipse), L'Herbier's Rose-France (Gaumont), and Dulac's Le Bonheur des autres (A.G.C.). Unfortunately, it is still unclear whether any were actually screened. In 1919, for instance, Abel Gance discovered that Path6-Exchange did have a print of J'Accuse, but they had done little to sell it. Nothing drastic was done until 1921, when the French heard that German films were being shown with some success in New York. In October, 1921, Gance finally sold a second version of J'Accuse (for $192,000) to United Artists, who at least exhibited it in New York and Los Angeles. A month later, Louis Nalpas personally sold his two biggest films, La Sultane de l'amour and Mathias Sandorf, to First National. After successfully blocking United Artists from showing Douglas Fairbanks's The Three Musketeers (1921) in France in order to leave the market open for his own production, Les Trois Mousquetaires (1921-1922), Henri Dia- mant-Berger also went to New York and sold his film there in the spring of 1922. On his return, he reported on half a dozen French releases, including Feyder's L'Atlantide (retitled Missing Husbands), Mercanton's Miarko and Phroso (all three exported by Aubert), Poirier's Narayana (Gaumont), and Ermolieffs L'Ordonnance (Pathe-Consortium). A year later, Andre Capellani returned to France and reported on several more-Baroncelli's Le Reve, Henry Roussel's La Faute d'Odette Marechal and Visages voiles . . . cmes closes (retitled The Sheik's Wife).15 In those three years, however, only three films enjoyed any measure of success in the United States: L'Atlantide, Boudrioz's L'Atre (retitled Tillers of the Soil), and Feyder's new film Crainquebille (Old Bill of Paris).16 Simply put, the French films never really caught on because the French export efforts were 41 FRENCH FILM INDUSTRY poorly financed, independent of one another and uncoordinated, and because their American distributors did not push them. As Marcel Lapierre concludes, The Americans had no intention of establishing a current of exchange. They produced films in overabundance: therefore, they had no need of ours, even if excellent. They had no interest, either material or moral, in opening their doors to our film exports. Of course not, otherwise how can one explain the fact that they mutilated or bowd- lerized some of our films which were presented to the American public?" No more than a dozen French films were exhibited annually from 1920 to 1925, and few reached cinemas outside New York, Los Angeles, and a couple of other large cities. Even in New York, according to Diamant-Berger, there were only five or six small cinemas that consistently screened foreign (let alone exclusively French) films.18 Disillusioned by their repeated failures in the United States, the French turned their attention to Germany and, almost immediately, were rewarded. Operating independently, Diamant-Berger made Le Petit Cafe (with Max Lin- der) the first successful French film in Germany since before the war. In No- vember, 1922, Vandal and Delac of Film d'Art opened talks with Erich Pom- mer about distributing French films on a regular basis in UFA's cinema circuit. The following year, Film d'Art and Aubert films began appearing in German cinemas-e.g., Feyder's L'Atlantide, Baroncelli's Le Rve. By early 1924, Au- bert and Film d'Art had a lucrative contract with Pommer and UFA, and twice as many French films (forty-four) were being shown in Germany as there were German films (twenty) in France. Although the French figure included films produced over several years, it still represented a good percentage of overall French film production. Within a year, according to Tout-Cinema, the number of French exports (thirty) to German imports (twenty-three) was nearly reciprocal. 19 A similar arrangement also developed about the same time with the Soviet Union, probably stimulated by the Herriot government's recognition of the U.S.S.R. in 1924.20 Little is known yet about this export agreement-who arranged the sales, the number of film prints that were sold, where and how often they were shown. According to Russian sources, however, the flow of French exports to the Soviet Union rivaled and perhaps surpassed those going to Germany.21 Beginning with just one film, Judex, in 1921, the number increased to more than fifty by 1925.22 As might be expected, these included most of the films produced by Films Ermolieff and Films Albatros-the most interesting entry perhaps was Le Brasier ardent (soon after its release in 1923). Other important films included La Dixieme Symphonie (in 1922); L'Atlantide and Don Juan et Faust (in 1923); J'Accuse and Les Trois Mousquetaires (in 1924); L'Atre, Crainquebille, Coeur fidle, La Cit foudroyee, La Britre, and Le Diable dans la ville (all in 1925). Surprisingly, La Roue is not mentioned as being imported until 1926. What kind of impact might these particular films have had on the nascent Soviet cinema? The question of foreign film imports so divided the French film industry that it took quite a while to develop any kind of organized effort to control them. Although producers obviously favored controls, most cinema owners actually opposed them; and the distributors were somewhere in between. Con- sequently, writes Georges Sadoul, it was not until the threatening invasion of 42 German films that a debate over film imports finally surfaced in the National Assembly. Pressured mainly by Pathe-Consortium and the Confed6ration G- nerale du Travail (C.G.T.), Aristide Briand had decreed, early in 1922, an ad valorem tax of 20 percent on all imported films. So benign was it, however, that the new tax had little effect, since most French distributors and exhibitors could still make more money by marketing American films than French ones. Under continued pressure, in 1922, Marcel Bokanowski, the Minister of Fi- nance, included in his budget request to the Assembly a proposal that would have charged a surtax on cinemas whose programming consisted of more than 80 percent foreign films and that would also have raised the ad valorem tax to 50 percent. When the budget was rejected and Bokanowski removed, the original proposal was modified by Taurines in order simply to establish a quota system that would reserve 33 percent of all cinema programs for French films. The idea for this quota system had come from the Comite de defense du film franqais, which persuaded some of the distributors (notably Aubert) to accept a minimum quota of 25 percent. However, another faction of the distributors (with Gaumont as its spokesman) argued that they would be hard pressed to fulfill even the lower quota figure. That faction was joined by the national organization of cinema owners (led by Leon Brezillon) in even more vehement opposition. In the confused debate of the Assembly's last days, during which the Radical party's fears of creating a new bureaucracy seem to have carried the day, the Taurines measure went down to defeat. The Comite de defense du film franqais, together with various labor unions and professional groups (writers, actors, cameramen, etc.), protested in vain.23 French films would remain unprotected from the flood of American imports. More important, a schism had opened between two major factions of the French film industry: Path6-Consortium and Aubert on one side, Gaumont on the other. In 1923-1924, either responding to the French threat of import quotas or taking advantage of the weakness of that threat, several of the American film distributors in Paris made a further incursion into the French industry. Under Adolphe Osso, Paramount began producing and distributing its own French films-Roussel's Les Opprimes (1923), Boudrioz's L'Epirvier (1924), Perret's Madame Sans-Gene (1925). Besides, it allied itself with Cinegraphic to distrib- ute (and perhaps deliberately undersell?) L'Herbier's Rsurrection (unfinished), Jaque Catelain's Le Marchand de plaisir (1923), and, in the United States, L'Herbier's L'Inhumaine (released in 1926, finally, as The New Enchantment). Vitagraph also expanded its distribution practice by releasing Volkoffs Kean (1924), before its network was bought out by Warner Brothers.24 Although not directly linked to these transactions, Gaumont moved steadily into the American orbit, seeking financial security and even survival. In 1922, L6on Gaumont finally conceded defeat to the American cinema: The American market is completely closed to us while our screens are cluttered with American films. We can do nothing about it-it's a battle between the clay pot and the iron kettle.25 With the company's financiers now in control, production was cut back and contracts signed with Metro, Goldwyn, and others to distribute and exhibit more American films. In 1924, when even Louis Feuillade's serials began to lose their appeal, Gaumont tied itself more closely to Metro-Goldwyn as the DISTRIBUTION 43 FRENCH FILM INDUSTRY new American combine's exclusive distributor in France." Within a year, however, the larger of the partners easily gained control, and Gaumont became Gaumont-MGM. In all but appearance, the once French giant was now MGM's branch office in Paris. Throughout the first half of the 1920s, Path6-Consortium was the bulwark of opposition to any threat of American hegemony in France. Besides distrib- uting its own films, either the superproduction serials or the literary adapta- tions of S.C.A.G.L., it was also the major outlet for the smaller firms and independent producers. From 1921 to 1923, the following producers and film- makers channeled their work through Pathe-Consortium: Films Ermolieff, Films Tristan Bernard, Films Abel Gance, Productions Leonce Perret, Productions H. Pouctal, Films Andr6 Legrand, Films Luitz-Morat and Pierre Regnier, Films Andre Hugon.27 In 1923 and 1924, as the level of French production dropped, it took the lead in re-releasing older films, especially during the summer months. Then, as Jean Sapene rose to power with Cin6romans, the company began to consolidate its distribution. Cineromans serials and Films de France became the single most important source of Pathe-Consortium prod- ucts. The Cineromans serials, particularly, seem to have played an important role in the French film industry. Their regular, increasingly popular distribution through Path6-Consortium came precisely at the moment when, fresh from their conquest of the British film industry and just before their intervention in the German industry, the American film companies made an effort to gain control in France. According to Jean-Louis Bouquet, The large American firms really wanted to impose a blockbooking system, a kind of annual programming deal, thanks to which they had already gained control of the British exhibition market. The serials shattered these attempts at a monopoly. . . .28 His colleague, Henri Fescourt, continues the argument: the serial episodes offered exhibitors the guarantees of a long series of weeks of huge returns from a faithful mass public hooked on the formula. Thus assured of a program base for three-quarters of the year, the cinema owners could resist the foreign film salesmen. . . .29 Without the Cineromans serials, French film exhibitors might, like the Brit- ish, have turned their screens over entirely to American films. Consequently, during a crucial period, Path6-Consortium and Cin6romans shored up the French film industry. Perhaps anticipating the coming American investment in Germany and def- initely encouraged by the Cartel des gauches' foreign policy, Sapene began cut- ting down the number of American film imports that Pathe-Consortium still distributed (the company kept its options open) in favor of German imports and Franco-German productions. In this new strategy, Sapene was following the lead of the second largest French film distributor, Louis Aubert. In 1921, Aubert had the foresight to take over the distribution rights to L'Atlantide when Jacques Feyder's financiers became alarmed over its soaring costs. The film made a fortune for him, playing at the prestigious Madeleine Cin6ma alone for one year, a record eclipsed only by Ben-Hur in 1927-1928. Within a year, Aubert had taken over most of the distribution contracts from A.G.C. 44 and formed an alliance with Vandal and Delac of Film d'Art as well as with DISTRIBUTION Films de Baroncelli. Prospering from these alliances, from an expanded circuit of cinemas, and now from his own productions, Aubert opened two initiatives in 1924. One was the agreement with UFA in Germany for a reciprocal ex- change of French film exports and German imports-e.g., Violet's La Bataille and Hervil's Paris for Lang's Niebelungen and Murnau's The Last Laugh.30 The other initiative complemented Sapene's serial strategy at Cinromans. For the 1924-1925 season, Aubert organized what he called "Festivals of French films," a program of recent and current French releases (primarily Aubert/Film d'Art productions) to be distributed in the provinces over* a two- or three- month period. According to Aubert's own publicity sheet, the typical program ran as follows: Week 1 Frou-Frou Week 2 Les Premieres Armes de Rocambole Week 3 Le Secret de Polichinelle Week 4 Le Voile du bonheur Week 5 Le Crime d'une sainte Week 6 La Sin-Ventura Week 7 La Bataille Week 8 La Legende de Soeur Biatrix" Although Cin6-Information-Aubert used several of these festivals for promotional purposes (Cin6ma-Palace de Joigny {Yvonne, Grand-Theatre d'Antibes), their overall success was apparently mixed.32 For his next season, Aubert seems to have dropped the idea. Still, this festival concept was important because it represented one of the few French attempts at a form of blockbooking. With Pathe-Consortium and Aubert concentrating on their own films, and the once powerful A.G.C. nearly defunct, the many smaller production com- panies and independent producers created a demand for new distributors. In 1924, Kamenka took over the distributor E. Girard, which became Films Armor, to release his films from Albatros. Soon Armor was being hailed as the third largest distributor of French films.33 Another new company, Grandes Productions Cin6matographiques (G.P.C.), started out as a production effort to make French films along the lines of Vandal and Delac and Aubert; but all it could come up with were small films by Gaston Roudes. In 1924, it turned to distribution and released a half-dozen films, most of them Cin6graphic productions.34 Its development was bound up with that of an important re- gional producer-distributor, Phoc6a, in Marseille. On the strength of its box- office bonanza with the serial, Les Mysteres de Paris (1922), Phoc6a enjoyed a momentary prominence, even to the point of opening a foreign exchange of- fice.35 By 1926-1927, both Phocea and G.P.C. had become production units in a small consortium called Interfilm.36 Along with Phoca, which was orig- inally formed to distribute American films during the war, several other for- eign import companies began releasing French films. Another American im- porter, Georges Petit, for instance, distributed most of Julien Duvivier's early films; and a German importer, Mappemonde Films, agreed to handle Ren6 Clair's Le Fantome du Moulin Rouge (1925).37 One of the chief distribution problems during this period was the relative lack of independent distributors and their weak position within the industry. 45 FRENCH FILM INDUSTRY Although independent producers proliferated, their counterparts in distribu- tion grew more slowly. With the exception of Armor, none of them had much influence in the French exhibition market. Furthermore, they tended to op- erate separately and were often at odds with one another. The only one, ap- parently, to have sought some kind of power base to coordinate the release of independent films was Jean de Merly, Henry Roussel's producer on Les Op- primes (1923). Somewhat like Aubert, de Merly made a small fortune from handling a single film, Roussel's Violettes impiriales (1924).38 The following year, he sank most of his profits into the distribution of a supposedly balanced program of half a dozen independent films by Roussel, Feyder, and Gaston Ravel. When all but one of those films showed a net loss, de Merly's drive toward a consortium of independent distribution stalled. In 1926, he was forced to reorganize his company in association with Fernand Weill.39 All the other important independent distributors were geared to individual filmmakers or films. L6on Poirier, for instance, founded the Compagnie universelle cin6- matographique in order to distribute La Briere (1925) as well as Jean Epstein's first independent film, Mauprat (1926).40 Even smaller distributors were Georges Loureau (perhaps allied with Mappemonde?) who handled Clair's Entr'acte (1924) and Le Voyage imaginaire (1926), and Maurice Rohier, who released several films by Epstein and Dmitri Kirsanoff.4' In 1927, the French film industry received another shock that spurred re- newed efforts to impose some form of import quotas and that led to a further round of reorganization within the distribution sector. During the middle 1920s, the level of American imports declined steadily as films became big- ger-from 589 in 1924 to 368 in 1927.42 However, that decline was offset by an increase in the level of German imports. From 1924 to 1926, the number of French exports to German imports had decreased, but not drasti- cally, to a level that was just less than reciprocal. Suddenly, in 1927, German imports tripled, without any corresponding increase in French exports. In fact, more films were imported from Germany (ninety-one) than were produced by the entire French film industry (seventy-four). This jump in German imports might have been foreseen in the spring of 1926, when the Alliance Cinema- tographique Europeene (A.C.E.) was set up in Paris.43 Although linked to S.G.F., as the distributor of Dreyer's La Passion de Jeanne d'Arc (1928), for instance, A.C.E.'s primary objective was to distribute the films of UFA and other German firms in France. Bypassing Aubert and Path-Consortium, A.C.E. quickly became the chief French distributor of German films.44 The German drive (supported by American money) to infiltrate much of the French market was on the verge of success. Cinematographically, France was in danger of being colonized a second time over. The French film export market, by contrast, remained substantially un- changed. A good number of films were now being exported, especially to Germany and the United States, but they were having little impact on either market. In Germany, although certain titles were quite profitable, the French films never comprised more than 5 percent of the total distribution market. The situation was worse in the United States. The number of cinemas exhib- iting foreign films, even in New York, was still quite low. And many of the French films had a limited audience because they were distributed only through the Film Arts Guild.45 According to George Pratt, Feyder's Visages d'enfants 46 and Clair's Paris qui dort were shown for one evening only, 29 June 1926, at DISTRIBUTION the Cameo cinema in New York.46 Moreover, the American distributors some- times sabotaged the more commercial French films. The classic example is Gance's Napolon. In its shortened triple-screen format, Napoleon was exhibited in twelve different cities across France and in eight major European capitols (where, in just twenty-one days, allegedly, it nearly made back its colossal costs).47 When MGM paid $450,000 for distribution rights to the film in the United States, Gance was hopeful of a worldwide success.48 But the American company cut the film drastically (from seventeen to just eight reels), concen- trating solely on producing a coherent narrative.49 They re.duced the triple- screen images to miniatures on a single screen and, according to Kevin Brown- low, ended the film with the Ghosts of the Convention sequence-including among the shots of Marat, Danton, and Robespierre, a similar soft-focus insert of George Washington!50 Ignorant of the film's origins, exhibitors were baffled and infuriated.51 Metro's publicity campaign could not disguise the fact that they had turned Gance's powerful epic into a dull period film. Only recently, therefore, has Napolon been shown here in anything resembling its original form. Yet the French themselves sometimes undermined their own efforts. The marketing of Fescourt's Les Misdrables (1925-1926) offers a case in point. As- suming that the American distributor would automatically exhibit the film in its four-part format, Pathe-Consortium shipped all twenty-two reels of it (about six or seven hours) to the United States. "It required eight months of the most difficult and expensive efforts," wrote Howard Lewis, "to rework and remodel this product of French studios into a shape suitable for American audiences."52 Besides, by the summer of 1928, the American film industry was already entering the transition period to sound films; silent films could no longer attract audiences in the large cities where French films would have been shown. Acutely aware of its continuing import/export imbalance, especially in light of the now favorable ratio in the general French economy, the film industry organized an intensive campaign in October, 1927, to protect its film pro- duction once and for all through a quota system that would really control film imports.53 The campaign was led by influential men like Sapene and Aubert, representing most of the French producers and distributors, and was supported by the unions and professional groups (including many of the same people as in 1922-1923). This time they were victorious, if only for a short time, and the new government adopted the Herriot decree of February, 1928: for every seven foreign films imported into France (four American, two German, one British), one French film would be purchased and exhibited overseas. More- over, in an effort to indirectly subsidize French film production, a system of import permits was instituted that favored French distributors at the expense of the American branch offices in Paris. The American film industry quickly retaliated by calling a boycott on fur- ther distribution contracts in France; and, in an unprecedented move, Will H. Hays sailed to Paris to lobby personally for the industry. The boycott divided the French again, with exhibitors, especially in the provinces, loudly opposing the new decree. After a series of joint conferences between repre- sentatives of the two industries, along with a timely objection from the League of Nations, within two months, the French capitulated. The seven-to-one 47 FRENCH FILM INDUSTRY quota system would continue in force, but the import permit subsidy was dropped. American films could still comprise up to 60 percent of the French market, while French films exported to the United States would remain at a level of ten or twelve per year. There were even those, such as Leon Moussinac, Jean Tedesco, and Hubert Revol, who argued that, since the original Herriot decree contained enforcement clauses that were clearly unworkable and that favored the major French producers (including Paramount), it may not have been in the best interests of all segments of the industry anyway. One year later, at a most inopportune moment, the French again tried to reduce the import-export ratio (this time to 3:1), but to no avail. Confronted by another American boycott, from April to September, 1929, which coincided with a growing demand for sound films, the government retreated to the seven-to- one ratio once more.54 By the end of the silent film period, the distribution sector of the industry was divided into three major factions. The American faction, led by Para- mount and Gaumont-MGM, controlled probably the largest share of the French market (certainly in numbers of films released). Included in that number were more French films than ever before, since First National, Fox-Film, Universal, and United Artists had all joined Paramount to produce and/or distribute French projects in 1928-1929. The German faction, headed by A.C.E. and Luna-Film, had an influence above and beyond its small size. Securely within its orbit now were two of the largest French companies, Aubert and Armor, whose programs increasingly were given over to German films. The embattled French faction still stood its ground. Pathe-Consortium continued to release more French films than most of the other firms put together. Aubert and Armor were next in line, but much of their profit now came from German releases. Jean de Merly and Fernand Weill joined together in a loose consor- tium to distribute Marie-Louise Iribe's Hara-Kiri (1928) as well as some of the bigger historical reconstruction films at the decade's end: Bernard's Joueur d'6checs (1927), Poirier's Verdun, visions d'histoire (1928), and Fescourt's Monte- Cristo (1929). Another new firm, Jean de Venloo, established itself by first distributing Le Miracle des loups abroad and then releasing several of Baroncel- li's films-Pecheur d'Islande (1924), Veille d'armes (1925), Le Rveil (1925)-as well as a new shortened version of Diamant-Berger's Les Trois Mousquetaires." By 1927, de Venloo was in a position to bargain with several important independent producers: Charles Dullin-Jean Grmillon's Maldone (1928), the Societ6 des Films Historiques-Roussel's Le Valse de /'adieu (1928), and Lutece Films-Cavalcanti's Le Capitaine fracasse (1929). Finally, through N6o-Film, Pierre Braunberger began to interest himself in the distribution of French avant-garde films (both narrative and non-narrative) in France and abroad. There is little doubt that the distribution sector, through the efforts of Path6-Consortium and Aubert, helped save the French film industry from complete capitulation to the Americans, especially during the period from 1922 to 1925. However, that sector also proved to be the weakest component of the industry. It failed to develop a systematic or coordinated pattern of film booking, other than the serial format, to counter the glut of American films. It failed to come up with an organized consortium or network to distribute the large numbers of independent films. And it wavered, divided, in its atti- 48 tude toward a government system of controls on foreign imports, eventually always siding with the exhibitors, whose profits depended on those imports. The consequence of these failures was that the Americans and Germans had a secure foothold within the French film industry at the crucial moment of the transition to sound films. The French cinema owners often seemed to operate quite separately from the other two sectors of the film industry. From the industry's beginnings, most of the cinemas constructed in France were independently, even individ- ually, owned (the figure consistently hovered around 80 to 90 percent). Dur- ing the war, as American films replaced French films in number and popular- ity, imports became the chief source of the exhibitors' revenues. Because that condition persisted throughout the 1920s, the exhibitors repeatedly opposed any form of government control on imported films, especially from the United States. Their position was strong enough, most importantly with the govern- ment, to defeat all proposals offered by the French film producers and distrib- utors. Compared to the rest of the industry, furthermore, the exhibition sector was unusually secure financially. The degree of that security is suggested by the steady rise in box office receipts, in Paris alone, throughout the decade: EXHIBITION Exhibition: "We're in the Money" 1923 1926 1929 85,428,746 francs 145,994,959 francs 230,187,461 francs' Although the 1926 figure may be deceptive because of the high inflation, the 1929 figure represents an astounding jump in the cinema's popularity. Despite this prosperity, sometimes at the expense of the rest of the industry, the French cinema owners had their share of problems. The most serious problem, in the minds of the exhibitors, was taxation. If the French government showed scant concern for film production and distri- bution (except on the matter of imports and exports), it quickly included film exhibition among the sources of added and increased taxes that the war seemed to have demanded. As audiences flocked to the cinemas during the first year of the war, the National Assembly debated various proposals to institute a national tax on spectacles. What was finally agreed to, on 30 December 1916, imposed a fixed tax rate on theaters and music halls and a progressive tax rate on the cinemas. 5 percent of monthly receipts under 25,000 francs 10 percent of monthly receipts of 25,000 to 50,000 francs 20 percent of monthly receipts of 50,000 to 100,000 francs 25 percent of monthly receipts over 100,000 francs This was in addition to the ancient "poor tax," which was set at 9.5 percent for all three spectacles. By 1920, the government had grown so used to this new tax base that it was sanctioned in peacetime as well. Lobbied by the theaters and music halls, which wanted to reverse the wartime advantages enjoyed by the cinemas, on June 25, 1920, the National Assembly adopted a new tax law which reduced their tax rate and increased the rate on the cinemas even more. 49 FRENCH FILM INDUSTRY 10 percent of monthly receipts under 15,000 francs 15 percent of monthly receipts of 15,000 to 50,000 francs 20 percent of monthly receipts of 50,000 to 100,000 francs 25 percent of monthly receipts over 100,000 francs The new law also authorized municipalities to levy a local tax on cinema admissions up to 50 percent of the national tax. After all, the Finance Minister is reported to have said, "Because of its current vogue, the cinmatographe can bear taxation more easily than any of the other spectacles."2 The French exhibitors, especially the smaller owners, were outraged by this new increase, particularly since it came during a period when high unemploy- ment and a general strike threatened to erode their audience. Moreover, they began to object to the rental rates demanded by the French film distributors, which they considered draconian compared to the cheap American rates. In 1921, a good number of the smaller cinema owners even threatened to close their cinemas for the summer. This "strike" turned out to be poorly organized, however, and had little effect. Simultaneously, the larger exhibition circuits organized a committee (headed by Edmond Benoit-Levy and Edgar Costil) to work with Marcel Bokanowski in getting the National Assembly at least to lower the tax. Although Bokanowski's measure failed, a second proposal to reduce the national tax was introduced by Taurines and passed (without its original import quota restrictions) in the National Assembly on 1 July 1923. 6 percent of monthly receipts under 15,000 francs 10 percent of monthly receipts of 15,000 to 30,000 francs 15 percent of monthly receipts of 30,000 to 50,000 francs 20 percent of monthly receipts of 50,000 to 100,000 francs 25 percent of monthly receipts over 100,000 francs Yet less than a year later, most of these small gains were lost when the Cartel des gauches government increased the taxes on all spectacles by 20 percent. Opposition to the tax began to fade, and the exhibitors accepted an overall tax rate of from 17 percent to 40 percent of gross receipts in Parisian cinemas and from 15 percent to 31 percent in the provinces. Throughout the 1920s, the cinemas contributed 50 percent of all spectacle tax revenues in France; by 1930, according to Andre Chevanne, their share had grown to over 300 mil- lion francs per year.3 Although the French film exhibitors were financially well-off, in spite of a heavy tax load, the scope and mode of their operations were quite restrictive for the production and distribution sectors of the industry. In 1918, according to Georges Sadoul, there were only 1,444 cinemas in France, nearly 200 of which were located in Paris.4 By 1920, their number had almost doubled to 2,400. Even so, the French exhibition market still ranked behind England (3,000), Germany (3,730), and the United States (18,000).5 By 1929, the number of French cinemas had grown to 4,200, but their position worldwide was the same as before.6 As if that were not bad enough for French distribu- tors, "less than 40 percent operated daily," reported Film Daily Yearbook (1929), "and only 900 had a capacity of 750 seats or more."7 Many of the rest in the provinces, adds Sadoul, "offered only three or four screenings a week."8 The insufficient number of cinemas, argued Andre Delpeuch, resulted from the French public's relative lack of appreciation for the new spectacle.9 And Andre 50 Chevanne's statistics seemed to confirm that: despite the stereotypical notion that France is a nation of cinephiles, only 12 percent of the French population regularly attended the cinema in 1928 (up from 7 percent in 1919), just half the percentage of regular customers in the United States. 10 Yet the French publicity strategies did little to entice that public into the cinemas to see French films. In 1919, Henri Bousquet writes, only a dozen or so of the eighty French films produced that year received special advertising campaigns. Path6, for instance, devoted an unusual effort to Gance's J'Accuse and especially to Pouctal's Travail, lavishing on the latter a full year of pub- licity before its first episode even premiered. Most others, including Feuillade's Tih-Minh, oddly enough, were advertised the week of their opening, and that was all. The situation improved slightly the following year, with the films of Baroncelli, Poirier, Le Somptier, Mercanton, Hervil, and Roussel singled out for special attention. In 1921, however, French publicity contracted to focus on less than a dozen films, including Feyder's L'Atlantide, Diamant-Berger's Les Trois Mousquetaires, Bernard-Deschamps' L'Agonie des aigles, and Leprince's L'Empereur des pauvres." Although much more needs to be known about French film advertising during this period, it seems consistently to have performed below the level of American advertising. This may account for the rapid rise of such men as Robert Hurel, Pierre Braunberger, and Bernard Natan (all associated with the Paramount publicity office); and it may explain why the specialized film journals-e.g., Cinia and Cina-Cin6-pour-tous-considered it so important to sponsor film poster contests. French film programming practices were determined in part by a system of film rental contracts which was standardized by the early 1920s. Cinema own- ers could rent a new feature film according to one of three basic contracts: Premiere vision confers on the lessee this privilege-that no other cinema owner can screen the film before him in a predetermined city or zone. But the premiere vision rental can be made simultaneously with many cinema owners. Priorit6 (or anteriorite) grants to the lessee the privilege of screening a film before all other competing cinemas in a predetermined city or zone, and of designating his choice of cinema locations. Exclusivit6 is understood as a booking strictly reserved to one cinema for a predeter- mined zone or period of time, along with the exclusive right of related publicity. 12 Although each year after 1921, more and more French films were released en exclusivite, only a limited number of cinemas in Paris and in the provinces (Lyon, Marseille, Bordeaux, Toulouse, Nice) could exhibit them.'3 In Paris, these were five of the most prestigious cinemas: the Marivaux, Madeleine, Aubert-Palace, Max-Linder, and Cam6o. 14 An exclusivit or priorit6 film usually played several weeks to a month in one cinema and then shifted to premiere vision status for its general run. The majority of films, however, were released under premiere vision contracts. As early as 1919, this meant a release pattern of anywhere from several weeks to several months. A film would begin playing one week at, say, several dozen major cinemas and then, if successful enough, week by week, would work its way through the smaller urban and rural cin- emas.15 Programming schedules for the larger cinemas were prepared from several months to a year in advance but usually could be juggled to accom- modate the exceptionally popular film-e.g., L'Atlantide (one year at the Ma- EXHIBITION 51 FRENCH FILM INDUSTRY deleine), Mathias Sandorf (six months at the Cirque-Hiver), Ben-Hur (over one year at the Madeleine). An exception to this practice ruined Renoir's Nana (1926). After a surprisingly good late summer run en exclusivit6 and en priorit6, it was withdrawn to make way for Aubert's fall season of films. 16 The serial and the film d epoques or multipart film provided a unique product for cinema programming. By 1918, Louis Feuillade had standardized the French serial format at twelve episodes of about a half-hour each. This meant that a cinema could run a single serial film title for nearly three months, screening one episode a week. In 1923, Jean Sapene modified this format when he ordered the Societ6 des Cineromans to produce serials of eight episodes each so that they could be released at two-month intervals, from September to May. Although Sapene's format dominated the rest of the decade, public opinion apparently forced some distributors eventually to market two versions of the same serial-a "condensed" version for the larger cinemas and a complete version for the smaller urban and provincial cinemas. 18 The unusually sus- tained popularity of the serial in France seems to have spawned a peculiar hybrid form, the film a epoques or multipart film. These films were made up of two or more parts or "acts" (lasting one or two hours each), none of which, taken separately, could be accepted as complete; and they were intended to be screened in single sances or performances over a period of consecutive weeks. Perhaps the earliest example of this format was Louis Feuillade's war film, Vendimiaire, whose two parts of 1,680 and 1,350 meters, respectively, were released on 17 and 24 January, 1919.19 Several other films were shown simi- larly within the next year-e.g., Gance's J'Accuse (four parts), G6rard Bour- geois's Christophe Colomb (two parts), Pouctal's Travail (eight parts)-and the format saw rather extensive use throughout the decade, generally for expensive historical or biographical films.20 Perhaps the most prominent films were Henri Diamant-Berger's Les Trois Mousquetaires and Vingt Ans apr6s (twelve parts each), Gance's La Roue and Napoleon (four parts each), Henri Fescourt's Les Misrables (four parts), and the same director's 1929 remake of Monte-Cristo (two parts). Apparently, each part would play just one week in a cinema until the whole film was recycled for another screening. Here, too, was a fairly successful strategy that assured the exhibition of French films in the face of the American film onslaught. Programming practices were probably determined as well by the location of cinemas and thus by the clientele they served. Unfortunately, there is no study yet of early French cinema locations to compare with the rigorous analyses of early American cinemas performed by Russell Merritt, Douglas Gomery, Rob- ert Allen, and Charlotte Herzog.21 However, Andre Antoine offers a place to begin. In 1921, Antoine did an informal survey of the cinemas in Paris ac- cording to the seven or eight quarters (and stereotypes) that comprised the city.22 The Left Bank cinemas then were frequented by students, teachers, lawyers, and civil servants who preferred "films marked by the qualities of dispassionate observation and truthfulness." The cinemas of Grenelle, on the other hand, catered to the bourgeois audience that enjoyed the serials and historical films, which swelled them with pride and respect. The elegant cin- emas of Passy, Auteil, and Etoile were becoming social playgrounds for the idle rich to carry on affairs and talk of affairs. The cinemas of the working- class quarters (Rpublique, Bastille, Pointe-Saint-Paul) were thronged by peo- 52 ple who, though passionately interested in the crime serials and in sentimental EXHIBITION melodramas, could also discern an exceptional film when they saw one. The new cinemas of the Boulevards were supplanting those around the Place de Clichy as "the center of cinema life," for they were crowded with shoppers from the nearby Printemps and Galeries Lafayette, day and night, weekdays and weekends.23 It was here, where cinema-going was becoming linked to the consumption patterns of the big department stores, that the industry could determine how successfully a film could attract the general public. Or whether, as Andr6 Delpeuch so blithely put it, a film could satisfy "this need that we sometimes feel to escape, if only for a brief flight, the conditions of our bour- geois existence."24 A typical French cinema program just after the war, according to Louis Delluc, looked like a "programme-salade."25 The recipe consisted of a half- dozen different kinds of films: a newsreel, a comic short, a serial episode, a feature-length drama or comedy, and either a travelogue, another serial epi- sode, or a second comic short. Here are some examples from Le Journal du Cin6-Club (30 April 1920): Gaumont-Palace La Fete espagnole (short drama) La Bretagne pittoresque (cinematographic study) Barrabas, 9th episode (serial) Charlot apprenti (comic short) Les Environs du Caire (documentary in color) Path6-Palace Pathl-Journal (newsreel) Houdini, 10th episode (serial) Le Secret d'Argeville (police drama) Fritzigli (comic short) Le Duel de Max (comic short) Folies-Dramatiques Dernieres actualits (newsreel) Barrabas, 9th episode (serial) Bigorno cireur (comic short) La Revelation (Rio Jim western) Les Chansons filmles (film songs) These were mixed in various combinations and presented en bloc, beginning at a certain hour of the afternoon or evening, with no indication of exactly when a specific film would appear or even if it would be repeated. The only thing one could be certain of was that the course would last from two to three hours, since the films' total length usually ran from 3,000 to 3,500 meters.26 The young Dadaist and Surrealist poets loved the chance encounters this system created, but people of more regular habits complained.27 Filmmakers and critics complained about this system as well, because ex- hibitors exercised a right to alter films to suit their needs. Feature films were already measured on the bed of Procrustes, according to Antoine, since dis- tributors often cut them to a standard length of 1,350 to 1,700 meters each, whether the story demanded it or not.28 Then the cinema owner could tailor any film to fit his particular program-cutting out scenes that might displease his clientele or shortening it in such a way that he could add the special attraction of an extra film. Even the "enlightened" manager of the Cameo and Artistic cinemas, Lucien Doublon, defended the exhibitor's need to heed the demands of his patrons. 53 FRENCH FILM INDUSTRY When the manager of a "permanent" cinema cuts a film, he knows what he is doing; he is eliminating useless things, repetitions in a scene, evocations, etc. But, I must add, he pays the same metrage rate to the distributor. There is not one single manager of a "permanent" who cannot lighten a program in such a way that the spectator doesn't notice it.29 Another irritating exhibitor practice, which was just as common in the United States, was to project films in the cinemas at faster than their "taking speed" (initially, 16 to 18 frames per second), presumably to squeeze in an extra daily program.30 This, along with other factors, tended to slowly increase the "tak- ing speed" of films during the decade. Yet the speed of French films, according to Barry Salt, still lagged behind most others-whereas American films, for instance, had reached a speed of 22 frames per second by 1924, the French films did not reach that speed until 1929.1 By the middle of the decade, programs were being organized around the exhibition of one feature-length film and a serial, with perhaps the addition of a newsreel and a short. Now they were standardized at from two to three hours in length and were presented regularly according to an announced sched- ule. The only change in this format was Paramount's introduction of double- bill programming (which the French also used sparingly) and the same com- pany's 1928 decision to adopt a policy of continuous screenings.32 To suggest the range of programs, the following examples are drawn from Cinmagazine (16 March 1928): Madeleine-Cinema Ben-Hur Omnia-Path6 Le Chauffeur de Mademoiselle Sevres L'Esclave blanche La Sirine des tropiques Poker d'as, 1st episode (serial) H6tel-de-ville Princesse Masha Gribouille veilleur de nuit (comic short) Imperial Paname n'est pas Paris Paris i y a vingt ans (documentary) In the provinces, the practice of changing programs twice a week was intro- duced; and, in Perpignan, at least, this included reruns of many films released earlier in the decade.33 Exhibitors everywhere, however, continued to alter films as they saw fit. Although most French cinemas were owned and operated independently, several large circuits controlled important segments of the French market. Their association with the major film producers and distributors went back to the industry's formative years before the war. Nearly fifty cinemas were oper- ated by Path6-Freres/Pathe-Cinema, through six regional circuits-five in France and Algeria, plus another in Belgium and Holland. Edmond Benoit-Levy had an administrative interest in four of them. The most important of the com- pany's early Paris cinemas were the Pathe-Palace (600 seats) and the Omnia- Pathe, "in the purest Louis XIV style" (1,000 seats), the latter of which probably initiated the exclusivit6 screening practice in 1916 with The Cheat. Gaumont controlled the next largest circuit of cinemas in Paris and the prov- inces. The flagship of the company was the Gaumont-Palace, formerly the Hippodrome Theater, on Place Clichy. Under the management of Edgar Cos- 54 til, the huge theater (nearly 6,000 seats) was transformed with a mammoth moveable screen that was surrounded by "an immense architectural frame in a Greco-Roman style." The orchestra area was remodeled to seat eighty musi- cians, and two great "Aeolian" organs were installed on each side. Opening in 1911, the Gaumont-Palace became the first French cinema to provide au- tomatic projection equipment. Louis Aubert's circuit was quite small at first, comprising just five cinemas in Paris. In contrast to the major companies, however, Aubert was the only one during the war period to construct new cinemas: the Electric-Palace (500 seats) and the Nouveaut~s-Aubert-Palace (800 seats), which was laid our in the shape of a drawing compass* Consequently, by the end of the war, despite these few exceptions, the French cinemas were in lamentably poor condition." All that began to change in 1919. While the production and distribution sectors of the film industry hesitated or reorganized in the face of American film imports, the exhibition sector embarked on a three-year binge of cinema construction (nearly doubling the number of cinemas in France). If anyone was going to make money, it would be cinema owners. They were encouraged in this boom by the nationwide adoption, in 1919, of an eight-hour workday and the 'English week" (five and a half days of work per week>." The undi- minished growth in cinema attendance fueled the industry's desire for more and better cinemas. Path6-Cin~ma/Path6-Consortium expanded its circuit in the outer districts of Paris and in the provinces, constructing the irst grandes 17. Omnia-Path6 interior z I 18. Gaumont-Palace fagade (left) and Madeleine Cinema faqade (right) salles in many regions. Aubert continued to expand its circuit with new cin- emas in Paris, Lyon, and Brussels. Soon nearly every Parisian quarter had a 1,500- to 2,000-seat cinema., However, what most fascinated the French film press was the marvelous apparition of a half-dozen or more luxurious "super-cinemas" in Paris. The center of the cinema district in Paris now was the department store and theater area of the Right Bank Boulevards, and most of the new super cinemas were constructed there from 1919 to 1921. The first and grandest of them all was the Salle Marivaux (1,050 seats), which opened in April, 1919, with Griffith's Intolerance.'' Built by the independent Compagnie generale fran aise de cinematographie circuit (whose founder, Benoit-Levy, linked it briefly with Vandal and Delac's grand scheme of a vertically integrated French film company), this new cinema was a peculiarly satisfying blend of sober lines, comfortable elegance, and ostentatious decor. Much of its atmosphere was "calculated on the play of light from three ranks of 5,000 lamps." But its decor was decidedly byzantine: Side by side with Ionian columns are vaguely Egyptian ornaments, all of which are illuminated in old rose and ochre, with green and gold motifs. One has the impression 56 . . .! A e h .. ' i [ .a "'" ljEf ' $1A7f* ^ rY ' r , s 1 iF b e r^,iY X 19. Salle Marivaux interior of swimming in pistachios.0 a gigantic ice cream mixture of strawberries and vanilla adorned with Related in style to the classical hardtop cinemas that were flourishing in the United States, with generous touches of the atmospheric, the Salle Marivaux seems an early example of what Dennis Sharp has called the Continental search for a distinctive and modern architecture in cinema design." Under Francis Aron's management, it quickly became the most prestigious of all the exclu- siviti cinemas in Paris. The Marivaux was soon joined by another luxury cin- ema, the Cine Max-Linder (900 seats), which had been a showcase for the French comic's own films and for the earliest American film imports during the war. In 1919, Linder himself spent two million francs to completely ren- ovate the cinema in order to exhibit a series of films that he planned to make in France.' Although the Cin6 Max-Linder reopened with the smash hit, Le Petit Cafe, Linder failed to get a second film project under way. Sold to an independent circuit, the cinema bearing his name soon became one of the few exclusivite cinemas in Paris. Another cinema that opened with a bang was the 850-seat Madeleine-Cinema (also financed by Benoit-Levy), which started L'Atlantide on its way to a worldwide success. 'l It, too, became an exclusivit6 57 FRENCH FILM INDUSTRY cinema. Finally, Louis Aubert kept expanding his Paris circuit with a half- dozen cinemas, including the rebuilt 2,000-seat Tivoli-Cinema. While this construction boom went on, several exhibitors and distributors experimented with special programming formats. For instance, the only cin- ema on the Champs Elys6es at the time, the Castillan-owned Colis6e, began to devote its programming regularly to German films, to re-releases, and to the work of filmmakers such as Delluc, L'Herbier, and their "circle." The Mogodor-Palace followed suit with a series of re-releases inaugurated by The Cheat. The 4,000-seat Trocadero theater had film projection equipment in- stalled for several gala presentations by Path6-Consortium-for Bernard-Des- champs's L'Agonie des aigles (April, 1921) and Diamant-Berger's superproduc- tion serial, Les Trois Mousquetaires (September, 1921).4 This screening of Les Trois Mousquetaires became the first highly publicized premiere in French cin- ema history. It was organized for the benefit of the men wounded and hand- icapped by the war, and its presiding hosts were five government ministers and the war hero, Marshal Foch. Path6-Consortium even consigned Jean Sa- pene (then advertising editor of Le Matin) to direct a massive publicity cam- paign for the film's premiere as well as for its general release in sixty cinemas two months later." One year later, on 6 November 1922, Path6-Consortium organized a similar special premiere for Diamant-Berger's Vingt Ans apres at Rolf de Mare's Theatre des Champs-Elys6es, where the Ballet sudois and Diaghilev's Ballets russes performed.46 At the same time, Universal held a spe- cial preview for Metro's The Four Horsemen of the Apocalypse at the Theatre de Vaudeville.47 From 1922 to 1925, the construction of new cinemas in France fell off a bit. Although construction continued in the provinces-e.g., Aubert-Palaces in Lille and Marseille-the mushrooming of big new cinemas in Paris ceased. Instead, the major circuits began to consolidate their holdings. In 1924, Au- bert bought the Mogodor theater and converted it into a cinema; Gaumont, shifting its concern to distribution and exhibition, bought out the Madeleine. Path6-Consortium formed an alliance with a new circuit based in Paris, Lu- tetia-Fournier. The highest-grossing cinemas in Paris, and in France, were nearly all controlled by one of the major chains: the Salle Marivaux (now 1,400 seats), Gaumont-Palace, Cameo (formerly Path6-Palace), Aubert-Palace, Max- Linder, Madeleine, Lutetia-Wagram, Omnia, Palais de Fetes (Brezillon), and Tivoli-Cinema. Finally, the industry reached a new stage of public acceptance when the management of the Paris Opera agreed, after several years of debate, to premiere special French films. The first film to be so honored was Bernard's historical epic, Le Miracle des loups, in November, 1924. The premiere became a national event when the president of the Republic presided over a select audience of ministers, diplomats, and celebrities in the arts and sciences from around the world. A year later, the Op6ra initiated a series of such premieres with Aubert's Salammb6.48 During the last four years of the decade, the stabilization of the French economy and the continued high risks of film production led to another round of super cinema construction, which paralleled a similar expansion in Germany and slightly preceded an even larger boom in England. Among the French circuits, Path6-Consortium opened the era with two new Paris cinemas in 58 1926-the Empire, where it now screened all of its new releases for the ex- hibitors and critics, and the Imp6rial, which soon became one of the top- grossing Paris houses. Its circuit was augmented further when Lutetia con- structed three Paris cinemas in 1927. Aubert kept pace by purchasing the prestigious Cameo and Artistic cinemas. In September, 1927, the Theatre des Champs-Elys6es became a combination concert stage, music hall, and cinema. A month later, finally, a new independent super cinema, the Rialto, was dubiously inaugurated with Jean Sapene's tribute to Claudia Victrix, La Prin- cesse Masha.49 All this activity was probably also spurred by Paramount's decision to in- vade the French exhibition market. In 1925, Adolphe Osso approved the purchase of the Th6atre de Vaudeville and its reconstruction as a super cinema. During the next two years, the American company either built or bought outright eight major cinemas in the French provinces. And it completed the Paramount-Palace (2,000 seats), located on the same spot as the old Th6tre de Vaudeville, just in time for the Christmas period of the 1927-1928 film season. The spacious art deco interior, the unusually comfortable main floor and loge seating, and the sophisticated underground electrical generating sys- tem-all forced the French to admit that the Paramount-Palace was the best of the super cinemas in Paris. Paramount also knew how to advance its own cause. It paid for a special supplement in the Cinmatographie franfaise (13 August 1927), and it regularly stunned the French distributors and exhibitors by spending up to one million francs to advertise a single film. By 1928, the Paramount-Palace accounted for 20 million francs or almost 10 percent of the total cinema receipts taken in Paris.50 The record grosses taken in by the exhibitors and the publicity given to this cinema construction boom helped provide the French film industry with sufficient capital, when the Americans and Germans forced the issue, to make the transition to sound films. As the decade began in a state of crisis for the French film industry, so did it end in a crisis of a slightly different sort. To the conditions that determined the industry's ambiguous position throughout the 1920s, another was added to create the crisis of 1928-1929: the failure to encourage industrial research that could produce a marketable technology for the sound film. The technology for sound film production and exhibition was developing very quickly by the early 1920s, particularly in the United States and Ger- many, but few perceived or pushed its commercial value.1 In France, Gaumont had experimented with various synchronized sound systems from 1900 to 1913, but none of them became marketable for lack of economic and technical fea- sibility. By the 1920s, the company was in retreat and projected its occasional short sound films only as novelties. This lack of commercial foresight and financing was all too characteristic of the French film industry. Of the basic technologies forming the cinema's base, the French bridged the gap between invention and industrial manufacture and marketing in only a single one with any kind of consistency-camera and projection equipment. In the case of electronic sound recording, they faced an added drawback-the slow devel- SILENCE IN THE FACE OF SOUND Silence in the Face of Sound 59 FRENCH FILM INDUSTRY opment of their electrical industry. Here the United States and Germany had a distinct advantage. For the incentive to convert to sound film production came from the powerful new electrical industries. By early 1928, ERPI (Electrical Research Products Incorporated-a subsid- iary of Western Electric) and RCA (an affiliate of General Electric) had ac- quired control of the American rights to sound film systems, signed contracts with the major film producers, and were beginning to push into the European market.2 Within six months, two large patent groups, Tobis (controlled by Dutch financiers) and Klangfilm (in the hands of the German electronics in- dustry) had emerged to challenge the Americans. The French were caught completely off guard in this confrontation. Gaumont was the only company to come up with an alternate sound film system, one developed by two Danish engineers working in France. Although technically advanced, the Gaumont- Petersen-Poulsen system proved commercially unfeasible, at least in its initial format. In March, 1929, Tobis and Klangfilm pooled their assets as a German corporation, and opened a year-long war with ERPI and RCA over control of the world market. When an agreement was finally reached in July, 1930, France officially became an open territory where the three corporations contin- ued to vie for control. Consequently, the French film industry was ill prepared for the sound film revolution. To survive, to make the conversion to sound film systems, it would have to depend on the monopoly of American or German technology. Cinia- Cin6-pour-tous reported that Western Electric was charging up to 600,000 francs to install its sound projection equipment in a single cinema.3 And that would be costly, doubly costly, since the money would not circulate back through the economy. Furthermore, the American boycott of film exports to France (from April to September, 1929) delayed the installation of sound systems in the cinemas.4 The industry also found itself strapped by other costs. In 1926, PathW sold its film-stock factory at Vincennes to Eastman-Kodak. Although the sale netted Path6-Cin6ma a tidy profit, the French film industry now had to depend on either Eastman-Kodak or Agfa film stock. Given such bleak material conditions, the French came through the transition period better than one might have thought. The sound film turned out to be an unexpected godsend-for it reintroduced the national barriers of spoken language. The subject of the new sound films came up in the French press as early as 1926. By 1927, film journals were printing reports on the various formats that were being tried out in the United States-Paramount's singing film shorts with Raquel Meller, Fox-Movietone newsreels (using the ERPI system), and Warner Brothers' projected series of "feature-length talkie productions." They were also debating the relative merits of films sonorisds and films parlants or talkies. The consensus was, according to Marcel Lapierre, that the sonorized film would eliminate poor orchestral accompaniment (or monstrosities like the "mechanical brouhaha" used for sound effects in the Paris screening of Ben- Hur) and, therefore, was preferable to the less familiar talkie.6 By 1928, more and more reports came in, saying that the new sound films, especially the talkies, were setting box office records in both the United States and England. While passing through Europe, Rene Clair recalls drolly, Jesse Lasky con- firmed the rumors.7 French producers, directors, and scriptwriters immedi- ately took off across the Channel to London to study the "savage invention" 60 firsthand. All complained about the quality of sound reproduction and the static nature of the films, but none could deny that the monster would be victorious. For, as one put it, "[if} silence is silver . . . the word is golden."8 It was not until late in 1928, two years after Warner Brothers introduced its Vitaphone sound-on-disc system and just as the entire American film in- dustry was deciding to convert to sound, that feature-length sound films began to appear in Paris.9 Drawing on its long experience and its new optical sound- on-film process (Petersen-Poulsen), Gaumont was able to present the first French sound film before any American films could be released in France. The film was Marcel Vandal's colonial adventure, L'Eau du Nil, projected at the Cameo on 19 October 1928. Although the sound track of L'Eau du Nil consisted only of music and songs, the French press realized a "revolution" was at hand. Cinimatographie frangaise even compared Gaumont's presentation to the first public projection of Lumiere's L'Arrivee du train and Sortie des usines Lumiere. A month later, at the Madeleine-Cinema, MGM screened an American sound film, Ombres blanches (White Shadows by Van Dyke and Flaherty), which used a Movietone optical sound accompaniment of Hawaiian guitar music. At the same time, Paramount released a newly sonorized version of Les Ailes (William Wellman's Wings). In early January, 1929, at the Cine Max-Linder, Path6- Consortium hesitantly joined the advance by releasing Marcel L'Herbier's L'Argent, accompanied at one point by phonograph recordings of Bourse crowd noises and an airplane roar. For a few heady weeks or months, it seemed as if the French film industry was on the cutting edge of the sound film revolution. But the euphoria proved illusory. As in the United States, the turning point came with the release of Warner Brothers' The Jazz Singer (1927), on 26 January 1929, at the Aubert-Palace. Although presented in English (with one scene of dialogue) and restricted by the Vitaphone sound-on-disc recording system, The Jazz Singer was a smash hit. It played at the Aubert-Palace for almost a full year and doubled the cinema's previous year's receipts. In March, at the Theatre de l'Apollo, Tobis- Klangfilm premiered the first German sound film, a popular music-hall film, La Revue Nelson, using its excellent optical sound process. At the same time, Paramount and Fox-Movietone began to exhibit their sound newsreels. In June, Paramount finally had its first genuine success (using Movietone optical sound) with Richard Wallace's A Song of Paris, starring the popular French singer, Maurice Chevalier. The demand reached the point where French film producers were forced to initiate sound film projects as a regular feature of their production schedule. In the process of changeover, the level of French film production fell steeply from ninety-four films in 1928 to fifty-two in 1929. Taking advantage of the five-month American boycott of film exports, however, the industry unloaded most of the last big French silent films (to limited competition) during the spring and summer months and prepared as best it could for the new fall season. 10 Because French studios were still equipped for silent film production only, the earliest French sound films were simply sonorized in France or made in England or Germany. The 1929 fall cinema season included a hodgepodge of French sound films. " In October, at the Cameo, Aubert screened its first sound feature, a period film by Gaston Ravel and Tony Lekain, Le Collier de la reine, whose lone sequence of dialogue--a comic finale between Marat and SILENCE IN THE FACE OF SOUND 61 FRENCH FILM INDUSTRY Robespierre-was postsynchronized, using the Tobis-Klangfilm process. At the Marivaux, Natan presented his first talking film, Andre Hugon's lackluster re-make of Les Trois Masques, which was recorded (with RCA optical sound) in England. The real hit of that program, however, was not Charles Mr6's creaky bourgeois melodrama, but Walt Disney's energetic cartoon, Mickey vir- tuose (Opry House. Pierre Braunberger contributed La Route est belle, which Robert Florey also recorded in England. Path6-Consortium weighed in with Marcel L'Herbier's Nuit de princes, which was shot in France and then sonorized by Sequana Films in Berlin. And Abel Gance was sent to London by the Societ6 de l'Ecran d'Art to shoot the sound sequences for his expensive new project, La Fin du monde. By the end of the year, French talkies were finally coming out of French studios.12 Natan led off a series of productions with L'Herbier's melodrama, L'Enfant de l'amour, and Pierre Colombier's boulevard comedy, Chique. Tobis- Klangfilm responded with a series headed by Henri Chomette's Le Requin and Ren6 Clair's Sous les toits de Paris. In December, Pathe-Journal began to appear as a sound newsreel. Within a year, the changeover was nearly complete: of the ninety-four French films produced in 1930, seventy-six were talkies. i3 And most of the others were silent films re-released with awkwardly synchronized sound tracks. Although silent film production in France did continue almost to the end of 1929, the transition seemed much sharper than it actually was because most of those films-e.g., Jean Epstein's Sa Tate (1930)-were vir- tually ignored in the French press. 4 The major corporations in the French film industry suffered a crisis of lead- ership just before the sound film conversion got under way. 15 Charles Pathe wanted to retire again from Pathe-Cinema and Path6-Consortium, where he still wielded some power. Jean Sapene was disillusioned after his personal intervention at Cineromans, especially by the failure to make his wife a star. When Louis Nalpas finally quit to organize his own production company once more, Cineromans seemed to lose momentum. In 1929, after a period of sporadic activity, L6on Gaumont arranged for Edgar Costil to take over the directorship of Gaumont. Louis Aubert was elected to the National Assembly in 1928 and soon began to seek a buyer for his company. After its disastrous international films, S.G.F. was acquired by Serge Sandberg in the summer of 1928. Although Abel Gance and Carl Dreyer were retained nominally as di- rectors, the only film S.G.F. produced (perhaps at Gance's insistence) was Jean Epstein's Finis Terrae (1929). Alexandre Kamenka abandoned Albatros's studio in Montreuil and closely allied his company with Sequana Films at Billan- court. With American and German interests poised to begin dividing up the French market, several spectacular upheavals seemed to reorder the French film industry structure completely. The principal impetus for this upheaval came from Bernard Natan, Robert Hurel, and Edgar Costil. The leadership crisis at Cineromans and Pathe-Con- sortium presented Natan with a speculator's dream, especially since his own company apparently was secretly in debt. In February, 1929, he bought out Path6's controlling interest in Path6-Cinema and had it simply absorb his debts. With the active support of the Banque Bauer et Marchal, plus that of the Banque Conti-Gancel, Natan soon had control of Pathe-Consortium as well. In August, he pulled into his new conglomerate the Lutetia chain of 62 cinemas. The giant consortium that emerged from this systematic takeover, Path6-Natan, was hailed, with some irony, "as the witness of a renaissance in the French cinema!" That same summer, Franco-Film picked up the option it held on Aubert, and Hurel renamed the combined companies Franco-Film- Aubert, Half a year later, in February, 1930, with the financial backing of the Credit Industriel et Commercial and the Swiss electrical industry, Gau- mont absorbed Franco-Film-Aubert. A second consortium was born, Gau- mont-Franco-Film-Aubert (G.F.F.A.), and Edgar Costil had himself ap- pointed its director. According to Henri Diamant-Berger, there was even an attempt to merge the two giants, but Hurel's financiers considered Natan a "primitive" (a racial slur?) unworthy of sitting at the same conference table with them. Even separate, however, the two looked formidable. As Jean Mitry remarked, the French film industry seemed to have reconstituted itself as the powerful force it once was before the war. 16 However, the resemblance proved ironic. In five years, through poor management and speculation, both would be bankrupt. By the end of 1929, the French film industry had been pared to a half- dozen companies with the resources and equipment necessary to begin full- scale sound film production. 17 Immediately after taking over Pathe-Consor- tium, Bernard Natan enlarged the studios at Joinville-le-pont from four to six stages, equipped each with RCA sound, and turned them into some of the best in Europe. Similarly, he also converted his studio at rue Francoeur to sound. At Paramount, Jesse Lasky and Walter Langer decided to concentrate their European operations in Paris and named Robert Kane to take over direc- tion of their Paris office from Adolphe Osso (who was soon to set up his own production firm). In a grand splurge, they acquired the old Reservoir studios in Saint-Maurice and built six new sound stages (using Western Electric equipment) that surpassed even those at Joinville. Tobis-Klangfilm set up an affiliate in Paris (under Georges Loureau), purchased the Eclair-Menchen stu- dios at Epinay, and converted them with its optical sound system. Jacques Haik built a new sound studio north of Paris at Courbevoie, while Franco- Film reequipped the Victorine studios near Nice. For its part, Gaumont con- verted its huge studio to sound. Finally, the independent producer Pierre Braunberger merged N6o-Film with a new production company set up by Roger Rich6b6, a regional film distributor in France. Films Braunberger-Ri- ch6be took over the former Abel Gance studios at Billancourt and converted them with Western Electric sound equipment. The other major companies- Albatros, S.G.F., S.F.H., Louis Nalpas-were assimilated, reorganized, or ceased production. The co-production "internationalism" that so marked the latter half of the 1920s in France carried over into the early practice of sound film production. 18 Following the pattern quickly established in the United States (by MGM) as well as in England and Germany, nearly all the French studios produced not one but several versions of each film, in the languages of their principal mar- kets: French, German, English, and sometimes Spanish and Italian. Occasion- ally, translated dialogue was simply postsynchronized, but the standard prac- tice at first was to hire a separate cast and director for each different version of the film. Marcel L'Herbier's La Femme d'une nuit (1931) was an extreme case: in French, it was a psychological drama-in Italian, a light comedy-in SILENCE IN THE FACE OF SOUND 63 FRENCH FILM INDUSTRY German, an operetta! But Paramount, under Robert Kane, was the center, the fabulous Babel, of multiple-version film production. Beginning in the spring of 1930, Paramount's Joinville studio became a veritable "League of Nations" of directors, writers, actors, and technicians (including such French filmmakers as Louis Mercanton, Julien Duvivier, Alberto Cavalcanti, and Claude Autant-Lara), sometimes operating on a twenty-four-hour factory schedule, churning out some 100 features and 50 shorts in up to fourteen languages. This strategy was hopelessly costly, compared to the practice of dubbing and subtitling, initiated by MGM again, that eventually became standard. Con- sequently, after a loss of three million dollars over less than two years, Para- mount converted its Joinville studio into Hollywood's dubbing center for Eu- rope. Although the practice of making several complete versions of individual films disappeared, the American and German production/distribution compa- nies retained a strong presence within the French film industry. A similar form of dependency, attrition, and consolidation marked the ex- hibition sector.' Throughout 1929, French cinema owners experimented with various kinds of sound film systems (Melovox, Synchronista, Idal-Sonore), and those that chose sound-on-disc systems often projected silent films accom- panied by their own choice of phonographs, without regard for synchroniza- tion. However, most soon followed the Paramount, Madeleine, Cameo, and other big cinemas in accepting the standardization provided by Western Elec- tric, RCA, and, to a lesser extent, Tobis-Klangfilm. Although the major cinemas in Paris and the larger provincial cities could afford to make the costly changeover to sound (and were forced to do so quickly), many of the smaller cinemas could not. By 1932, sound projection equipment had been installed in 95 percent of the Paris cinemas but in less than 50 percent of French cinemas nationwide. The number of small provincial cinemas declined as a result, and the larger circuits strengthened their hold on the exhibition mar- ket. The consortium of Path6-Natan controlled a total of 166 cinemas (69 in Paris, 35 in the outer districts of the city, 91 in the provinces, and 2 in Brussels). G.F.F.A. controlled 46 cinemas (20 in Paris and 26 in the prov- inces). Together these two concerns controlled the best cinemas in the country. "By the end of 1931," according to the figures of Andre Chevanne, "France had nineteen different circuits, whose 630 cinemas comprised [the most im- portant] 15 percent of the total number of cinemas. "20 Jacques Feyder, Henri Diamant-Berger, Henri Fescourt, Henry Roussel, Augusto Genina, and others within the French film industry saw the sound film revolution as an opportunity, a source of renewal.21 Despite the presence of Paramount and Tobis-Klangfilm as major French producer-distributors and the modernization of most French cinemas with RCA and Western Electric synchronized sound systems, the triumph of the talkies gave the French one distinct advantage. They could make films in French and about French culture more easily and more effectively than anyone else. And that was what the French public was beginning to demand. When Fox Follies 1929 was shown in English at the Moulin-Rouge-Cinema, in December, 1929, for instance, the near riot that ensued was a clear sign of that demand.22 So was the high profit margin (five to six times production costs) on most of the first French sound films-e.g., Roussel's La Nuit est d nous (Tobis-de Venloo), Robert Florey's La Route est belle (Braunberger-Richebe), Pierre Colombier's Le Roi de 64 resquilleurs (Path6-Natan), Julien Duvivier's David Golder (G.F.F.A.), and Marcel Pagnol's Marius, directed by Alexander Korda (Paramount).23 The French cin- ema screens could be theirs again, for the first time since 1914. And they would be regained, not by any new stratagem of the French film industry, but by default. Ironically, by promulgating the sound film revolution, the American film industry divested itself of some of the French market. Further- more, the French were helped by the relatively slow reaction of the country's economy to the Depression that staggered the United States and Germany. To the optimists, therefore, the French film industry seemed to have reached a position analogous to that of the American film industry in'1921-consoli- dation into several large corporations that could monopolize the horizon of their concern. There were differences, of course-less vertical integration, smaller markets, fewer capital resources, the presence of foreign producers and distrib- utors, an approaching Depression. But the French industry appeared to be on the verge of realizing some of the hopes it had first articulated at the end of the Great War. It seemed closer to resolving its persistent state of crisis. Others in the industry, particularly the formerly independent filmmakers, were pessimistic. After seeing The Jazz Singer, Marcel L'Herbier confessed, "When the cinematographe said to us all of a sudden: 'Look here, I, too, am going to talk,' I said to myself, 'So the catastrophe is coming.' "24 Gance, Dulac, Clair, and L'Herbier, at one time or another, all thought it was "the end of the world." And they had cause for alarm, not about the new technol- ogy itself, but, as Clair put it, about "the deplorable use our industrialists will not fail to make of it."25 "We are attending a death or a birth," concluded Alexandre Arnoux26-but which would it be? As it turned out, the sound film did not really break with the silent film; the "perfection" of the latter was not entirely lost. But many important people in the industry and a whole way of filmmaking were adversely affected. The men who had shaped the industry for fifteen years or more were displaced- Pathe, Aubert, Sapene, Nalpas, and the unheralded Russian emigr6s (Ka- menka, Bloch, Grinieff). For every filmmaker who was able, financially and technically, to carry on and improve his or her work, there were far more whose work was abruptly cut off, constrained, or deflected. The reason for this was simple: the higher costs of film production and increased standardization. The kind of small- or medium-budget project financed by independent pro- ducers quickly became impossible. As a consequence, the entrepreneurial in- dependence so characteristic of the 1920s abated, while the debilitating inter- est in speculation increased all the more. Ironically, as if still slavishly imitating the American cinema, a uniformity of style and social vision spread throughout the industry. As Bernard Eisenschitz has written, The talkie, by imposing a clumsier, more expensive infrastructure, by putting in question the narrative modes of silent film, cut short their elan-for it represented a knockout blow to all the avant-gardists and their advanced practices. . . .27 Only in the middle and late 1950s would the industry return to the rather diversified, artisanal conditions of the 1920s. And the New Wave would arise out of conditions similar to those of the First Wave.28 SILENCE IN THE FACE OF SOUND 65  The Commercial Narrative Film I It No, the cinema is not an art. . . . It is chemistry, physics, optics, mechanics, industry. And, above all, it is commerce.-Jacques de Baroncelli (1923) Neither an art nor a business, the cinema is a craft. -Alexandre Arnoux (1945) What is the best way to come to grips with the hundreds of commercial Introduction French narrative films, from 1915 to 1929? That question depends on another, prior question: What, to the French, exactly was a commercial film? The one anticipates the answer to the other. To be commercial, paradoxically, a film need not have been profitable. If profitability were the issue, then Rene Clair's Paris qui dort (1924) would be commercial whereas Aubert's extravaganza, Salammb6 (1925) would not. Similarly, the means of production financing hardly provided a clear-cut criterion. Many of the films one could include as part of a narrative avant-garde, for instance, were actually initiated or under- written as commercial projects by the major production companies-e.g., Abel Gance's La Roue (1922) and Napoleon (1927), Marcel L'Herbier's El Dorado (1921) and L'Argent (1929), Germaine Dulac's La Souriante Madame Beudet (1923), Jean Epstein's Coeurfidele (1923), and even Carl Dreyer's La Passion de Jeanne d'Arc (1928). Several filmmakers from the period offer a way to resolve these contradic- tions. "In cinema jargon," quipped Jean Renoir, "a commercial film is not one which necessarily makes money, but one which is conceived and executed according to the principles of businessmen."' But what were those principles? In one of the earliest books on French film, Le Cinema (1919), Henri Diamant- Berger argued that, first, the producer, not the filmmaker, made the crucial decisions that determined a film's production.2 Henri Fescourt, one of Cine- romans' most consistently successful directors, agreed: "With few exceptions, it was not the filmmaker who chose the subject. In most cases, that was the prerogative of the producer....Second, the essential element of a film, for the producer, was the scenario or story. In fact, said Diamant-Berger, "it was the film."4 And what kinds of stories did he tend to choose? Here is Fescourt again: "The story most crammed with events or spectacular elements-a story with a recognizable title, which had already garnered accolades in the theater or in the bookstalls."5 Diamant-Berger was more precise. Even if "roughly [and childishly defined," such stories were formulas or genres that could be duplicated for mass circulation, exactly as were the products in "a grocery or lingerie shop."6 As a site of pleasure, therefore, the French cinema seemed to attract and hold its audience, not with the images of popular stars (which was much more common in the United States), but with the familiarity or noto- riety of a film's subject or story.7 These statements suggest that the best way to analyze the commercial French narrative films of the period is to consider them within the framework of a loosely defined genre system in French film production.8 Such an approach offers more than the convenience of labeling; it imposes certain questions. What film genres dominated the French cinema? What were the material conditions of each genre's development? What were the privileged conventions of its form-its fundamental structural components of character, setting, ac- tion, style? What was the social or ideological function of its tacit contract with the audience? Which specific films were most characteristic of a genre, which were most popular and profitable, and which were especially effective either in resolving a genre's representation of cultural conflicts or in changing its conventions? Finally, which genres in particular provided the context of accepted conventions within which and against which the narrative avant- garde tended to position itself? 69 COMMERCIAL NARRATIVE FILM The distribution of film genres within the French film industry seems to have had its origins in the kinds of films produced on a regular basis prior to the Great War, when the cinema was establishing itself as a form of spectacle entertainment. According to Georges Sadoul, by 1908 Pathe-Freres had "de- fined, classified, organized, and developed" a catalogue of film forms that would consistently attract a mass clientele. The most important of these were "the newsreels, 'dramatic and realist scenes,' 'Biblical stories,' and . . . comic films with their 'wild and woolly' chases. "9 Several others (associated more with Georges Melies) had already declined in popularity and nearly disap- peared-the trick film and the fairy tale or fantasy.10 During this period, two of Pathe's products underwent significant changes. The comic film, of course, came to focus on the misadventures of a single clown or comic type--Andre Deed, Prince Rigadin, Max Linder. The "dramatic and realist scene," on the other hand, became a label for several different kinds of films that reflected divisions within French society. There were "social" films about the working classes, "sentimental dramas" about the bourgeoisie, and crime films ("dramas of passion") involving either class as well as bands of outsiders." These latter films provided the basis for the rapid emergence of the series film and the serial. About the same time, one last format was tested successfully-the Film d'Art and S.C.A.G.L. adaptations of literary classics, especially those from the theater. During the war, in the face of competition from the American cinema, which threatened the viability of much of their production, the French aban- doned the comic short almost altogether-to Sennett, Chaplin, Arbuckle, Lloyd, and Keaton-and concentrated most of their production efforts on the serial and particular kinds of feature-length films. The popularity of the serial quickly led to a standardized format of half-hour episodes projected weekly over a period of several months, a format that could handle stories of crime, senti- mental melodrama, and historical adventure and intrigue. In contrast to the serial, the term grand fllm designated a film of at least an hour in length, designed to be shown complete in a single sance or performance. By 1920, the average French feature was anywhere from four to six reels or 1,200 to 1,800 meters long."12 Since the format encompassed a variety of genres that shifted in priority as the decade advanced, their development demands more specific attention. Prior to the war, the first feature films had been literary adaptations, many of them drawn from theatrical productions. However, by the war's end, the plays that now were turned into film scenarios tended to be modern or con- temporary-principally boulevard melodramas-typifying a shift away from the classic dramas, which had already been pretty well rummaged through. Simultaneously, a further shift was underway to film adaptations of novels and short stories, especially nineteenth-century realist fiction. During the early years of the twenties, therefore, the bourgeois melodrama and the realist film, followed by minor genres such as the fantasy, the Arabian Nights adventure story, and the boulevard comedy, had precedence. These were superseded rather quickly, however, by larger and longer superproductions. After the phenom- enal success of Les Trois Mousquetaires (1921-1922), Les Opprimds (1923), and Koenigsmark (1923), the French turned more and more to large-scale historical reconstructions, which soon became the "top line" of production. As the na- 70 tion rebuilt the northern provinces devastated by the war, so too did the film SERIALS industry reimagine and thus give representation to ideologically significant figures and events in certain periods of French history. This genre was chal- lenged in turn, around 1926-1927, by the development of studio spectaculars in a generalized modern European decor. These films of contemporary life played out in the new chic restaurants, nightclubs, theaters, and resorts sought to position the French within an emerging worldwide consumer society. When the industry finally changed over to talking films, in 1929-1930, the prefer- ence for adaptations from fiction ebbed once again in favor of contemporary theatrical productions. This brief overview of French film genres provides a point of departure. The classification schema I offer is obviously tentative, in the absence of much prior research; and it is perhaps less than consistent-but then my purpose does not include a theoretical consideration of genre. As Henri Diamant-Ber- ger confessed, "most of the really successful films prove the instability of any rigid classification."13 Still, the schema can be valuable if it helps to mark out areas of intertextual relations among disparate films and to suggest ways that the films may be read contextually within the structures and processes of French culture. In one sense, the serial was simply a transformation or retooling of the Serials prewar series film. Initially, there were several kinds of series films-some concocted in imitation of story serializations in the popular magazines, others drawn from imported dime novels and their French equivalents. They included westerns (Riffle Bill, Arizona Bill) and burlesque (Onsime, Bout-de-zan), but the most lasting was the police or criminal series. 1 Its period of formation was 1908 to 1914, from Victorin Jasset's Nick Winter, Le Roi des detectives (Eclair, 1908)-which soon had imitators in Britain, Denmark, and the United States- to Jasset's Zigomar (Eclair, 1911), Zigomar contre Nick Winter (Eclair, 1912), and Zigomar, peau d'anguille (Eclair, 1913); Georges Denola's Rocambole (Path6- Freres, 1913); and Louis Feuillade's Fant6mas (Gaumount, 1913-1914).2 The criminal heroes of these films were especially popular, which made Arquilliere, as Zigomar, and Ren6 Navarre, as Fant6mas, early French film stars. It would be imprecise to call the series film simply a serial in another guise. Each film in the series format was complete in itself, and the series was re- leased, more often than not, in an irregular pattern. For example, Fantomas was really five separate films involving the same major characters (Fant6mas, Inspector Juve, the journalist Fandor), and they were released over a one-year period at intervals of two to four months.3 There were exceptions to this format-the most important probably was Leonce Perret's L'Enfant de Paris (Gaumont, 1913)4-but nothing to anticipate the abrupt change that occurred late in 1915 with the conjunction of two famous films. First came Feuillade's Les Vampires (starring Musidora and Edouard Math6), which opened at the Gaumont-Palace, 13 November 1915. From then until 30 June 1916, its ten semi-independent parts were released anywhere from three weeks to two months apart.5 Although quite popular, its success was quickly superseded by Path6-Freres' Les Mysteres de New-York (starring Pearl 71 COMMERCIAL NARRATIVE FILM 20. An early poster for Fantmas (1913-1914) White), which opened three weeks later, on 4 December 1915. This French- American import caused a sensation not unlike that caused by Eugene Sue's novel, Les Mysteres de Paris (1842), from which it slyly drew its title.6 To distribute the film, Path6 compiled episodes from the first three Pearl White serials which its New York counterpart, Pathe-Exchange, in partnership with William Randolph Hearst, had made as part of a popular new trend in Amer- ican films.' Following the American format, Les Mysteres de New York was made up of linked episodes, twenty-two in number, released in a continuous weekly series until 29 April 1916. Furthermore, its story was transposed by Pierre Decourcelle and printed daily, weeks in advance, in one of Paris's largest newspapers, Le Alatin, as well as in several provincial papers.8 La Renaissance du livre then published Decourcelle's novelization, episode by episode, in a series of "volumes" that together could constitute a book-the first of their Collection des Romans-Cinemas.9 From that moment on, the French adopted the practice with enthusiasm, and the series film became the serial, the cini-roman, or the "film in episodes." For nearly four years, the epidemic of serials developed according to the parameters established by FantOnas, Les Vampires. and Les Mysteres de New-York. These included a contemporary milieu with many shifts in setting (some ex- 7? otic); a series of ingenious crimes, searches, entrapments, and escapes; and a SERIALS clutch of characters distinguished by their fascination with (or revulsion from) evil (sometimes defined sexually) and by their position in a social class or ethnic group. Nearly every French producer turned out several such films on crime and/or detection annually, and distributors imported even more from the United States. In 1916, for instance, Abel Gance made several thrillers (Les Gaz mortels, Barberousse) for Film d'Art in order to take advantage of the genre's popularity. The following year, also for Film d'Art, Henri Pouctal filmed Arthur Bernede's patriotic diatribe, Chantecoq, as a serial (published in Le Petit Parisien); and Germaine Dulac directed Ames de fous (starring Eve Francis) in six episodes.10 But there were some differences among all these films, differences which were most apparent in the products of the two major companies, Gaumont and Path6-Freres. Most of the serials made by Pathe-Exchange and imported into France had women as their central characters. The major heroines were Pearl White-Les Exploits d'Elaine (The Perils of Pauline}, Le Masque aux dents blanches (The Iron Claw}, Le Courrier de Washington (Pearl of the Army}, La Reine s'ennuie (The Fatal Ring}-and her chief rival Ruth Roland-Who Pays?, The Red Circle." Usually the heroine was the victim or threatened victim of an outlandish scheme-the assassination of her father, her own murder for inheritance money, corruption and criminal implication by her fiance-but she was also the one who tracked down or outwitted the villains and defeated them. Pearl White especially was a resourceful, agile, intrepid adventuress, willing to take on a gang of toughs as easily as a cream tart. In her optimistic blend of naivete and natural strength (in this she resembled another favorite actress of the French, Mary Pickford), she may have seemed to the war-pressed French pop- ulace a kind of modern-day Jeanne d'Arc (her villains often were German actors or had German names). Francis Lacassin suggests another, darker reason for the success of her serials and of the others imported by Path-Freres. It was less her triumphs than her persecutions that drew the crowds-persecu- tions so often inflicted in subterranean worlds that they reminded Lacassin of the Marquis de Sade and his notorious chateaux. Was she a kind of Justine who "fell into the arms of vice all the while believing that she was protecting herself?"12 It is difficult to decide whether the heroine of these serials is a "new woman," a model of emulation, or simply a reconstituted image of the "clas- sical" figure of violation. The serials that Louis Feuillade made for Gaumont during this period shared a kind of schizoid interest in women characters, but the scandalous success they acquired came initially from their fascination with crime. The black-clad criminals of Fant~mas and Les Vampires are almost magically adept at manip- ulating, to their own advantage, the spaces in and around Paris. In Les Vam- pires, they can commandeer automobiles, scramble about moving trains, dis- appear around corners and down manholes, and creep unseen over the rooftops of the city. They can anesthetize a whole party of wealthy Parisians and rob them of their jewels or pluck a voman from an apartment window and drop her into a car waiting below. In the long-take, deep-space, long shots that characterize these films, Feuillade and his cameraman, Guerin, come close to celebrating a revolutionary underworld force against society. At the time, some people were disturbed particularly by the close correlation between Fan- 73 COMMERCIAL NARRATIVE', FILM _ 21. A publicity photo of the ballet sequence in the first episode of Les Vampires (1915- p t6mas or the Vampire gangs' exploits and the activities of the anarchist Bonnot band that terrorized Paris in 1912."' Yet, as Richard Roud has observed, in Les Vampires, "the battle is not only political and social, but also sexual."' The Vampire gang goes through several leaders during its twelve episodes-the Grand Vampire, Satanas, and Vene- nos-but the original French screen femme fatale-Musidora as Irma Vep- outlasts them all. In episode after episode, this sorceress appears and vanishes in a different disguise: as a demure family maid, a young male secretary, a laboratory assistant, a wealthy widow, and a street tough. Gradually, the reporter hero, Guerande (Edouard Math6), becomes infatuated with her, es- pecially when she helps kidnap his own wife. But Musidora's end already is figured in the opening episode when a black bat-winged dancer (Stacia Na- pierkowska) hovers threateningly over a white-gowned woman, only to col- lapse and die of poison. Sure enough, in the final moments of the film, in another ballet fantastique, the kidnapped wife guns down Musidora in a rescue attempt, and Guerande lingers longingly over her dead body. All this is conspicuously absent from Feuillade's most popular serial, Judex (1917), and its successor, La Nouvelle Mission deJudex (1918). Perhaps fright- ened by the temporary banning of several episodes from Les Vampires, Gaumont and Feuillade deliberately gave Judex an "uplifting moral tone." Although Musidora was revived as the villainess, Diana Monti, the real star now was the detective hero (Rene Crest ), whose status was epitomized in the transfor- 74 SERIALS mation of the Vampire gang's sinister black costumes into Judex's black cape. "That majestic cape that he threw over his shoulder in such a noble gesture . . .asfudex," wrote Bardeche and Brasillach, "the rest was of little impor- tance.' " Louis Delluc, despite his admiration for Feuillade's technical skill, was considerably less charitable: "3Jidex [andi La Nouvelle Mission dejudex . . . are crimes certainly more serious than those which have been condemned as traitorous by the Military Tribunal." 2 The only relief came in Marcel Le- vesque's antics as Judex's confidant, Cocantin. It is worth pausing a moment on Feuillade's crime serials. Although they were less popular in their time than was the police serial, Judex (which did not protect them, nonetheless, from criticism by the young film critics and cinastes around Delluc), the crime serials have had their champions from the Surrealists in the 1920s to more recent figures as diverse as Alain Resnais and Annetre Michelson. The Surrealists were struck by Feuillade's vision of "a defamiliarized reality." The conjunction of fantastic acts in recognizably real spaces induced, for them, a peculiar state of disorientation, evoking the yar- velous. "There is nothing more realistic and, at the same time, more poetic than the serial," wrote Breton and Aragon, "in Les Mystres de New-York and Les Vampires, one discovers a real sense of our century."" For Resnais, con- versely, "Feuillade's cinema is very close to dreamsnand therefore ... per- haps the most realistic kind of all." '' For Michelson, according to Roud, his films reveal the architectural structure of bourgeois Paris as everywhere dan- 75 COMMERCIAL NARRATIVE FILM 23. Suzanne Grandais in L'Essor (1920) gerous, threatened with being undermined or subverted. 19 I myself wonder if, in their conjunction of the real and the unreal, the banal and the unexpectedly terrifying, the films also convey, through displacement, the French experience of the war-the absurd proximity of normal life to the ghastly horrors of trench warfare.20 From the beginning, the serial was an easy target for parody. In fact, Jacques Feyder did his filmmaking apprenticeship on one of the earliest and best of these parodies, Le Pied qui etreint (1916).-1 In this four-episode film, the gang's signal of recognition is a raised wriggling foot (ridiculing episode one of Les Mysteres de New-York: "The Clutching Hand"), and the gang boss is carted about in a baby carriage. In the fourth episode, he even imitates Chaplin in a robbery scheme. Finally, the hero and his boy sidekick watch themselves pur- sue the gang on a cinema screen, followed by their models-Marcel Levesque, Suzanne Grandais, Musidora, and Edouard Mathe. Such parodies, however, only seemed to whet the public appetite for the genre. From 1919 to 1922, the serial's continued popularity began to form a bulwark against the onslaught of American films that threatened to displace French films from the cinemas. In the process, the formula underwent some changes. Perhaps these changes came about because, in the postwar society, a central commission now licensed and thus exercised some control over all commercial films. Perhaps they paralleled, through imitation or example, the development of subjects in feature films. Whatever the cause, although the format had become standardized at a dozen episodes, each from 600 to 900 meters in length, the parameters of the serial shifted and were redistributed in several different modes.2 As a backdrop to these changes, plenty of films still held to the familiar formula of the war period. From its New York branch office, Path -Cinema and then Path -Consortium continued to import serials starring Pearl White, Ruth Roland, Marguerite Courtot, Irene Castle, and even Harry Houdini. Path6-Consortium also distributed film adaptations of popular novels by Pierre SERIALS 24. The Vampire gang leaders in Tih-AlInh (19 19) Decourcelle and Jules Mary: Pouctal's Gigolette (1921), Burguet's Baillonne (1922), Etievant's La Pocharde (1921) and La Fille sauvage (1922). " Through his own personal company, Societe des Cinfromans, Ren6 Navarre starred in a half-dozen films of espionage, such as Edouard Violet's La Nouwelle Aurore (1919), in which he tried vainly to recapture his old fame as Fant6mas.21 For Phoc a, the small Marseille production company, Suzanne Grandais appeared in several serials, regaining her earlier status as France's first female star." She was the only French film actress who could rival Pearl White. Even Louis Delluc admired her: "Oh, how charming she is. Not all of her films may be good, but she is always good in them, that is to say, smiling and amusing."27 In 1920, her career was cut short when, working on a film by Charles Burguet in Alsace, she was killed in an auto accident. The most promising of these films, however, was Alexandre Volkoff's La Maison du mystere (1922-1923), produced by Films Ermolieff. Its chief surprise, according to Jean Mitry, was the multiple disguises of Ivan Mosjoukine. His "erudite, concise, synthetic" acting "contrasted sharply with the exaggeration of most French actors" and made critics compare him favorably with Sessue Hayakawa.28 A real break in the genre, interestingly enough, came in the work of Louis Feuillade. His first two serials after the war, made in and around the Nice studio he had taken over for Gaumont, surprisingly returned to the criminal world. In Tih-Minh (1919), a French explorer (Ren6 Crest6 as Jacques d'Athys), his servant (Biscot as Placide), and an English diplomat (Edouard Math6 as Sir Francis Gray) are in search of a secret document that will lead them to an immense war treasure. Their antagonists include a Doctor Gilson (actually the German agent, Marx, played by Gaston Michel), a Hindu fakir, and the res- urrected Vampire gang, who have taken over a villa named Circe along with its "living dead." Jacques falls in love with an Indochinese princess, Tih-Minh (Mary Harald), a young woman (and orphan) who has taken refuge with his family; and the villains kidnap her (several times) in hopes she can direct them to the treasure. " Although Francis Lacassin has described the world of Tih- 77 COMMERCIAL NARRATIVE FILM 25. A kidnapping on the grounds of the estate in Barrabas (1920) - Alinh as a kind of tourist's nightmare of exotic locales," the schematic narra- tive seems most striking for the way it replays the nationalist and imperialist conflicts of the Great War. In Barrabas (1920), the criminal gang is even more international in scope, operating behind the cover of an important banking corporation. Rudolph Strelitz (Gaston Michel) is the head of this double organization whose main bank leads directly to a health clinic, a luxury hotel, and a simple guesthouse on the grounds of a lost chateau-which is really a fantastic universe ruled over by the mysterious figure of Barrabas. The intrigue develops slowly from the guillotining of an innocent man to the revelation of Barrabas as an im- poster and to his condemnation. For Lacassin, the title suggests the continu- ing, ineradicable existence of evil, for it is "the password that allows the gang members to move back and forth from the world of appearances to the parallel nightmare world." Could Barrabas bear comparison with Fritz Lang's Dr. Mabuse. d/er Spieler (1922)? In 1920, either because Feuillade's last two serials declined in popularity or because the Commission du Contrle des Films was censoring film violence, Gaumont decided that the public no longer wanted to see crime serials, no matter how fantastic, a heladcl n t along with him.oe Where previously he had shown either a fascination for the criminals or a moralizing interest in the "dispenser of justice," now, in three straight serials Les Deux Gamines (1921), L'Ophe/ine (1921) and Parisette (1922) Feuillade redirected his at- 78 SERIALS U 26. Rend Clair and Sandra Milowanoff in Parisette (1922) tention to the victims, all young women. These victims, however, were a far cry from the robust, combative characters of Pearl White or Suzanne Grandais. As embodied in the Russian emigre actress (and former St. Petersburg balle- rina), Sandra Milowanoff, they were closer to the exquisite purity and fragile innocence of a Lillian Gish.'3 To match this usually orphaned ingenue heroine (doubled in Les Deux Gamines), the figure of the detective gave way to a "sentimental hero" (one played by Rene Clair). j' The narrative became less an accumulation of bizarre and baffling intrigues and more a character-oriented, dramatic structuring of separation, adventure/misfortune, and reunion. The former exotic or urban settings were replaced by family dwellings, villages, and convents in the provinces. Older people and children became prominent. And in the end, the long-suffering characters realized their undying hopes in marriage. "A simple shift in the poles of attraction," writes Lacassin, "was sufficient to transform a powerful magnetic field into a pool of tears."" In their frankly romantic and sentimental appeal, Feuillade's serials had become lengthy bourgeois melodramas. Although the few film critics and historians who nave seen these melodra- matic serials find them much less interesting than Les Vampires, Tih-Minh, and Barrabas, the French public of the early 1920s, most film reviewers, and Feuillade's own friends were delighted. The three films coincided with a wave of sentimental popular novels that held French readers enthralled just after the war-Les Deux Gamines even drew its title from two of them, D'Ennery's Les Deux Orphelines and Pierre Decourcelle's Les Deux Gosses.36 Often these novels were stories of orphans, and Feuillade's new serials were among the first films- along with Pouctal's Travail (1919), Bernard's Le Petit Cafe (1919), Baroncel- li's Le Secret du Lone Star (1920), L'Herbier's El Dorado (1921), and Poirier's Jocelyn (1922)-to establish as a dominant subject in the French cinema of the 1920s the situation of the abandoned child or of the lone adult faced with abandoning or adopting a child/orphan." The collective French interest in, even obsession with, this subject may have a simple historical basis-the ter- rible loss of men in the war as well as the decline in marriages and births during and after the war, both of which perpetuated a situation that was well 79 COMMERCIAL NARRATIVE FILM 27. Leon Mathot (left) in Le Comte de Alonte-Cristo (19 17- 19 18) established by 1900.8 To the film industry, however, Feuillade's postwar serials also served as models in their desperate attempts to set off the "French- ness," the cultural superiority or sensibility, of French commercial films from the American competition. In the words of Aladin, Ah! how ridiculous and infantile the American serials seem when we attend the pre- view of a serial such as Les Deux Ganines. . . . Here, in a word, is no silly story, but one governed by rational imagination, by logical dramatic development, rigorous, well-designed, and magisterially executed.' That concern for Frenchness was equally apparent in the films resulting from a second shift in the parameters of the serial. This change probably had its origins in Henri Pouctal's 1918 version of Alexandre Dumas's Le Comte de Monte-Cristo (starring Leon Mathot in his most popular role), produced by Louis Nalpas at Film d'Art and released in eight episodes.-0 Till then, the serial had been restricted to current cinero,;ans and to the subject of criminal activity in the modern, mostly urban world. Suddenly it expanded to include "a pseudo-historical subject, full of costume adventures embroidered with a facile fabulism."' "Monte-Cristo is very good," wrote the usually reserved Louis Delluc, "very well conceived in its fabulous action and its dramatic interest. I have never before seen such fine understanding of just what a popular film should be like. "-' Georges Sadoul was also complimentary: "The story is sim- ple, clear, direct. The editing is concise and intelligent, the lighting advanced [the cameraman was Guerinj but without an excessive precision. A lovely sense of natural landscapes gives a poetical quality to many episodes. . . ." In due time, especially after a rerelease late in 1920, the way opened by Le Conte de Mlonte-Cristo was seized on by two newly independent producers. Louis Nalpas had left Film d'Art in 1918 to set up his own production company in Nice. One of his most ambitious projects was Henri Fescourt's Alathias Sandorf (1921), an adaptation of Jules Verne's transposition of the same Dumas novel. A story of intrigue and adventure involving a Hungarian independence hero, laithias Sandorf was released early in 1921 in nine weekly episodes. It became so popular, however, that it was recut later in the year 80 and rereleased as a single-seance film that ran for seven months at the Cirque SERIALS d'Hiver.44 Francis Lacassin speaks of it in terms similar to those he used to praise the Frenchness of Feuillade's serials: An "adventure film," certainly, but one that differs from the stale and uneven resale items that the Americans dispense-because of its strong dramatic structure, the qual- ity of its acting based on a well-established psychology, the intelligent choice of landscapes, and the splendor of its photography. . .4. At the same time, an even more ambitious project was initiated by Henri Diamant-Berger. His plan was to film the most popular French adventure novel, Dumas's Les Trois Mousquetaires, complete in twelve unprecedented hour- long episodes. The role of D'Artagnan was offered first to Douglas Fairbanks, but Fairbanks refused to work in such a vulgar genre as the serial.46 Stimulated by the idea, nonetheless, he counteroffered by inviting Diamant-Berger to Hollywood to direct him in a two-hour American version of the novel. Refus- ing to cut his scenario so drastically, and with Charles Path's financial bless- ing, Diamant-Berger went ahead with his plan (as did Fairbanks), using an entirely French cast (Aim6 Simon-Girard, Edouard de Max, Armand Bernard, Henri Rollin, Vallee, Charles Dullin) and shooting much of the film on lo- cations, particularly in the old fortified city of Perouges in the hills above Lyon.47 Initially projected for three consecutive gala evenings at the Trocad6ro in September, 1921, Les Trois Mousquetaires was shown over a three-month period all over France, accompanied by its publication in Comoedia and many provincial newspapers.48 Its phenomenal success-17 million francs against a cost of 2.5 million-convinced Diamant-Berger and Pathe-Consortium to make a second adaptation from Dumas, Vingt Ans apres, which was released in ten episodes late the following year." The success of these two films, in fact, also helped establish the genre of historical reconstructions. By 1922-1923, the serial had become both an economic strategy and an umbrella concept covering various kinds of narratives of adventure and in- trigue: police (Fescourt's Rouletabille chez les bohemiens, 1922), criminal (Feuil- lade's Le Fils du flibustier, 1922), sentimental (Feuillade's Parisette, 1922), cos- tume (Burguet's Les Mysteres de Paris, 1922, starring Huguette Duflos), and, for want of any better label, a lackluster hodgepodge (Ren6 Leprince's L'Em- pereur des pauvres, 1922).50 There were even more parodies: L'Herbier's "hu- moresque," Villa Destin (1921), and Mosjoukine's Le Brasier ardent (1923). More and more critics and producers realized that the serial had reached a dead end. There were complaints about the length of time required to see one serial, about their monopolization of cinema screens to the disadvantage of other French films, about the stagnation of repeated formulas.51 In 1922, Pathe- Consortium stopped importing American serials when Path6-Exchange was sold to a group of Merrill Lynch investors in New York. In September, trou- bled by less than anticipated grosses, the company abandoned its expensive costume serials.52 In the summer of 1923, after Pathe-Consortium had rere- leased Le Comte de Monte-Cristo (a second time) and Gaumont had rereleased Judex in a reduced, single-seance format, Louis Feuillade admitted to physical and artistic exhaustion.53 During this period, Cin-Journal, Mon-Cine, and Ci- nmagazine engaged in a lengthy public debate on the discredited serial format. Mon-Cine even conducted an extensive survey among its readers and concluded 81 COMMERCIAL NARRATIVE FILM that the serial indeed was desperately in need of reform.54 What the public wanted, wrote Pierre Desclaux, was a serial of six to eight episodes on an historical subject.55 And that is precisely what they got when Jean Sapene, the advertising editor of Le Matin, turned film producer. Sapene's prescription for reviving the genre depended chiefly on organiza- tion.56 After taking over the Societ6 des Cineromans from Rene Navarre in 1922, he built up a double-barreled distribution system of cind-romans that involved Path6-Consortium in association with the Lutetia cinema chain and the major Paris newspaper consortium of Le Matin, LeJournal, L'Echo de Paris, and Le Petit Parisien.57 His new production team was led by Louis Nalpas (chief producer or "artistic director") and Feuillade's former scriptwriter, Ar- thur Bernede (scenario department head).58 Their task was to engage reputable directors, scriptwriters, and actors for an annual production schedule of four serials, standardized in eight episodes, to be released evenly throughout the year. They were aided in their efforts by a spirited defense of the serial format by Pierre Gilles in the pages of Le Matin.59 From 1923 to 1927, outside of an occasional film by Gaumont (Feuillade's last films), Aubert, Phocea, and Albatros, the serial became the exclusive property of Sapene's Cineromans. The films themselves, beginning with Henri Fescourt and Gaston Leroux's Rouletabille chez les bohemiens (1922), adhered basically to the parameters already established-with one important difference. Some, such as Germaine Dulac's Gossette (1923) and Rene Leprince's L'Enfant des Halles (1924), continued the pattern of combining elements of the police and sentimental serials. These still made much of children and orphans. Others exploited the exoticism of adventures in the colonies-e.g., Le Somptier's Les Fils du soleu (1924). How- ever, they were all outnumbered by costume adventures drawn from original scenarios rather than from popular literary classics. Now, more often than not, the hero was an historical figure. War heroes and adventurer-brigands from the period of 1750 to 1850 were especially popular-e.g., Gaston Ravel's Tao (1923), Jean Kemm's Vidocq (1923) and L'Enfant-Roi (1923), Fescourt's Man- darin (1924), Luitz-Morat's Surcouf (1925) and Jean Chouan (1926).60 In their return to a rich French world and an optimistic, valiant hero, both belonging to a supposedly more glorious past, these latter films can be read as part of a collective attempt at an ideological restoration or redefinition that is even more visible in the expensive historical reconstruction films. At least two of the Cin6romans productions survive and allow us more than a passing glance at the last important serials. Mandarin (1924), one of the most well-received of Cinromans' serials, had as its hero a kind of Robin Hood of tax-collecting (Fairbanks's Robin Hood had been very popular in France the year before), whose exploits were based on an historical figure operating in the Dauphin6 region around 1750.61 Claude Beylie and Francis Lacassin describe his particular brand of derring-do: Declaring himself the enemy of the landowners, attacking their collectors to whom he sold contraband tobacco in exchange for their cashboxes, benefiting from the com- placent indifference of the local gendarmes, Louis Mandarin, in less than three years, disrupted the French fiscal system and escaped all the soldiers ordered to pursue him.62 Perhaps in anticipation of a sequel, Bernede and Fescourt allowed their hero to avoid capture and the guillotine, through the intercession of Voltaire, no 82 less (who actually did support him in several writings and letters). Instead of SERIALS emphasizing complicated intrigues or the sentimentality of suffering, accord- ing to Beylie and Lacassin, Mandarin seems to have gone for a deft combina- tion of action and atmosphere: It is a fast, concise film, intercut with spectacular horseback chases and explosive confrontations. In the exterior shooting, Fescourt has used the ravines and gravel slopes of the Niqois back country to remarkable effect. . . . [It] remains a model of the French adventure film, full of a sweeping epic inspiration whose secret is perhaps lost.61 A critic writing at the time of its release, however, was much less enthusiastic: ". . . the technique of Mandarin-especially that which involves 'montage'- is regressive by several years. . . . If you wish to follow innovations in the serial, look again at Gossette and La Maison du mystere.""6 Indeed, Gossette (1923) does differ markedly from Mandarin, but its avant- garde status is highly suspect. The story that Dulac was assigned to film was in the tradition of Feuillade's L'Orpheline or Parisette: the sentimental hero, who is falsely accused of murder, disappears; the ingenue heroine'(already an orphan) is taken in by the hero's parents who then are killed, abandoning her again to misfortune; the hero and heroine team up to discover the "truth"; the real villain is unmasked; the old parents' spirits bless the young couple's union.65 This tearful tale is complicated by frequent flashbacks (the lack of intertitles in the only existing print does not help), but several choices that Dulac makes in the narrative are of interest.66 Most of the film was shot in the old Pathe studio at Vincennes and then at the Levinsky studio in Joinville- le-pont, so it eschews the natural landscapes that function so prominently in Mandarin. Instead, the film relies heavily on close shots (MSs to CUs) of the characters-often starkly or flatly lit against dark backgrounds (a lighting technique which had become rather standard practice in French silent films). It also occasionally uses a subjective montage, including soft-focus images (e.g., the hero's hours of half-drunken peregrinations before he is accused of murder) and a montage of objects (e.g., the narration of the initial murder by means of a glass, a cigar, hands, and shadows). However, these supposedly avant-garde devices are awkwardly integrated into the film, and Dulac herself felt deceived and unduly compromised by Cineromans after making it.67 It was during this period, also, that one of the best parodies of the genre appeared in Jean Epstein's Les Aventures de Robert Macaire (1925), produced by Albatros. For his scenario, Epstein drew on the legendary character of the fictional Dauphine highwayman created by Frd6rick Lemaitre as well as by the famous series of caricatures by Daumier.68 The narrative reminds one of Fescourt's in Mandarin-five separate intrigues and adventures (beginning in 1824) involving Macaire (Jean Angelo), his sidekick Bertrand (Alex Allin), and a bevy of women and victims.69 Epstein's concern for atmosphere, how- ever, is much more pronounced than Fescourt's. In fact, the authentic and sometimes exquisite Corot-like landscapes (photographed by Paul Guichard), together with the realistic interiors (designed by Lazare Meerson), tend to counterpoint the adventures and reduce their significance. But there was method in Epstein's apparent mania. According to Pierre Leprohon, 83 28. Alex Allin and Jean Angelo in Les Atzentures de Robert Maclair (1925-1926) In choosing this picaresque hero of popular literature, the filmmaker could-by jux- taposing the intelligence and cunning of his two comrades Robert and Bertrand with the stupidities of the other characters-both amuse the general public as well as amuse himself by parodying the genre that he seemed to serve.7" Jean Angelo plays Macaire with a blithe hauteur that belies his often well- worn and changing costumes. He saves aristocratic ladies from death, rescues good people from the gendarmes, and steals fortunes to set up young couples in marriage; but he is not above picking Bertrand's pocket. Bertrand, on the other hand, fails at nearly everything he tries, but that does not deter him. Begging for food, he is scared off by a farm family that includes a tough old woman, a mangy dog, and a boy picking his nose. After Macaire rescues Mlle. de Sermeze (Suzanne Bianchetti) from drowning, Bertrand whistles for her horse and watches it scamper away. Lovesick over a maid, who keeps reap- pearing with a different man, he tries to drown himself in an inches-deep fountain and, then, with a huge blunderbuss, succeeds only in blasting his hat. Finally, he is defeated in a duel with the sentimental hero (Nino Costan- 84 BOURGEOIS MELODRAMAS 29. Sarah Bernhardt inJeanne Dore (1915) tini) and then is beaten up for good measure. Macaire and Bertrand, however, sometimes do make a good team, as when, in their hilarious robbery of a farmer's wife, they disguise themselves as St. Anthony and a half-blind, purse- snuffling pig. In the end, on a ridge road overlooking a picturesque valley, they swear to become honest men and stride off over a hill, picking one another's pockets. By 1927-1928, the serial had all but disappeared, as it had even earlier in Germany and the United States. Its once prominent position in Ciniromans' production schedule had passed to the Films de France series and to the ex- pensive spectaculars the company was co-producing with Cin6-France and others. As more and more publicity was given to these bigger films and as American and German films consolidated their hold on the French public, cinema pro- grams increasingly were geared to the single-sance films and even to the Amer- ican double-bill format. As a consequence, the serial declined in popularity and profitability. Here the history of the French film serial seems to close off. But the study of that history, despite the fine work of Francis Lacassin and Francois de la Breteque, has hardly begun. As the grand film developed into a major production strategy during the war, it defined itself initially within the tradition of literary adaptations es- tablished by Film d'Art and S.C.A.G.L. Producers at Path6-Freres, Eclipse, and Film d'Art could now more easily mobilize the reputable actors of the Comedie Frangaise as well as their slightly less illustrious colleagues from the boulevard theaters and music halls to lend some measure of respectability either to their patriotic films or to their condensed versions of the current theater. Thus, the appeal of these films, for a short time anyway, derived from the presence of famous stage actresses and personalities. The music hall star, Mistinguett, for instance, had her greatest success in Fleur de Paris and Chi- Bourgeois Melodramas 85 COMMERCIAL NARRATIVE FILM gnon d'or (both 1915).1 Rejane made one of her rare film appearances in Pouc- tal's Alsace (1916), which reprised her prewar theatrical performance and ran for two consecutive (and unprecedented) weeks at the Gaumont-Palace. Sarah Bernhardt posed before Jeanne d'Arc's statue and the ruined Rheims cathedral in Hervil and Mercanton's Les Mires franfaises (1917); and in Mercanton's ad- aptation of Tristan Bernard's Jeanne Dor (1915), she gave an unusually re- strained performance that is affecting even today, partly because of her ravaged face and nearly immobilized body.2 At Pathe-Freres, Gabrielle Robinne, who had performed in the original version of L'Assassinat du duc de Guise (1908), was now starring in her own series of films (most of them directed by Ren6 Leprince) in which, according to Ren6 Jeanne, she did a lot of loving and suffering, with dignity.3 In 1918, she was still representative enough for Louis Delluc to hold up her mannered style of acting for ridicule against the robust spontaneity of Pearl White.4 Most of these early feature films were drawn from a particular form of drama, the popular melodrama. There had been instances of the genre in the prewar cinema, say, in Path6-Freres' adaptations of several Henri Bernstein plays.5 But the genre did not come into its own until after 1916, once the ground had been prepared by the Italian melodramas starring Francesca Bertini and by De Mille's The Cheat with Sessue Hayakawa (whose success, in part, depended on its French bourgeois melodrama origins).6 The bourgeois melo- drama had long been the favorite of the boulevardiers, whose life revolved around the theaters, nightclubs, and caf6s on the Grand Boulevards of Belle Epoque Paris. Motivated perhaps by the war restrictions on theater performances in Paris, the film industry eagerly expropriated the genre to itself and began to expand its audience with a new clientele. The bourgeois melodrama film generally took its subject from the socioeco- nomic context of the high and middle bourgeoisie.7 The boulevardiers and Opera-season ticket-holders could find familiar settings here: Renaissance din- ing rooms, Louis XV bedrooms, and Empire offices.8 There were the familiar plots, full of lurid intrigue and violent action usually predicated on the "eter- nal love triangle." And at the center was the question of a woman's loyalty or social function as wife and mother as well as the consequent affirmation of the family as the locus of moral and spiritual value and, hence, of social stability.9 The attraction of the bourgeois melodrama, according to Jean Mitry, was that it produced the world of "a society drawn from the theater . . . not the reflection of the actual bourgeoisie, but the image of the privileges that they no longer enjoyed except in the fictions that were made for them and which they greedily devoured."10 Several films preserved at the Cinematheque fran aise provide a good index of the early feature-length bourgeois melodrama. Camille de Morlhon's Maryse (1917) tells the story of a young woman (Maryse Dauvray, Eclair's star) who accepts a brief liaison with a doctor in order to keep up an apartment for her mother." Later she becomes the model and mistress of an up-and-coming young painter, who just happens to be the doctor's son. The climax comes in a recognition scene between the two men; the painter attempts suicide, un- successfully, and father and son are reconciled. Yet the last shot is of the young woman alone once more. The narrative sacrifices passion for family harmony (between father and son), but the surviving print makes it uncertain 86 whether the woman is an immoral temptress or the real victim. Maryse uses a limited number of flatly lit sets (several apartments, the painter's studio, the Opera) but divides up many of its scenes, in the manner of American conti- nuity editing, into a range of shots from LS to ECU (with most of them falling between FS and MCU). However, the rhythm of the editing is not always smoothly paced; and the acting, though quite restrained for a French film of that time, relies on theatrical clich6s at highly emotional moments. In com- parison, Jacques Feyder's early two-reeler about a double love triangle, Ttes de femmes, femmes de tite (1916), looks even more restrained and original-it is played out almost entirely in MSs.12 Abel Gance's Le Droit a la vie (1917) tells a similar, though less ambiguous, story. A young woman (Andr6e Brabant) is in love with a stockbroker (Leon Mathot) but is forced to marry an older financier (Paul Vermoyel) to pay the debts of her dying mother. The financier discovers the young couple's love and then an embezzlement by one of his assistants. When the assistant shoots and wounds him, the enraged financier pins the blame on Mathot. Finally, in the midst of a heated trial, he confesses Mathot's innocence and, soon after, conveniently dies so the young couple can be reunited. Gance's film is distin- guished from Maryse by much more sophisticated lighting and editing, both of which probably derive (as does some of the story) from The Cheat. 13 In several key sequences, the characters are isolated in close shots and sculpted by side lighting against black backgrounds. This technique, mastered by Gance's young cameraman, L.-H. Burel, also appears in other bourgeois melodramas such as Aubert's Notre Faible Coeur (1916) and Germaine Dulac's Venus victrix (1917).14 Significantly, the moral and psychological condition of the central characters (but only the men) is emphasized through several subjective images. The suspense of the shooting climax is heightened by a strategy of parallel editing (using crude wipes) between the confrontation of men in one room and a progressively more unrestrained masked ball in another."15 The trial is articulated almost entirely in MCUs that depend on eyeline matches and shot/ reverse shots; in fact, the film as a whole is marked, even more than is Maryse, by frequent MCUs and CUs. Finally, there is one fascinating fetishistic gesture whose repetition almost turns it into a rhetorical figure. In linked sequences, Mathot lovingly caresses the young woman's hair, and then the financier dreams of caressing her in the same way as he prepares his marriage plan. After the wedding, however, as she stands by the bedroom window, the young woman refuses to let him touch her. The potential of the bourgeois melodrama is probably most clearly dem- onstrated in Gance's Mater Dolorosa (1917), said to be the most popular French film of the 1917-1918 season. 16 Although an original scenario, Mater Dolorosa follows the dramas of Paul Hervieu, Henri Bernstein, Charles Mere, and, perhaps more specifically, Henri Kistemaecker's L'Instinct.17 Much like Le Droit d la vie, it also owes a good deal to The Cheat. The narrative is predicated on a love triangle involving Dr. Gilles Berliac (Firmin Gamier), a pediatrician, his wife Marthe (Emmy Lynn), and her lover, his brother Claude (Armand Tallier). 18 Overcome with guilt for her affair, Marthe tries to commit suicide and, in a struggle with Claude, accidentally shoots him. When Gilles is in- formed of his brother's death (through a humpbacked blackmailer!), he be- lieves Marthe guilty of murder and leaves her, taking their young son with BOURGEOIS MELODRAMAS 87 30. (left) Emmy Lynn and Armand Tallier in Alater Dolorosa (1917) 31. (right) Emmy Lynn at the window him. Uncertain of his fatherhood, Gilles agonizes over caring for the boy when suddenly the latter falls ill. Finally, a letter comes that exonerates Marthe, and Gilles recognizes her suffering. The couple is reunited; the son, restored to health. As in the other films belonging to the genre, the emphasis through- out this narrative is on the psychological condition and moral state of the two central characters. But here, quite clearly, the health of the individual as well as the society literally depends on the health of the family or the condition of a marriage. Besides the social or ideological implications of its narrative, Mater Dolorosa exhibits most of the stylistic conventions of the bourgeois melodrama. The settings are confined to the lover's study, several rooms in the couple's com- fortable parkside house, the hospital ward, the doctor's apartment (only these last two deviate a bit from convention-the hospital ward, for instance is unusually realistic in its detail) and a few exterior shots. The range of shots used in the film's dozen or so segments-from LSs to CUs, from straight-on shots to various angled shots is not all that different from Maryse or Le Droit a la tie. Nor is the number of shots deployed uncommon. Most of the se- quences are composed of many separate shots, yet they do not add up to the nearly "400 shots" in Andre Antoine's comparable crime melodrama, Le Cou- pable (1917). " In Gance's film, however, the cutting within and between sequences is exceptionally clear and economical, the more so since there are quite a number of inserts, short flashbacks, and brief subjective sequences, along with several sequences of parallel action. By contrast, Maryse and Le Coupable, the latter of which tries to narrate a past story within the context of a trial, display much less economy, smoothness, and clarity. Yet Alater Do/orosa transcends its genre conventions in several ways. Com- menting on its mise-en-scene, Jean Mitry neatly sums up the judgment of French film historians: Alater Dolorosa . . . surprises, astonishes, by means of lighting effects, the knowing use of light and shadow to intensify dramatic scenes, the intimate fidelity of the decors, singling out particular details, and a thousand unusual qualities for a French film. o 88 Gance and Burel used a good deal of sidelighting and low-key spot lighting on faces and parts of figures against dark backgrounds, creating even softer images than in The Cheat or Le Droit a la vie. Here the effect serves not only to model the characters in a three-dimensional space but also to heighten their emotional expressiveness. Figures are framed in silhouettes at windows and behind curtains-a technique that owes a debt not only to The Cheat but to earlier French films such as L6once Perret's fascinating L'Enfant de Paris (1913), photographed by Georges Specht.21 As Colette first noted, the film also evi- dences "a new use of the still life, the poignant use of props, as in the fall of a veil to the floor."22 The slightest movements within the frame resonate metaphorically: Emmy Lynn backing off into the darkness after the shooting, a single window opening outwards (with no person visible) in a LS of the apartment building, a MS of the doctor behind a curtain as he touches his wife's veil, which echoes the earlier CU of the same veil falling beside the window through which she watches him leave the house. More obvious met- aphors are evoked in a shattered mirror (in the sequence of doubted fatherhood) and in a painting of the mater dolorosa which provides the prior image or figure recognized by the doctor and replicated in his wife's suffering. Here, in em- bryo, is the possibility of a sustained pattern of rhetorical figuring. Film adaptations of the bourgeois melodrama continued to be an important asset to the French film industry into the early 1920s. Especially popular were the plays of Bernstein, Kistemaecker, and Mer6. When Gaumont, for in- stance, wanted Marcel L'Herbier to prove himself commercially after the total failure of Rose-France (1919), he told him to consider the success that Louis Mercanton and Gaby Deslys had made of his previous scenario for Bouclette (1918); and he assigned him to direct Bernstein's Le Bercail (1919). L'Herbier agreed, as he put it, "to practice his scales. "23 The result was that Gaumont made a tidy profit and L'Herbier could go on making films. Film d'Art pro- duced several more melodramas starring Emmy Lynn, that "victim of passion and mother love": Gance's La Dixieme Symphonie (1918) and Henry Roussel's La Faute d'Odette Marichal (1920).2 When the company determined to make films marketable in both France and the United States, it chose as its first projects Bernstein's La Rafale (1920) and Kistemaecker's Le Secret d Lone Star (1920), both directed by Jacques de Baroncelli and starring the American actress, Fanny Ward, and Gabriel Signoret of the Comedie Fran aise. Despite risking most of its capital on expensive serials and multipart films, Path6- Consortium also distributed popular melodramas such as Henry Krauss's ad- aptation of a Mere play, Les Trois Masques (1921). However, much like the serial, the genre was under a great deal of pressure that was pushing it in several different directions at once. Abel Gance, for instance, was reshaping it with his technical experimentation as well as his philosophical pretentions. His original scenario for La Dixime Symphonie (1918) tried to transform a bourgois melodrama plot of marital fidelity by posing the problem of artistic creation-an idea, Kevin Brownlow writes, that was in- spired by a quotation from Berlioz: "I am about to start a great symphony in which my great sufferings will be portrayed."25 The composer Enric Damor (Severin-Mars), a widower with a grown daughter, Claire (Elisabeth Nizan), marries Eve Dinant (Emmy Lynn), who has an untold, compromised past- she is being blackmailed by a former lover, Frdric Ryce (Jean Toulout) for BOURGEOIS MELODRAMAS 89 COMMERCIAL NARRATIVE FILM accidentally killing his sister. When Ryce courts Claire and Eve opposes their announced marriage (without explanation), Damor believes his wife is secretly in love with the suitor. In despair, he writes and performs his symphony, and Eve prepares to sacrifice her own happiness if Ryce will leave Claire alone. Claire discovers what is happening, and, after a flurry of threats and counter- threats, Ryce shoots himself. Claire defends Eve to her father (without telling all), and Damor relents and forgives. Two years later, Andre Hugon would rework this dual problem of marital fidelity and artistic inspiration, with Severin-Mars as a poet cursed by alcoholism, in a much more conventional melodrama, Jacques Landauze (1920).26 What is admirable in La Dixieme Symphonie extends the techniques of Mater Dolorosa: the mysterious elliptical opening of the past accidental shooting, the repetition of gestures for psychological effect, the lighting and placement/ movement of characters, the rhythmic clarity of the action. But Gance's at- tempts to relate his artistic intentions to the classics (e.g., comparing Eve to the Winged Victory of Samothrace, citing numerous literary quotations) and to represent the act of artistic creation itself (for Damor's performance, he films an "Isadora Duncan" dancer [Ariane Hugon of the Opera} in a series of natural settings, masked by a horizontal vignette of Greek vase designs) pro- voked critics to misleadingly brand the film "inaccessible to the public."27 Today, these moments merely seem embarrassingly simplistic. Furthermore, Eve's suffering is much greater, ironically, than that of her composer husband, and Damor never discovers the full extent of her discontent. J'Accuse (1919), Gance's first superproduction for Path6-Cinema, trans- formed the genre almost completely by thrusting the melodrama plot into the disturbing context of the war. Here, even more than in La Dixieme Symphonie, the love triangle intrigue is but a pretext for a fiercely emotional personal statement and for further cinematic experimentation. Paradoxically pacifistic and nationalistic at the same time, J'Accuse was a stunning commercial success, especially considering its release several months after the Armistice. As will be demonstrated in Part IV, it was also technically and rhetorically much in advance of any other French film of 1919. The same blend of family melo- drama, personal statement, and experimentation would also mark Gance's next epic, La Roue (1922-1923). In contrast to Gance's "epic symphonies," Raymond Bernard, a protg6 of Jacques Feyder, oriented the genre toward the "chamber music" of a more intimate drama. Drawing primarily on a series of plays and scenarios by his famous father, Bernard concentrated on the psychological possibilities of the bourgeois melodrama-in Le Secret de Rosette Lambert (1920), La Maison vide (1921), and Triplepatte (1922). The subject of Le Secret de Rosette Lambert was a conventionally "cruel and violent story," adapted from Tristan Bernard's Coeur de Lilas.28 In it, a businessman sets a trap for the wife of his partner in order to ruin him. When the villain dies, his proxy keeps up the intrigue, only to be unmasked by a family friend of the victims. Unlike previous bourgeois melodramas, however, this film situated its story in quite modern decors de- signed by the young master architect, Robert Mallet-Stevens. Mallet-Stevens's sets simply and starkly validated Bernard's psychological interest as well as creating a suitable atmosphere for informal dancing (which was then undergo- ing a new wave of popularity in Paris).29 For the first time, apparently, mod- 90 BOURGEOIS MELODRAMAS 32. A Robert Mallet-Stevens decor for Triplepatte ( 1922) ern interiors were constructed specifically for a film and not merely chosen from the stock of traditional studio decors.'( Despite fine acting from Henri Debain, Sylviane Grey, and Charles Dullin (in debut), the film was compro- mised by its star, an unknown American actress, Lois Meredith." Still, Ber- nard's direction, according to Louis Delluc, resulted in "one of the most splen- did manifestations of photogenic plasticity in the cinema. "32 La Maison vide, the first film to come out of Bernard's own independent production company, was an original scenario that he devised from a banal love story. In his unpublished memoirs, Bernard has confessed that his ambition in the film was to make the audience aware of subtle mental states of which the characters themselves were not always conscious." Henri Fescourt remem- bers the film vividly: An impalpable story of grays. Hesitations, subtle nuances, slight incidents. A timid entomologist, suspended over his precious collections, falls in love with his secretary. Scarcely anything happens. A bouquet on a table, displaced by an angry kick, falls to the floor. A microscope that reveals darkened images. Is it the fault of the lens? No, the eye that observes, an eye darkened by a tear. . . . No hands clutching the breast, no long sighs. Very few intertitles. It was enough to watch the images and Henri Debain. Unfortunately, Bernard's choice of a title, La Maison vide, failed to attract audiences to the cinema, and the film ended up appealing to only a few 91 33. (left) Eve Francis, Jean Toulout, and Gaston Modot in La Fte espagnole ( 1920) 3=4. (right) Ginette Darnys in Le Silence ( 1920) filmmakers and critics.'' In Triplepatte, Bernard fared much better by retreat- ing to one of his father's stock of well-known plays, which allowed Henri Debain to mildly satirize a similar "hero of indolence and indecision."1 Bernard was not alone in creating "chamber music" films during this pe- riod, for the analogy (it is Henri Fescourt's) also fits a small group of films produced by filmmakers involved in the narrative avant-garde. Several of Louis Delluc's earliest films, for instance, develop a refined psychological irony from rather conventional bourgeois melodrama plots. Le Silence (1920) can serve as an example, even though it survives only in the form of a published, decep- tively simple scenario. Pierre (Signoret) is awaiting his lover Susie (Eve Fran- cis) alone in his rooms. Suddenly, in a chance comparison of letters, he realizes that she was responsible for an anonymous letter in the past that inflamed his jealousy and incited him blindly to kill his young wife. When Susie finally arrives late, Pierre is dead in his chair, the victim of the same revolver with which he shot his wife. Since the action takes place almost entirely within Pierre's consciousness, Le Silence has been aptly described as "a monologue in images. "" The "dramatic theme" (to use Delluc's own phrase) unfolds through the alternation of Pierre's memories and his perception of certain key objects in his possession. Elliptical and full of ironic twists, this alternation seems to have been, in contrast to the usual bourgeois melodrama, more intellectual than emotional in its effect. Germaine Dulac shared Delluc's interest in the psychological states of char- acters, and not only in their collaborative film, La Fete espagnole (1920)-where two suitors kill one another over a bored Spanish lady (Eve Francis again) who ignores them and, drunkenly awakened to life once more in a festival dance, gives herself to a third man. " The rather dense narrative structure of this short film will be discussed later; for now it is worth mentioning that Eve Francis's costume was one of the first designed specifically for film.-'0 La Mort d soled (1922) is perhaps more characteristic because its scenario, written by Andre Legrand, resembles that of Mater Dolorosa. While it celebrates the struggle 92 of Doctor Lucien Faivre (Andre Nox) against tuberculosis-epitomized in re- curring shots of supplicating diseased children in some barren place-it also focuses on the conflict dividing Marthe Voisin (Denise Lorys), who is dedicated to helping Faivre but is equally devoted to her family, especially her young daughter." Given this scenario, Dulac wanted in her words, "to describe the inner workings of the mind or soul, within the theme of the action. "'2 Con- sequently, she uses her plot, with its several misunderstandings, to explore the subjective life of both the woman and the doctor, separately and in com- bination. The most unusual of these moments is a strange conjunction of the two characters as the doctor lies near death early on in the film. Images of Faivre's delirium and of Marthe's inner thoughts are intercut to produce a "communion of souls" whose significance oscillates between escape, shared passion, and awakening dedication to a scientific discipline. Unfortunately, according to Charles Ford, "this eminently cinematographic scene was cut in nearly all the cinemas [because} the spectators fand the exhibitors] would not let the action of a film be encumbered with psychological notations. "B By 1922, the bourgeois molodrama had been so transformed--and separated from its theatrical origins-that it no longer existed as a distinct genre. Per- haps this transformation was due to an apparent rivalry (or lack of cooperation) between the cinema and theater as much as it was to the pressure of the interests of the narrative avant-garde.-" Perhaps, too, Charles Path6's warning about adaptations from the theater had proved true: . . . the author and director should consider not only the title of the play they want to adapt, or the intensity or violence of the series of well-linked actions, but also-- and this is not current practice the nature of the emotions that are expressed. The extreme plot situations of prominent playwrights, unless substantially amended, will fail to pass censorship, particularly in the Anglo-Saxon countries. . . . The audiences of these countries will not appreciate the spicy, risque plots that Bernstein and, preem- inently, Bataille have established as the norm." In any case, this change was actively supported by a new film journal, Mon- Cine. Within months of its appearance in February, 1922, Mon-Cine embarked on an educational campaign to encourage the development of a "cinema of quality" in France. Its purpose was twofold and implicitly ideological: to forge a link between film and fiction and to dignify the cinema by displacing the serial or cinfroman, in the popular taste, with the roman-cin{ or something very much like the bourgeois melodrama.46 To that end, each issue of Mon-Cine was devoted to a serialized novel and to the novelization of a current "serious" film, which novelization soon became as important as the film itself. Largely ignoring the work of the narrative avant-garde as well as the serials, Mon-Cine singled out for praise such filmmakers as Bernard, Roussel, Feyder, Fescourt, Robert Boudrioz. and Leon Poirier." Their films, and especially Poirier's Jocelyn (1922) and Genevieve (1923), came closest to realizing what Maurice Roelens has identified as an "aesthetic of emotion"-"a [logical] series of naturally melodramatic situations, apt to arouse emotion and especially 'the voluptu- ousness of compassion' that one usually associates with melodrama. "4 Al- though terminologically vague-and thus capable of infusing various genres, as we shall see-this was one answer to the question of what was peculiarly French about the French cinema. 35. Three shots from the delirium sequence of Doctor Faivre (Andre Nox) in La Mort du soleil (1922) 93 COMMERCIAL NARRATIVE FILM Realist Films In the course of Mon-Cin6's polemic, most of the filmmakers who had come into the cinema from the theater before or during the war turned exclusively to adapting fiction or writing their own scenarios. And those theatrical mel- odramas that did become films were usually subsumed in one of several other genres.50 Some, such as Ren6 Hervil's Blanchette (1921) or Fescourt's Les Grands (1924), evoke the atmosphere and ambience of natural location shooting and are best seen as realist films. Others, such as L'Herbier's "melodrama," El Dorado (1921), or Julien Duvivier's Maman Colibri (1929), became part of the cycle of exotic or colonial films. Still others, such as Dulac's Ame d'artiste (1925), L'Herbier's Le Vertige (1926), or Perret's La Femme nue (1926), create elaborate sets and an international milieu that make them early examples of the modern studio spectacular. Otherwise, the only truly theatrical adaptation to replace the bourgeois melodrama in the 1920s was the boulevard comedy, especially as represented by the plays of Labiche and company. But the boul- evard comedy is important enought to be treated separately. The exceptions to these changes were few and far between. Germaine Du- lac's Antoinette Sabrier (1927), produced by Cineromans from a play by Romain Coelus, which R6jane had made popular, is one of the sole surviving exam- ples.51 Its subject is a busy industrialist (Gabriel Gabrio) who cannot decide which of two women he loves, his wife (Eve Francis, who had not made a film in four years) or a younger woman "whom he believes to be the incarnation of Love" (Yvette Armel).52 According to Charles Ford and Rene Jeanne, the film was too close to being a Comedie Fran aise production.53 Julien Bayart pin- pointed one of the reasons: "the decors . . . produce an impression of artifi- ciality, an impression of coldness accentuated even more by the way they are lit. ..."Antoinette Sabrier, he concluded, "is a theatrical, a terribly theatrical . . . story.. .""The genre had come back to its origins. As the Great War drew to a close, a number of French films appeared- Andre Antoine's Le Coupable (1917) and Les Travailleurs de la mer (1918), Louis Feuillade's Vendimiaire (1919), and Jacques de Baroncelli's Ramuntcho (1919)- all of which challenged the prominence of the class-conscious, studio-bound evasions of the bourgeois melodrama. These films mark the beginning of a broad genre of feature films that developed in the French cinema of the late 1910s and early 1920s. At the time, they were identified in the press under different labels-e.g., "simple dramas," "atmosphere films," or just "plein air films." Despite the problematics of the term, let me call them realist films, for the designation seems to have some historical basis. First of all, these films were not without precedents in France. Before the war, after all, as a counteraction to the Film d'Art literary adaptations, Louis Feuillade had produced a series of sixteen short films under the high-sounding title, La Vie telle qu'elle est (Life as It Is). In a statement introducing the series, in April, 1911, Feuillade spoke of his films as "slices of life"-"they eschew any fantasy and represent men and things as they are, not as they should be."1 As Francis Lacassin has shown, despite Feuillade's invocation of Zola and Mau- passant, most of these sketches were violent, intimate melodramas, stripped down to a handful of characters in a few simple, stereotypical settings.2 An- other, perhaps unlikely, precedent can be seen in some of the prewar crime 94 films and serials. LUonce Perret's L'Enfant de Paris (1913) and Georges Denola's La Jeunesse de Racombale (1913), for instance, already exhibited the same sen- sitivity to natural light and open air that would characterize the later films.3 Like several of Perret's early films, L'Enfant de Paris was shot on location (by Georges Specht), in the back streets of the Paris faubourgs as well as in and around a villa above Nice.4 In these early serials, however, the conjunction of fantastic crimes and deceptions with recognizably real spaces induces a pecul- iar state of disorientation or dislocation that differed radically from the effect of the postwar realist films. Georges Sadoul has argued that Feuillade and his colleagues at Gaumont were influenced in these efforts by the Vitagraph films then being imported into France and by the verism that characterized French literary naturalism.5 Even more marked by that verism, however, were similar films by Victorin Jasset and Maurice Tourneur at Eclair and by some of their colleagues at S.C.A.G.L.6 The best of Jasset's Les Batailles de la vie series, Au pays des tinebres (1912)--inspired by Zola's Germinal-was no less melodramatic and conform- ist in its moralism than were the Feuillade films. But it did seem to have a surface verisimilitude as a result of being shot in the northern coal-mining regions.7 At S.C.A.G.L., Zecca assigned Ren6 Leprince and Camille de Morl- hon to produce a comparable series, Scenes de la vie cruele,8 but the company's major realist productions were two early feature-length films. The first was Andre Capellani's adaptation of Hugo's Les Mistrables (1912), which Antoine later cited as one of "the first accurate realizations of the cinema."9 The second was Gerard Bourgeois's remake of Les Victimes de /'alcool (1911), based on Zola's L'Assomoir. This film stuck more closely to Zola than did any of the others, although it depicted alcoholism as an individual vice rather than as a social phenomenon. Its real interest, however, according to Sadoul, lay in its mise- en-scene.10 Each sequence was filmed in the old-fashioned style of the tableau (and in studio decors), yet the movements of the actors often carried them from LS to MS within the space of a single take. Here was one of the earliest sustained instances of a realism based on deep space. The man probably responsible for this initial influx of verism into the French cinema was none other than Andre Antoine. His work in translating the aes- thetics of naturalism from fiction to theater-through the Th6tre Libre (1887- 1896) and then the Theatre Antoine (1897-1906)-had provided an incentive, particularly for the filmmakers at S.C.A.G.L.11 This should not be surprising since some of them-Tourneur, Capellani, Desfontaines, Denola, and Krauss- originally had been members of his theater companies. In 1914, Antoine's position of influence increased dramatically when, out of financial exigency, he accepted a contract to direct films for S.C.A.G.L.12 Apparently, he thought, as did Charles Path6 for a time, that he might transform the French cinema the same way he had the French theater a generation before. Although that mission proved unsuccessful-and his age (sixty), abrasiveness, cantankerous nature, and sense of self-importance alienated many of his colleagues-Antoine did articulate ideas and institute practices that were fundamental to the emer- gence of the French realist film. The first thing Antoine insisted on was shooting on location. As early as 1917, in response to a question on the crisis in the French cinema, he argued that REALIST FILMS 95 COMMERCIAL NARRATIVE The cinema would make real progress if it abandoned the studios to work in nature FILM just as the Impressionists did. Instead of improvising an artificial milieu for the cam- era, we should transport the cameraman and his instruments into real buildings and interiors, as well as develop mobile electric generators for lighting. 1 In part, these words were provoked by the economic and material conditions of the wartime French film industry. The studios were operating at less than full capacity, and they were clearly now inferior to the American studios, especially in lighting and set design. Furthermore, neither Pathe nor Gaumont seemed willing or able to provide the capital needed to renovate their facilities. So filming on location was a viable response to the industry's limitations. Antoine's own production of Les Travailleurs de la mer (1918) seems to have set a precedent. Within two years, many of the French film companies, particu- larly the independents, had dispersed into the countryside, using mobile or on-location studios for their productions. More important to the specific development of the realist film was Antoine's concept of location shooting and the new acting style that it demanded. When motion pictures first appeared, Antoine was fascinated by the filmed docu- ments and then by the newsreels. 14 What struck him was the plasticity of these images, the natural landscapes and real figures in those landscapes, ren- dered in relief through natural gestures and movements as well as through the direction, intensity, and texture of natural light. For his naturalistic theatrical productions, it should be remembered, Antoine had often taken photographs of their alleged locations, photographs which his set designer then used as an index of verisimilitude. Now the cinema seemed to offer the possibility of enacting and recording the story or scenario in those very locations. Just as he opposed the painted sets and reusable decors of the studios, so did he object to good theater actors becoming grimacing film stars: "these performers must be exclusively plastic. . . ."1 Because of the difference between theater and film, Antoine also advocated the formation of a special troupe of actors who would work only in the cinema. 16 And he suggested that, wherever possible on location, "real types" among the local inhabitants enact the smaller roles- in effect, play themselves. Finally, he had the novel idea that, instead of making the actors play to the camera (which was usually immobile), the cam- era should be free to follow the actors. "[It should follow them step by step, to surprise their looks, from whichever angle they present themselves."17 The last thing Antoine stressed was the importance of the scenario, the subject of the film. Despite his interest in the plasticity of the film image and in what he called "impressionistic tableaux," he believed that "photography was no more than a means."18 The principal factor in a film's appeal and success was its subject. For Antoine, the best subject was either an existing realist or naturalistic work of fiction or an original scenario by a like-minded contemporary novelist. This would allow the narrative cinema to represent life "as it really was," to hold up a window or mirror for the spectator. Antoine's own choice of scenarios is significant. Those that he chose, rather than those he was assigned, were stories of peasants and working-class people, adapted from major nineteenth-century narrative poems and novels. His production of Hugo's Les Travailleurs de la mer (1918) and his decision, in 1919, to film Zola's La Terre (1921) established precedents for representing something other 96 than the life of the French bourgeoisie in Paris or in the southern coastal playgrounds. Thus, in its exploration of geographical areas and provincial cultures, the French realist film developed a mode of representation which differed some- what from that of realist and naturalistic fiction. The critical, even fatalistic, representation of the social order so crucial to that fiction was tempered in the realist film by several influences, among them French landscape painting, reaching from the Impressionists back to Millet, his more conventional con- temporaries, and even Corot. Instead of analyzing the relationships among individuals or groups in the social order, the realist film tended "to record the verities of nature en plein air" and to celebrate natural landscapes as a presence that encompassed and affected the characters, usually with a sense of gentle melancholy.' The genre seemed to subordinate social analysis to a concern for the pictorial and even the picturesque. Were these French landscapes displayed as so many scenes or documents for the disinterested aesthetic pleasure of a spectator, a touring bourgeois spectator, for whom, in John Berger's words, "the landscape [wasi his view, the splendor of it his reward"?20 Or was that pleasure perhaps more than disinterested-did such landscapes accurately rep- resent a France that was still predominantly rural (until 1931)21 or did they deflect attention from the harrowing reality of the war-devastated regions of northeastern France? Whichever, in contrast to the serial and the bourgeois melodrama-whose love triangle plots it sometimes shared-the realist film tended to invest less in narrative and more in description. More precisely, it emphasized the emotional and connotative relationship between landscape and character, sometimes to the point where a particular landscape or milieu be- came the central character of the film. It is tempting to characterize the realist film as having two broadly different subjects. The distinction is a simple one-life in the city, usually Paris, and life in the country. Something of the deep social/ideological division in France between Paris and the provinces, between a bourgeois society dependent on a state apparatus, industry, and a proletariat (as well as the unemployed) and one dependent on the land and a peasant class of farmers and artisans, between the locus of the "modern" and the locus of the "traditional," does seem to undergird the genre. Yet the division is perhaps less significant than one that distributes the more numerous films set in the provinces according to different landscapes and regions-the seacoasts, the mountains, the rivers and canals, the agricultural plains. Along with the modern urban milieu, each of these regions and their local customs is represented by some half-dozen films in the early 1920s. Interestingly, they correlate somewhat with the regions that had been the subject of a folklorist concern of the nineteenth-century Realist art, what Linda Nochlin calls, "picturesque regional genre painting"-e.g., Pro- vence, the Pyrenees, and especially Brittany.22 Whatever the geographical area or culture, however, the realist films, much like their forerunners in painting, tended to turn to the past or to more traditional ways of life in presenting positive images of France and of the French social order. One of the earliest areas to be explored was the coastal region of Brittany. After making Les Freres corses (1916) and Le Coupable (1917), Antoine refused REALIST FILMS 97 COMMERCIAL NARRATIVE to shoot another film in Paris. Deliberately, he chose to adapt Victor Hugo's FILM Les Travailleurs de la mer (1918) so that he, his assistant Georges Denola, and cameraman Paul Castanet could work almost entirely on the western tip of Brittany.23 Recently, his son talked about the production: Antoine had a villa at Camaret. He loved that spot very much. It is there that he shot the essential parts of the film . . . it was an occasion to photograph the sea. For the real character of the film is the Sea. 24 The stills that remain of Antoine's film evidence his obvious love for the "filmed document . . . this quality of ubiquity and presence, of the cinema's omnipotence."25 Yet the film did not always achieve the naturalness that An- toine desired. Delluc criticized its scenario (for which trick shots had to be used in a climactic battle between the hero and an octopus!) and pointed to the contradiction of Comedie Fran aise actors (Romauld Joub6, Armand Tal- lier, Andree Brabant) playing Breton fisherfolk in the midst of the real thing.26 Ah, how I wish he had a story of his own, a scenario that was alive, new, modern. Say, a story about workers or, even better, about peasants. . . . Perhaps the characters could be acted by their real-life counterparts. I would like a peasant to be played by a peasant.27 That was a practice, however, which S.C.A.G.L. apparently refused to let Antoine indulge in. Still, Les Travailleurs de la mer was unusually popular- the readers of Comoedia, for instance, voted it one of the top five films of the decade-and it seems to have stimulated the production of a number of coastal realist films.28 Two of the most interesting of these early films were Marcel L'Herbier's L'Homme du large (1920) and Louis Mercanton's L'Appe du sang (1920). Mer- canton, in collaboration with Ren6 Hervil, had directed L'Herbier's first sce- nario, Le Torrent (1918), whose theme, L'Herbier says, was "a raging river in the middle of France, in a village at the water's edge, with thickets of little dramas that are nourished by the torrent."29 Their strategy, however, was to film it with Comedie Fran aise actors, such as Henry Roussel and Gabriel Signoret (with Jaque Catelain in debut), against the rather tame gorge of the Loup river on the C6te d'Azur.30 Although L6on Gaumont thought highly of the film when it previewed at the Cin6-Opera in November, 1917, L'Herbier himself did not even recognize his story and characters in Mercanton and Hervil's images.31 Three years later, the young cineaste returned to the theme of Le Torrent with a Breton film that caused a brief scandal. L'Homme du large (1920) was a rather free adaptation of a short philosophical story by Balzac, Un Drame au bord de la mer.32 Following his habit of desig- nating the genre of his films, L'Herbier christened it "a seascape." The nar- rative, told in a long flashback, caused a stir by contrasting two children of a Breton fisherman (Roger Karl-the daughter Djenna (Marcelle Pradot) is pious and obedient; the son Michel (Jaque Catelain) is profligate and afraid of the sea.33 Infatuated with a new singer in a local cafe and egged on by his friends (including young Charles Boyer), Michel refuses to attend his dying mother and, in a drunken fight, stabs the singer's lover. Then he tries to steal the money which his mother has entrusted to Djenna-he threatens both his sister and father with a knife but is entangled in a fishing net as he attempts to 98 36. (left) Roger Karl and Jaque Catelain in L'Hlomme du large ( 1920) 37. (right) The cabaret singer escape through a window. The father punishes him by putting him, bound and alone, into a small boat and pushing it off into the sea. Then he and Djenna separately take vows of silence-the one living in a sea cave, the other becoming a nun. Michel survives the ordeal and, one year later, returns a changed man; and in turn his father and sister are freed from their vows. Another kind of notoriety attached to the film, soon after it opened in Paris on 3 December 1920," when several sequences in the rough cafe were censored as excessively violent and sexually provocative. Much of L'Homme du large was shot by Georges Lucas and Jean Letort in the early summer along the southern coast of Brittany, from Quiberon to Penmarch.11 Like Antoine, L'Herbier's intention was to have "the sea as a protagonist." As he himself wrote of the experience, "the awesome subject of elementary forces, when they reach a high level of pure expressiveness, has been, is, and will be the privileged subject of the cin6matographe. However, just as in Les Travailleurs de la mer, there is an obvious contradiction in the film between the overwhelming images of the seacoast and village and those of the actors, who simply are not convincing as a Breton family. More sur- prising is the artificial nature of the frequent wipes, irises, superimpositions, and stylized intertitles supered over shots of the sea-taken at the time as a sign of L'Herbier's concerns as a literary aesthete, carried over from Rose-France (1919) and Le Carnaval de veritfs (1920).7 Despite these real and apparent limitations, L'Homme du large does succeed sometimes in creating an atmos- phere that is both convincing and aptly connotative-through its choice of seascapes and village landscapes (with Djenna's white figure, in LS, gliding along the dark walls and across the rocks), its low-key interior lighting (es- pecially during the mother's death and during Michel's pathetic theft), its views of actual Breton life, and its exotic seething cafe (where L'Herbier first seems to have used a juxtaposition of soft and sharp focus within a single 99 COMMERCIAL NARRATIVE FILM 38. Gabriel die Gjravone, man. :wm Phyllis Terry, Le Bargy, andL T-KD Ivor Novella (background1) in _____________________ shot). Perhaps Juan Arroy sums up the film best: "It produced exceptional visual harmonies, through the alternation of great waves of images, animated and aerated by the sea, and the feverish rhythm emanating from a sailors' cafe."''A After Antoine abandoned film production in 192 1, the leading proponent of a realist cinema in France, according to Rena Jeanne, was Louis Mercan- ton. Mercanton had been one of the first filmmakers, after Antoine, to adopt location shooting in earnest. By 19 19, he had developed a model mobile studio "of four motorized trucks and two trailers (carrying a generator capable of producing 12,000 amps) along with eighty lamps of all sorts. .. ."...With this newly devised equipment, he shot several films in England and France; for one, he even ventured out across the Mediterranean. L Appe! dii sang (1920), adapted from a Robert Hitchens novel, tells a story similar to L'Homnme dii large, but with a different class of characters, in a very different locale-the dry rocky shores of Sicily. It also combines a lot more narrative complications with a much less exciting visual style. The film was a resounding commercial and critical success-Cn'-pour-tous judged it one of the three best French films of the 1919- 192) season. 4 Even Louis Delluc, who appreciated "the open air" of Mercanton's films, found it more than the usual 'remarkable album of photographs" and praised the acting of its international cast.' The narrative (articulated in frequent intertitles) begins with Hermione Les- ter (Phyllis Neilson Terry), an Englishwoman who lives in Rome, deciding to marry a younger man, Maurice Delarey (Ivor Novello), unaware that her good friend, the novelist Emile Artois (Le Bargy), is deeply in love with her.' The 100 couple go to live at Hermione's villa in Sicily, and Maurice takes a liking to her servant, Gaspare (Gabriel de Gravone). Enchanted by the culture and by the sea (his grandmother was Sicilian), he is soon drawn into an affair with an island siren, Maddelena (Desdemona Mazza). Her fisherman father, Salvatore (Fortunio Lo Turco), begins to blackmail him; and when Maurice decides to break off the affair, the old man kills him and throws his body in the sea. Hermione discovers what has happened only when she confronts Maddelena grieving at her husband's grave. Released from his promise of silence to Mau- rice, Gaspare tries to take revenge on Salvatore; their struggle in his island cottage is interrupted by a knock at the door, and the old man accidentally shoots his own daughter. Hermione finally realizes Emile's love for her, and the two are engaged to marry-even though the novelist priggishly repri- mands her for transferring some of the flowers from Maurice's to Maddelena's grave. Viewed today, L'Appel du sang looks more like a slightly exotic bourgeois melodrama than a realist film. Although it was shot by Emile Pierre entirely on location, the images rarely suggest the ambience of life in Sicily. Many shots of the characters posed against the Sicilian hillside or seacoast look as if they could just as easily have been filmed on the Cte d'Azur or even in Southern California. Only occasionally is there an inkling of what attracts Maurice to the culture-the men dancing the "Pastorale"; the torchlight fish- ing (perhaps one of the earliest sequences of location shooting at night); the village festival, which ends in a huge fireworks display (culminating in a superimposed image of a flaming religious icon, smokestacks, wheel frames, and tiny silhouetted human figures). Maddelena, on the other hand, is at first more Aldonza than Dulcinea, but she gradually earns our attention and even our sympathy (she really does love the bloody Englishman). For brief mo- ments, both Maurice and Emile are given subjective images, yet none captures the emotional state of the characters as well as do either the shots silhouetting the illicit lovers in a small boat (framed by a sea cave) and, later, against the fireworks display (framed by a hotel room balcony) or the shot of the sea and rocky shore where Hermione first catches sight of Maurice's body in the dis- tance. Overall, the film seems more interested in clarity of action than in the evocation of an atmosphere or in the representation of emotional states. Yet despite its conclusion in a proper bourgeois marriage, L'Appe du sang remains ambiguously drawn to the tragic passion that Sicily awakens in its unfortunate hero. Mercanton also seems to have initiated a short-lived vogue for films shot in the region of Arles and the Camargue in southern France.44 There, in the early summer of 1920, he filmed Miarka, la flle a l'ourse (1920), with the famous actress Rjane, Ivor Novello, and Charles Vanel.45 Although the technical facilities of the on-location production work were not the best, Miarka seems to have been quite successful-largely because of Rejane's understated, ago- nizing performance which exacerbated her illness and led to her death just two weeks after the shooting ended.4G Soon after, Andre Hugon directed a string of low-budget films with Jean Toulout and Claude M6relle in the Camargue- Le Roi de Camargue (1921), Diamant noir (1922), and Notre Dame d'amour (1922).47 Even Antoine agreed to shoot his last film there, an adaptation of Alphonse Daudet's novel (not the popular play), L'Arl/sienne (1922).48 His interest was REALIST FILMS 101 COMMERCIAL NARRATIVE less in the story of passion and its attendant action on horseback than in the FILM landscapes of flat marshes and huge skies. But one decision directed audience attention elsewhere. The legendary Arles girl would actually be seen on the screen, played by Fabris, a famous nude dancer at the Casino de Paris!49 Some colleagues and critics were aghast, but L'Arlsienne proved to be one of An- toine's more profitable films.50 Another series of landscapes to be explored by the realist film was, first, the mountains of the Pyrenees and then, the French Alps. After at least one suc- cessful film set in the provinces-Le Retour aux champs (1918)-Jacques de Baroncelli seems to have made the first of the cycle of mountain films with his adaptation of Pierre Loti's 1897 novel, Ramuntcho (1919). The film was shot by Alphonse Gibory in the Basque region of the Pyrenees, in the area where the action of the novel was located. There, the engagement between Ramuntcho (Ren6 Lorsay) and Gracieuse (Yvonne Annie) is broken when he is sent off to Indochina to do his military service; subsequently, she enters a convent. When he eventually returns, in a final encounter, Gracieuse rejects her vows in order to renew their engagement.51 Henri Fescourt was especially struck by Ramuntcho when he returned from the war to begin working for Films Louis Nalpas.52 But Louis Delluc was even more enthusiastic: the film was successful precisely in the way that, unfortunately, Antoine's films were not. Here is a metteur-en-scene who has perceived the harmony between things and living creatures, the sensibility of a landscape, the distinctive light of a sky that is at once awesome and finely nuanced.3 Baroncelli has taken a great leap forward. Perhaps the entire French cinema will take that leap forward with the same conviction. We approach here that animated impressionism which will be, I believe, the unique property of the French cinema- the day when the French cinema will fully merit being called French.54 If the terms impressionist or impressionism meant anything to Delluc and his colleagues in the French cinema in 1919, they defined the emerging realist film. And if the French cinema was going to distinguish itself from the Amer- ican cinema that threatened to overwhelm it, here was one way to do so. Delluc himself attempted to follow Baroncelli's example in Le Chemin d'Ernoa (1921), whose narrative of criminal intrigue becomes secondary to a documen- tarylike description of hauntingly empty Basque landscapes." But Ramuntcho alone was not enough to interest the French in their moun- tain regions. That took the appearance of several Swedish landscape films, such as Sjbstrom's Les Proscrits (1918) and Stiller's Trisor d'Arne (1919) and Dans les remous (1919)-all shown in Paris during the 1919-1920 season-which much impressed them with the evocative power of their snowscapes.56 As a conse- quence, several filmmakers began to explore the mountainous regions of the French Alps, apparently unaware of the early German mountain films-e.g., those of Arnold Frank, who would become Leni Riefenstahl's mentor. In Jo- celyn (1922), one of the first French films to be shot in the Alps, Leon Poirier took advantage of the snowy wastes around Mont Pdlat for a number of dra- matic and rhetorical effects.57 Although Henri Fescourt called Jocelyn merely an illustration" of the famous Lamartine poem of the same title,58 and its 102 REALIST FILMS e , ~ \"~'~' 9. Mile. Myrga and a nMOLuntain village in Gei Cfle''u' story is set in the period of the French Revolution, the film is firmly rooted in the realist genre. Its narrative depends upon an unequal opposition between the revolutionary turbulence of Paris and the tranquil solitude of a mountain retreat, which allows for "a dramatic scherzo," as Poirier conceived it, with "a largo responding in echo. "5 In a mountain cavern, Jocelyn (Armand Tal- lier), who has fled the Terror, gives refuge to an orphaned boy, whom he later discovers is a disguised young woman (this Mary Pickford type is played by Mlle. Myrga). He falls in love, but that love is denied him by his religious superiors. Thinking she has been abandoned, the young woman throws herself into a life of pleasure in the Paris of Napol6on, as a way of forgetting Jocelyn. Years later, their separation is recapitulated in a single poignant encounter. Under her windows, the hero espies the woman he no longer is permitted to love; she appears on her balcony, leans there a moment and drops a rose which falls in the pebbled stream at Jocelyn's feet; he reaches for it, but the flower is carried off by the current, escapes the hand that pursues it." When she finally returns to the mountains, exhausted, to die with Jocelyn at her side, in Poirier's delicate tableaux, their souls are joined as one in his simple room and then in adjacent graves on a high mountainside. Impressed by Jocelyn's commercial and critical success (the conservative So- ci6t6 des Auteurs des Films voted it the Best French film of 1922), Gaumont asked Poirier to put together another film based on Lamartine. " Drawn from a lesser-known novel, Genevive (1923) was even more clearly a realist film. Uncommonly, unrelentingly pessimistic, the narrative is little more than a 103 COMMERCIAL NARRATIVE FILM series of episodes that chronicles the life of a pious orphan girl (Mlle. Myrga).62 She is refused in marriage; raises her sister only to have her die in childbirth; is unjustly imprisoned for abandoning the child; wanders in the mountains as a beggar; becomes the servant of her old fiance, who then dies in an epidemic; is restored by Jocelyn; and finally becomes head of a hospital where she redis- covers her sister's grown child. For the four months of location shooting in the region of Dauphin6, Poirier used nonactors consistently in the secondary roles.63 He also seems to have aimed for a painterly quality in his exterior and interior tableaux, which reviewers much remarked on at the time. "Betrothed, Genevieve looks like a Sargent; as a servant, a Holbein; as a traveler entering the village, a Millet."64 The overall effect, wrote Cinemagazine, was to produce a Corot-like "study of Nature's magnificence: wooded landscapes, fields, and mountains" that tempered the tragic experience of "the poor peasant girl. "65 Leon Moussinac, usually a harsh critic of Poirier's films, found in Genevive, a unity of direction, a very successful equilibrium. . . . Certain passages are remark- able: the mother's death in the beginning . . . the story of the [abandoned child and the servant's wanderings in the snow, among the most terrible landscapes imagina- ble.66 As an "atmosphere film," Genevieve may bear comparison with Henri Fescourt's Mandarin, also shot in the Dauphin6 region several months before, as well as his Les Grands (1924), which documented the life of several schoolboys aban- doned to their own devices during a vacation period in Aix-en-Provence.67 The best of these mountain films, and arguably one of the best realist films of the decade, was Jacques Feyder's Visages d'enfants (1925). As I mentioned earlier, this film was the victim of production disagreements and distribution caprice that caused a delay of two years in its release and hence a limited commercial run of scant success.68 "Simple, intimate, lacking any special at- tractions, stars, or prestigious sets," to quote Georges Sadoul,69 it deftly, even subtly, narrated a well-acted story of sustained emotional poignancy. French critics generally thought highly of it, and Feyder proudly noted, in his slim autobiography, that, in 1926, the Japanese press called it the best European film of the year.70 Visages d'enfants was Feyder's first original scenario since his apprenticeship days at Gaumont during the war. He wrote it specifically to use the talents of Jean Forest, the young actor he had discovered for Crainquebille (1923), as well as the location of the Haut Valois region of Switzerland. Basically the scenario integrates the character study of a young boy who must learn to accept a new stepmother and stepsister with the social study of an isolated Catholic com- munity's rituals and customs, in a landscape that alternately separates, endan- gers, and forces people closer together.71 In the village of Saint-Luc, in the Haut Valois, the mayor's wife (Suzy Vernon) dies unexpectedly, leaving her husband, Amsler (Victor Vina), with two young children, Jean (Jean Forest) and Pierrette (Pierrette Houyez). Jean is especially affected by his mother's death, and, when his father marries a widow (Rachel Devirys) with a daughter his age, Arlette (Arlette Peyran), he bitterly opposes their presence in the house. One day, while the family is returning home in the sleigh, no one notices as Jean maliciously tosses Arlette's favorite doll into the snow. When the girl goes to search for it alone in the evening, a sudden spring avalanche 104 REALIST FILMS 40. Jean Forest and Victor Vina in the village cemetery in Visag e d'en/ants (1925) buries her in a tiny mountain chapel. Although she is rescued, Jean is beside himself with remorse. He writes a note to his father and goes off to drown himself in the nearby river. Reading the note, Arlette tells her mother, who catches him just as he is about to be swept over a waterfall. Recognizing their concern and love for him, Jean finally relents and accepts his stepmother and stepsister. The first thing that, even now, impresses one about Visages d'enfants is the unusual authenticity of its natural and social milieu. It has one of the most beautiful and most efficient film openings of the decade. Several ELSs intro- duce the village, nestled beside a waterfall in a boulder-strewn mountain val- ley. There is an LS of the mayor's long house, constructed of hewn logs and ornately-carved dark wood, and then several shots of villagers walking up to it. In the large central room, the major characters are singled out (Amsler, Jean, Pierrette), and the occasion is finally identified-the funeral of the may- or's wife. Her coffin is carried down a flight of wooden stairs (a CU of shoes on the steps conveys the difficulty), and a HA shot describes its shadow pass- ing across the bare wooden floor. Outside, the funeral procession (the men and women are in separate groups; Jean and his father precede the priest) moves slowly down the narrow, stone-cobbled streets, past rows of log buildings whose first stories begin at the height of the winter snows. The cemetery is in a grove of trees some distance away, but several LSs there situate the village clearly in the background against the mountains. So consistent is the texture of the lighting and decor here (the cameraman was L.-H. Burel), it is difficult to distinguish the studio from the location footage. Smoothly, succinctly, and 105 COMMERCIAL NARRATIVE FILM 41. Arlette Peyran in Visages d'enfantsr (1925) in some detail, the sequence establishes a particular natural landscape, a social community drawn together in an important ritual, a family in that commu- nity, and, as we now see, one figure who is already given some psychological depth. The only thing missing from this description is Jean's subjective experience of the funeral. As soon as he is introduced, the sequence singles him out by linking his figure with POV shots and with various optical effects. Before the coffin even appears, a shot of his morose face is followed by a MS that blurs in as it tilts up to reveal his father. Other shots also seem to be his POV: that of his little sister blowing bubbles, the CU of shoes descending the stairs. In the HA shot of the coffin's shadow, he is the one who follows its path across the floor. Even the procession through the street is described in several han- dheld moving shots that either take his POV or dolly ahead of him. Finally, at the cemetery, when they start to fill in the grave, he is isolated in several CUs until a brief series of rapidly cut swish pans erupt. At the end, he faints, and his father has to carry him back to the house unconscious. This judicious pattern of subjectivity quickly determines Jean as the film's central character and his acute sensitivity to his mother's death-his inability to cope with her loss-as the film's primary subject. The overall narrative structure of the film is a model of rigor and balance. It divides into four major segments of alternating sequences, connected by transitions, all of which produce a narrative line that ascends steadily through two sets of complications to a doubled climax. The first segment of alternation separates father and son. At Amsler's request, the village curate takes Jean over the mountains (in a series of breathtaking shots) to Voissay in the next valley. In the meantime, Saint-Luc celebrates Amsler's wedding with a ritual dance. Finally, at the end of several alternating sequences, while the two rake hay one late afternoon, the curate tells the boy about the marriage. The mag- nificent mountain spaces, through which Jean first seemed to move harmoni- ously, now offer little solace. The mountains now stand as a barrier, dividing him from his family and community, and the lush hayfields of Voissay lie open to the sun in mute contrast to his misery. He is out of place, in both society and nature. As Jean returns alone now to Saint-Luc, the sequences that had narrated his father's wedding shift focus to the stepmother and stepsister Arlette. Here the pattern of alternation juxtaposes Jean and the two intruding women (the father is deemphasized) in different parts of the house. Jean soon finds himself dis- placed in his own home-Arlette uses his crayons and takes over his room. While she and her mother go about their work, Jean spends his time clinging to his mother's memory through such things as a portrait and some clothing stored in the attic. Jean's hostility to Arlette grows until it climaxes in the avalanche segment. After the sleigh ride, Arlette runs out into the snow to look for her lost doll while Jean turns restlessly in bed. The avalanche now separates her from the family/community, replicating the mountain barrier that had cut off Jean earlier. As Jean and Arlette, separately, pray to figures of the Madonna, the villagers search the darkness with lanterns and discover the chapel's cross projecting from the snow. This miraculous conjunction, across distant spaces, restores Arlette, but Jean still considers himself an out- cast for his 'criminal" act. 106 Three short sequences with all the characters together in the large central room serve as a bridge from this first climax to the concluding one. The distance that still separates Jean from the family is articulated in several HA/ LA shots, with the boy looking down from the top of the stairs to the others around the table. As he prepares to do away with himself, the final segment of alternation begins, intercutting the stepmother milking cows on the hill- side, Amsler driving off in a cart to the local lumber mill, and Jean saying goodbye to Arlette and asking her to give the note to his father. Appropri- ately, Jean heads for the stream, to drown himself in the run-off from the snow that buried Arlette earlier. When he reaches his destination, the alter- nation shifts to exclude first his father and then Arlette. In a final paroxysm of rapid cutting, Jean falls from a tree limb, and his stepmother wades into the swollen current above the waterfall to snatch him up as he is carried by. The final image of Jean and his stepmother, restoring the lost mother-son relation, is prepared for in an interesting pattern of rhetorical figuring. That figuring associates Jean with a number of objects, as if forcing him to come to grips with reality. His dead mother's possessions especially obsess him, either as a means of sustaining his closeness to her or as a sign of his separa- tion. At first, the most important of these is the portrait of his mother, which he takes into his new room and to which he prays daily. Early on, her image had even seemed to smile down on him momentarily. Then, over Sunday dinner, during the stepmother's first week in the house, Jean notices a brooch at her neck. When he breaks down crying, his father (even more ignorant than we are) orders him out of the house. Only as he lies in the fields do we understand the threat literalized in this object-a single subjective CU of the brooch tilts up to reveal his mother's face. Several days later, Jean tries to revive the presence of his mother through one of her dresses which he finds in the attic. In a brief, eerie scene (LS, FS, MCU), he literally fetishizes the dress, draping it over a chest of drawers against the wall, caressing it, and finally laying his head in its lap. When the stepmother finds the same dress and makes plans to use it for smaller articles of clothing, Jean savagely hacks it to pieces. Again the father shows surprising insensitivity and orders him out of the house. Either this lack on the father's part is so exaggerated as to be a projection of the son's (consistent with the Oedipal relation at the center of the film) or it raises disturbing questions about his actual feelings and attitudes. Jean's destruction of the fetish, to deny the transfer of desire, raises the possibility of his doing actual violence to one of the intruding women or even to himself. This comes to pass through another object that appears just prior to the mother's dress-Arlette's doll. Arlette takes the doll with her whenever she tends goats on the hillside. One day, when Jean brings her some lunch, he grabs the doll and attaches it to the horns of one of the goats, which begins to prance about wildly. His teasing has a disturbing edge-with the doll serving as a metaphorical substitute for Arlette, it raises the specter of vio- lence, even rape. The conjunction of this action with the fetishization of the dress that follows produces an extraordinary mirror image-of desire for the mother versus repression of desire (through violence) for either the stepmother or daughter who would replace her. Several days later, to punish Arlette for reporting his cutting of the dress to her mother, Jean throws away the doll, REALIST FILMS 107 COMMERCIAL NARRATIVE FILM as if to erase her image, her presence. The avalanche then literalizes the met- aphor by burying Arlette with her doll. Suddenly pushed to the extreme, the sexual symbolic of this rhetorical figuring is resolved, or displaced, by a reli- gious one. That displacement is articulated in the miraculous conjunction that restores Arlette. At that moment, Jean exchanges one mother for another, as the object of his praying. From the faded portrait of his dead mother, he turns to a small statue of the Madonna in his room. The disturbing sexual antago- nism within the family is suppressed, even in Jean's decision to commit sui- cide. For there in the stream the stepmother appears as a deus ex machina, the Madonna come to life and made flesh. The transformation is complete. In the raging waters, Jean is baptized and born again. Nature, the community, and the family are in harmony once more. And the portrait of the dead mother smiles down on them in the end, blessing the restoration. Feyder's achievement in Visages d'enfants was diminished by its commercial failure and then almost obliterated by another provincial realist film very much like it which was highly successful less than a year later. That film was Julien Duvivier's Poil de carotte (1926), whose scenario was adapted from a Jules Renard novel-ironically, by Jacques Feyder.72 Because it was a small produc- tion for Phocea, Duvivier shot the film in the Morvan mountains of central France rather than in the Hautes-Alps (which lessened the impact of the land- scape considerably).73 The narrative is complicated, focusing not only on a young misunderstood boy, Carrot-Top (Andre Heuz6), but on his deceitful brother (Fabian Haziza), his frightful stepmother (Charlotte Barbier-Krauss), his rather ignorant father (Henry Krauss), a cabaret singer who seduces the brother, and a family maid who befriends the outcast.74 Greed, cruelty, ug- liness, and a hypocritical piety are concentrated in this most stereotypical of stepmothers, one of the most offensive images of women in the French cinema of the decade. Furthermore, the film's resolution is oddly fractured: father and son are reunited, as are the stepmother and stepson; yet the reunions are separate, and the family does not seem to be restored. Despite all the conflicts and contradictions, this narrative is located in a rather detailed village setting, so that Duvivier is at least partially correct in calling Poil de carotte an "at- mosphere film."75 And, in 1932, he had the commercial acumen to make a sound film version of his scenario that was even more popular. The last two geographical areas to be represented by the realist film genre may well have been initiated by one of Louis Feuillade's more original works, Venddmiaire (1919). It starred two of his chief serial actors, Rene Creste and Edouard Math6, in a contemporary story divided into a "prologue and three parts (the vineyard, the vat, and the new wine)". 76 The narrative is predicated on the displacement of refugees (with two disguised German spies among them) from the war zone to the vineyards of Bas-Languedoc. The film stresses the allegorical, almost Biblical, elements of this narrative and operates accord- ing to an alternation of sequences that obsessively juxtapose war and peace. Yet, at times, it is surprisingly realistic, even lyrical-perhaps, as Francis Lacassin suggests, because Feuillade chose to set the film in the landscape of his own childhood.77 Particularly moving are the prologue that follows a boat of refugees down the Rh6ne river and a documentarylike sequence on the sense of community that develops among the peasants during the grape harvest. 108 REALIST FILMS r4 . A 1rrne still trom These two sequences seem to lead into two fairly distinct sets of realist films- the peasant film and the river or canal film. The many peasant films that were produced in the early 1920s tended to cultivate a feeling for landscape in the agricultural regions of western, central and southern France (the north and east having been destroyed by the war). Most of these films narrated rather simple stories and tried to document as economically and authentically as possible the land-holding bourgeois and peasant milieu.'8 Once again it was Baroncelli and Antoine who spearheaded the effort. Ba- roncelli seems to have initiated this cycle of films with Le Retour aux champ (1918), which Deliuc praised for bringing to the screen the peasant and th "truth" of his way of life with a characteristic 'impressionism." The foliow- ing year, Antoine decided to film an adaptation of Zola's La Terre 119;_ This time Antoine was uncompromising. As was his practice, he and L Burel shot the film entirely on location, in the Beauce region below Chartr, (presumably, in the summer and winter of 1919). The results Paul de la bore described as meticulously painterly, almost Millet-like, in their realism'( Thee were repeated HALSs which literally positioned the peasants as part ol tr earth, carefully composed "deep space" LSs sensitive to natural and man-ma. patterns in the fields and pastures, and many CUs of animals and rarmyar_ machines and objects that gave the film a material weight. Although Comeci'- Franqaise actors once again played the principal roles, the acting was uniformi% excellent, with Armand Bour particularly effective as Old Fouan. Above a'_ Antoine refused to dilute the unrelenting pessimism of Zola's story of green 109 COMMERCIAL NARRATIVE FILM 43. A publicity photo of the threshing sequence in None (1924) and deception among small landowners. An old farmer, Old Fouan, is slowly robbed of his land, his farmhouse, and his money by his own sons, "Buteau" (Jean Herv6) and Hyacinthe (Milo). At the same time, a young newlywed neighbor woman, Franqoise (Germaine Rouer), whose sister Lise (Jeanne Briey) has married "Buteau," discovers the plot against Fouan and is brutally at- tacked one day in the fields. In the final sequence, these stories come together in a cross-cut series of grimly poignant images. In a bitter snowstorm, Fouan drags himself across the parched earth that was once his until he collapses, while Fran oise dies alone in her simple husband's house, without denouncing or implicating anyone. Rene Hervil contributed two of his best films to the genre, both shot by Am dee Morin, L'Ami Fritz (1920) and Blanchette (1921), the latter adapted from a play by Eugene Brieux." L'Ami Fritz told the story of Fritz Kobus (Leon Mathot), a rich and propertied Alsatian, who falls in love with the daughter (Huguette Duflos) of one of his tenant farmers (de Max). Set in his bachelor ways, Fritz, after a long struggle, grudgingly relents to marriage. Disturbed by all the attention to what he considered a poorly crafted film (there was no electrical illumination and scant time to film), Hervil himself preferred Blanchette.82 According to Fescourt, Blanchette combined impressive winter landscapes and unusually natural acting (by Maurice de F raudy, Leon Mathot, Pauline Johnson, and Therese Kolb) with a reactionary subject-"the danger to a peasant family that comes from giving their daughter an education above their social level."" Much like his mentors, Antoine and Mercanton, however, Hervil was criticized for a rather monotonous mise-en-scene, an ab- 110 sence of rhythm in the editing, and a gaping discrepancy between the intrigue and "the depiction of a way of life."84 Several other peasant films seem to have avoided these shortcomings. One is only marginally about peasant life-Louis Delluc's L'Inondation (1924)-and its rhetorical and narrative complexity carry it well beyond the conventions of the genre. Others include Protazanoffs L'Ombre du pchd (1922) and a recent rediscovery that has yet to be fully reappraised-Rene Le Somptier's La Bite traqude (1923), adapted by Michel Carr6 from Adrien Chabot's novel, Marielle Thibaut.85 The best known, at least at the time, was Baroncelli's Nine (1924), probably because it was drawn from a Prix Goncourt novel. by Ernest Perro- chon.86 In it, Sandra Milowanoff played a peasant girl who cares for the house and two children of a widowed farmer named Corbier (Van Dadle). Threatened by a jealous hired hand, a scheming village seamstress (France Dh6lia), the drunken vengeance of her own brother Jean (Gaston Modot), and the blind egoism of her master, Nene is finally told to leave the farm; and, heartbroken, she tries to drown herself in a stream. Corbier rescues her and comes to his senses, accepting the woman whom his children now consider their mother. Some of the best moments in Nne depend on well-chosen natural locations. Having searched out the most suitable landscapes in the Vendee region of western France, Baroncelli lodged his cast and crew (including cameraman Louis Chaix) for a month of shooting in a tiny hamlet, "in the midst of actual peasants," where they took "their meals in a communal house."87 The results were remarkable, especially in the hay-baling sequence (where Jean loses an arm in the threshing machine) and in several dramatic encounters in the re- gion's vast wheatfields and pastures. And they were complemented by interior night scenes of minimal lighting and by poignant exchanges of looks in CU. Here, as in Ramuntcho, Baroncelli seems to have produced the synthesis of realism and impressionism that so marks the genre. Perhaps the best of these peasant films was L'Atre (1923), an almost for- gotten work by Robert Boudrioz, who had once worked before the war as a scriptwriter for Maurice Tourneur.88 With the financial backing of Abel Gance's new production company, Boudrioz (and his cameramen, Gaston Brun and Maurice Arnou) shot the film during the spring and summer months of 1920.89 However, the edited film was withheld from distribution for almost three years, perhaps because of the double whammy strategy of serials and super- productions which then gripped Pathe-Consortium.90 Although L'Atre was only marginally successful commercially, Cinia-Cin6-pour-tous's readers in- cluded it among the ten best films of 1923.91 The scenario for the film was drawn from La Chevauchee nocturne, a "peasant tragedy" written by Alexandre Arnoux.92 The story begins with a desperate young woman who, on Christmas Eve, abandons her baby girl at the farm- house window of an old peasant couple (they already have two grandsons to raise) and then drowns herself in a nearby stream. Several years pass. The younger boy gives the girl one of his whittled wooden dolls, and the two boys fight over her on their way to school. Days later, all three watch puzzled and terrified as their grandmother (Renee Dounis) dies in her chair. Again time passes. The elder son, Bernard (Charles Vanel), now does most of the work on the farm while his brother Jean (Jacques de Fraudy) loafs about and carves clay sculptures-for instance, an ugly bust of the cruel, oafish servant boy. REALIST FILMS 111 -14. (left) Maurice Schutz and Renee Dounis in L Atre (1923) 45. (right) A landscape publicity photo Jean has fallen in love with Arlette (another Mary Pickford type played by Renee Tandil), and Bernard watches them sullenly at a village dance. When Bernard threatens to leave the farm, his grandfather (Maurice Shutz) suddenly decides to make him his principal heir and marry him to Arlette; then he orders Jean off the farm. Heartbroken, Jean goes off to the city, where his sculptures eventually win acclaim at an exhibition. Arlette reluctantly agrees to marry Bernard, and the couple takes over the farm when the grandfather dies. However, Arlette's continuing love for Jean enflames Bernard's jealousy, and the two fight bitterly one day when he discovers that she has written to his brother. Learning that Jean is returning to the farm, Bernard spirits Ar- lette away to an abandoned mill. On reaching the farm, Jean threatens the servant boy (who has sided with Bernard earlier) and, when he rides off toward the mill, is shot in the back. He struggles into the mill just as Arlette is about to turn a knife against herself. Bernard's fury dissolves into shame at the sight of his brother's condition, and Jean "remarries" Arlette to his con- trite rival with his dying breath. Viewed today, in a 35mm nitrate print at the Cinfmatheque fran aise, L'Atre is something of a revelation. "Simplicity," wrote Ren6 Clair, "that is the quality, the chief quality of the best parts of L'Atre."' Indeed, its lovingly composed images (often framed in oval or arched iris masks) have a meditative, lyrical power that remind one of Tourneur as well as of F. W. Murnau's Sunrise (1927) and City Girl (1929). At the beginning of the third time period of the film, for instance, the grandfather admires his farm in a slow montage of images that ends with shots of a bullock plowing a field (including a CU of the plow slicing through the soil) intercut with shots of the old man picking up a clod of earth to hold in his hands. Just before he turns the farm over to Bernard, they move to the farmhouse window-where his vision becomes Ber- nard's-and the transfer is sealed in a CU of their clasped hands. This natural relationship between the peasant and the land even marks the sequence of the grandmother's death (handled in a few restrained shots of her sitting by the l 12 window, the children standing in the doorway, and the grandfather called back from walking across the fields and into a wood). Her funeral procession is described in just one long-take LS-stretched out along a country road, it is silhouetted against an immense sky as well as reflected in the still waters of a marsh. Jean's relationship to the land is strangely one of ignorance-even though he reworks the earth for his sculptures. Only when he is forced to leave does he really seem to see it-in one shot, he looks out sadly (from some trees in the foreground) over an unplowed field stretching off into the distance. But the image he carries with him and which finally inspires his art is a LS of Arlette, as she crosses the stream on a wooden bridge, carrying a water jar on her shoulder. Significantly, the figure seems to derive more from classical sculpture than it does from his life on the farm. Indeed, the most moving moments in the film are those articulating a sense of loss or tragic confrontation. The sequence that follows Jean's expulsion from the farm is particularly eerie. From the barn Jean longingly watches Arlette at her bedroom window, and his hand twists and retwists a rope at his side. When her light goes out, he puts a ladder against the house and climbs up and inside. The servant boy spots him and tells Bernard, who steals up the inside stairs and opens the bedroom door to see Jean crouched over Arlette (what was he going to do?). Stealthily they pull knives and face one another across her bed. When she moves in her sleep, they stare at her; the tension ebbs, and they go off silently to argue downstairs. At least two later sequences involve the poignant conjunction of separate characters' heightened emotional states. At the precise moment, for instance, when Jean is inspired in his work by the memory of Arlette carrying the water jar, she is looking through a storage drawer filled with the wooden figures he carved as a boy. The sight seems to rekindle her love; but when Bernard discovers her, he angrily tosses the figures into the hearth fire and, in CU, crushes one under his boot. After their fight in the kitchen (the staging is surprisingly brutal), Bernard leaves Arlette and walks out to the stream. For the first time, we begin to share his dejection as he sits (with his back to us in HALS), tearing up her letter to Jean and dropping it in the water. Simultaneously, Arlette writes another letter to Jean and starts when a cat suddenly appears in the doorway. Their separate miseries are ironically matched in back-to-back shots: a HALS of Bernard slouched beside the stream; a HALS of Arlette going through an arch in the village street, on her way to post the letter. Despite its power, the resolution of L'Atre is less than satisfying. "The sculptor hesitates" / "The farmer never does" / "The earth remains most sig- nificant"-so read the final intertitles. But the opposition has not been that simple. Jean and Arlette's passion for one another has been less responsible for the tragic chain of events than has Bernard's and his grandfather's blind ad- herence to the land and to its possession. In fact, the earth that feeds and protects the peasant also produces the blind passion that divides and destroys. The hearth fire that brings the children together in the beginning (where the boys replace the Christmas manger doll with Arlette) also serves later to shat- ter the bonds that join them. Although the film attempts to celebrate the earth and its bounty, it gives more attention to the loss and sacrifice that precludes its continued sustenance. Boudrioz once said in an interview that "the ideal purpose of the cinema" for him, was "to record as simply as possible REALIST FILMS 113 COMMERCIAL NARRATIVE the simplest, yet most powerful things."94 In L'Atre, he achieved that sim- FILM plicity with a degree of artistry that renders it darkly ambiguous. The landscapes and peoples of the French rivers and canals were the last to be explored by the provincial realist film. Less than a year after Feuillade's Vendimiaire, Antoine conceived the idea for a film devoted entirely to the life of the boatmen on the canals. His friend, Gustave Grillet, was persuaded to write a scenario, L'Hirondelle et la msange, which simply "followed the journey of a barge," from the canals of Flanders all the way to Paris.95 In the winter of 1920-1921, Antoine finally got permission from S.C.A.G.L. to film the scenario in Belgium, integrating his main actors, Henry Krauss and Pierre Alcover, with the crews of two canal barges.96 In an interview with Andre Lang, he gave this account of the experience: We left on our barge from Anvers and went up the Escaut to Bruges. . . . Magnifi- cent! . . . Since everything was shot during the trip, all the footage was in sharp contrasts. . . . The story was solid . . . a very simple drama . . . it ended with a man sinking out of sight in the cargo hold one night . . . and the next morning the barge threaded its way up river, quietly, in the sunlight and silence. . . . It was very beautiful. . . ."97 Yet the film was never released-for one of two reasons. Antoine told his son that the new directors of Path6-Consortium were indignant over the footage and refused to let him finish editing the film for distribution.98 Grillet's son, however, reports that, because of the constant fog and rain that hampered shooting, Antoine was not satisfied with the lighting-it was he who refused its release.99 Whatever the case, Antoine's son concludes, L'Hirondelle et la msange was to have been "a poem in images."100 Several years later, a number of films more or less fulfilled Antoine's am- bitions. In Le Carillon de minuit (1923), Baroncelli also went to Belgium to document the milieu of a canal town and its religious mission work. Rene Jeanne and Charles Ford especially praised Baroncelli's half-tone and gray mon- ochrome images for the way they perfectly complemented the story's discreet feeling of resignation. 101 In La Fille de l'eau (1925), Jean Renoir used the banks of the Loing river near Montigny to provide a painterly context for the begin- ning and ending of a picaresque series of wild adventures involving Catherine Hessling. But Renoir's film was much more than a realist film, and its exu- berant experimentation demands further discussion later. The most sustained evocation of barge life on the canals occurs in Jean Epstein's La Belle Nivernaise (1924). This charming film was adapted from Alphonse Daudet's popular 1886 novella-a simple, even banal, staccato-styled tale of an orphan boy adopted by a river barge family. In his scenario, Epstein updated the story to the present, altered the narrative structure a bit, added some dramatic conflict at one point, and generally "poeticized" his material.102 Here, in brief, is his version of the plot. On one of his biannual runs to Paris, Louveau (Pierre Hot) finds an abandoned child and, with police consent, takes him back to his barge, "La Belle Nivernaise," and his unsuspecting wife and daughter. Ten years later, the boy, Victor (Maurice Touze), has become "Pere Louveau's right arm" on the barge and is close to being in love with his daughter Clara 114 REALIST FILMS 46. The benediction ceremony at the end of La Belle Nvernaise (1924) (Blanche Montel). Called to Paris by the authorities, Louveau discovers that Victor is actually the son of Maugendre (David Evremont), one of his charcoal shippers on the Nivernaise canal (connecting the Loire and Seine rivers in central France). Sometime later, as he and his wife make arrangements to return the boy, Victor defends Clara from an attack by the barge mate (who has become jealous of his position) and barely saves the barge from crashing into a lock. Maugendre sends him off to a lycle-"to make something of him"-but Victor pines away from loneliness. Recognizing his error, Mau- gendr6 reverses his decision and offers Victor and Clara a river barge of their own, "La Nouvelle Nivernaise." Unfortunately, La Belle Nivernaise had only a modest success with the French public and disappointed many critics who were expecting to see some spectac- ular moments of rapid cutting which Epstein had become known for."'" Wrong- headed though that expectation may have been, there was reason for disap- pointment. For instance, the sequence that depends most on rapid cutting turns out to be one of the least effective in the film. This is the moment when the barge mate attacks Clara and the barge drifts dangerously close to a canal lock. Yet the motivation for this attack is hardly plausible, and the rapid cutting does little more than sustain suspense. Perhaps the sequence would work halfway effectively if projected with the music intended to accompany it; but, as is, the moment seems expanded to an exhausting length, to the point where discrepancies between the position of the barge in different shots are surprisingly obvious. 115 COMMERCIAL NARRATIVE FILM When La Belle Nivernaise does work effectively, which indeed is much of the time, it does so by refining the conventions of the realist film genre. Of all Epstein's early narrative films, it most privileges description over narrative, as if it were simply documenting a way of life or a character's emotional state in the context of a particular space and time. One sign of this is the frequency with which Epstein and his cameraman Paul Guichard employ LSs. Nearly all of the sequences on the barge include them, either positioning the characters in a deep space stretching from stern to bow or in a somewhat flattened space (shot from a 90° angle) as they walk along the barge edge. As a young boy, Victor is initially defined in LSs-as he walks back and forth in front of an 6picerie or sits by a lamppost in a narrow street (where Louveau happens on him). The best of these moments is a marvelous long-take ELS of an empty square with a small central park of bare trees, through which the boy slowly meanders in the early morning air, as if wavering between fear and awe. The combination of wide stretches of river and grassy fields, in the area where Maugendr6 lives, have a very different ambience in LS. Repeatedly, the char- acters come and go through the high grass, moving in harmony with its undulations. When Maugendre decides to return Victor to the Louveaus, the sequence of their agreement concludes in a magisterial LS of the group on the barge silhouetted against the fields and sky. The film finally ends in the same area with a benediction ceremony for "La Nouvelle Nivernaise," which in- cludes the implication of Victor and Clara's marriage as well (a bird cage passes from one to the other to be hung in a small arbor of vines and flowers). A HALS of the barge and surrounding countryside, as a priest comes on board, produces an image of harmony that is almost paradisiac. As early as Bonjour Cinema (1921), Epstein had echoed Delluc and others in writing, "A landscape can embody a state of mind. Especially a state of re- pose."104 La Belle Nivernaise gave him the chance to capture the spirit of one particular landscape-the Seine river between Paris and Rouen. This section of the river, along which most of the film was shot (in the late summer months of 1923), Epstein later described as "the greatest actor, the strongest person- ality that I have known intimately."'05 There can be little doubt that "the heart and soul-the living axis and artistic center" of the film, in the words of Paul Ramain, is the Seine.106 Its languid current provides the "landscape dance" on which the film's slow, limpid rhythm depends. It is a rhythm accentuated by the LSs, by the gentle landscapes and character movements, by the dissolves and graphic match cuts (e.g., a short sequence in the fog), but principally by a host of smooth tracking shots, taken from a camera mounted on the barge or on one close by moving parallel to it. Fixed and flowing at the same time, this camera movement has a descriptive rather than a narrative function, operating as an extension of the river itself. Those critics who ini- tially praised La Belle Nivernaise noted the rhythmic fluidity generated by its landscape and mediated by Epstein's camera; and they compared it favorably with the early masterworks of the Swedish cinema. 107 Ten years later, a similar landscape and rhythm, much inspired by it, would animate Jean Vigo's L'A- talante (1933). Despite the peaceful harmony of the film's ending, the river's presence is not without some ambiguity. In the beginning, for instance, it seems a rather neutral backdrop to the quarreling that initially marks the Louveau family. 116 Briefly, in the fog sequence, it even poses a physical threat. Perhaps the best evidence of this ambiguity, and of the hypnotic shifting perspective so char- acteristic of Epstein's films, is the first subjective sequence in the film, the family's departure from Paris after Louveau has learned of Victor's identity. The sequence opens with a fluid series of tracking shots from the barge (linked by dissolves): the river's edge, smokestacks, trees, cottages among the trees. Then comes another series (linked by dissolves again): a shot of open water, a CU of Victor's face, a MS of Victor at the tiller, a LS of Louveau (from behind) walking toward the barge bow, and a tracking shot (from the bow) of the river ahead. This measured pattern of movement seems to elide the time pass- ing and produce a continuous moment interrelating man and nature, with Victor easily controlling the barge in the river's current-except for one thing. Between the two series of shots comes an intertitle describing Louveau's emo- tional state: "Troubled, Louveau had hastened his departure from Paris." The intertitle makes the shots problematic. Are they Victor's POV shots (juxta- posing a visual to a verbal subjectivity) or an omniscient description which gradually becomes anchored in Louveau? Or are they both? Victor's relation to the river turns ambiguous-he is both subject and object, ironically una- ware that his movement on the river will separate him from it. As shots of the prow and water now alternate with shots of Louveau in the bow, the latter wavers between duty and love (in an intertitle): to give the boy up or keep him as his adopted son. He turns to look at Victor and Clara, who has joined him, and the backward-looking POV shots suddenly look almost like memory shots (since they echo the first images of the young couple together). When he turns once more to face the poplar-lined river ahead, he turns away from an image that is already past (it is the last he will see of Victor until the end). Although Louveau's look seems to dominate the sequence (as his choice seems to determine Victor's destiny), he, too, is determined-by the knowledge he bears and by the very motion of the barge and the river. His subordinate position within a larger design is confirmed in the final shots of the sequence: a LS of the barge bow, with Louveau as a small figure (frame right), and a FS of the barge mate (who has just written to Maugendr6), squatting by the rail. As in Jean Renoir's films, the lovely, seductive rhythm of the river sometimes carries with it an ominous undercurrent. By 1924, the realist provincial film reached a kind of apotheosis in two large-scale productions which combined the French love for the ambience of river or canal landscapes with their interest in the culture of Brittany. One was Leon Poirier's La Briere (1925), which was adapted from a recent Prix Goncourt novel by Alphonse de Chateaubriand and starred two of his popular actors from Gaumont, Armand Tallier and Mlle. Myrga. Although an inde- pendent production, its commercial status was confirmed by an exclusive month- long run (following L'Herbier's L'Inhumaine) at the prestigious Madeleine-Ci- n6ma.108 Yet the narrative was predicated on a bitter dispute over whether or not to drain marshland for a brick factory in La Briere, a famous salt marsh at the mouth of the Loire river in lower Brittany. Poirier shot most of the film around Saint-Joachim in the Briere region over a period of several months and chose an obscure actor named Jose Davert-who had appeared opposite Rene Navarre in La Nouvelle Aurore (1919)-for his central character, Aoustin, REALIST FILMS 117 I 47. (left) Jose Davert in La Briere (1925) 48. (right) A landscape publicity photo the old watchman of the marshes, who opposes the drainage plan." The result was a solemn, impeccably crafted, but somewhat uneven film which had a very successful run. For the most part, La Briere's rhythm is slow, magisterial, in correspond- ence with the calm, flat landscape of the salt marsh and the lives of its rustic inhabitants. With his large solid physique and expressionless weathered face, Aoustin becomes an icon of the people's determined resistance to change. His quiet, stolid figure is in perfect harmony with the serene marsh landscape he slowly poles his skiff across; and his silence intensifies the conflicts he has with his less obstinate wife and daughter, Theotiste, and her peasant fiance, Jeanin. At the film's preview in August, 1924, one viewer was heard to remark, "Doesn't it seem that you can actually see the silence.""') This documentary- like diegesis, as well as the film's technical facility, is disturbed, however, by several misplaced rhetorical conventions. For example, there is one sequence of accelerating montage to bring the daughter and fiance together (by means of a stampeding horse) and another to intensify the stalking climax between Aoustin and Jeanin. At three points, the film also abruptly inserts into the narrative a strange young woman (a variation on the Arles's girl?) in order to give support to Aoustin. In the end, as Leon Moussinac argues, the basic conflict underlying the film is simplified into a nostalgic contrast between "modern mechanics" and a "rough mysticism.""' The socioeconomic conflict is displaced almost completely into the family, where it is resolved tragically in a romantic, moralistic plot that turns the old man's rebellious vision into a form of blindness. Theotiste bears a dead child illegitimately; Aoustin loses a hand that is replaced by a wooden one; the salt marsh turns into bricks all because a daughter won't obey her father. 118 49. (left) Charles Vanel in Nscheur d"Ildande ( 1924) 50. (right) Sandra Milowanoff Even more prestigious than La Briere was Jacques de Baroncelli's Pfcheur dIslande (1924), which preceded Raymond Bernard's Le Miracle des loups at the Salle Marivaux and then continued for another month at the new Mogodor cinema."2 Pierre Loti's novel had been set in Paimpol on the northern coast of Brittany, and Baroncelli took his cast and crew there (again his cameraman was Louis Chaix) for several months of shooting in the spring and summer of 1924." For a while, the sailing ship Marie became a veritable studio in the harbor and on the seas; and the people of Paimpol joined the production to enact the story of what they considered to be their book. The story is a simple one. Gaud Mevel (Sandra Milowanoff), a young woman from Paimpol, falls in love with a fisherman, Yann Gaos (Charles Vanel), who is wary of reciprocat- ing her love. Only after she is orphaned in poverty does he relent and marry her, just days before he must sail on the Marie for a summer of fishing in the North Atlantic. Gaud waits for him into the late autumn; but the ship never returns. Jean Mitry remembers Ptcheur dislande merely as "an album of photo- graphs.""4 But Henri Fescourt recalls images "the crucifixes on the hills, the crosses in the cemeteries, the slopes of heather, the low fishermen's houses under sad cloudscapes, a vast horizon of waves"-which embody "a particular kind of sensibility as well as the social customs of the epoch when the novel was written that have persisted up to the beginning of the twentieth cen- tury.""5 Frame stills taken from the few surviving prints (which are all I have seen of the film) confirm its "carefully composed photography of the Breton coast and its affectionately observed details of the customs of Brittany. . . ."6 Several critics at the time wrote of its psychological acumen in depicting the characters, especially through the performance of Charles Vanel; while another was struck by the omnipresence of the sea, which, even more than in L'Homme du large, actually seemed to be the film's central character. " Baroncelli him- self saw the film as "the story of a fisherman and not the collective drama of all fishermen," acted out before "the deep enveloping backdrop" of the sea." 119 COMMERCIAL NARRATIVE FILM His words describing Loti's novel aptly describe his own film: "the views, the visions that follow one another like waves, are bound up, commingled, mul- tiplied in a 'simultaneism,' that expresses a profound psychological and mental truth. ..119 Much like Visages d'enfants and La Briere, this film was a quasi- documentary of a remote French society as well as a slowly accumulating, impressionistic tableau of character and landscape co-existing in harmony. So popular was Pcheur d'Islande that the readers of Cinia-Cini-pour-tous voted it the second best film of 1924.120 This success led Baroncelli to embark on a cycle of film adaptations of sea stories-Veille d'armes (1925), Nitchevo (1926), Feu (1927)--that one critic personified as "the various faces of the sea."121 But these latter are no longer realist films. Feu was a sea adventure about a gunboat captain during the recently ended colonial war against the Rifains in Morocco. And Nitchevo was a drama about men trapped in a submarine off the French naval base at Bizerte, Tunisia. As Baroncelli shifted away from the genre of the realist film, he found himself working on films set up to advance the careers of young actresses as well as relying on the facilities of Cineromans' new studios. The representation of Brittany's seacoasts and the life of its fish- ermen was left to younger filmmakers such as Jean Epstein and Jean Grmil- lon. On the basis of this survey of the genre so far, the French realist film offers an interesting contrast to the realist cinema that developed in Germany about the same time. Perhaps because the German film industry modernized its studios more quickly and on a grander scale than did the French, their films tended to be shot in studio sets that reproduced reality. There were exceptions, of course-Lulu Pick's Shattered (1921), for example, and Gerhard Lemprecht's Children of No Importance (1926-but nearly all, whether shot on location or in the studio, gave representation to a distinctly urban milieu. 122 Influenced by Kammerspiel theater, perhaps especially through Carl Mayer, and to a much lesser extent by Expressionist art, architecture, and drama, these realist films created city spaces that were enclosed, intimate, and sometimes claustropho- bic.123 Their preoccupation with rendering Stimmung (mood) seems to have depended heavily on the use of artificial light. 124 By contrast, the French films tended to open out into expansive (if no less deterministic) landscape spaces and to rely on recording the nuances of natural light. The German realist films were tied more to the lower and middle classes of the city (often a limited number of characters living in an everyday ambience), which gradually led to an important subgenre of street films during the last half of the decade. 125 The French realist films, on the other hand, found their locus more often in the natural landscapes and villages of the provinces and in their bourgeois, peas- ant, and artisan classes. This does not mean that there were no French realist films whose environ- ment was the city, but they were less numerous and, with a few exceptions, not as popular as the provincial realist films. Toward the end of the war, at least three films grounded the genre in the milieu of the city, in the modern. The earliest of these was Antoine's Le Coupable, which was interrupted in its shooting by the battle of the Somme (during the summer of 1916) and then not released until 1917.126 Because the surviving print of the film lacks inter- titles, its complicated narrative, which alternates a court trial with a past story 120 REALIST FILMS 5 i . A frame still from L C oupable (1) 17) in which an abandoned woman (a flower girl) bears a child and later turns to crime, is now close to incomprehensible.1" Yet Le Coupable still holds some interest for its grim images of wartime Paris (the cameraman was Paul Cas- tenet) and for its deliberate contrast of the bourgeois and little-known prole- tarian milieux. Ren6 Predai, in fact, calls it "the first authentic {film} docu- ment about Paris.""125 Another quasi-realist film set in the newly built industrial belt around Paris was Henry Roussel's adaptation of a propagandistic war novel, L'Ame a'u bronze (1917). According to Georges Sadoul and others, most of the film was "an abominable melodrama." "9 Harry Bauer playeci an engineer who let a riva fall to his death in a crucible of molten bronze. After the metal is forged into a cannon, the engineer is called to serve in the war; there he is forced to usi the very same cannon in a climactic battle. Despite this heavy-handed irony, the film's initial sequences created a sensation. They were shot entirely in a huge ironworks factory, by Paul Castenet again, in an almost documentary fashion. Louis Delluc was particularly struck by the evocative authenticity o; this setting: The most powerful character in L Ame du bronze is not an actor; it is a facrory. Astonishing images reveal to us the modern marvels of metallurgy. It is as beautiful as a page of Verhaeren. Here is a real poetry, the best that the cinema can aim fo:. . . ." Probabiy the most important of these early city films was Henri Pouctai'z eight-part adaptation of Zola's Travai', the most publicized and probabiy tne most successful French film of the 1919-1920 season. ' According to jean 121 52. Leon Mathot in a factory location in Trcvail (1919- 1920) Galtier-Boissiere, Pouctal smoothly resolved the problem of integrating two different film modes-"the documentary and the scripted or dramatic film.""' The workers looked and acted like workers, whether in the factory or in the bistros and poorly paved streets surrounding it. And the chief actors (Leon Mathot and Huguette Duflos) seemed to blend in with them. Most impressive was the factory itself, where much of the film was shot by Louis Chaix: the factory in action, with its tall smoking chimneys, with its pulsating entrails . . . the blast furnaces, the streams of incandescent iron, the blind steam pistons, all this gigantic machinery, around which scurry, sometimes illuminated by the light of a purple flame, the human pygmies, the galley slaves of industry.' Jean Mitry argues that, with its well-composed images, its detailed atmosphere, its relative authenticity, Tra- vail helped orient the French cinema toward subjects inspired by everyday life, illus- trating the social problems of the proletariat as opposed to the pleasures of an idle class.' 122 Viewed today, at least in the well-preserved tinted print of chapter one, Tra- vail's tableaux sequences of human figures struggling to achieve parity in an industrial landscape are still impressive.135 Yet the film's explicit though muted class conflict and resolution (already sentimentalized in the manner of Lang's Metropolis [1927}) is masked by the melodramatic story of a victimized young woman, an orphan, and her eventual marriage into wealth. Reformist through- out in its ideology, Marcel Oms writes, Travail articulates its sense of national reconciliation, capitalist-worker accord, and social justice in the context of a moral rejuvenation. An even more simplified ideology marked Rene Le Somp- tier's La Croisade (1920), in which a "reborn" factory owner (a returning war veteran) was pitted against an anarchist worker who conveniently goes mad at the end. 136 On the strength of Travail's success, several producers tried to turn other works of nineteenth-century urban fiction into profitable realist films. After its American-style productions failed to find American distribution in 1920, Film d'Art assigned Baroncelli to direct an adaptation of Balzac's Pere Goriot (starring Signoret), which was released in 1921. Although some critics praised Pere Goriot for its detailed reconstruction of the shabby Vauqier pension, the scenario did not allow Baroncelli to exercise his sensitivity to natural land- scapes.137 The film was but the first of a series of studio-bound adaptations of Balzac that included Gaston Ravel's Ferragus (1923), Jacques Robert's Cousin Pons (1924), and Max de Rieux's Cousine Bette (1924).138 Even less distin- guished film adaptations were made of several Zola novels-Bernard-Des- champs's La Nuit du 11 septembre (1920) and Marsan and Maudru's L'Assommoir (1920).139 Yet the latter, Louis Aubert confessed to Andre Lang, was one of his most profitable films that year. 140 Several realist films from this period actually straddle the distinction I have been making between the provincial and urban environments by locating their stories in a major city in the provinces. Louis Delluc's Fievre (1921) and Jean Epstein's Coeur fidele (1923), for instance, are both set in the down-and-out waterfront sections of old Marseille. Drawn from original scenarios, both have simple, rather melodramatic narratives, predicated on love triangles; and both make much of the atmosphere of their milieu, though by different means. Delluc was confined to a Gaumont studio but had a finely detailed set designed by Robert Garnier. Epstein was able to film in several deserted areas of the Marseille harbor and created a famous sequence of rapid cutting out of a local carnival. In rhetorical complexity, however, the two films deviate quite a bit from the standard realist film. Certainly the most impressive of these genre deviations was Abel Gance's La Roue (1922-1923). The scenario was inspired initially by Pierre Hamp's working-class novel, Le Rail, but it soon took off from its source in several different directions. At first glance, La Roue seems like an epic, highly mel- odramatic modernization of the Oedipus story. Its subject is a locomotive engineer named Sisif (Severin-Mars) who falls in love with his adopted daugh- ter, Norma (Ivy Close), whom he had rescued as a girl from a train wreck in Nice. To protect her from his obsession, and even from his son Elie (Gabriel de Gravone), he marries her to a wealthy railroad inspector. Still maddened by guilt, he fails in several attempts to commit suicide and then is blinded REALIST FILMS 123 53. (left) Gabriel de Gravone and Ivy Close in the railyard house in La Roe (1922-192 3) 54. (right) Severin-Mars by an accidental blast of steam from his own locomotive. Demoted to running the funicular railway on Mont Blanc, the aging man is confined with his son to a mountain cabin. When Norma and her husband visit a nearby resort, Elie is drawn into a fight with the husband, and both are killed. After blaming her for his son's death, Sisif finally relents and allows Norma to live with him. In the end, while she joins in the peasants' spring dance in the snows, Sisif dies peacefully at his cabin window. There are several features which distinguish La Roue-its moments of sub- jectivity, its techniques of rapid cutting or accelerating montage and their multiple function, its patterns of rhetorical figuring, its doubled narrative structure-all of which carry it well beyond the standard French film. But one feature is of particular interest here. That is the hybrid form of its realism, combining elements of both the urban and provincial realist films. In the first part of the film, this realism stems largely from the location shooting in the Gare Saint Roch railyards near Nice, in which Gance tries to reproduce the proletarian ambience of the railway workers' milieu. In the final section, how- ever, that realism depends on the location shooting in the French Alps. Here, the mountain landscape, which initially dwarfs Sisif and accentuates his suf- fering, is finally transformed in the peasants' dance in the vast expanse of the snow fields. Somewhat as inJocelyn, a human community has become one with a natural landscape. Most of the commercial urban realist films, however, were different from La Roue. For one, they focused exclusively on the French capital of Paris. In some, the idea of documenting the daily life of the city became as important as narrating the story. Leon Poirier had wanted to make just such a film as early as 1921. I dreamed of a great symphony about Paris, modern Paris, the magnetic pole, the human crucible which produced both geniuses and crimi- nals. The subject was vast."- 1 Poirier's prologue had already been shot when Gaumont decided that he could not release two films with the same title- 124 55. A street scene publicity photo for Paris (1924) and Louis Feuillade's Parisette would certainly make more money. Perhaps the first film to actually employ the city's name alone in its title was Rene Hervil's Paris (1924), which was based on an original scenario by Pierre Hamp. Ac- cording to Cinia-Cine-pour-tous, Hamp's scenario constituted a sort of "synthe- sis of modern Paris, the city of work as well as the city of amusement."" It followed several different characters-a dressmaker's assistant, a fruit-seller, a mechanic, an industrial magnate, a music hall star, a fashionable young woman, and an inventor as they circulated around Paris. " "The cinema too often forgets its real goal," Hamp said during its production, "which is life itself and the reproduction of life. "4 Unfortunately, though quite profitable in Aubert's cinemas, Hervil's film seems to have sacrificed its documentary mode and exaggerated its plot that called for the hero (Alibert) to foil the theft of the inventor's designs for a new locomotive and then to convalesce side by side with his fiancee (Dolly Davis). 14 By 1925, the urban realist film, much like its provincial counterparts, was undergoing a transformation. The change becomes clear if one examines several 125 56. (left) Jean Forest in Crainquebille (1923) 57. (right) Maurice de Feraudy films directed by Jacques Feyder and Jean Epstein-Crainquebille (1923), L'Af- fiche (1925), and Gribiche (1926). With Crainquebille, Feyder was taking more than the risk of independent financing for a small film. The film trade press considered the classic Anatole France story on which it was based too interiorized, too subjective, for the cinema; and they thought that audiences would remember only too well Lu- cien Guitry's popular performance of the old man's role in its theatrical ad- aptation. li6 Crainquebille is "a simple human story" (says the American version of the film) about an old vegetable cart vendor (Maurice de F6raudy) who has worked the rue Saint-Antoine area of Paris for nearly forty years. '- One day, a gendarme misunderstands the old man and arrests him, thinking he has been insulted. Powerless in court, Crainquebille is sentenced to several weeks in jail, which he quickly adapts to-as he does to most things. After his release, however, his old customers (petit-bourgeois shopkeepers) reject him, and he takes to drinking. Finally, one rainy night, when on the point of throwing himself in the Seine, the old man is befriended by "The Mouse" (Jean Forest), a young orphan who lives in an abandoned shed near the river. This is a neat reversal of the conventional orphan plot denouement-the plucky kid (lighting a cigarette butt at the end) saves the old man from suicide and restores his desire to live. Crainquebi//e opened in March, 1923, and, despite the odds, soon was suc- cessful-worldwide-and even influential. - For instance, it helped persuade Albatros to produce a modern adaptation of Le Chiffonnier de Paris (1924), starring the Russian emigr6 actor, Nicolas Koline, in one of his best roles as Pere Jean (a role that Fred6rick Lemaitre had made famous in 1847).' No doubt Crainquebi//e's success depended to a great extent on the performances of F6raudy and Forest, who was making his acting debut. When the film was previewed for a special audience of vendors from Les Halles, Lucien Doublon exclaimed, "That's Crainquebille himself!"'"" Even the vendors acknowledged 126 the authenticity the two actors gave to their characters. But Crainquebille was exceptional, too, in L.-H. Burel's location shooting around Les Halles and in the Marais (then still one of the oldest communities in Paris). 151 Particularly evocative of this milieu are the sequences of Fraudy pushing his cart through the narrow streets, an early one of the boy selling newspapers, and another of the old man drinking at a cafe while outside the door vendors gather in the huge market area. In the film's opening and closing, some of the shooting was done at night-for the sequence of market wagons rolling into the city and for Crainquebille's suicide attempt. This combination of naturalistic acting and location shooting gives Feyder's film adaptation of the story a significance that synthesizes the social and psychological-the naivet6 and supposed infe- riority of the main character balances the indifference and inhumanity of the social order that victimizes him. Some film historians have accepted Crainquebille as an avant-garde film, because, in several sequences, it resorts to subjective fantasy in depicting the old man's victimization. Although these moments deviate from the film's dominant mode of representation, they do so in a way which is somewhat different from the narrative avant-garde film practice. Both moments occur during sequences when Crainquebille is in unfamiliar surroundings-in the courtroom and in jail. There is some logic and consistency, consequently, in the fact that they are set off from the representation of his everyday life. The first moment represents his perception of the trial-articulated through blurred, distorted shots (of the judge and prosecutor), multiple exposures (of a gigantic gendarme and a tiny defense witness), mise-en-scene tricks (a statue bust turns to look down on him), and camera movement (an erratic dolly shot through the courtroom crowd). The second moment represents a nightmare that as- saults him after he has been convicted. Though the rhythm of its editing is ragged, this sequence is quite unique since it includes not only slow-motion shots and superimpositions of the tiny defense witness and judges (who leap down from their benches) but also negative images (of the chief judge) and a brief flurry of rapid cutting. There is some question whether this second se- quence actually was part of the original film. 152 Even if it was, however, the nightmare serves little function in the film's narrative. It simply repeats, in different images, the terror Crainquebille experienced during the trial, and it has absolutely no effect on nor parallel with any later sequence in the film. The subjective moments of the trial may function with slightly more justifi- cation. Yet they serve less to reveal something about Crainquebille (his igno- rance of the legal system and his terror before "superior beings") than to narrate the course of the trial by aggrandizing one side at the expense of the other (even across social lines-for instance, the defense witness is a gentle- man). Consequently, both sequences seem roughly imposed on the film. All in all, however, Crainquebille is probably the best of the 1920s realist films about the French capital. Upon seeing it in New York, D. W. Griffith is said to have remarked, "I have seen a film which, for me, precisely sym- bolizes Paris."53 A different kind of near tragedy and amalgam of modes marks Epstein's L'Affiche (1925), which was drawn from a rather sentimental scenario by his sister, Marie. 154 Albatros was pleased enough with its success to ask the pair to duplicate their efforts six months later in Le Double Amour (1925).1155 L'Af- REALIST FILMS 127 58. (left) Nathalie Lissenko in the opening of L A/ich (1925) 59. (right) The riverside cabaret fiche tells the story of Marie, a Paris flower girl (Natalie Lissenko), who is disturbed to realize that her youth is passing and so spends the night with a young man (Genica Missirio) she meets one evening at a riverfront cabaret. He turns out to be a rich playboy who disappears the next morning, leaving her pregnant. Once her child is born, she is desperate for money and enters its photo in a real estate company's contest for Paris's "most beautiful baby." She wins the 15,000 francs prize, but, soon after, the baby sickens and dies. Her grief becomes all the more poignant when an advertising campaign covers the walls of Paris with thousands of posters of her baby. Once the company director (Camille Bardou) recognizes her plight, it is revealed that the man who abandoned her is none other than his son, who now realizes his "sacred debt of love" and proposes marriage. Compared to Cr inquebi//e. L'Affiche integrates the subjective experience of its central character into its diegetic flow almost effortlessly. The opening creates a simple emblem of Marie's desire: a series of superimpositions in which she lifts her arms (in FS) through a revolving floral design and then (in CU) looks at something just off-screen and starts. Thereafter, the subjective appears at crucial moments, as in the distorted shots of a cross when she visits her child's gravesite; " but the most effective moments occur when she con- fronts the multiple images of her dead child, from which the film derives its title. In these sequences, Marie's own memory becomes visible, concrete; and her inner torment literally echoes, as in a nightmare, throughout the real world of Paris's streets. Even more crucial, however, to the development of the realist film genre is the vivid, if facile, dichotomy that L'Affiche sets up between proletarian and bourgeois milieux. The one is characterized by the simple decors of Marie's workplace and apartment and, especially, by the riverfront cabaret. For there, in a leisurely montage of LSs, natural landscape and social community seem to co-exist in harmony. " The other is characterized by stark, spacious offices, restaurants, and townhouse interiors. For these interiors, Boris Bilinsky's black- 128 60. (left) Lochakov's office decor for L /\ithe (1925) 61. (right) Jean Forest in a Lazare Meerson interior in GrIlhhe ( 1926) and-white decors, which are echoed in the repeated motif of checkerboard floors, seem cold and sterile, almost antagonistic. Yet it is here that the nar- rative is resolved, in a highly conventional fashion, exactly as in the senti- mental serial or bourgeois melodrama. Could Epstein be concocting a delib- erately false happy ending? A similar dichotomy marks Feyder's Gribiche, which opened en exdusiviti at the Aubert-Palace in late March, 1926, and which soon became even more popular than Crainquebille in France. 1" This time, the film belongs almost entirely to Jean Forest, whom one viewer described as a French synthesis of Charles Ray and Jackie Coogan. l5" The narrative, based on a short story by Frederic Boutet, is an interesting variation on the orphan subject. '" A young boy accepts adoption by a rich American woman (Fran oise Rosay) so that his working-class mother (widowed in the war) can remarry. Puzzled and intim- idated by his new environment, and then repulsed by it, the boy escapes during the Bastille Day festivities and wanders through Paris at night to his old apartment, where he is discovered and reunited with his mother and her new husband. Somewhat facilely, the film juxtaposes the working class and the wealthy, the French and the American-and clearly celebrates the one at the expense of the other. The most effective articulation of this idle fantasy comes in the story of the boy's returning the American woman's lost purse, which sets up the adoption. The woman tells the story several times to differ- ent friends, and each time (presented in flashback) it becomes more exagger- ated, more melodramatic, until finally the boy is dressed in rags, and his mother is nearly dying of starvation in a garret! Many film critics praised the film's sensitivity in conveying the boy's aware- ness of his own and his mother's situation. " Except for one awkward section of rapid montage in a fairground sequence (imitating the carnival sequence in Epstein's Coeur fidele), the initial and concluding parts of Gribiche, shot mainly on location, still confirm that judgment. The sensitivity falters, and the ide- ological contrast is nearly reversed, in the long middle half of the film, which 129 COMMERCIAL NARRATIVE is played out in the modern decor of the wealthy American's apartment, de- FILM signed by Lazare Meerson. As photographed by Maurice Forster and Maurice Desfassiaux, that decor is both sumptuous and sterile, fascinating and off- putting. The reason seems to lie in an apparent contradiction between the mise-en-scene and the narrative. Although the film celebrates Meerson's craft (its release coincided with the famous architectural designs at the Paris Ex- position des Arts Dcoratifs), it also condemns the fictional world his decors represent. In its heavy reliance on studio sets rather than on location shooting, Gribiche, even more than L'Affiche, clearly evidences the enervation of the ur- ban realist film in the face of growing interest in the modern studio spectac- ular. After 1925, the simple realist film nearly disappeared from the French cin- ema programs. Adaptations from nineteenth-century fiction continued to be made, of course, but now they assumed a different form. Some, such as Jean Renoir's adaptation of Zola's Nana (1926), were expensive, finely detailed historical reconstruction films as well as examples of the growing pattern of French-German co-productions. Others, such as Marcel L'Herbier's L'Argent (1929) and Julien Duvivier's Au bonheur des dames (1929), both contemporary adaptations of Zola novels, were important films in the new cycle of modern studio spectaculars. Several films, however, despite their conception as large- scale productions, attempted to maintain some degree of simplicity and some attention to either natural landscapes or bourgeois and proletarian or peasant milieux. The most prestigious of these was Cin6romans' big film for the 1925- 1926 season, Henri Fescourt's adaptation of Les Misrables. Initially, Cin6romans wanted to reduce Victor Hugo's novel to a single film like the rest of its Films de France series. But Fescourt, who had been assigned to direct it, fought determinedly to win acceptance of his own adaptation "in four parts, each of which would correspond to a complete film."162 Those four parts closely followed the divisions of the novel: "L'Evasion de Jean Valjean," "Fantine et Cosette," "Marius," and "L'Epoque de la rue Saint-Denis." Fes- court was also allowed complete freedom in choosing his large cast. The only difficulty arose when he selected a rather unknown actor, Gabriel Gabrio, for the leading role. However, Gabrio proved to be "an excellent peasant at heart," wrote Fescourt, "he was Jean Valjean, Jean Valjean completely. . . . Victor Hugo would have been pleased."163 Finally, Cin6romans budgeted the project at the considerable sum of six million francs, which allowed Fescourt to shoot much of the film on location. 164 Four months into the shooting, however, the sudden bankruptcy of the German company, Westi, led to a serious curtail- ment in finances. In order to finish the film, Cinromans frantically cut corners everywhere, which is why, according to Fescourt, certain sequences in the last two parts seem sketchy. "Authentic town centers, authentic streets were to have served as a frame for the riot scenes: we recreated them in the studio. . . . That is the reason for the inequalities in the film."16 Most audiences did not seem to mind, for its success, which was worldwide, exceeded all expectations; and it won a 1926 Gold Medal from Cinmagazine's Amis du Cin6ma.166 The narrative of Les Miserables covers almost a twenty-year period, from the end of Napoleon's reign to the Paris insurrection of 1832.167 During that time, Jean Valjean, a peasant and convicted thief in Provence, becomes a small 130 textile factory owner and the mayor (by changing his name to Madeleine) of the northern town of Montreuil-sur-Mer, and then is exposed and reduced to factory work and thievery again in Paris. Paralleling his story is that of Fantine (Sandra Milowanoff) who, abandoned with a child after the battle of Waterloo, has to turn to factory work and, eventually, to prostitution in order to pay the money-grubbing Thenardiers (Georges Saillard and Renee Carl) to raise her daughter Cosette (Andree Rolanne). Once she dies, Cosette is protected and supported by Valjean until she grows into a young woman (Sandra Mi- lowanoff again) who falls in love with Marius (Franois Rozet) and is taken into his aristocratic family. Valjean's chief antagonist during much of this time is the grim soldier-turned-lawyer, Javert (Jean Toulout), while his spir- itual benefactor is Monseigneur Myriel (Paul Jorge), a priest in Digne who inspires him initially to renounce his criminal past. In contrast to most previous realist films, Les Misrables is concerned pri- marily with the slowly building action of its narrative and with the conflicts that logically determine its course. "For Fescourt," write Beylie and Lacassin, "the cinema remained essentially the art of narrative; all the embellishments (the decors, the photogenic quality of the actor, the montage) should only serve, and never supersede, the narration."168 As a consequence, the film's mise-en-scene (decors, costumes, lighting) is consistently simple, even prim- itive, in design. Interiors (whether they be the priest's rectory, Javert's office, or Marius's townhouse) are spare and unadorned; the lighting tends to be full and even. The editing relies heavily on a conventional continuity system of establishing shots and then closer shots of the characters' interactions. Long shots and full shots predominate, while middle shots and close-ups are reserved for moments of tension or conflict and for individual character reactions (as when Madeleine/Valjean saves Fantine from arrest by Javert). For such a lengthy film, Fescourt keeps the narrative under rather tight control, which is one reason Jean Mitry still thinks highly of it. 169 For instance, Fantine's descent in Montreuil-sur-Mer closely parallels Valjean's systematic rejection by the townspeople of Digne after he gets out of prison; and her rescue by Madeleine echoes Valjean's rescue earlier by Monseigneur Myriel. Even in the flashbacks, which economically narrate certain stages of Valjean's life, actions are paral- leled: Valjean frees a mired wagon in the prison rock quarry; later as mayor he lifts a wagon that has tipped over on a man. Finally, in the, climactic last reels, Fescourt and his editor, Jean-Louis Bouquet, intercut several different actions together as Valjean's story merges with the Paris insurrection: the fighting at the barricades (including the death of a boy named Gavroche and the wounding of Marius), the arrest of Javert as an Orkanist spy, and Valjean's flight through the Paris sewers with the injured Marius on his back.170 Despite the primacy of its narrative, Les Miserables does evidence some con- cern for atmosphere and for the evocative power of certain landscapes. The opening sequence functions much like a documentary of a quiet, peaceful southern French provincial town. But when Valjean first appears, he is framed against bare trees that are "bowed down and splitting like the weight of his own past."'171 Ironically, the town rejects him, and, after he has stolen from the priest and been forgiven, he is sent back out into the same mountain landscape to redeem himself. Sitting by a gnarled tree on a mountain top, he undergoes a long night of penance. If natural landscapes provide a simple REALIST FILMS 131 ) 62. (left) A street in Montejil-sur-mer in Les is 1/rah/es ( 1925-1926) 6 (right) Gabriel Gabrio in the opening of Lfa Lisrer< Needless to say, Carmen was Raquel Meller's first commercial flop. The most successful of these big films to reconcile atmosphere, narrative, and star apparently was Feyder's Thirise Raqiuin (1928), which opened at the Cin~ Max-Linder in September, 1928."'8 The film was produced by DEFU- First National and put together entirely in DEFU's Berlin studios. But Feyder made it truly a French-German collaboration with a contract that allowed him to rewrite the prepared scenario adaptation from Zola's novel and to use a half- French, half-German cast: Gina Manes (Th rese), Jeanne-Marie Laurent (the mother), Wolfgang Zilzer (Camille), and Hans-Adalbert von Schlettow (Lau- rent)."'' Feyder chose a structure, according to L on Moussinac, that allowed him in the first half, to create an atmosphere, to resurrect, while critiquing it, the milieu where the heroine is formed, to evoke the daily life that thoroughly conveys the petit- bourgeois spirit against which Th6rese is in continual revolt. . . .18 This sordid, oppressive atmosphere was chiefly the work of Andre Andrejew, a Russian emigr6 set designer (who would later do Pabst's Three Penny Opera), and the German cameramen, Frederick Fugelsang and Hans Scheib. 's" In the stills that remain of the film, the Expressionist/Kamnierspie/ elements of the decor and lighting are carefully blended according to an overriding concern 134 66. (left) Gina Manes and Wolfgang '/ilzer in Ther6e Raquin ( 1928) 67. (right) The domino party, with Jeanne-Marie Laurent in for the ambience of a certain social environment. Marcel Lapierre remembers the center several striking sequences and images: the dimly lit Pont-Neuf bridge; the Raquin apartment, the domino parties on Tues- days, solemn meetings whose ennui is analyzed by a high angle camera; the languid life of Thterse compared to the slow hopeless turnings of a goldfish in a glass bowl. . . .182 And Victor Bachy describes "a strange black and white symphony" in the bridal bedroom: Black: the dressers, the bedboard-and most sinister-the husband's nightgown. White: the wife's makeup table, the veil, the crown of orange blossoms, the bed sheets and the handkerchief with which Therese moistens her husband's forehead. Several vials of medication are arranged on the bedstand; and the light plays with stunning, morbid reflections on the polished black wood and the icy satin of her dress. '' These images were all the more striking because Threse Raquin was one of the few French films in the 1920s to be printed exclusively on black-and-white rather than on color-tinted filmstock.18-4 In the second part of the film, wrote Moussinac, "Feyder comes to the melodramatic action, without trying to falsify and reduce the odious and cruel nature of the characters."'s5 According to Charles Ford, "the spectator ac- cepted everything that could be repulsive in the manner of this woman because Feyder had placed her so precisely in the milieu which explained her ac- tions. ""' And Gina Manes made the character all the more credible by giving one of the best acting performances of the decade-wrote Pierre Leprohon, "This is what is called living a role."I' For Jean Mitry, "the culminating scene {was] the one where Therese and Laurent, after the murder, experience the faint beginnings of anguish and remorse as they face the silent accusation of the old infirm mother."'88 The characteristic motif of Andrejew s decor now became the dirty windows covering the passageway between the flower shop 135 COMMERCIAL NARRATIVE FILM 68. Le Petit Jimmy imitating Maurice Chevalier in Peau de P che (1929) -- -- - - and the tiny apartment. In this unhealthy greenhouse, this "sort of glass jar a trio of petit bourgeoisie festered and decomposed." 89 As a work of sustained power that depended on a synthesis of much of the narrative avant-garde practice, Therese Raquin was one of the more uncompro- mising of the realist films and, for some, a masterpiece of the French silent cinema. '" Yet the critics as well as the public were far from unanimous (the film was a commercial failure in France). " Some, such as Moussinac, re- proached Feyder with showing "nearly too much elegance in a theme which demanded, it seemed, less circumspection and a more willingly developed vulgarity."" The social criticism, he argued, was muted and tamed in an artful period piece, as if within its own glass jar. Others, such as Jean Dreville, praised Feyder for transposing Zola "according to the principle of an accu- mulation of psychological details." 1 For them, the film was a relentless, totally plausible, psychological study. Unfortunately, we may never be able to decide for certain because, despite the high praise it won from both cin6- philes and cineastes, Therese Raquin remains one of the most famous of lost films. Bucking this trend toward large-scale productions were a few scattered low- budget films. There was Fescourt's adaptation of La Glu (1927), set in "a milieu of rough Breton fishermen" and shot in a freakish month of sunshine around the Pointe de Croisie (just west of La Briere). "' There was Duvivier's Le Mariage de Mademoiselle Beulemans (1927), which apparently achieved some degree of atmosphere through location shooting in the streets of Brussels. 136 Perhaps the most interesting of these small films, however, were done by the new writer-director team of Marie Epstein and Jean Benoit-L6vy. Ames d'enfants (1928) and Peau de peche (1929), both produced by Aubert, followed the format of films about children developed by Feyder and Duvivier. The more successful of the two was Peau depche, which opened at the Electric- Cinema in March, 1929. 196It is a rather conventional moral tale about a Poulbot orphan from the streets of Montmartre whose natural generosity is allowed to flourish in the open air of the French countryside. 197 As a reward for returning a necklace that she lost on her wedding day, Mme. Defleures. (Denise Lorys) sends Peau de peche (petit Jimmy) out to her cousin's farm in Charmont-sur-Barbuise where he grows up to fall in love with his "cousin" Lucie (Simone Mareuil). But his best friend, La Ficelle (who once saved him from drowning), seems to desire Lucie as well, and Peau de peche (now Mau- rice Touze) sacrifices his love and returns to Paris. Once she learns of his despair, Mme. Defleures counsels the young man and arranges his marriage to Lucie. Cinea-Cin6-pour-tous found Peau de pche quite refreshing: Finally, here is a film that leaves behind the stifling, miserable limits of the studio, with its conventional decors and illusory techniques, a film where one can breathe the vigorous outside air . . . a film where there are no dance clubs, music halls, or worldly salons, but where, by contrast, there are admirable and restful perspectives of fields, of lovely rivers, and farm houses, so marvelously peopled that one would like to remain there the rest of one's days. 198 More specifically, the film's charm is suggested in the sequence where petit Jimmy does a classic Maurice Chevalier imitation, in a circle of delighted village children (the film's only color tinting appears here in a slight reddening of the boy's cheeks-hence the name, "Peachskin"). And its poignant evoca- tion of desire is evidenced in the opening wedding ceremony (emphasizing the impression the bridal veil makes on the boy) and in the sequence where the spring run-off fills the dry river bed of the village and where Lucie's image, reflected on the water's surface, comes between the two young men. Several years later, the emotional poignance, economical storytelling, and generally "poeticized" discourse that so marks Peau de pche would return in Epstein and Benoit-Levy's highly acclaimed La Maternelle (1933). Otherwise, the simple realist film survived, interestingly enough, in the work of the narrative avant-garde. And there it flourished in a series of major works to be examined later-Dmitri Kirsanoffs Menilmontant (1926), Alberto Cavalcanti's Rien que les heures (1926) and En rade (1927), Jean Gr6millon's Tour au large (1927) and Gardiens de phare (1929), and Jean Epstein's Finis Terrae (1929). Most of these films, however, were scarcely seen outside the network of specialized cinemas and cine-clubs. Based on this survey of the genre, the French realist films of the 1920s can be seen to have a significance that rivals that of the more highly budgeted genres of the historical reconstruction and the modern studio spectacular. For one thing, they provided the basis for the more publicized achievements in the genre in the 1930s. Their concern for the ambience of working-class Paris and for the relationship between environment and character fed into the im- portant cycle of city films by Clair, Gremillon, Duvivier, Renoir, and Carne- REALIST FILMS 137 COMMERCIAL NARRATIVE FILM Fairy Tales, Fables, and Fantasies Prevert. Their interest in the different milieux of the provinces and in the relationship between landscape and character was resurrected in such films as Jean Epstein's L'Or des mers (1932) and Chanson d'amour (1934) and Jean Re- noir's Toni (1934) and Une Partie de campagne (1936) as well as transformed in the enormously popular films of Bas- and Haut-Provence by the Marseille dramatist, Marcel Pagnol. But the 1920s realist films were more than precursors, either to the French films of "poetic realism" or to the Italian "neo-realist" films after World War II. For the genre produced a body of work that constitutes a unique achieve- ment on its own terms. It was the earliest consistent attempt to ground film art in reality, an attempt unfairly neglected by nearly all cinema histories, perhaps because it was superseded by the seemingly more radical concept of reality in the Soviet films later in the decade. Furthermore, the realist genre provided the context for most of the work of the French narrative avant-garde. Delluc, Epstein, Kirsanoff, Cavalcanti, and Gr millon-all have their roots in the genre. And Gance, L'Herbier, Feyder, and Renoir produce some of their most interesting or successful films from within it. That work, as will become clear in Part IV, consisted of extending the genre parameters, especially through experiments at the rhetorical and narrative levels of film discourse. In contrast to the realist film, the fierie-the fairy tale or fantasy-had been one of the earliest and most popular of French narrative films. Around 1900, Georges Melies virtually created the genre with his magic shows, trick films, fantastic adventures "a la Jules Verne," and adaptations of classical and not- so-classical fables. Melies's films were widely copied and imitated, not only in France but also in England and the United States. Their chief characteristic was the creation of an other world by means of uniquely cinematic techniques (superimposition, stop motion, fast motion, reverse motion, etc.) used in con- juction with deliberately unreal, often mobile sets. This was a world of the marvelous or the uncanny, in Tzvetan Todorov's terms, usually distinct from but connected to reality, a world whose laws were often determined by the camera apparatus itself.1 Although interest in the genre soon waned, in com- parison to the comic shorts, police and crime series, and literary adaptations, it never disappeared entirely in France.2 When the war ended, the fantasy returned as a minor genre, but transformed in several different ways. Although the industry generally considered it "demod6' and unpopular in the 1920s, the fantasy is at least important for providing a point of departure for several filmmakers involved in the narrative avant-garde. From 1918 to 1922, most of the French fantasy films were produced at Gaumont, especially in its Series Pax, where they coincided briefly with Feuil- lade's exotic and fantastic serials, Tih-Minh (1919) and Barrabas (1920). Most of the S6ries Pax fantasy films were predicated on the juxtaposition of two different worlds, usually a real world and an imaginary one or sometimes an imaginary world and one even more imaginary. And they tended to be marked either by a Symbolist aesthetic or by the attempt to represent an individual character's mental or hallucinatory experience. Gaumont's major successes in the genre seem to have been the films of L6on Poirier. One of the most discussed French films of 1920, for instance, was 138 Poirier's Le Penseur, "a story of the fantastic," based on an original scenario by the Nouvelle Revue Franfaise writer, Edmond Fleg.3 Cin6-pour-tous, in fact, voted it one of the top three French films of the year.4 The scenario was inspired by the cult that was then forming around Auguste Rodin's famous statue. Pierre Dartigue (Andr6 Nox) is an artist who, one day meditating beside The Thinker, discovers a means of seeing into people's souls. Soon this leads him to develop a method by which he can paint the inner states of his models. "But that which he apprehends and represents," writes Henri Fescourt, "shocks him deeply, forces him into isolation, and finally drives him completely mad. Having discovered the universal deceptiveness of appearances, he dies in his studio, paralyzed in the posture of The Thinker." According to various critics, Poirier compensated for the inherently static nature of his subject with a great number of dissolves and superimpositions, with superbly composed images that involved a seductive play of light and shadow (the work of his cameraman, Specht), and with finely nuanced acting from Andre Nox.6 However simplistic its irony, Le Penseur did seem to develop useful strategies for representing and integrating several different modes of perception. Poirier followed Le Penseur with a more realistic fantasy film, L'Ombre ddchire (1921), in which a mother (Suzanne Despres), "who foresees what would be- come of her sick daughter if she committed suicide," is tormented by conflict- ing desires.7 Jocelyn (1922) also continued this pattern, but in a different framework, with astonishing commercial success (as discussed previously). Here the other world is an historical past of opposites-the upheavals of the Revo- lution and the peaceful solitude of the mountains.8 These opposites are me- diated by the repeated image of Jocelyn's deathbed (a visual motif not unlike the rocking cradle in Intolerance) and resolved in a secret mountain cavern, if only temporarily. The whole film is tied rather conventionally to Lamartine's poem, on which the scenario is based, and to the figure of the poet himself, who sits reading Jocelyn's memoirs beside the deathbed.9 It is as if the artist in Le Penseur had been replaced by Lamartine, and the earlier frightening images of duplicity and suffering had given way to images of a spiritual od- yssey toward eternal peace. There were other more obviously symbolic fantasy films, ranging from the sublime to the ridiculous. One was a lost experimental work, Le Lys de la vie (1920), directed by Gabrielle Sorere and Loie Fuller, the famous though aging dancer who was best known for her illuminated fire-dance. Leon Moussinac called their film "the first attempt to transpose a poem to the screen."10 In this fairy-tale ballet of a princess and a marvelous lily, Fuller employed the magical lighting techniques she had developed on stage with an original con- junction of such cinematic techniques as slow motion and the alternation of tinted positive and negative images." Another sort of transposition was Rene Le Somptier's La Monte vers l'Acropole (1920), one of Louis Aubert's first major film productions. 12 Le Somptier's story seemed overtly political: a socialist journalist by the name of Lesieur (Van Dale) contests and defeats a reactionary bourgeois politician, Labrousse (Andre Nox). And the French censors were offended enough to force Le Somptier to delete any mention of "the proletar- iat" or "the working class."13 Drawing its title from Renan's famous Priere sur l'Acropole, the film sought to dramatize, in Marcel Oms's words, "the conflict between generations in the succession of political power."4 The overall effect, FAIRY TALES, FABLES, AND FANTASIES 139 69. (left) Philippe Hriat and Vanni Marcoux in Autant- Lara's costumes for Don jean et Faust (1923) 70. (right) Marcelle Pradot however, came across as pseudophilosophical: "the Parthenon looked down from the heights of eternity, at several smartly dressed French film actors at its feet (including a miscast France Dhelia}. "" Henri Fescourt summed up its mongrel nature well: it was "ambitious, psychological, sometimes boring, seething with lighting effects, and clotted with what seemed to be an infinite number of intertitles. Besides being out of step politically in 1920, La Monte lers l'Acropole was overly static and pretentious (the cameraman was Amed~e Morin), and Le Somptier turned to direct less controversial films. Several of Marcel L'Herbier's early films for Gaumont oscillated wildly be- tween these extremes, at least according to his boss. Anticipating Loie Fuller, whom he much admired, L'Herbier conceived his first film, Rose-France (1919), as a poetic work and subtitled it "a cantilena in black and white."" Rose- France has a narrative that is as slight as it is highly symbolic: a young man (Jaque Catelain) discovers that the pure French woman he loves (Aiss6) has been writing poems of passionate love to someone else. Finally, she reveals that her love is for France, menaced by the war. The film's discourse, however, is marked by a strange combination of effete actors and decadent decors, col- lages of optical effects, verse intertitles signed by Charles d'Orleans and Charles Peguy, and experimental forms of narrative simultaneity. The combination proved a commercial disaster, and L'Herbier's third film for Gaumont, Le Carnara/ des d irites (1920), was much more conventional. ' He called it a "realistic fantasy," since it combined a Symbolist mask drama with a melo- dramatic intrigue enacted in the Villa Madone on the Basque coast. " In order to reverse their financial losses, the Comtesse Della (Suzanne Despres) and her 140 lover (Paul Capellani) attempt to seduce and blackmail a rich neighbor, Jean (Jaque Catelain), who is in love with a naive young friend of theirs, Clarisse (Claude France). The plot ultimately fails; the Comtesse commits suicide; and her lover re-covers her face with its mask. The Symbolist drama was carried out by means of superimpositions and color contrasts as well as Claude Autant- Lara's decors, especially a fantasmagorical garden where the masked ball is held. Louis Delluc appreciated it all the same: "The paradoxical charm, both deft and deep, that enlivens this modern fresco, so audacious in its halftones, pleases me no end. "20 L'Herbier's Don Juan et Faust (1923) was the most ambitious of these films. This was no literary adaptation but an original scenario of two juxtaposed stories and conflicting symbolic forces. Faust (Vanni Marcoux) and Don Juan (Jaque Catelain)-the opposed figures of libido sciendi and libido sentiendi-vie with one another over Dona Anna (Marcelle Pradot), the daugher of the Com- mandant of Castille.21 With the help of his devilish servant, Wagner (Philippe Heriat), Faust carries off Dona Anna to his castle while Don Juan is tricked into a duel with her father whom he finally wounds mortally. Unable to locate Dona Anna, Don Juan and his sidekick Colochon (Lerner) embark on a quick series of adventures in which he returns to his old ways, seducing nuns, servant girls, and aristocratic brides. Meanwhile, Dona Anna is locked in some kind of time warp with Faust until, with the assistance of Wagner (who by now is disappointed in his master's scientific endeavors), she stabs him to death and returns to her dying father. At the end, in the midst of an orgy he has arranged for all his women, Don Juan is attacked en masse by their husbands, lovers, and assorted male relatives. But Dona Anna saves him, and together they repent and take vows of humility and chastity within the church. This juxtaposition of contending forces determines, in part, the visual style of the film, which L'Herbier himself has described as a meeting of "Velasquez and the Gothic. "22 The interiors for Faust's castle (designed by Robert Garnier) and the costumes for Faust and Wagner (designed by Autant-Lara) were marked by a peculiar mixture of Cubism and "Caligarism"-a rare instance of the German film's influence in the French cinema.23 Yet the other costumes and interiors as well as the location exteriors were straight out of the Spanish period of Philip II, and they were often composed (by cameraman Georges Lucas) in ensemble tableaux inspired by the classical school of Spanish paint- ing.24 The friction between these modern and historical styles, though ad- mired by critics such as Canudo and Clair,2 proved disconcerting to French film audiences, and still does today, partly because the conflict is resolved so easily by the elimination of Faust's world and then by a pat religious conver- sion. But there may be another, simpler reason. The version of Don Juan et Faust released by Gaumont was incomplete and rather obviously patched to- gether near the end; as a consequence, L'Herbier apparently had his name removed from the credits. Furthermore, the company delayed the film's pre- views until October, 1922, aid then downplayed its exhibition run early in 1923.26 Despite its flaws, let alone its commercial and critical failure, Don Juan et Faust is still provocative on at least one point. Once Don Juan learns that Dona Anna is with Faust, he sets out to prey on the nuns in the convent where he has gone to seek her. To seduce one, interestingly, he uses the letter FAIRY TALES, FABLES, AND FANTASIES 141 COMMERCIAL NARRATIVE and medallion (from Wagner) which has just convinced him that Dona Anna FILM went off willingly with Faust. To intensify the perverseness of this moment, L'Herbier alternates shots of Don Juan seducing the nun in a country inn with those of Faust hypnotizing Dona Anna (apparently in an alchemist's attempt to produce gold). While Faust seems to invest Don Juan's passion with a magical power, reciprocally, Don Juan seems to invest Faust's passion with sexual malice. The alternation concludes in the convent courtyard with a dreamlike sequence of softly-focused, eyeline-matched MSs and CUs, in which Don Juan and Dona Anna seem to gaze longingly at one another-as if they have emerged from parallel dark recesses. Do lust and passion merely mask a transcendent pure desire? As if in answer, a similar transformation occurs at the film's climax when Dona Anna rescues Don Juan. The threatened orgy of female bodies around one man quickly changes into a formal ceremony in which the same man confirms his faith in the company of veiled nuns. As Don Juan finally recognizes Dona Anna, through an intercutting of CUs that echo the earlier dream sequence, the film ends in a chaste coupling that is, if in no other way, at least aesthetically satisfying. When Gaumont reorganized its production in 1922, the genre lost its pri- mary support within the industry. For the next several years most of the French fantasy films-excepting the short puppet animation films produced independently by the Russian emigre, Ladislas Starevich27-can be attributed to four men: L'Herbier, Jaque Catelain, Jean-Louis Bouquet, and Ivan Mos- joukine. L'Herbier's Cingraphic briefly provided a haven for the genre with two radically different films. Le Marchand de plaisir (1923), Jaque Catelain's first film for his mentor's company, was a fable that owed less to previous French fantasy films than it did to Chaplin and Griffith.28 Its schematic story brought together Mary-Ange (Marcelle Pradot), a wealthy young woman staying at a seaside resort, and Gosta (Catelain), a simple "seller of novelties," who lives with his poor family in a shack in the sand dunes. Gosta's love, however, is no match for Mary-Ange's engagement to a young adventurer named Donald (also played by Catelain); and it is compromised by a plot perpetrated by his drunken father (Philippe Heriat), who robs Mary-Ange and incriminates his own son. The climax is brutal-Gosta has to kill his father as the latter is trying to strangle his mother. Arrested and then released (when his sanity is questioned), in the end, he and his mother are sent off on a train. Compared to such films as The Vagabond and Broken Blossoms, Le Marchand de plaisir, unfortunately, is overlong and lacks an emotional center (Catelain is no Chap- lin or even a Richard Barthelmess); and in style, it combines, all too predict- ably, American continuity editing, simple subjective superimpositions, and unmotivated moments of rapid cutting. L'Herbier himself was much more ambitious in L'Inhumaine (1924), an early science fiction film that juxtaposes two worlds much more successfully than did Don Juan et Faust.29 The infamous singer, Claire Lescaut (Georgette Le- blanc) lives in an ultramodern mansion (designed by Mallet-Stevens) that in- cludes a cavernous, geometrically furnished, moat-encircled banquet hall and a winter garden of Melies-like artificial plants. Einar Norsen (Jaque Catelain) has a spacious scientific laboratory (designed by Fernard Lger) with an exper- imental television system and an elaborate process for resurrecting the dead. 142 Both places seem to lie on the outskirts of Paris, overlooking the city from FAIRY TALES, the vantage point of the near future. The narrative builds to Einar's apparent FABLES, AND suicide (out of unrequited love for Claire) and his resurrection, which is then FANTASIES replicated inversely with Claire's apparent murder and her resurrection. The final climactic sequence of the film, in which Einar brings Claire back to life, employs a barrage of cinematic techniques (superimpositions, rapid montage, brief flashes of pure color film stock) to represent the impossible as possible. Here the display of technical virtuosity is so awesome as well as self-reflexive that it places L'Inhumaine well beyond the limits of the conventional fantasy film. Jean-Louis Bouquet was an architecture student who came to the cinema as a scriptwriter, editor, and critic.30 His first scenario was directed by Luitz- Morat for Pathe-Consortium, as La Cite foudroyee (1924), which, simultane- ously with L'Inhumaine, created a different kind of ancestor for the science fiction film.31 Its narrative involves a young engineer, Richard Galke (Daniel Mendaille) who hopes to win the hand of his cousin, Huguette Vr6court (Jane Marguenet).32 Her father has been ruined financially, and Richard concocts a plan to reverse his misfortunes. He makes a pact with the neighborhood "Bad Man," who then constructs a strange factory on his property. The project turns out to be an elaborate scheme by a master criminal who threatens the destruc- tion of Paris unless he is paid a huge ransom. In the end, Richard publishes his confession as a successful novel (the strange factory is a printing press) and marries Huguette. But what about Paris and its apparent destruction? The film depicts the apocalyptic disaster in several sequences and stages, and Rob- ert Gys's miniatures and Daniau-Johnson's slow-motion cinematography pro- duced quite believable effects for the period-all the major Paris monuments, including the Eiffel Tower, collapse in varying stages of ruin. The irony of the film is that, through the delay of information as well as through the cross- cutting of real and fictional action, the audience is deceived into thinking that the novel (and its images) is a confession of something that has already hap- pened in the past-that the engineer who goes about his publishing scheme is really the master criminal! In La Citifoudroyee, therefore, fantasy and reality are ambiguously, bafflingly intertwined. Bouquet attempted a similar confu- sion of the real and the fantastic for satirical effect in his scenario, Le Diable dans la ville (1925), but Cinromans persuaded Germaine Dulac to change its setting from the modern world to the fifteenth century.33 The change resulted in a considerable loss of force, since the film's uncanny events now had a simple explanation in witchcraft. Ivan Mosjoukine, Albatros's principal star, also dabbled in the genre, for instance, in his scenario for Protozanov's L'Angoissante Aventure (1920), like- wise for purposes of irony and satire. His eccentric film, Le Brasier ardent (1923), engages the fantastic at several crucial points.35 Its opening is a stun- ner-an inexplicable adventure of violence and eroticism in which a woman (Natalie Lissenko) flees a man (Mosjoukine) through a half-dozen very different settings, only to encounter him each time in a different guise. An abrupt cut to a bedroom sequence reveals that this strange journey into the marvelous has been a nightmare, inspired by a detective novel. This shift from the fan- tastic to the comic announces the general drift of the film, and these nightmare episodes provide a loose framework for the narrative that follows. Later, the 143 71. (left) A poster for La Cite fdlropCe ( 1924) 72. (right) [van Mosjonkine doubled in Feu ,Uathia Pasca/ (1925) nightmare is even reenacted as comedy when the woman is chased by her husband (Nicolas Koline), who literally falls into a detective agency where he, too, meets Mosjoukine and ironically hires him to investigate his wife. From there on, the fantastic disappears, and the narrative turns to satirize other genres-e.g., the police serial and the bourgeois melodrama. For its mixture of styles and genres alone, Le Brasier ardent is worth considering separately. Late in 1924, L'Herbier and Mosjoukine teamed up to make perhaps the biggest fantasy film of the decade and the first film adaptation of a work by the Italian writer, Luigi Pirandello-Feu Mathias Pascal (1925). In comparison to the two men's previous films, Feu Mathias Pascal reversed the shift from the fantastic to the comic in Le Brasier ardent and anchored the fantastic in the narrative much more securely than did L'Inhumaine. L'Herbier himself called it "a fantasmagoria with a realistic premise. "v' The narrative premise is clearly Pirandellean: a falsely reported death seems to free the central character Pascal (Mosjoukine) from a dull married life in the provinces for a second life of adventure and romance in Rome. Some of the film's fascination, as in Le Bnsier ardent and L'Inhunaine, comes from its fanciful play with settings, especially the interiors designed by Cavalcanti-from the dark, cluttered rooms of the provincial home to the hallucinatory open spaces of an apartment build- 144 ing in Rome. It is here that the fantastic finally becomes dominant. In a long imaginary sequence that includes several abrupt shifts, Pascal contends not only with a rival and his brother but also with a gigantic double of himself. The disconcerting uncertainty and ironic twists that characterize this second half of the film place Feu Mathias Pascal as well beyond the limits of the conventional fantasy film. After Feu Mathias Pascal, Mosjoukine and L'Herbier were caught up in the tide of expensive period films and modern studio spectaculars, and the fantasy genre was left momentarily to one man-Ren6 Clair. Clair was opposed to what he saw as aesthetic pretension in Poirier and L'Herbier, among others. Instead, he saw his first films as a conscious renewal of the early French comic shorts and of Melies fantasy films. Although independent productions, they were certainly conceived as commercial projects. Paris qui dort (1924), which was released in the wake of La Cit foudroye and L'Inhumaine, makes an interesting contrast to them.37 It, too, is an ances- tor of the science fiction film. But where the other two films opt for spectacular effects and rhetorical or narrative complexity, Clair's film chooses the simple and direct (also partly, of course, because of budget limitations). His scientist's laboratory looks like a cardboard set for a Keystone comedy. The shot explain- ing how the airplane passengers and the Eiffel Tower watchman escape the scientist's invisible, immobilizing rays is the most rudimentary animation imaginable. Yet for the crucial idea of a paralyzed city, Clair's simple strate- gies of slow-motion footage, deserted streets, and frozen human figures are remarkably effective. Besides being a comedy of manners and situations, Paris qui dort opens onto another, marvelous world. The stillness of the city strangely sets off the slightest movement of the six characters who wander through it. And their movement seems to reproduce that magical moment when photo- graphs first "came to life" in the movies. In Le Fant6me du Moulin Rouge (1925), which followed L'Inhumaine and La Briere at the Madeleine-Cin6ma, Clair was inspired even more directly by M6lies: "Trick shots amused me, interested me, a lot during that period."38 The original scenario allowed him and his cameraman Louis Chaix a great deal of play with superimpositions-which eventually does get repetitious. Through a misunderstanding, Julien Boissel (Georges Vaultier) believes that his fiance, Yvonne (Sandra Milowanoff), no longer loves him-actually, her father (Mau- rice Schutz) is being blackmailed by a rival (Jose Davert). In despair he gives himself over to a doctor (Paul Olivier) who is experimenting with a "Cartesian machine" that separates the spirit from the body. The experiment works, and Julien cavorts about Paris as a disembodied apparition, invisible to all but the doctor (and the audience). He plays tricks on people on the streets (lining up top hats beside a lamppost), at the Moulin Rouge (stealing coats), and at the Louvre (painting a mustache on the Mona Lisa-in homage to Marcel Du- champ). Finally, he refuses to come back to his body and takes up residence at the Moulin Rouge and at a cafe with a distinctly Cubist decor. Then comes the ironic twist. Suddenly made aware that Yvonne really does love him, Julien discovers that the doctor has been jailed for murder and that his own body has been confiscated for an autopsy. There ensues a frantic race to release the doctor (only he can reunite body and spirit) before the fatal operation. Success comes at the last moment. Julien's body revives in front of the aston- FAIRY TALES, FABLES, AND FANTASIES 145 73. (left) Sandra Milowanoff and Georges Vaultier in Le Fantome du Moulin Rouge (1925) 74. (right) Albert Pr6jean and Jim Gerald in Robert Gys's fairyland set for Le Voyage imaginaire (1926) ished doctors; he clears up the intrigue; he and Yvonne are reconciled. The real world proves a safer, more pleasant place after all. As a metaphysical fantasy, Le Fantme du Moulin Rouge exhibits a mood and style characteristic of the French films which contrast quite neatly with those of their German genre counterparts." In The Cabinet of Doctor Caligari (1919), Destiny (1921), Nosferatu (1922), or The Hand of Orlac (1924), the phantom is a creature of menace, ultimately threatening death. Here, he is a mischievous character whose gaiety compensates for a deep-seated sadness and whose con- dition of freedom and power ironically turns illusory. In the German films, the settings tend to be shadow-laden with tension, claustrophobic, palpably ominous. Here, the sets (interiors by Robert Gys) are light and simple, open to Julien's playfulness as well as blandly normal in their indifference to his suffering. The overall style is lighthearted, exuberant in its technical facility, and given to rapid shifts and ironic twists. Perhaps the only thing lacking, as Leon Moussinac first observed, is "an emotional center . . . the sense of a director attentive to express that which is humanly poignant in the simplest and most banal of our actions, just as in the greatest dramas."' More pre- cisely, whereas the central character of Le Fantome du Moulin Rouge achieves a balance of sorts-the union of body and spirit-the narrative resolution seems to contradict the film's free-wheeling, exuberant discourse. Clair's next film, Le Voyage imaginaire (1926), was written expressly for Jean Borlin, the leading dancer of the Ballet su6dois, whom Rolf de Mare wanted to star in a film (he had already appeared in Clair-Picabia's Entr'acte, 1924)." Jean Mitry has argued that its subject was influenced by Harold Lloyd's Grand- 146 ma's Boy (1922)-a long dream miraculously invests a timid hero with the strength and courage to win his love.42 But the imaginary voyage itself is clearly inspired by Mlies. If the film as a whole is only half successful, the fault may lie not in the scenario but in some of the acting (especially Borlin's) and in the limited budget that forced skimping on some of Robert Gys's sets as well as Amed&e Morin's camerawork.43 Jean (Borlin), Albert (Prjean), and Auguste (Jim Gerard) are all clerks in love with their typist, Lucie (Dolly Davis). One afternoon, while everyone in the office is napping, Jean dreams of rescuing an old woman from two toughs. She tells him that his kiss has restored her powers; and she leads him down an Alice-in-Wonderland tunnel, through castrating doors of jagged teeth, over the ceiling of a room (done with an upside-down set), and into a fairyland. This fairyland, however, is no ordinary place. It's a rest home for old fairies (no one believes in them anymore), and the set is appropriately "dmod6": a tawdry baroque collection of seashells, serpents, coral reefs, bijoux, and papier- mache flowers-an outright parody of Melies's fairy kingdom sets. With his special kiss, Jean transforms all the old hags into lovely young women (an apt metaphor for Clair's own transformation of the genre); but, unfortunately, the decors change scarcely at all. The grateful fairies bring Lucie to their world so Jean will be happy, but an evil fairy (a racist stereotype in blackface) compli- cates matters by bringing Albert and Auguste as well. Intrigues develop- Lucie turns into a white mouse and is chased by Puss-in-Boots (a tongue-in- cheek play on sexual appetite), and the fairies give Jean one of their store of wishing rings so that he and Lucie can return safely to the real world. All four characters somehow materialize on the towers of Notre Dame, where Jean promptly gets himself changed into Lucie's pet bulldog (an ironic twist on the sexual chase), and the other two men fight over the ring. The struggle carries over into the Musee Grevin, where Lucie recognizes Jean among the mannequins and falls over in a faint. At night, the Revolutionary mannequins put both of them on trial, threatening the guillotine (a mannequin hand gets chopped off and the bulldog begins chewing it up); but just then a deus ex machina intervenes-in the form of Charlie Chaplin and the Kid! Jean and Lucie are freed; Albert and Auguste are dismissed; and Jean is restored-end of dream. Now bursting with self-confidence, Jean breaks through Lucie's resistance, and he sharply heels her dog as they walk off down the street together-in a parodistic image of the petit-bourgeois couple. Placed in the service of bourgeois conventionality, the marvelous simply empties that con- ventionality of substance. Neither Le Fantme du Moulin Rouge nor Le Voyage imaginaire fared very well with the French public so Clair, much like his predecessors, turned away from the fantasy genre to more assuredly commercial projects. His first film for Albatros, La Proie du vent (1927), contained elements of fantasy, but they were embedded in an adventure narrative based on a popular novel, Armand Mer- cier's L'Aventure amoureuse de Pierre Vignal, which could have been done just as easily as a serial.44 An airplane pilot (Charles Vanel) loses his way and crash- lands near a mysterious castle where a madwoman (Sandra Milowanoff) and a suspected spy (Lillian Hall Davis) involve him in an intricate series of puz- zling, half-developed intrigues.45 Critics and historians praised the film's tech- nical facility in creating a finely detailed atmosphere (this was the first collab- FAIRY TALES, FABLES, AND FANTASIES 147 COMMERCIAL NARRATIVE FILM oration between Clair and set designer Lazare Meerson) and a light, balanced rhythm overall. However, subordinating the fantasy to an almost American- style adventure was what apparently made La Proie du vent Clair's most prof- itable silent film.46 As the French film industry turned almost exclusively to period films and modern studio spectaculars, often in co-productions, the genre of fantasy films passed largely into the hands of the narrative avant-garde as an alternative to the realist film. They, too, would account for some of the uniquely French films produced during the latter half of the decade. For instance, Jean Renoir's Charleston (1926) and La Petite Marchande d'allumettes (1928) carry on Clair's interest in a Mlies-style fantasy. Jean Epstein's La Chute de la maison Usher (1928) transforms the genre according to an eccentric fantastic mode and a very personal form of pantheism. Germaine Dulac's La Coquille et le clergyman (1928) and Luis Bunuel's Un Chien andalou (1929) revolutionize it according to the Surrealist aesthetic. All three of these latter films are remarkable in creating a fantastic world whose events cannot be explained naturally. But more about them later. Only two of these filmmakers could be said to have worked commercially as well as experimentally in the fantasy film genre. Jean Epstein did so in the first film he made for his own production company. Mauprat (1926), loosely adapted from a Georges Sand novel, is a curious blend of generic elements.47 Its subject is a mysterious intrigue among the eighteenth-century provincial aristocracy in the upper valley of the Creuse River, on the northern edge of the Massif Centrale.48 Epstein shot his exteriors there around several different chateaux, especially the ruins at Chateaubrun. Some of this footage has a freshness and charm undeniable even today: for example, the deft rapid mon- tage of Bernard's flight into the woods after being confined in the civilizing chateau; a MS of the train of Edm6e's gown gliding across tiny wildflowers by a pond; a LS, after she has been wounded (shot while on-a hunting party), of Bernard proclaiming his innocence to his friend under a huge arching tree. Marie Epstein has even called the film "a forest ode. " The interiors are strikingly different: the Mauprat chateau is soberly conventional, but the Roche Mauprat ruins contain immense dark spaces with a rich gothic atmosphere. In the sets for these latter spaces, Epstein and his designer, Pierre K6fer, relied heavily on a limited number of arc lights: for instance, in one shot, the motif of "two converging pencils of light concentrate attention on a strange canopy bed."50 This obvious concern for natural landscape links Mauprat closely to the provincial realist film, but the film also works as a modest historical reconstruction, faithfully rendering the surface of eighteenth-century pro- vincial life. The narrative, however, along with the gothic atmosphere, situate Mauprat within the fantastic. It is predicated on a simple but rather interesting di- chotomy. The Mauprat family has two branches, a good one and bad one. That the one darkly mirrors the other is clear in the decision to have Maurice Schutz play both fathers, Tristan and Hubert. Bernard (Nino Constantini) is the bad family son who is awakened by the beauty of Edm&e (Sandra Milo- wanoff), the good family daughter. In exchange for helping her escape his treacherous family (as in a fairy tale, she loses her way and simply wanders in), she promises to give herself to no one but him (the sexual sense here is 148 75. Alex Allin and Nino Constantini in the forest in lauprat ( 1920) as strong as the marital). After falling in love with Edmee, Bernard is taken from his original family, civilized, humiliated (she reinterprets her promise), honored (through service in the army), accused of trying to kill her, and finally exonerated when a trial reveals that his evil father, Tristan, actually uses Ber- nard's gun to perpetrate the deed. Edmee recovers and eventually admits her love for Bernard, and the two are united at last. This dichotomy of families and fathers that converges in the double-natured son creates a curious psychoanalytic fable. Is the conflict one of fathers over possession of the son' It would seem so, if one considers the flashback sequence where, beside the strange bed that holds the corpse of his mother, the or- phaned boy moves back and forth between the two brothers, only to be wrapped in Tristan's huge cloak and spirited off. Or is the conflict an interior one, involving a son who does not really know his nature. Aligned with one family, he means to rape and destroy a woman; aligned with the other, he falls in love and is repeatedly denied her. Can the repression of love so easily produce the return, displaced through the father, of the son's murderous desire' The 149 COMMERCIAL NARRATIVE final shots seem to evade an answer with a conventional ending: the young FILM couple, by a window, look out on a pair of swans on a pond. Yet the previous sequence demands to be read in conjunction with it. Bernard's friend, deter- mined to do away with Tristan "with his own hands," journeys to the ruined chateau only to watch the old man cavort about the heights and finally walk out onto a rock outcropping over a lake and, with a wave, dissolve away. If a different figure of desire still exists repressed and dispersed in nature, can that pond of swans really be so idyllic? An abyss gapes in Mauprat, just as it does, even more disturbingly, in Murnau's Nosferatu (1922). Germaine Dulac also worked both sides of the fantasy genre. Two of her films were produced independently and were consciously organized as visual ballets or musical compositions. La Folie des vaillants (1925) was adapted from a Maxim Gorky story about a wandering violinist and Bohemian woman with whom he falls in love. According to Charles Ford, their story was paralleled throughout by the Russian fairy tale of the eagle and the snake.51 Still, Harry Alan Potamkin found it full of a "sentimental podsie."52 L'Invitation au voyage (1927), on the other hand, was based on just a few lines from the famous Baudelaire poem. In a sailor's bar, a woman seems on the threshold of a dream romance until her "inviting" officer notices that she is married. Most of the film's attention is focused on the milieu of the bar and on the imaginary escape, each of which is articulated in a contrasting montage of visual motifs. Its rhetorical complexity will warrant further scrutiny later. Dulac's third film in the genre was a commercial venture for Alex Nalpas and Louis Aubert. The fantasy in La Princesse Mandane (1928), adapted from a Pierre Benoit novel, sounds similar to that of the Arabian Nights world of Salammb6 (1925).53 But the subject and spririt of the film are closer to Clair's Le Voyage imaginaire (and perhaps also to Keaton's Sherlock Jr. [19241). Dulac herself summarizes the story this way: In my film, Benoit's hero became a victim of the cinema. An obsession with all the glories of the screen persuades him to abandon his peaceful life and seek through the world. He thinks, at one point, that he is transported into a wonderful country, a marvelous kingdom over which a fairy princess reigns.54 As in the Clair film, this fantasy kingdom is enveloped, Dulac continues, "in a fine web of comedy; a constant obsession with adorning the ways of reality in dream is interwoven with a wish to transform the gravity of any conflict with laughter."5 Despite Clair's failure with Jean Borlin, Dulac chose to emphasize the ethereal balletic quality of this fantasy by also starring dancers in two of the film's leading roles: Ernest Van Duren and Edmonde Guy. However, La Princesse Mandane ends with "a moral" that has none of Clair's irony: "After his many adventures, my hero chooses to find happiness in sim- ple things."56 The conservative, self-congratulatory ending, Charles Ford con- cludes, helped to make the film a profitable one for Aubert at the end of the 1928 election year. Throughout the next decade, the French film industry continued to refuse to exploit the fantasy genre. There were exceptions, of course. L'Herbier re- turned to the "realist fantasy" for one of his best early sound films, Le Parfum de la dame en noir (1931), but it was little more than a romantic idyll. Clair, 150 of course, used a fantasy format for one of his best social satires, A nous la libertd (1931). And, if the Surrealist fantasy culminated in Luis Bunuel's L'Age d'or (1930), it was also conventionalized, mythified, in Jean Cocteau's Le Sang d'un poete (1930-1932). Only the adverse conditions of the German occupation during World War II forced the French to revive, and to see the advantages of reviving, the genre once more-and on a much larger scale. Throughout the 1920s, in at least one important genre, the French cinema displaced its concern with its own culture through fictional adventures in a different kind of other world-one outside France, especially in North Africa and in the eastern Mediterranean. At first, the genre drew on the conte arabe tradition of A Thousand and One Nights, recently made fashionable, for in- stance, by Diaghilev's Ballets russes, especially by his Schihrazade (1910). It also drew on Biblical stories, which had fascinated many French writers of the nineteenth century. Exotic tales of romance and adventure offered an accept- able escape from the economic and ideological problems wracking postwar French society. But the French colonies of North Africa (especially the well- settled territories of Tunisia and Algeria), according to Pierre Boulanger, pro- vided appropriate settings for similar "stories of bloody conflicts, sensual pleas- ure, and mystery." The genre quickly reoriented itself, and a number of melodramas and romances that could just as easily have been set in France instead took advantage of the local color and exoticism of North Africa. The earliest successful adaptation of the conte arabe originated with Louis Nalpas, one of the men who later sustained the French production of serials. This is less than surprising when one considers that the producer had been born and raised in Smyrna. When Nalpas established his own film production company in Nice in 1918, according to Henri Fescourt, "he thought of the 'Thousand and One Nights' stories [and decided to make a film that would resurrect a fabulous Orient. "2 The result was La Sultane de P'amour (1919), directed by Rene Le Somptier and Charles Burguet from a scenario by the French translator of Omar Khayym, Franz Toussaint. After a sensational spring preview, the film's fall release was extremely successful. So popular was Nal- pas's first production that Pathe-Consortium included it among its re-releases in 1923.3 The distribution of stereotypical characters in La Sultane de /'amour suggests that the film may have been a hybrid outgrowth of the bourgeois melodrama, the serial, and the exotic "Arabian Nights" tales. For there was "Marcel Le- vesque with his comic finesse; Gaston Modot whose strong features took on a ferocious character under his [soldier's] helmet; Sylvo de Pedrelli, the lovesick Emir, imposing and handsome; France Dhelia, the fiery sultaness; [and] Ver- moyal with his pathetic glances."4 Louis Delluc called it "the best French film" of the season-"a remarkable synthesis . . . because the film is composed of many films and its diversity [including documentary inserts as well as magical trick shots] could just as well be called complexity."5 The film's technical limitations-it was shot without any artificial lighting at the Villa Liserb outside Nice, using painted sets (by Marco de Gastyne) and a large diorama with miniatures in order to represent vast expanses-were obviously out- weighed by the gusto of the acting (especially the athletic prowess of Modot), ARABIAN NIGHTS AND COLONIAL DREAMS Arabian Nights and Colonial Dreams 151 COMMERCIAL NARRATIVE FILM the complicated intrigue, and the unusual decision to use the stencil-color Pathe process (which was normally reserved for newsreels) in the final prints.6 Capitalizing on Nalpas's success, Leon Poirier produced a series of Oriental films (also set in Persia, or farther east, and shot in Nice) for Series Pax at Gaumont: Ames d'orient (1919), Narayana (1920), and Le Coffret de jade (1921). All three films narrated tales of brutal, mysterious, tragic passion; and the Oriental woman was consistently typecast as the deadly seductress. Le Coffret de jade, according to Ren6 Jeanne, affirmed the dubious maxim: "Life is a rosebush, and woman is the thorn."7 Narayana (a transcription of Balzac's La Peau de chagrin) Louis Delluc described as "an enticing dream-more sugges- tive than complete."8 To the genre's racial and sexual misogyny, here religious idolatry was added in a sequence where a shot of the hero (Van Daele) kissing the neck of the Oriental seductress (Marcelle Souty) is followed by another of a Buddha statue over which a skull is superimposed.9 What fascinated French audiences as much as these tales of forbidden, destructive desire, however, wrote Fescourt, was the "exotic luxury" of Poirier's "Persian 'tapestries.' . . . They oohed and aahed over the composition of the decors, the arrangements of the furniture, the wallpaper, the knickknacks, the paintings."10 The sets for these films, designed by Robert Jules Garnier, were obviously an advance over La Sultane de l'amour, but the genre quickly passed on to the real thing." The popularity of Nalpas's and Poirier's films encouraged the Russian emigre colony that had settled in Paris and begun to produce films. The first major success for Films Ermolieff, in fact, was Tourjansky's Les Contes des milles et une nuits (1922). 12 In this three-part film, the Princess Goul-y-mar (Natalie Ko- vanko) is traveling to the neighboring palace of her sister when she is captured by barbarians and carried off. Accused of heresy, she is prepared for torture by Prince Soleiman (Nicolas Rimsky), who instead falls in love with her. Together they survive a number of misadventures in his land and eventually return to the princess's father's kingdom where the prince is accepted as the sultan's legitimate heir. In contrast to the earlier films, Kovanko's Oriental woman exhibits a benevolent power that transforms uncivilized barbarians. Yet, as before, what apparently appealed to French audiences was the location shooting (done in Tunisia by Bourgassov and Toporkov) as well as the rich, coordinated decors (created by Ivan Lochakoff) and costumes (designed by Tourjansky himself). Cinemagazine, for instance, was mesmerized by "images that evoke the lush splendor that we imagine must constitute the mysterious world of Islam."13 Two years later, for Films Albatros, Tourjansky returned to his version of the conte arabe formula with Le Chant de P'amour triomphante (1923), again starring Natalie Kovanko (as Val6ria) with Jean Angelo, Nicolas Koline, and Jean d'Yd.14 Here Lochakoffs decors were even more stunning, especially during the sequence of Valeria's dream. The success of these films led to Tourjansky's first big film for Cin6-France, Le Prince charmant (1925).15 Although set in the present or recent past, the narrative basically offered a pretext to explore (or fabricate) the exotic and sinister world of the Orient. A European prince (Jaque Catelain) and his fi- anc6e (Claude France) are sailing a yacht through Middle Eastern waters when they witness the shooting of a young man trying to reach a high window in the local caliph's palace. The prince soon discovers a beautiful young woman (Natalie Kovanko) hidden in the palace and falls in love with her. Through a 152 chain of intrigues and counterintrigues, the prince and his servant (Nicolas Koline) succeed in freeing the young woman. Returning to his own country (where his father has died), the prince ascends to the throne, and defies pro- tocol by taking his new-found love for his bride. Again love conquers all- with the aid of brute force and treachery in the Orient and civilized diplomacy in Europe. Since most of the narrative takes place in the Middle East, Lo- chakoffs spectacular decors for the caliph's palace, especially the huge recep- tion hall, are the film's chief showcases. 16 Cinia-Cin6-pour-tous praised them unequivocally for creating "a marvelously consistent world of complete fan- tasy. "17 The crown jewel of these conte arabe films was to have been Aubert's Sa- lammb6, which premiered at the Paris Opera on 22 October 1925.18 This was the most expensive of the productions that Aubert had arranged to make, during the heady early days of international production strategies, in the film studios of Vienna.1 Pierre Marodon was assigned to direct, but soon discov- ered that he had to serve as scriptwriter, set decorator, costume designer, and God knows what else.20 Despite his problems, Marodon continued to be re- spectful of his author; his express purpose, he said, was, "to illustrate Flau- bert's novel."21 Perhaps recalling the formula of Quo Vadis? Aubert encouraged him to turn his scenario into a Biblical film in the Italian operatic manner. Consequently, the tragic story of the Carthaginian princess, Salammb6 (Jeanne de Balzac), her father King Hamilcar (Victor Vina), and her two rivals, Matho (Rolla Norman) and Narr'Havas (Raphael Livin), apparently was narrated almost exclusively in LSs that focused on mammoth decors (both exteriors and interiors) and on great masses of people. The central narrative action-Matho's attempt to seize Salammb6, which ends in death for them both-seems to have gotten lost in an endless series of marches, battles, and celebrations.22 "Though shorn of any poetry," Ren6 Jeanne and Charles Ford write, "Maro- don's formula did achieve, at moments, a sort of wild grandeur that one could not find in the Italian films, even in Cabiria. . . ."23 Flaubert, of course, would have been appalled by all of this (given his distaste for "illustrations," if nothing else), but Aubert probably believed that he was righting a wrong done the novel during the war by an Italian film adaptation, in which Mtho succeeds in marrying Salammb6 at the end.24 Critics generally were not kind to the film, and some even suggested that only Florent Schmitt's specially composed score saved it from being an absolute disaster.25 Its luster consid- erably tarnished, Salammb6 marked an end rather than a beginning. Thereafter, the French abandoned ancient history and the Middle East to the Americans- e.g., Ben-Hur (1926) and King of Kings (1927). The initial success of the conte arabe film stimulated several different varia- tions in the genre. The first might be called the Spanish film, which enjoyed a brief vogue in the early 1920s.26 Louis Nalpas again produced the first of these, Germaine Dulac's La Fete espagnole (1920), from a Louis Delluc scenario about a femme fatale (Eve Francis). Others included Musidora's Pour Don Car- los (1921) and Dal-Film's L'Infante a la rose (1921).27 But the cycle peaked in Gaumont's production of El Dorado (1921), by Marcel L'Herbier, which pro- voked an outburst of praise and outrage. While critics unanimously accepted El Dorado as "an advance" in film art, some viewers booed what they took to be virtuoso technical effects.28 Yet its poignant tale of fatalistic love (in which ARABIAN NIGHTS AND COLONIAL DREAMS 153 COMMERCIAL NARRATIVE Eve Francis commits suicide behind the stage curtain of a cabaret), its vivid FILM use of tinted film stock, and its location shooting within the famous Alhambra (previously off-limits to filmmakers-all won for it a considerable audience and big box-office grosses, at least in France. After 1922, however, especially after the commercial failure of L'Herbier's symbolic fantasy set in Spain, Don Juan et Faust (1923), the Spanish film-with the exception of Feyder's Carmen (1926)-nearly disappeared. The other variant was much more important, for it led to a redefinition of the genre into something best called the colonial film. The key film in this transformation was Jacques Feyder's L'Atlantide (1921). At least one film sparked Feyder's initial interest in North Africa-Luitz- Morat's Les Cinqs Gentlemen maudits (1919), which was produced independently from a scenario by journalist Andre Heuz.29 The subject was a mixture of sleuthing, adventure, and the occult. One of five French tourists visiting Tunis tears the veil from a young woman's face, and a beggar accompanying her prophesies their deaths within a month. One by one, four do disappear, but the last discovers that the beggar is actually a bandit plotting to acquire his fortune. With the villain punished, the hero can fall in love with the niece of a colonist. Although Luitz-Morat did not insist on an oriental touch, accord- ing to Pierre Boulanger, he shot most of the film on location in Tunis. Costing but 137,000 francs, Le Cinq Gentlemen maudits brought in more than a million, enough to impress Feyder and several others.30 Early in 1919, Pierre Benoit had published a novel which rapidly became the first best-seller in modern French literature. L'Atlantide described the lost paradise of a beautiful Circe-like queen in the Sahara Desert, where she took European explorers as her lovers and then turned them into golden statues. After the success of Ames d'orient, Leon Poirier got Benoit's permission to adapt his best-seller for film. But Gaumont refused Poirier's stipulation to shoot on location in the Sahara-Fontainebleau had been good enough for Feuillade, and besides, what about all that sand?31 Feyder then snapped up the rights to the book (with only a vague notion of how he would adapt it), just weeks before Leonce Perret tried to buy it for Paramount in Hollywood.32 After securing independent financing (which had to be resecured time and again), early in 1920, Feyder marched his crew and cast-Stacia Napierkowska (An- tinea), Jean Angelo (Capitaine Morhange), Georges Melchior (Lieutenant Saint- Vit), Mary-Louise Iribe (Tanit Zerga)-into Algeria for eight months of shoot- ing around Touggourt, in the Aures Mountains, and finally at Djidjelli, on the Mediterranean coast. Instead of returning to Paris for interior shooting, he had the Italian painter Manuel Orazi improvise a huge tent studio at Ba- bel-Oued just outside Algiers.33 The production's ordeal was well publicized in the press; and when all 4,000 meters of the completed film were previewed at the Gaumont-Palace, in June, 1921, it caused a sensation. Released in Octo- ber en exclusivit6 first at the Gaumont-Palace and then at Aubert's newly opened Madeleine-Cin6ma, L'Atlantide ran for a full year and was shown around the world.34 Rereleased by Aubert in 1928, perhaps because of Andre Delpeuch's accolades to it in his survey of the French film industry, the film had a similar success.31 Jean Mitry has called it "the first really successful postwar French film. "36 The femme fatale figure, the idealistic passionate Frenchman (or his stand- 154 '~:'~! J //i r 76. (left) Georges Melchior and Napierkowska in L A lanthde ( 1921 ) 77. (right) Antinea's robe displayed in front of a poster for L'Atlantide, both designed by Manuel Orazi in), the narrative of desire that leads to death-L'Atlantide shares all this with films such as Poirier's earlier Persian tapestries. There is even a short sequence that echoes Narayana: a shot of Saint-Vit dissolves into a gong over which, as it is struck, are superimposed first the face of Morhange, then a cheetah (An- tinea), and finally a skull. What L'Atlantide adds is a sense of nature not found within France itself (or in its studio properties). "It began with a sense of torpid heat and rhythm that was lacking in La Sultane de lanour," Delluc pointed out." "Here," wrote another critic, "the Aures mountains provided unusual rock formations and wind-eroded foothills; and, even more strikingly, the desert offered the slow undulation of its rippled dunes.""' "The one central character in L'Atlantide," Delluc concluded, "was the desert sand," which "spoke with {an} undisguised eloquence. . . ." In contrast to the harmony of character and landscape in the realist film, in L'Atlantide, the desert be- comes the locus of a test. The infernal sun, the expanse, the solitude provoke an encounter with pure beauty, mystery, and death. Before this landscape, the colonized North Africans vanish like a mirage, only to reappear transformed in a perverted paradise in order to measure the power and idealism of the French. The testing produces a guilt to be expiated (maddened by Antinea, Saint-Vit kills his friend, Morhange) through suffering (he wanders in the desert with his rescuer, Tanit Zerga) and revenge (in the final shot, he returns to the desert in search of Antinea and her kingdom). As the first important colonial film, L'Atlantide establishes the landscape of North Africa as a special 155 COMMERCIAL NARRATIVE FILM site for the resolution of specifically French crises of the individual, of com- rades-in-arms, and, later, of the family. Viewed today, L'Atlantide retains some of its fascination.40 The early se- quences of the caravan traveling across the desert and entering Timbuktu, the exploration of caves in the mountain canyons of the Hoggar, a corridor se- quence during the Frenchmen's first night in Antinea's kingdom-these still hold one's interest in their choice of location, their composition, their light- ing, their ambience (much of this is due to the chief cameraman, Georges Specht). However, the film seems even longer and more slowly paced now than it did then. Its rhythm is steady, even monotonous, interrupted only by a few sequences with frequent intertitles. Feyder structures the film to narrate the story of Antinea in a long flashback which ends in such a way (emphasizing Saint-Vit's powerlessness and suffering) that the final sequence of his imme- diate return to the desert loses much of its credibility and force. The sets for Antinea's palace, in the words of Georges Sadoul, are a puzzling concoction of styles-"Carthaginian, Black African, President Fallieres, and Kaiser Wil- helm II"4 '-as if someone had visited fin-de-sidcle France and Germany and then reproduced them from a nightmare twenty years later. Although most of the acting is more than acceptable-even quite good, especially given the shooting conditions-Napierkowska's Antinea nearly ruins the film. An inter- nationally known dancer and actress before the war, Napierkowska had been imposed on Feyder by his financiers (he himself had wanted Musidora).42 Un- fortunately, no one realized that she had gained thirty pounds; and, instead of diminishing her appetite, as Feyder hoped, the Sahara whetted it all the more.43 Consequently, in contrast to her pet cheetah, Antinea lounges around much like a comfortable, extremely well-fed tabby. Out of L'Atlantide's phenomenal success, which was complemented by Val- entino's romp in The Sheik (1921), came a whole series of French films that were shot in Tunisia, Algeria, and even Morocco.44 In Franz Toussaint's Inch' Allah (1922), directed by another Italian painter, Marco de Gastyne, Napier- kowska appeared as the desirable dancing daughter of a sultan, for whom a suitor kills the requisite number of seven rivals.45 In Luitz-Morat's Le Sang d'Allah (1922), Gaston Modot plays a Frenchman who rescues Yasmina, the rebellious wife of a sultan, and then is saved by her suicide in the desert (Yasmina seems a reformed version of Antinea).46 In Andre Hugon's Yasmina (1926), a French doctor (Leon Mathot) returns the favor by saving a Tunisian (half-French) princess (Huguette Duflos) from the horror of marrying a rich old Moslem.47 Jean Vignaud, editor of the tabloid, Cin6-Miroir, specialized in whipping up popular romances set in the colonies: Mercanton and Hervil's Sarati-le-Terrible (1923), Fescourt's La Maison du Maltais (1927), and Mercan- ton's "a la mode" Venus (1929).48 Finally, Pierre Benoit tried to repeat L'At- lantide's success with an adaptation of La Chdtelaine du Liban (1926), an inter- national co-production, directed by Marco de Gastyne and starring Arlette Marchal, for Bernard Natan and Paramount.49 Even the subject of the bourgeois melodrama began to infiltrate the genre. In Dmitri Kirsanoffs Sables (1927), a separated husband and wife (Van Dale and Gina Manes) are reconciled over the sickbed of their daughter (Nadia Sibirskaia), who has driven alone across the Tunisian desert in a sandstorm to find them.50 Kirsanoff himself condemned his film as "terrible, childish, stu- 156 pid, merely amusing . . . an imbecile wrote the story."51 In Julien Duvivier's adaptation of Henri Bataille's old sex play, Maman Colibri (1929), an unhap- pily married woman (Maria Jacobini) falls in love with the best friend of her eldest son, a young lieutenant of the Spahi. After following him to sunny Algeria, where he abandons her for a younger woman, she returns, chastened, to snowbound Paris, to be pardoned.52 La Revue du cinma called it "a senile platitude, an absolute void of thought and feeling, a total incomprehension of cinema. "5 In this catalogue of mayhem, villainy, suffering, and sentiment, at least two films made slightly more serious attempts to und'erstand the largely Is- lamic Arab population. Visages voiles . . . imes closes (1921), written and di- rected by Henry Roussel and partly shot in Algeria, had the honor of being shown to the French Chamber of Deputies as an official example of national film production. The daughter of a high colonial administrator (Emmy Lynn) and a mighty caliph (Marcel Vibert) fall passionately in love, but their cultures separate them. One critic wrote: Here is one of the best psychological studies that I have seen on the very different mores of the Europeans and the Arabs. Besides the intimate scenes between the prin- cipal characters, there are the beautiful frescoes of the nomadic life of the southern Algerian tribes. They remind one of masterful paintings, of the picturesque frescoes of our best orientalists. Harka's attack on the French fort and his defeat particularly recalls Horace Vernet's famous painting, the capture of the family of Abd-El-Kader.51 The Arab culture is unable to escape French stereotypes, however, for, as another critic wrote, "where love, for us, logically should triumph, there the Koran triumphs over it" (an opposition that conveniently ignores one of the medieval origins of romantic passion).55 By articulating "the irreducible an- tagonism that exists between the Orient and the Occident," in Pierre Boulan- ger's words, Visages voiles . . . aimes closes easily assents to French superiority. 56 Henri Fescourt's ill-fated L'Occident (1928), based on Henri Kistemaecker's scenario for his own play, reverses the characters in a story of love between a young Moroccan woman (Claudia Victrix) and a French naval officer (Jaque Catelain). Here, the subject is doubled: Can an Oriental woman become an Occidental? And will the officer's passion or duty triumph in the end?57 The first question is undermined by the fact that the Moroccan woman is already a French actress (and a bad one at that). The second is resolved against love once again, but the rulebook that thwarts it this time is hardly the Koran. Complaining of "the fundamental banality of L'Occident," On Tourne also re- marked "that the characters confirm our hypocrisy. "58 In the climactic battle between the French legions and the rebellious Chleuhs-filmed "in the best western style"--the leader of the legionnaire extras even complimented the director: "Monsieur Fescourt, your manner of treating the Berbers has done a good deal for their pacification."5 Years later, Fescourt realized that the idea never occurred to any of the directors of this warlike spectacle that, among the crowd of retreating Moroccans, perhaps there were some who were angry at the role that we asked them to play. . . .60 The genre would not begin to redeem itself until Marie Epstein and Jean Benoit-Levy's Itto (1934). ARABIAN NIGHTS AND COLONIAL DREAMS 157 COMMERCIAL NARRATIVE FILM An interesting feature of the colonial film in the 1920s was its general lack of attention to the French foreign legion and the Moroccan war, which con- cluded with the surrender of the Riftains in 1926.61 Perhaps the negative public reaction to the war, which provoked increasingly frequent demonstra- tions in the middle 1920s, made the film producers wary of the subject-as happened to the American film industry during the Vietnam War. In any case, an American film, Beau Geste (1926), became the first film devoted en- tirely to the legion, and the Americans devoted more films to the subject than did the French, even into the 1930s.62 Except for L'Occident, the French seemed more drawn to the sea dramas associated with the Moroccan war, as in Ba- roncelli's Feu (1927) and Nitchevo (1926). However, it was in a film recon- structing an earlier period of the Moroccan war that the subject of individual redemption in the North African colonies was first articulated. The film was one of Cineromans's more lavish serials, Rene Le Somptier's Les Fils du soled (1924), which turned the Moroccan independence fighters into a particularly villainous Emir and his hordes, who team up with an unscrupulous French financier. One of its plots involved a young man falsely accused of theft who is exonerated through service in the legion.63 Despite so-called documentary sequences of Moroccan life and its major festivities, Les Fils du soled celebrated "France and its sons as friends come to collaborate in a great civilizing mis- sion."64 The subject of redemption comes to the fore again in Rene Hervil's Le Prince Jean (1927), adapted from a Charles Mr play, which was set in the milieu of aristocratic gamblers.65 But it assumes its clearest expression in Jean Renoir's Le Bled, which had a prestigious opening at the Salle Marivaux in June, 1929.66 Produced by the Societe des Films Historiques, Le Bled (1929) was com- missioned for the French government's centennial celebration of the conquest or pacification of Algeria. Its purpose was to encourage tourists and immi- grants by propagandizing "the productive energy of Algeria in all its manifes- tations as well as the beauty of its landscapes. "67 With the Algerian govern- ment's full cooperation, Renoir's crew filmed in some of the country's principal tourist and commercial centers: the port of Algiers, the Bardo Gardens, the Mitidja Plains, a model farm at Staoueli, and the desert around Sidi Ferruch.68 In fact, Le Bled opens with a documentary prologue of brief sequences (each tinted differently) depicting tourist sites, industrial plants, and various agri- cultural lands before focusing on Algiers, where the hero and heroine arrive on a ship from France.69 The exploration of a large coastal farm allows an imaginary re-creation of the French colonization process: beginning with the 1830 landing of troops, who dissolve into marching soldiers in more and more modern uniforms (done in-camera with tracking shots), who in turn become plows and finally tractors that cover the fields (and overrun the camera) in a choreography of machines quite similar to that in Eisenstein's The General Line (1929).70 This civilizing process is not complete, for the later sections of the film document the archaic but exciting falcon-hunting practices of the no- madic Arabs. The scenario, written by Henri Dupuy-Mazel and Andre Jaeger-Schmidt, narrates the healthful (and economically enriching) effect of the country on a dissolute young Frenchman. Having dissipated his inheritance and his health in Paris, Pierre Hoffer (Enrique de Rivero) comes to Algeria to borrow money 158 ARABIAN NIGHTS AND COLONIAL DREAMS 78. Arquillicre and Enrique de Rivero in Le Bled (1929) from an uncle (Arquilliere) who has made a fortune in farming. The money is promised on the condition that Pierre work the farm for six months. Claudie Duvernet (Jackie Monnier), a young woman Pierre meets on the boat, simul- taneously collects an inheritance, only to fall victim to a plot by her envious, perfidious cousins (Manuel Raabi and Diana Hart). Pierre rescues her; they marry and plan to remain in Algeria. As in L'Atlantide, the desert becomes the locus of an ideological testing: the technical skills of French industry and commerce and the moral/physical measure of the individual Frenchman. The indigenous Arab population conventionally makes up part of the decor, with two exceptions. Pierre's closest friend is Zoubier (Berardi Aissa), an Algerian he knew in the army; and in order to finally rescue Claudie, Pierre has to resort to nomadic falconers whose birds blind the camel on which Manuel is fleeing. Although the villains are not Algerian, but lazy, disgruntled French- men, the Algerians in the film all assent to this best-of-all-possible French worlds. Yet, in 1929, the struggle for Algerian independence was but a few years away. There is a tongue-in-cheek quality about Le Bled that disconcerted some critics at the time-and perhaps some audiences, for its success was only moderate-a quality that now makes the film a bit interesting. Pierre is set off from his uncle's milieu by a comedy of costumes: while everyone else sits down to dinner in peasant garb, he comes down in Parisian evening clothes; for a tour of the farm, he dresses in a sporting jacket, jodhpurs, and tie. Through the simple juxtaposition of a ladle of mush and a cocktail shaker, he 159 COMMERCIAL NARRATIVE FILM is even associated with the villainous cousins. His romantic involvement with Claudie is as comic as it is serious. On the way to his uncle's farm, Pierre is nearly hit by Claudie's car-the accident throws them together in a kind of parodic embrace. Their actual discovery of mutual love is handled just as unconventionally. It happens out in the fields one day when, suddenly drenched by rain, they take shelter in a hut with several Algerian shepherds. Jacques Rivette has spoken of this moment as if it were already glistening with drop- lets from Renoir's later film, The River (1951),72 but the realistic detail (and the allusion to Dido and Aeneas) creates, for me at least, a peculiar balance of the lyrical and the farcical. In the climactic chase, done with all the verve of American action films (and lots of tracking shots), the pursuit uses up cars, horses, camels, and finally falcons; and Pierre is miraculously transformed into an athletic Douglas Fairbanks. Then he and the villain flail about absurdly over the fainted Claudie much like McTeague and Marcus over the dead horse and the money at the end of Stroheim's Greed (1922). The grotesquerie turns to comedy again when a shot of a horizontal bar swaying back and forth near the rear end of a horse is cut into the final sequence, where hero and heroine sit together coyly on a barn swing. One could argue that Le Bled is just as much satire as celebration. Although several of these films attempted to illustrate or undermine the division between cultures in the North African colonies, generally, the colonial film was ideologically quite comforting. As Pierre Boulanger has argued, the rigid hierarchy and repressive contradictions of the French social order were replaced by a fictional society simplified into "a caste of masters (large or small) and . . . a multitude of lesser men' '73-who were either villains or just part of the decor. Here was a world where the dropout or prodigal son (who would become the dominant hero of the genre) could find "the magic ladder of assistance that would permit him to climb back into the social hierarchy that had been hostile or inhospitable to him in the metropolis."74 Even in battles or intrigues with the villainous natives, a Frenchman could still find some measure of glory (which the Great War had obliterated with ghastly irony in France itself). By the end of the decade, the subject of the colonial film had coalesced into the "myth of redemption."75 That myth would find its apoth- eosis in French films of the 1930s-films such as Feyder's Le Grand Jeu (1934) and Duvivier's La Bandera (1935). One might expect that the French, with their rich legacy of history and many extant chateaux, along with their achievements in the genre of historical painting, would be pioneers in the development of period spectacle films or historical reconstructions. But such was not the case. In comparison to other national cinemas, in fact, the French came rather late to the genre. Before the war the French had made just one briefly sustained attempt to develop a genre of historical reconstruction films. That was between 1910 and 1912, when Film d'Art, Film biblique, and S.C.A.G.L. (the latter two affil- iates of Pathe-Freres) produced a small number of feature-length, tableau-style period films among their literary adaptations. Three of the most successful of these exploited crucial periods in French history: the late medieval period- Historical Reconstructions 160 Andreani's Siege de Calais (1911); the French Revolution-Pouctal's Camille Desmoulins (1912); and the mid-nineteenth century-Capellani's Les Mistrables (1912).' The most famous, however, had a non-French subject-Desfontaines and Mercanton's Queen Elizabeth (1912), starring Sarah Bernhardt, for which Adolph Zukor and Edwin S. Porter raised sufficient money to distribute it widely across the United States.2 Perhaps because its share of the world market was declining, the French film industry had already begun to scale back this kind of production when the war broke out in 1914 and curtailed it alto- gether. It was the Italian cinema which virtually created and then dominated the historical reconstruction film genre before the war, with the earliest feature- length films that resurrected the exploits of ancient heroes against the back- drop of mass crowd movements and mammoth, elaborately constructed sets. Especially notable were De Liguoro's Inferno (1909), Guazzoni's Quo vadis? (1912), and Pastrone's Cabiria (1913), with its hero, the giant Maciste.3 Un- der the influence of these films (as well as several of the French), the American cinema took up the genre in such films as Griffith's Intolerance (1916) and De Mille's Joan the Woman (1917) and The Woman God Forgot (1917). Under the same Italian influence, as well as that of Max Reinhardt's theatrical stagings, the German cinema produced a flood of historical films from 1919 to 1923- 1924.4 Especially prominent were Lubitsch's Kstumfilmen: Madame Du Barry (1919), Sumurun (1919), and Anna Boleyn (1920). All of these genre examples finally began to stimulate French interest again by 1919-1920. Cabiria and its sequels had an extended run in Paris after the original version opened at the Vaudeville Th6atre in 1915;5 Intolerance was shown in Paris in the spring of 1919; in the United States, the German films were seen and remarked on by visiting French producers and filmmakers. The French interest, however, took a course rather different from that of the German and American. There seem to have been at least two major reasons for what eventually became a heavy French investment of energy and capital in the historical reconstruction genre. One was frankly economic. Most of the prominent fig- ures in the film industry were quickly convinced by the end of the war that the American superproductions now determined the course of world film pro- duction, distribution, and exhibition. In order to survive, even in their own cinema circuits, the French would have to engage as much as possible in large- scale productions-they would have to go beyond the budgets of their serials, bourgeois melodramas, and realist films. Such large-scale productions might also create the breakthrough that would allow them to reenter the lucrative American exhibition market. In his typically blunt fashion, Charles Path6 summed up the challenge: "the Americans will accept nothing from Europe but costume films, historical films."6 And according to a Cinimagazine story (early in 1922), the success of the German historical films in breaking into the American market provided the French with a model.7 The other reason was ideological. Much like the realist film, but to an even greater degree, the historical reconstruction answered a postwar collective French need. If I may simplify matters, the war had brutally divested France of its ideological trappings,.and the country found itself questioning its own collec- tive identity.8 By resurrecting past historical moments of French glory, and tragedy, the historical reconstruction film contributed to the process of na- HISTORICAL RECONSTRUCTIONS 161 COMMERCIAL NARRATIVE FILM tional restoration and redefinition. The effect, as one might expect, was deeply nostalgic and often escapist. As Marc Silberman has argued with regard to the German costume film, the dominant tendency of these films was toward a mass spectacle that displaced the contemporary historical process.9 This escap- ism did not sit easily with some in the narrative avant-garde wing of the French film industry. But for most, the period spectacle film provided an answer to the question, first raised by Louis Delluc, of what was peculiarly French about the French cinema. The fact that most French films in the genre were about French historical subjects clearly distinquished them from the more international subjects of their German and American counterparts. When the first French period spectacle films appeared after the war, how- ever, this was not yet apparent. Gerard Bourgeois's Christophe Colomb, for instance, which began production in 1916, was done in imitation of the Ital- ian films. Shot mostly in Spain, supposedly using 40,000 extras (is that pos- sible?) and costing 800,000 francs, it was finally released in two parts during the summer of 1919-and found a rather indifferent reception.o Louis Nal- pas's six-part Tristan et Yseult (1920), directed by Maurice Mariaud from a Franz Toussaint scenario, met a similar response." But, then, Nalpas had simply dispensed with the Atlantic mists and the oaks and granite of Cornwall and done it in the Italian manner on the C6te d'Azur.12 The only major film with a French subject was Pouctal's serial adaptation of Dumas's Le Comte de Monte-Cristo (1918), which Film d'Art had been ready to put into production when the war intervened. Four years later, the project was revived and com- pleted (with L6on Mathot starring) to become the most popular French film of the 1918-1919 season. Two years later, it was re-released to similar ac- claim. 13 By 1921, then, partly because of the success of Monte-Cristo, the French film industry began to settle on specifically French historical recon- structions. Two films that year established some standards for the genre. They were Dominique Bernard-Deschamps's L'Agonie des aigles and Henri Diamant-Ber- ger's Les Trois Mousquetaires (the latter, though distributed as a serial, was premiered as three separate, sequential four-hour films). Both were produced by Pathe-Consortium, at unusually high budgets of two and two-and-a-half million francs, respectively. And both were given special gala premieres at the Salle de Trocadero (which could hold 4,000 people), with high government officials in attendance. The publicity for and national sanction of these pre- mieres were unprecedented in the French cinema. Moreover, Denis Richaud, the new director of Pathe-Consortium, boasted that these films would serve as an important instrument of education and socialization. "How excited we are to represent our history cinematographically, to reconstruct each period's ar- chitecture, costumes and manners, and, at the same time, recall the high points of our National Unity."" Accordingly, L'Agonie des aigles was an explicitly political film. Adapted from the novel, Demi-Solde, by Georges d'Esparbes (who was the chief curator at Fontainebleau), it espoused the cause of Napoleon's followers in a conspiracy to restore his son as the Emperor of France (and as ruler of most of Europe). 15 The first half of the film narrates a secret meeting between the conspirators, led by Colonel Montander (Sverin-Mars), and Napoleon's son, the "Eaglet," in Austria, where the latter is confined. This meeting prompts a long flashback 162 . +rii 41 tw y i ... -......y. M 41 to ea " t.< 79. A publicity photo of Napoleon's farewell at Fontainebleau in L Agonie do aglo ( 1921) chronicling the last reigning years of Napoleon (also played by Sfverin-Mars), which focuses on his farewell address to loyal troops at Fontainebleau. The second half narrates the betrayal of the conspiracy by an actress who has caught Montander's eye (Gaby Morlay, in one of her first major screen roles). In the end, the conspirators go proudly to their execution by a firing squad. There are at least two points worth making about L'Agonie des aigles. One involves its royalist, ultra-conservative ideology which was attacked by the Socialist and Communist press and by some film journals, with justification, for at least two of Pathe-Consortium's financiers at the time were Bloc national members in the Chamber of Deputies.'1 But the ideology of the film's narra- tive is most interesting in the context of France's victory in World War I. In the film, the authority figure responsible for Napoleon's son's confinement is the German minister, Metternich. And the French fail to liberate him (once more, a woman is the traitor), to restore him as a symbol of the national spirit (historically, the boy wanted only to die in peace).'" All the conspirators can 163 COMMERCIAL NARRATIVE do is die nobly, defiantly-one combs his hair at the last moment, another FILM blows his nose nonchalantly, a third turns to the wall so that his face (once kissed by the emperor) won't be disfigured. It is as if the conflicts of the Great War had been transposed a hundred years earlier, and the victory turned into a defeat. The old heroic gestures, rendered absurd by the real war, return in this reconstruction of past suffering (Henri Fescourt remembers this finale as very moving at the time). 18 L'Agonie des aigles, therefore, seems to represent, through a kind of displacement, the French mood of defeatism after the war. And it indirectly supported the Poincare government's intransigent policy to- ward Germany, epitomized in the empty slogan: "Germany will pay!" The second point concerns the film's discourse. Much of L'Agonie des aigles seems to have been shot in Pathe-Cinema's old studio at Vincennes. In contrast to later historical reconstruction films, the studio work here is undistin- guished, probably because of inadequate facilities. The sets are skimpy and uniform; the arc lighting is simple and disposed according to rather static character positions (for instance, Gance's Mater Dolorosa [19171 is much more sophisticated). In fact, the most interesting sequences are those shot on loca- tion-the farewell at Fontainebleau (involving mass troop movements before the famous chateau), the almost realistic executions at the end. Overall, the film has a repetitious, monotonous rhythm in which images often illustrate the frequent and lengthy intertitles, which actually produce most of the nar- rative. Sequences rely heavily on long shots-even extreme long shots-cre- ating a static, tableau style of representation that evidences few of the advances made in filmmaking (even in France) during the previous five years. Cutting to and between medium shots and (rarely) close-ups is done without regard for consistent matches in eyeline or even in character position. Conventional literary symbols provide the film's only use of metaphor: the fallen eagle for Napoleon's defeat, a crashing oak tree for his death. In contrast to the other French film genres, consequently, the historical reconstruction seems quite regressive, at least initially, in its deployment of narrative film discourse. At least one shot, however, raises a question about the conventions that then marked imaginary or subjective sequences. As Napoleon's son pores over a map of the world and dreams, there is a fade to a negative image of a group of cavalry riding across a plain. Finally, in the battery of fades, irises, and masks that mark off many shots and sequences, one moment at the end stands out: on the execution wall, above the heads of the conspirators, there appears the superimposition of a cavalry charge led by Napoleon-a reversal of the famous shot in Gance's 1919 J'Accuse, where (on the screen's top half) the French army marches through the Arc de Triomphe while (on the bottom half) the dead rise up from the battlefield in anger. Here, the soldiers in the firing squad hesitate at this apparition, and then put down their muskets; ironically, the Swiss Guard is ordered up to perform the executions. The gesture will reappear, in a very different conflict, in Eisenstein's Potemkin (1925). Les Trois Mousquetaires offered another form of national inspiration and eva- sion with the mock-heroic rapiers of d'Artagnan and his companions. For three long months during the winter of 1921-1922, as this mammoth serial of seventeenth-century adventures swept through the country, the French could divert themselves from their mounting unemployment and inflation and from their government's constant bickering with England and the United States 164 t HISTORICAL RECONSTRUCTIONS -o. Martinelli, Ain Simon- Girard, Henri Rollan, and Marcel Valle (left to right) in .Los ?Ie rois Alouoquetairro 19 1) - about German reparations. 19 For the development of the period spectacle film, neither the serial format nor the staid, monotonous rhythm and theatrical mis- en-scene of Diamant-Berger's film (which complemented L'Agonie des aigles only too well) was particularly crucial. Slightly more important were its nar- rative of intrigue piled upon intrigue and its well-chosen cast: Aime Simon- Girard (D'Artagnan), Armand Bernard (Planchet), Henri Rollin (Athos), Mar- tinelli (Porthos), Marcel Vallee (Aramis), Edouard de Max (Cardinal Riche- lieu), Charles Dullin (Pere Joseph), Claude Merelle (Milady de Winter). How- ever, Les Trois Mousquetaires differed sharply from its sister production in at least two ways-ways that would much affect the genre. As if he were following the precepts of Antoine, Mercanton, Delluc, and others, Diamant-Berger decided to shoot most of his film on location.2" In the once-fortified city of Pfrouges, near Lyon, he and his cameraman Maurice Desfassiaux found streets and alleys that could represent old Paris. The cha- teaux of Perigueux, Chenonceau, Azay-le-Rideau, and Chissay served as royal quarters for the English and French. The church at Montbazon held the wed- ding of Milady and Athos; Chartres cathedral provided the backdrop for a major duel; and the port of Coisic functioned as both Boulogne and Dover. Finally, on a tract of land near Path6-Cinema's studio at Vincennes, Diamant- Berger constructed Richelieu's camp at La Rochelle and th-_ fortifications of Saint-Gervais. The result was that Les Trois Mousquetaires gained, along with the picturesqueness of its landscapes, some measure of authenticity that turned it into a gigantic pageant celebrating seventeenth-century France. The effect would not be lost on later filmmakers. Furthermore, Diamant-Berger assigned 165 COMMERCIAL NARRATIVE Robert Mallet-Stevens--"the uncontested master of the young school of French FILM architecture"-to construct decors and costumes for the interior shooting at the Vincennes studio.2' Diamant-Berger himself remembers that Mallet-Stevens . . . was enthusiastic about the Musketeers; we stylized the decors and costumes, deciding on a clean, simple, unadorned figure, quite the opposite of the Louis XIII style, which was rather well known and "unappetizing." For the court costumes, we used furniture velour because its stiffness gave them an imposing look.22 All this was designed to be spectacular and also to allow for some fast-paced action. There were those, of course, who were not impressed. Ricciotto Ca- nudo, for one, ridiculed the film as a travesty of "four swaggering petty officers caught up in drunken revelries, badly mounted horseback rides, and an un- speakable naval battle allegedly before La Rochelle but actually filmed in the pools of the Luxembourg Gardens!"2 In his second adaptation from Dumas, Vingt Ans apres (1922),24 Diamant- Berger added a new element to this combination of exterior and interior shoot- ing, again with Desfassiaux and Mallet-Stevens. This was an enormous set construction outside the studio, of which the filmmaker is still immensely proud. In the huge Niepce and Fetterer factory yard at Billancourt, which furnished us ply- wood for the decors, we "constructed" the facade of Notre-Dame de Paris, up to the height of the King's Gallery. . . . I was going to film the Te Deum given by the queen in honor of the battle at Lens; on the occasion of this ceremony, the Paris mutineers tried to disrupt the royal cortege in order to seize the King. . . . I collected 3,000 extras [for the scene]. . . . I placed ten cameras along the course. A telephone system allowed me, while perched on a platform, from which I could control the action, to order in sequence the different groups of demonstrators to shout and throw rocks . . . made of cork.25 From the technical point of view, Diamant-Berger told Andre Lang, Vingt Ans apres was even better than Les Trois Mousquetaires.26 But both films were im- portant. In their emphasis on numerous location sites, on expensive breath- taking set constructions and costumes, on ritualistic pageantry involving great masses of people, on intrigues or conflicts that could be resolved in the action of battle, they laid the foundations of the French historical reconstruction film genre. Still, the French film industry hesitated to commit itself completely. Small- scale adaptations of classic nineteenth-century fiction continued in favor. Pathe- Consortium, for instance, assigned Antoine to film an adaptation of a Jules Sandeau novel, Mlle. de la Seigliere (1921), starring Huguette Duflos, Ro- mauld Joub6, and Charles Lamy.27 His son says that Antoine accepted only because he was interested in re-creating, in a chateau near Paris, "a sense of daily life in the eighteenth century in a certain [aristocratic) milieu."28 Cin6- magazine found the result meticulously accurate in its details, certainly lovely to look at, but rather static and old-fashioned in style.29 Film d'Art, likewise, assigned Baroncelli to direct low-budget films drawn from Zola and Balzac novels. Antoine roundly condemned Le Reve (1921) because Film d'Art con- structed a cheap cathedral facade for Baroncelli rather than let him shoot on location.30 But Le Reve was unusually popular-it played for almost a year in and around Paris-and briefly made Andr&e Brabant a star.31 Less successful 166 was Baroncelli's Pere Goriot (1921), starring Signoret. Still, Georges Sadoul remembers that it reconstructed quite accurately the novel's shabby Vauquier pension.32 As I mentioned earlier, Baroncelli's film was the first of a series of low-budget adaptations of Balzac novels. Of these, one deviates radically enough from its genre conventions to be set aside for later discussion-Jean Epstein's L'Auberge rouge (1923). Perhaps the reason for what seemed to be industry hesitation lay in the internal changes that forced the scaling back of film production at both Film d'Art and Path6-Consortium in 1922. Certainly one direct consequence was that Diamant-Berger's third adaptation of Dumas never got beyond the prep- aration stage. But other companies quickly took up the slack, and at least two films seem to have convinced the industry to pursue the large-scale production of period spectacle films. Both were released early in 1923, in conjunction with the Paris premiere of Douglas Fairbanks's Robin Hood (with its magnifi- cent sets by Wilfred Buckland), and that film's extraordinary popularity in France must have confirmed the industry's commitment.33 When Aubert plunged into film production in 1922, his company's biggest project was Rene Le Somptier's La Dame de Monsoreau (1923).3 This six-part adaptation of yet another Dumas novel was calculated to exploit the interest generated by Diamant-Berger's two films. It was released close on the heels of Vingt Ans apres and-received an unusual publicity buildup that included a full issue of Cinemagazine." Furthermore, it was distributed in a stencil-tinted version that often processed two or three colors within a single frame. The print of La Dame de Monsoreau that survives at the Cinematheque fran aise is a revelation-it is both breathtakingly gorgeous and incredibly dull.36 Le Somptier and his cameraman Amedee Morin rely heavily on LSs of landscapes, chateau exteriors and interiors, and a parade of costumes to produce one spec- tacular tableau after another. And he sustains the tableaux with a simple continuity editing similar to that in L'Agonie des aigles. Finally, he tells his story almost completely in the intertitles, and their absence in the surviving print quickly renders the film incomprehensible-Genevieve F6lix, Gina Manes, Jean d'Yd, and the other actors become so many mannequins drifting about in a brightly colored museum. Story and spectacle are kept strictly separate; the latter does not merely illustrate the former, sometimes it literally over- powers it with a nostalgic resurrection of the genre of historical painting. Even more crucial in spurring the industry's commitment to historical re- constructions was a Paramount production, Henry Roussel's Les Opprims, one of the most popular films of the 1922-1923 season.37 Roussel's story of a spunky sixteenth-century Flanders girl apparently was not unlike opera, with flamboyant gestures, elaborate costumes, and passionate characters placed in schematic oppositions.38 Wrote Canudo, The austerity and implacable arrogance of the Duke of Albe; the goodness of the provost marshal whose daughter, Conception, falls in love with Philippe de Hornes, a young Flemish patriot; the diplomatic generosity of Don Luis de Zuniga, the King's ambassador and suitor to Conception, who in the end marries her to the young pa- triot-all are worked skillfully into an intrigue of persecution and passion embroidered onto a background of suffering and thundering revolt by an oppressed people.39 The film's real surprise, however, was the Spanish singer-actress, Raquel Meller, HISTORICAL RECONSTRUCTIONS 167 81. Huguette Duflos in a Bone Soeurs gown in Koenigsmark (1923) whose performance as Conception made her an immediate star, almost on the level of an Asta Nielsen. According to the French film journals, Les Opprims was a magisterial reconstruction of Flanders during the reign of Philip II and demonstrated to the world that the French, too, could make successful large- scale films.-" Still, that it should be made with American money piqued the French a bit. Besides, its appearance also happened to coincide with the be- lated release in France of Lubitsch's "scandalous" treatment of French history in Madame Du Barry (1919).,l Within a year, the French film industry produced no less than three major films that consolidated the conventions and confirmed the new status of the historical reconstruction genre. The appearance of these films coincided with Cineromans' revival of the costume adventure serial that celebrated historical figures as their heroes. However, the characteristics of the historical recon- struction films set them off distinctly from the smaller-budgeted Cineromans' productions. Those characteristics can be summarized as follows: 1) a subject calling for the reconstruction of an historical period (usually French and aris- tocratic); 2) elaborate, sumptuous, authentic decors and costumes; 3) a narra- tive that emphasizes climactic set pieces-of dazzling tableaux and/or sensa- tional sweeping action; and 4) major stars in the leading roles. 168 The most successful of these films was Leonce Perret's Koenigsmark (1923). An independent production (though Paramount may have been involved in its financing behind the scenes), Koenigsmark was a very calculated project.2 Per- ret capitalized on the success of L'Atlantide by adapting his scenario from another best-selling novel by Pierre Benoit which had just been translated into several languages. At the film's premiere at the Salle Marivaux, in December, 1923, he predicted confidently that it would have a worldwide success-and it did." The subject was a tragic French-German romance between a young poet-tutor, Vignerte (Jaque Catelain), and Princesse Aurore (Huguette Duflos) in a small German court on the eve of World War I (displaced slightly by involving the Austrian, not the Prussian, aristocracy). After discovering a mysterious crime committed by the grand duke (her father) years before, the couple flee to France where Vignerte enlists as a pilot in the war and is killed in combat. Perret carefully suppressed all the later war scenes in the novel, in order to emphasize the prewar court spectacle and intrigue and to allow for a bittersweet epilogue at the French tomb of the unknown soldier, where Prin- cesse Aurore (with "six drops of glycerine on the edge of her eyelids") could whisper, "Perhaps it is he!"4" Although certain elements align it with another film genre, the emerging 82. Jaque Catelain and Huguette Duflos in a Menessier interior in Koenig'nark (1923) 169 COMMERCIAL NARRATIVE FILM 83. Raquel Meller in Violettes imperiales (1924) ". modern studio spectacular, Koenigsmark is principally an historical reconstruc- tion, intent on recreating an aristocratic period that had just been obliterated by the war. Much of the film was shot on location in Bavaria, taking advantage of "the romantic decor of a feudal chateau"; but Perret and his set designer, Henri Menessier, also used Levinsky's new Joinville studio to create several lavish interiors." Baldly put, the film was nothing less than a showcase for grand spectacle (the discovery of the skeleton in the gothic fireplace, the burn- ing of the chateau, the Court Ball, the hunt, the marriage banquet on Mont- martre); extravagant costumes by the Bou6 Soeurs (demonstrating that Paris was indeed the capital of fashion); and France's number-one star at the time, Huguette Duflos." As Leon Moussinac admitted, along with Aubert's naval epic starring Sessue Hayakawa, La Bataille (1923), Koenigsmark provided the industry with a model for French superproductions: Such successes are encouraging not only for the producers of French films but also for the investors. Certainly these are super-films worthy of the name [they can say]. . . . They will re-invigorate [the industry) and give us mastery of the whole world.' The readers of Cinenzagazine confirmed its stature by naming Koenigsmark the best film, by far, of 1923.x' 170 Following Koenigsmark's three-month run at the Salle Marivaux came Henry Roussel's second big film with Raquel Meller, Violettes impriales (1924).49 Also independently produced, it, too, was a cleverly calculated operetta-style proj- ect, much like Les Opprimes. At the time, "Raquel Meller had made a popular song, 'Violettera,' so much her own that the mass public no longer saw her as anything but a melodious flower girl."50 So Roussel's scenario contrived to place the singer as a simple flower-seller named Violetta in the period of the Second Empire.5' In Seville, Violetta is rescued from a life of petty thievery by Eugenie de Montijo de Guzman (Suzanne Bianchetti), who, three years later, marries Napoleon III. Taken to Paris and given a career at the Opera, Violetta falls in love with Count Saint-Affremond (Andre Roanne) but must abandon her suitor to save Eugenie from scandal in a court intrigue. Later, she learns that her brother Manuel (San-Juana) has become part of a plot to blackmail and kill the empress. Refusing to betray him and unable to divert Eugenie, Violetta employs a ruse to take her place in the royal carriage and sacrifices herself instead. A narrative of regeneration and of almost Corneillan sacrifice, Violettes im- piriales was also a nostalgic return to a milieu of luxury and splendor long since past. The location shooting took Roussel, Jules Kruger (his chief cam- eraman), and his cast from the sun-drenched streets and peasant dwellings of Seville to the magnificent ceremonial receptions at Compiegne and in the Tuilleries. Critics were especially impressed with the scrupulous reconstruction of the pageantry: . . . the scenes that the filmmaker has shot in the actual locations, inspired at one moment by a famous painting by Winterhalter, are full of such freshness that they can be counted among the most gracious tableaux that we have been privileged to admire on the screen.52 "But beyond these photogenic seductions," added Henri Fescourt, "the lu- minescence that emanated from the film . . . derived above all from the pres- ence of Raquel Meller. "" The third film was an expensive production from Films Albatros, the com- pany organized around the Russian emigr6 film colony in Paris. Previously, Albatros had been known for its serials, Arabian Nights fantasies, and satirical comedies. Now came Alexandre Volkoff's Kean, released in February, 1924.54 Its subject was the last years of Edmund Kean, the illustrious English Shake- spearean actor, and the historical period was 1830 London. The star was one of the more popular French film actors at the time, Ivan Mosjoukine. A mer- curial actor, Mosjoukine made Kean one of his most successful film roles, playing him as a multifaceted character who envisioned and lived life-as-thea- ter. Like most French film historians, Ren6 Jeanne and Charles Ford consider this film "the high point of the collaboration between Volkoff and Mosjoukine [and cameraman J.-P. Mundvillerl as well as of the Albatros productions."55 Cinea-Cin6-pour-tous's readers agreed by voting it the third best film of 1924, just ahead of Violettes imperiales.56 The scenario for Kean was adapted from Dumas pere's play, written for the famous French actor, Fr6d6rick Lemaitre, just three years after Kean's death; unlike its source, however, it ends in tragedy or, more accurately, in bathos.57 In the opening sequences, during a performance of Romeo and Juliet, Kean falls HISTORICAL RECONSTRUCTIONS 171 COMMERCIAL NARRATIVE FILM 84. Nicolas Koline and Ivan Mosjoukine in the Coaly Hole tavern in Kean (1924) in love with one of his admirers, the Countess de Koeleld (Natalie Lissenko), the wife of the Danish ambassador to England. But he is in no condition to carry on such a love affair. Hounded by creditors, Kean and Solomon (Nicolas Koline), his good friend and servant, have to disguise themselves as a sailor boy and his mother (after Solomon has done a neat turn as a tiger) in order to escape to Hyde Park (where the countess rides by on horseback and says, tauntingly, she prefers him as Romeo). At the Coaly Hole tavern, the two men drink themselves silly, burn up the floor with several jigs and dances, and fall asleep until morning. Soon after, in back-to-back intrigues, Kean has to save the reputation of a young woman, Anna Damby (Mary Odette), who is infatuated with him, and then has to suffer an apparent rejection (of his roses) by the countess, while he is carousing at a men's club. When his rival for her love, the Prince de Galles (Otto Detlefren), appears with the count and countess in the theater box one night, Kean becomes enraged and breaks off his performance in Hamlet to insult them. The count protects his wife (and keeps her from reaching the actor), and when the crowd boos and yells at him, Kean collapses on stage. In a long coda, he retires from the theater, ill and penniless, holes up in Solomon's run-down cottage on the outskirts of London, and wastes away. There, of course, the countess comes to confess her love just before he dies. From its opening, Kean displays an especially effective classical continuity style of editing-the admiring glances of more than a half-dozen women in the theater audience are integrated clearly into the space of Kean's performance in Romeo and Juliet. And the performances of Mosjoukine, Koline, Lissenko, 172 HISTORICAL RECONSTRUCTIONS 85. Ivan Mosjoukine as Kean dying in Kean ( 1921) and Odette are strong and affecting throughout. It is surprising, therefore, that the film's handling of the two Shakespeare plays is so mundane. So un- distinguished are these moments that an intertitle has to tell us that Kean (and Mosjoukine) is a marvelous Hamlet. But the choice of the plays per- formed resonates aptly in the narrative. The actor first sees the countess-an exchange of privileged looks initiate his tragic desire-during the balcony scene between the star-crossed lovers. And he breaks down, mad with envy and grief, during Hamlet's tirade to Ophelia on marriage. What seems to have struck the French reviewers and audiences most about these moments, however, was the illusion of being in the old Drury Lane Theater. This, one of the major spectacles of the film, owed much to the work of Albatros's chief set designer, Ivan Lochakoff. In the old Montreuil studio, Lochakoff meticu- lously reconstructed the interior of the Drury Lane Theater (from designs and photographs he found in the Bibliotheque nationale)." "Its dimensions [were] enormous," wrote Cin6;nagazine, "twelve meters high, and twenty-five meters long. It [was] perfectly furnished: loges, boxes, plush seats; nothing was missing. The moments accepted as classic set pieces by the cin6-clubs, however, were the final sequence of Kean's death and the night-long carousing in the Coaly Hole tavern." Viewed today, Kean's dying seems to go on interminably- much like Hernani's, Jeanne and Ford confess.' Only Mosjoukine's restrained acting in MCU and CU, as well as the intercut exterior shot of several frail trees whipped about in the wind, save the sequence from falling completely into bathos. The Coaly Hole sequence is something else again. Kean's drunken dancing is described in ensemble shots (where several black sailors are prom- inent), hand-held CUs, shots Gf a dog and cat backing off, swish pans, all of which are edited together in a rhythmic montage that changes according to his emotional state, accelerating at several points into short bursts of rapid cutting. At the end, a pattern of subjective inserts culminates in several su- perimpositions of the countess on horseback riding (almost threateningly) to- ward him. Later, in a parallel sequence at the men's club, Kean's hallucina- 173 COMMERCIAL NARRATIVE FILM tions reach a feverish climax. Told of the countess's rejection (the long-take CU of his reaction is marvelously poignant), he drunkenly imagines a distorted version in which his gift of roses is scorned and crushed, and then he is confronted by superimpositions of his rival and the countess (culminating in ECUs of her mouth) that seem to fill the room. If anything in Kean deviates from the conventions of the historical reconstruction genre, it is certainly these sequences in the Coaly Hole and the men's club. The impressive interior decors of Koenigsmark and Kean raise an intriguing point about the development of the historical reconstruction film in France. Its elevation to prominence in French film production coincides with and probably derives from a major change in set design. According to L6on Bar- sacq, it was not until about 1922 that a technical evolution in film scenery freed the set decorators from trompe-d'oeil painting and other conventions of fin-de-sikcle theater so that a new group of architect-designers could enter the profession. "In general," Barsacq adds, "through their exacting demands and their achievements, the Russian designers [especially] contributed to raising the status of film designers to one of close collaboration with directors."62 Interestingly, the newcomers were not employed by the older production com- panies of Pathe-Consortium, Gaumont, Aubert, and Film d'Art. Instead they were made important members of recently established, often independent companies: Diamant Films had Mallet-Stevens; Films Albatros had Lochakoff, Boris Bilinsky, and later Lazare Meerson; the Societe des Films Historiques also had Mallet-Stevens. One can argue, I think, that the work of these ar- chitect-designers was a key factor in the success of the historical reconstruction genre, as it would be later in that of the modern studio spectacular. This change in set design, among other things, clearly differentiates the two major period films of the 1924-1925 season: Raymond Bernard's Le Mi- racle des loups (1924) and Germaine Dulac's Le Diable dans la ville (1925). Both, interestingly enough, are set in the fifteenth century. The source for Le Diable dans la ville was an original scenario by Jean-Louis Bouquet, entitled La Ville des fous.61 The story involved a case of sudden terror that seized an entire village. . . . This contagious fear was provoked by inexplicable phenomena that superstition soon attrib- uted to the devil. The mystery was explained by the presence of a happy band of robbers who, haunting the underground caverns of the old town, duped the inhabit- ants in their houses at nightfall . . . for their own profit.64 Initially, the setting and time were to have been unspecified (yet modern), but the Cin6romans producers convinced Dulac that a fifteenth-century French village would provide a properly plausible context. Because it was a moder- ately budgeted Films de France production at Cin6romans, the film was shot by an undistinguished cameraman, Stuckert, entirely in the studio, using rather old-fashioned stock sets and costumes by Path6-Consortium's chief set designer, Quenu.65 The resulting "demod6" look fails to mesh with Dulac's emphasis on cinematic rhythm (sometimes to the point of abstraction), and both vitiate the narrative's satire and social criticism. In fact, one could argue, as Dulac herself did, that Le Diable dans la ville is really not a period spectacle film at all.66 By contrast, the Russian-emigre-financed Societe de Films Historiques de- 174 cided to make Le Miracle des loups the most expensive historical reconstruction HISTORICAL ever mounted by the French film industry. The director, Raymond Bernard, RECONSTRUCTIONS confessed that he was stunned by the project:"I had the possibility of employ- ing means that had never been used up to that time in France."67 For example, in the newly renovated Levinsky studios at Joinville-le-pont, he and Mallet- Stevens reconstructed the spectacle of a medieval mystery play with "more than a thousand extras."68 For the siege of Beauvais, actually filmed in the fortified city of Carcassonne (which Andre Antoine apparently convinced the producers to use rather than studio decors),' Bernard said, every morning I had four thousand people ready in costumes; I had hundreds of horses in period harnesses; I shot with fourteen or fifteen cameras and usually one of my assistants [including Maurice Forster and Daniau-Johnson} was behind each cam- era. . . . I myself remained above the battlefield ... in the place where I could see everything.67 Le Miracle des loups became a national event when the producers arranged for its premiere, along with a special orchestral accompaniment composed by Henri Rabaud, at the Paris Opera, on 13 November 1924.71 Presiding over the event was the president of the Republic, and in the audience were his ministers, the diplomatic corps, and celebrities in the arts and sciences from around the world. It was "a stunning success," wrote one film journal, "a veritable official consecration of the cinematographe. "72 Within two weeks, the film opened a three-and-a-half month exclusiviti run at the Salle Marivaux.73 One year later, the readers of Cinia-Cin6-pour-tous voted it the best film of 1925 (topping Chaplin's The Pilgrim, Fairbanks's Thief of Bagdad, and De Mille's The Ten Commandments).74 The subject of Le Miracle des loups, adapted by Andre-Paul Antoine from Henry Dupuy-Mazuel's novel, took French audiences back to a period when their own national unity was being forged.75 The time was 1461-1472, the historic years of conflict between Louis XI (played by the famous Atelier actor, Charles Dullin) and his brother, the Duke of Bourgogne, Charles le Tmeraire (Vanni Marcoux)--a conflict that was mediated and resolved, according to legend, by the figure of Jeanne Hachette (Yvonne Sergyl) and "the miracle of the wolves." The film opens with an epigraph from Michelet: "History is a resurrection" (written in a stylized medieval script, as are most of the inter- titles). Then a prologue establishes Charles's desire for the crown and the mutual love between Jeanne and one of Charles's noblemen, Robert Cottereau (Romauld Joube). However, another of his retinue, de Lau (Gaston Modot), is also enamored of Jeanne and plots her abduction during the performance of the mystery play. Louis XI foils the plot and wins Jeanne's allegiance, but Charles breaks off with him, and Robert dutifully has to follow his master. This antagonism between brothers leads to an indecisive battle at Montlhery (during which Robert saves Charles from death) and an uneasy truce. Soon, however, Charles comes to believe that Louis is guilty of inciting an insurrec- tion against him at Liege and makes him a virtual prisoner at Perrone. Jeanne and her father set out to help Louis (with a document proving his innocence); but de Lau pursues them, kills the old man, and chases Jeanne through the snowbound mountains into a pack of wolves, who miraculously ignore her and attack him and his men. Upon reaching Perrone, she and Robert stay Charles 175 COMMERCIAL NARRATIVE FILM So. The 5lcizc of Beauvais III K :1rtI \lish lurrwqn192-4) I 14 S I 4* 176 from ordering Louis's death. Several years later, Charles's army, led by Robert and de Lau (who alone survived the wolves), attacks the city of Beauvais where Jeanne is now living. She rallies the peasants and townspeople, but they are driven back into the cathedral tower, which is set afire. Robert unknowingly wounds Jeanne and then defends her against de Lau in a duel that takes them to the tower ramparts, where de Lau falls to his death. Remembering Robert's service to him in the past, Charles refuses to kill the reunited couple, and Louis's army finally arrives to put him to flight. Louis, at last, is undisputed king. The legend of Jeanne Hachette bears a striking resemblance to that of Jeanne d'Arc (even Louis here corresponds to Charles VII, his father), and the film's phenomenal success probably inspired the production of two later historical reconstructions: Carl Dreyer's La Passion de Jeanne d'Arc (1928) and Marco de Gastyne's La Merveilleuse Vie de Jeanne d'Arc (1929). However, here the char- acters are probably more interesting in the context of earlier films such as L'Agonie des aigles and Violettes impriales. Except for his inadvertent foiling of de Lau's plot and his brief leadership during the battle of Montlhery, Louis is a strangely passive, meditative king, not at all like the ruthless, prevaricating historical figure. In one of the best film performances of the decade, Charles Dullin plays him like an older, laid-back Hamlet or a milder, gentler Richard II (in two sequences with Charles, he adopts a mocking half-supercilious, half- subservient attitude, and he is fascinated by objects of power and by his own powerlessness).76 Much like Napoleon's son in L'Agonie des aigles, his function in the film is not to restore but to be restored. Three times, Robert and Jeanne, separately and then together, save him from defeat and death. Even more than Louis perhaps, Jeanne at Beauvais incarnates the spirit of France- a feminine replication of Jeanne d'Arc or of Delacroix's goddess of liberty during the Revolution. But she, too, must be saved twice, first by a kind of divine intervention in the miracle of the wolves and then by her lover-antag- onist. No conflict of nations, of classes, here. Instead, a civil conflict of brother against brother, of fianc6 against fiancee, that is resolved by the code of suf- fering and sacrifice. If this is the French equivalent of the American Birth of a Nation, the royalist conclusion is unexpectedly muted, but historically pre- scient. Sitting alone before a chessboard, Louis grabs the king, impulsively sweeps all the other pieces away, and, smiling, plumps his piece down in the center of the board. But his final gesture is to raise one hand anxiously to his neck as his face takes on a quizzical, disturbed expression. Does he know where the monarchy will end? Much like Koenigsmark and Violettes impriales, Bernard's film was a showcase for spectacle. But here, I would argue, it is integrated even more effectively into the narrative. The first battle of Montlhery is brief but shockingly real- istic-and looks ahead to Gance's Napoleon (1927), Orson Welles's Chimes at Midnight (1966), and Kevin Brownlow's Winstanley (1976). In a succession of graphic shots, the film catalogues nearly a dozen different ways that the ar- mored horsemen and foot soldiers are wounded or killed. Its effect is registered immediately by Louis's hasty signing of the treaty. Although slightly extended in length, the attack of the wolves is also presented in realistic detail-with MCUs and CUs of the wolves gripping and tearing at the men's bloody necks and faces. Henri Fescourt, for one, criticized the sequence for failing to convey HISTORICAL RECONSTRUCTIONS 177 COMMERCIAL NARRATIVE FILM 87. Mallet-Stevens's interior for the Beauvais Cathedral tower in Le Mirade des loups (1924) the moment of grace; and the means are rather mechanical: horizontal frame maskings simply shift from black to white.' But there is a rhetorical rightness at the end in the juxtaposition of a CU of de Lau's bloody face (the predator has become prey) and an ELS of Jeanne running off across a vast snow-covered valley. The climactic siege of Beauvais is a set piece of spectacular action, staged on a scale clearly reminiscent of the Babylon section of Intolerance. The se- quence is orchestrated clearly, vividly, using an unusual variety of shots (from ELSs, looking past the defenders on the wall toward the attackers swarming over a small hill, to MCUs of a cannon firing or a woman protecting her child); and it shifts the narrative focus smoothly from army versus army, to small group versus small group, to Robert versus Jeanne, then to Robert versus de Lau (all the while intercutting Charles's soldiers pillaging the town as well as Louis's soldiers advancing). For the interiors of Louis's court, of Charles's rooms at Perrone, of Beauvais's cathedral tower, 'Mallet-Stevens eliminated all unnecessary details: splendid, shiny floors and high fireplaces are the only visually rich areas. "' The effect was to create a regal but almost monastic space for Louis and, in the Beauvais cathedral, to conjure up a kind of magical emblem-a LS of two armored knights poised over the wounded Jeanne in a high-ceilinged, shimmering, debris-filled chamber. The film employs metaphor sparingly but effectively-as in Louis's swift, then suddenly hesitant, gestures over the chessboard at the end or in his paradoxical behavior at the beginning-greeting Charles with open brother- 178 liness and then teasingly shooting a crossbow arrow into the shield of his coat of arms. The most extended sequence that relies on metaphor is that of the elaborately staged early mystery play, The Game of Adam (strangely reminiscent of a M6dies fantasy film). As the stage serpent seduces Eve who seduces Adam with the apple (and their clothes drop off), in a side room, Charles sends de Lau outside to bring Jeanne to him. Nimble, cavorting devils push Adam and Eve to the huge dragon mouth of hell, and Jeanne stands before Charles, who orders her to go with de Lau. As she struggles to resist, the king's crown which is nearby falls to the floor, and Louis (who has also been watching the play) suddenly enters the side room. In an economical, measured, dramatically charged montage of shots, Louis looks at de Lau, at the crown, at Charles, and then seizes the crown before his brother can reach it. As he does so, one of his retinue pulls back a curtain, and the crowd stands to face this new side stage that has displaced the mystery play. Slowly, pointedly, the king ad- dresses his antagonist (in an intertitle): "Crowns are like women." And as Charles kneels obediently before him, Louis mischievously raises the crown to his brother's head, only to grab it back. Thoughtfully, he rubs one of the jewels and laughs: "It's cracked!" Charles stares, and the people realize that war is at hand. It is a marvelous moment, worthy of the Shakespeare it echoes- Richard II (Act IV, scene i)-and suggests perhaps better than any other se- quence how effectively Le Miracle des loups works even today. After 1924, the genre of historical reconstruction seems to have developed along two different lines. In part, these developments were linked to the Cartel des gauches policies which encouraged the strategy of international co-produc- tions. But they were also determined by rival groups within the French film industry. One group had its base in the Paramount production center in Paris. It included L6once Perret and Henry Roussel (producer-directors who had already worked for Paramount), Robert Hurel (a former Paramount producer who founded Franco-Film with Perret), and Bernard Natan (an ambitious spec- ulator with ties to Paramount, who built his own studio and eventually took over Path6-Consortium). Most of the period spectacle films this group pro- duced had one thing in common-they focused on a major historical figure in the period just before or after the French Revolution. The first of these films was Perret's second blockbuster, Madame Sans-Gene (1925), which opened at the Salle Marivaux in December, 1925.19 This Par- amount production was made, during the winter of 1924-1925, largely be- cause of Gloria Swanson. The central role of the famous Sardou play, written originally for Rjane, appealed greatly to the American actress.80 She enlisted the aid of film critic Andre Daven to persuade the French film industry and government to accept the project, whose support in turn convinced Paramount to accept her wishes. Ostensibly, the subject of the film was not unlike that of Violettes impriales: a woman's ascent from washerwoman to marshal's wife and finally to duchess. Actually, write Ren6 Jeanne and Charles Ford, Para- mount gave Perret a 14-million-franc budget to "Americanize" the play and make "something appropriately gorgeous for Gloria Swanson."81 A good per- centage of that money went to the nouveau-riche costumes designed by Rene Hubert and to a month of location shooting at Compiegne and Fontaine- bleau.82 HISTORICAL RECONSTRUCTIONS 179 COMMERCIAL NARRATIVE All of France with its artistic and historical resources was mobilized to serve the star. FILM Having at his disposal the huge Henri II gallery at the Fontainebleau palace, how could L6once Perret resist the temptation to add a cortege of five hundred extras in court regalia and set off a fireworks display over a pool of carp, since they were there, awaiting the pleasure of the cameraman? . . . Here the rich, intricately balanced work of Victorien Sardou disappeared under a mass of ornaments. . . .83 Yet Madame Sans-Gene received a tumultuous reception at its Paris premiere; it won the Jury Grand Prix at the 1925 Exposition des Arts Decoratifs; and Gloria Swanson still considers it one of the best films in which she appeared.84 Although some French critics denigrated Perret's work as no more than "a charming armchair fantasy," others described it as the first great film of the French-American collaboration.85 Their reaction was complicated by the fact that the shortened American version of the film was released six months earlier than the French and received far less glowing reviews.86 Another quite similar historical tableau was Madame Rcamier (1928), which Gaston Ravel directed for Franco-Film. To mount the story of the most bril- liant salon hostess of the early Napoleonic years as well as of the Restoration period, Ravel took his large cast and crew to the actual historical sites of the action: the Folie de Saint-James park, the Coppet chateau, and, of course, Fountainebleau.87 The story begins during the Terror of 1793, when Juliette Bernard (Marie Bell) becomes the young wife of the wealthy M. Rcamier (Victor Vina).88 The marriage is a chaste one because Recamier is actually Juliette's father who has decided to atone for his past affair with her mother (Madeleine Rodriguez). Mme. de Stael (Fran oise Rosay) soon becomes a good friend of Mme. Recamier and spirits her off to Germany when Napoleon and his minister Fouche (Van Dadle) threaten her life. There she falls in love with a Prussian prince (Franois Rozet); but, when her aging husband refuses a divorce, she sacrifices her love and returns dutifully to him. This whole story is a long flashback, framed by an evening late in the life of Mme. R6camier (Nelly Corman) as she discreetly explains to her close friend, Chateaubriand (Charles de Bargy), why she cannot marry him. Ren6 Jeanne remembers Ra- vel's film as respectful, tasteful, but, unfortunately, not particularly satisfy- ing.89 Perhaps the most interesting thing about Madame Rcamier was its source and the timing of its production. Ravel drew his adaptation from a biography by Edouard Herriot, the prominent leader of the Radical-Socialist party, for- mer prime minister (in 1924), and current minister of art and education. Shot late in 1927, Madame Rcamier was premiered at the Op6ra, on 12 June 1928, just after the spring national elections-which the Radical-Socialists lost.90 It also followed the Herriot Decree (in February), which adopted a very weak quota system for the importation of films into France. This conjunction of events raises some interesting questions. Did either Herriot or Franco-Film consider the film as a political instrument before the elections? Was Franco- Film soliciting the minister's favor, for themselves or for the film industry in general? Whatever, the film was successful enough for Ravel to nearly repeat its subject in a celebrated, but mediocre, sonorized version of Dumas's Le Collier de la reine the following year.9' "The mystery of the queen's necklace" was also the subject of Cagliostro (1929), an Albatros-Wengeroff (French-German) production based on Dumas's 180 88. Gloria Swanson in a Fontainebleau interior in Madame Sans-Gene (1925) novel, Joseph Balsamo.92 Jules Kruger and Jean Dr ville shot the film in Paris, under the direction of the German metteur-en-scene, Richard Oswald. In it, Cagliostro (Hans Stuwe) was an Italian doctor at the court of Louis XVI (Van Dadle) and Marie-Antoinette (Suzanne Bianchetti). Implicated in the cele- brated affair with the Prince de Rohar (Alfred Abel), he was not exiled in the film, as Raymond Villette complained, but transformed into "the architect of the French Revolution. "" This was a blatant example of rewriting history in the service of film industry internationalism. Perhaps as a consequence, the film's great number of decors (some forty in all), designed by Lazare Meerson and Firenzi, were notable for their sobriety and balance. In style, they were much closer to the simple clarity of Mallet-Stevens than to the extravagances of Lochakoff and Bilinsky.5 Cagliostro was rushed into release in late June, 1929, and had the dubious fortune to share a double bill at the Paramount- Palace with the new talkie hit, A Song of Paris (starring Maurice Chevalier). ) The most successful period spectacle films, however, came from a second group within the French film industry. Their common denominator was a 181 COMMERCIAL NARRATIVE FILM loose network of financiers, producers, directors, and set designers whose cen- ter was the Russian emigre colony in France and Germany. One faction in- cluded the directors (Tourjansky and Volkoff) and set designers (Lochakoff and Bilinsky) initially associated with Alexandre Kamenka at Films Albatros and then with Noe Bloch at Cine-France (financed by Stinnes and Wengeroff) and the Societe G6nerale des Films (financed by Grinieff). The other was the So- ciet6 des Films Historiques (also financed by Grinieff), whose major director was Raymond Bernard. The films that these two factions produced tended, at least initially, to look outside France for their subjects-to eighteenth-century Italy and especially to eighteenth- and nineteenth-century Russia. Although generally ignored in film histories (partly because only incomplete prints seem to have survived), they deserve some attention. And not only because their assured direction, magnificent sets and costumes, and fine acting (led by Mos- joukine and Dullin) made them quite successful in France and the interna- tional market, but because they created benchmarks of spectacular action and sumptuous tableaux in the genre that provided a context for the work of Gance, Dreyer, and others in the late 1920s. And, perhaps equally important, they allowed the Russian emigre colony to celebrate-and criticize in their own way-the country and society from which they had fled when the Bol- sheviks set about transforming Russia into the Soviet Union. The first of these films was Tourjansky's Michel Strogoff, an epic Cine-France- Westi production which, after delays in financing, premiered at the Imperial Cin6ma in December, 1926.9' The scenario was drawn from the Jules Verne adventure novel about a czarist courier who carries out a dangerous mission in Siberia." The Tartars, aided by several Russian rebels, are threatening czarist control of the area between Omsk and Irkutsk in south central Siberia. From Moscow, Czar Alexander II (Gaidaroff) sends Strogoff (Mosjoukine), because he was born in the region, to order the Irkutsk governor to seize the rebel commander Ogolieff (Chakatouny) and rout the Tartars. Coincidental with this mission, two journalists, Blount and Joulivet (Henri Debain and Gabriel de Gravonne), decide to journey through Siberia to Irkutsk. As their path continually intersects with Strogoff's, they provide some comic relief (which may not be all that necessary) to the narrative. On his way down the Volga river, Strogoff befriends a woman by the name of Nadia (Natalie Kovenko), only to lose her to a Tartar band which seizes their boat, wounds him, and dumps him overboard. After his escape and recovery, Strogoff disobeys orders and visits the chateau of his mother (Jeanne Brindeau) near Omsk. Inadvert- ently, she betrays his presence to the rebels, and both of them are seized separately and taken to the Tartar encampment at Enofer, where a major festival is in progress and where Nadia is also held captive. When both mother and son continue to deny one another's identity, the Tartars temporarily blind Strogoff with a red-hot sword, and his mother dies from the shock. After the Tartars depart, Nadia and Strogoff struggle through the snowbound moun- tains to Irkutsk, only to discover that Ogolieff is impersonating the courier in the palace. Confronting him alone, Strogoff suddenly regains his sight and defeats his rival in a hand-to-hand combat. He then proves his identity to the governor and helps repulse the Tartars who have begun to attack the city. Returning to Moscow, he is rewarded by the czar and marries Nadia, his "dearest prize," in a richly detailed Russian Orthodox ceremony. 182 Although the ideology this narrative serves is quite regressive-a Russia ruled by the czar and by the Russian Orthodox Church, in which a heroic individual such as Strogoff can perform herculean tasks and be rewarded in kind-Tourjansky's film is much more interesting than it may seem. Most of the production was filmed in Latvia and Norway by Bourgassov, Toporkov, and chief cameraman L.-H. Burel, who had just finished shooting Feyder's two films in Switzerland and Austria, Visages d'enfants (1925) and L'Image (1926).9" The skirmish on the river boat, the Tartar attack on Omsk (with a magnificent HAELS of hundreds of white-robed soldiers rising en masse and charging up and over the slope of a hill), Strogoff s escape from Omsk (shot quite skillfully at night), the arduous mountain journey-all gain a natural- ness and rough immediacy from the authentic location shooting. Given the conditions and the mode of the narrative, Lochakoffs sets, especially the Tartar village at Enofer, are unusually restrained, almost realistic. Perhaps the prox- imity to their homeland curbed the Russian emigres' penchant for the fantas- tic, which usually characterized their films. Whatever the reason, even in the three-reel 9.5mm Pathescope version that alone seems to survive, the sober cinematography and swift, economical rhythm of Michel Strogoff are quite im- pressive. Both Tourjansky and Burel came to Michel Strogoff fresh from several months of work on the early stages of Abel Gance's Napoleion. That experience was partly responsible for the range of technical effects in the film, especially the use of unconventional camera movement. The most sustained example of tech- niques usually associated with the narrative avant-garde occurs in the opening 89. (left) Ivan Mosjoukine in Michel Strogo/f (1926) 90. (right) The Tartar encampment 183 91. The wedding ceremony at the end of Michel Strog~ol (1926) sequence of a grand ball at the czar's palace in Moscow. The Path6scope version of the film opens with a HALS looking straight down on the dancers and then a long dolly out from the orchestra box perched over the ballroom floor. The major set piece of the sequence, however, is a subjective moment involving the czar, after he has spoken with one of his generals about the Irkutsk crisis in a room adjacent to the ballroom. Alone now, he pores over a map, and a series of shots depicts the Tartar horsemen sweeping across the landscape (in several swift tracks and dollies) and rampaging through a village. Suddenly, to the intercutting between a MS of the worried czar and various shots of the Tartars is added LA shots of the ballroom dancers' feet and CUs of the cymbals in the orchestra. The rhythm accelerates into rapid cutting, and the sounds of the dance ironically seem to impress the Tartar threat on the czar's conscious- ness. The moment leads immediately to his decision to send off Strogoff on his mission. A similar moment occurs later to Strogoff at a way station when 184 a blustering Russian officer seizes the horses that have been reserved for his use. The camera dollies swiftly into a MS of their confrontation; then, after the officer strikes Strogoff, there is a LA dolly in from LS to MS on the czar as he stands in the shadows of his study. The subjective insert (in which one camera movement parallels the other) keeps Strogoff from returning the blow and perhaps revealing his identity. Compared to Michel Strogoff, Volkoffs Casanova, which also premiered at the Imperial, in September, 1927, is much less tightly constructed and much more fantastical in style.l100 Its narrative is drawn from Casanova's Memoirs and is little more than an episodic series of adventures, oscillating from comedy to tragedy, from melodrama to satire. It begins in Venice with Casanova (Mosjoukine) threatened by a bailiff who has come to take possession of his house. With a bit of phony magic, he delays the threat and then arranges a rendezvous with the Baroness Stormont (Olga Day). The baron surprises them, however, and takes his case to the governing council of Venice, which orders Casanova's arrest. When a trap to capture him on the Rio San Traverse Bridge fails (the prey makes a spectacular dive into the canal), Casanova enjoys a night of love with the baroness (while the baron sleeps peacefully in the next room). Sometime later, Casanova organizes a dinner for the famous Venetian dancer, Corticelli (Rina de Liguoro). An uninvited guest, the Russian Count Orloff (Paul Guide), tries to gain her attention (as he tosses roses at her, Casanova spears them neatly, one by one, with his rapier), and Casanova challenges him to a duel. After he has won easily, Corticelli joins their hands in friendship, and the orgy, presumably, continues. Still pestered by the council, Casanova escapes to Austria in the company of a black boy servant whom the baroness has given him. One night in an inn, he rescues a young woman, Carlotta (Jenny Jugo), whose mother has sold her to a rich brute of a man. However, the rescue proves ineffectual-the man's henchmen simply follow them and take her back again. Now penniless, Casanova meets the royal dressmaker of Paris, on his way to the court of Peter III in Russia. The adventurer promptly steals his papers, money, and clothes and goes off in his place. In St. Peters- burg, he meets Orloff again, who protects him from the emperor when he shows some interest in the Empress Catherine (Suzanne Bianchetti). After the coronation of Catherine (Peter is assassinated mysteriously), Casanova wavers in his affection between Catherine and Bianca (Diana Karenne), newly arrived with her father from Venice. Finally, he flees with Bianca for a brief tryst in her carriage (interrupted by soldiers) and leaves Catherine his black servant (which apparently pleases her). Returning to Venice at the time of the Carnival, Casanova finds Bianca among the revelers and spends the night with her in a covered gondola. The next morning, awakened by the screams of a young woman, he rescues Carlotta all over again. Pursued by the bailiff (who appears out of nowhere) and Carlotta's husband, Casanova hesitates between the two young women. When he chooses Carlotta, Bianca shoots and wounds him. From prison, he is forced to watch her execution. Just before he, too, is to die, Carlotta and several of his friends help him to escape and set off to sea on a sailing ship. At the last moment, he almost cannot resist taking a young gypsy woman with him. The character of Casanova was well suited to Mosjoukine. It allowed him to exploit his penchant for multiple personalities through disguises and play- HISTORICAL RECONSTRUCTIONS 185 COMMERCIAL NARRATIVE FILM acting, and he gave perhaps the best performance of his career, equaling his portrayal of Strogoff and even Edmund Kean. All in all, he plays Casanova like a Douglas Fairbanks adventurer, with more sophistication, irony, and cool sexual presence. His disguises range from the fantastical magician who de- ceives the bailiff early on (parodying the magicians in several German films- e.g., The Golem [19201, Waxworks [19241) to the clown and sailor costumes he uses to escape at the film's end. When disguised as the royal dressmaker in Russia, he even indulges in a bit of drag comedy, dressing up for Catherine in a wide floppy hat, corset, bustle, and fan. The character's sexual exploits also provide the film with several voyeuristic sequences that are rather unusual for the period. To check the bailiffs bluff of a threat, for instance, Casanova lets him gaze on Corticelli half-naked against the fine, cascading drapery of her couch. Later, at the dinner he arranges for her, she and a dozen other women perform a nude dance-described mostly in silhouette against a set of windows. The erotic nature of the dance is suggested in Corticelli's gestures and facial expressions (in MCU), intercut with Casanova's, in a shot of the men staring forward as the women's shawls descend over them, in another shot of them handing over their rapiers to the outstretched women's hands, in the silhouetted figures of the women dancing in pairs and playing with the rapiers. At the climax, Casanova carries Corticelli naked through the group of men and sets her like a draped statue on the steps above them. With its concern for spectacle and sometimes heroic adventure, Casanova also exploits some of the strategies and techniques associated with the narrative avant-garde film practice. The film opens, for instance, with a brief subjective shot. A HALS of swirling women dancers and a LS of a Venice palace are superimposed over a CU of Casanova's sleeping face-as a dream site of pleas- ures. Near the end, a similar subjective moment appears more ominously. In a MCU of the gondola interior, Bianca leans over to kiss him awake, and her image dissolves briefly into the image of Carlotta. Other sequences are marked (the cameramen were Bourgassov and Toporkov) by deformations, unusual camera angles, and esoteric lighting effects. In one, the magician is made grotesquely mammoth in several distorted shots and then deflates like a punc- tured balloon (cf. the balloon travelers on the metro in Clair's Entr'acte [19241). In another, Casanova and the Baroness Stormont are bathed in the most ro- mantic of lights. The CU of their kiss (she on the left, he on the right) floods her face with a key light from the right, while his is nearly silhouetted (still dripping beads of water from the canal) with a back light from the left. When they part later, it is in a FS of a highly decorative grille, sharply etched with a single light from the right rear. There is even an effective sequence of rapid cutting that employs several unusual CUs. The sequence occurs as Casanova paces his quarters in the Aus- trian inn and hears a struggle in the next room. Intercut with CUs of Casanova listening and looking one way and then another, before he leaps to the door, are a series of dimly lit shots-an ECU of a mouth opening, a CU of a woman's white shoes backing off to the right, a CU of a man's hands reaching in from the left, a HAMCU of shadows on the floor and then a woman's shoes crossing to the left, a CU of her hands clawing at a wood panel, a CU of boots pausing and then moving to the right, a CU of an arm struggling in a flash of white cloth, and an ECU of a hand coming away from the open mouth. 186 U V -1', 92. (left) Olga Day and Ivan Mosjoukine in Bilinsky's costumes in Casanova (1927) 93. (right) The Petersburg palace exterior Although these shots never last less than a second or two, they are perfectly selected to describe the near rape and, simultaneously, to evoke Casanova's subjective experience. The sequence operates more effectively perhaps than does a similarly experienced attack in L'Herbier's Feu Mathias Pascal (1925), where Mosjoukine is in a similar position. More than anything else, however, Casanova is a showcase for the spectacle of Lochakoffs decors, Bilinsky's costumes, and Venice itself. According to Barsacq, "Lochakoff and Bilinsky directed their efforts toward the spectacular aspect of scenery, toward an overall effect rather than atmosphere or the search for exact detail." "" An important element in Lochakoffs decors was the ma- quette plastique, an intricate scale model, usually representing the upper portion of the filmed image. He employed this latest technique from the German film studios in conjunction with sets constructed according to the designs of sev- enteenth- and eighteenth-century Italian operas. "0 The Venice and St. Peters- burg settings, consequently, took on the decorativeness of a Diaghilev ballet. Complementing these fabulous decors were Bilinsky's rich variety of costumes. Compared favorably with Bakst's work for Diaghilev, Bilinsky's costumes be- came the major attraction at the second Exposition du cin6ma (Galerie d'Art de la Grande Maison de Blanc) organized by the Cine-Club de France in the spring of 1927."" Finally, both sets and costumes were embellished by an exquisite stencil-color process that included as many as three or four sharply distinguished colors in a single frame." The most stunning effects were carefully withheld until the last two major 187 I pL 94. The bedchamber of Catherine 11 in Ca-anova (1927) episodes of the film-the sequences in Catherine's court and the Carnival of Venice. The principal exterior for St. Petersburg was a vast snowy expanse in front of the palace, produced by means of a maquette plastique and a simply decorated snow-covered landscape. It is here that Casanova's carriage collides with that of the Count, and he and Bianca can use the ensuing chaos to flee together. The best sets, however, are the interiors: the throne room where Catherine sits under the golden emblem of the czars and an immense white dome with statues projecting from the columns, the cavernous main hall where the banquet and the coronation ball are held (with the guests dressed in blue- greens, light reds, and golds). For her ascent to the throne, Catherine wears the most fabulous of Bilinsky's costumes. In a HAELS of the throne room, with people lining both side walls in an arching curve, she enters from the bottom of the frame, trailing a magnificent dark cloak (was it deep blue in the original?) emblazoned with the gold patterns of the czar's emblem. As she moves farther into the frame, the cloak stretches longer and longer behind 188 95. Casanova's gondola at the end of the Carnival sequence in Casanova (1927) her, supported by as many as two dozen servants, until it nearly fills the entire floor space of the room. The climactic sequences of the Carnival in Venice were apparently even more spectacular than these. Cinfa-Cin-pour-tous waxed ecstatic over the finale: The Carnival of Venice is, in one sense, the culmination of this historical and artistic reconstruction. It is also one of the masterpieces of set design in world film production. Jewels glittering everywhere, brilliant red costumes, gondolas passing rhythmically over the still canal waters, everything encompassed in the magnificent orderly movement of the seething crowd."0 Even Jean Dreville was impressed with "this remarkable ensemble of images- remarkable for its movement and the rhythm of that movement. "106 Unfor- tunately, the only surviving print of Casanova in the United States has suffered some radical recutting (probably for its distribution here), especially in the final reel. The Carnival is reduced to little more than a sketch, and Casanova's 189 COMMERCIAL NARRATIVE FILM escape from prison is a bit of a jumble. Perhaps the sequence that still works most effectively is the execution of Bianca (even despite a confusion in the plot). The sequence alternates MCUs of Casanova behind the bars of his cell and LSs of Bianca with the executioner and a priest silhouetted on a raised plaza in front of an immense arch. In the foreground of the LSs is the blurred silhouette of a bar pattern, which, together with Casanova's anguished face pressed to the cell bars, makes her death shockingly poignant. A similar mo- ment would be utterly transformed a year later in Dreyer's La Passion de Jeanne d'Arc. Much like Casanova, Raymond Bernard's second film for the Soci6t6 des Films Historiques, which cost six million francs and opened to box office records at the Salle Marivaux in January, 1927, also tended more to the dec- orative and spectacular. 107 Its title, Joueur d'6checs (1927), could be seen as a clever takeoff on the final sequence of Le Miracle des loups, but the subject was an intrigue involving legendary independence fighters in Poland and Russia in 1776. The narrative is worth summarizing.108 Two young Polish aristocrats in Vilno, Boleslas Vorowski (Pierre Blanchar) and Sophie Vorowska (Edith Jehanne), are pupils of the Baron de Kempelen (Charles Dullin), an inventor of life-size mechanical mannequins. Although both are leaders of the inde- pendence movement and Boleslas loves Sophie, she is attracted to Serge Ob- lonoff (Pierre Batcheff), a young officer in charge of the Russian forces in Poland. When Boleslas is wounded in a spontaneous insurrection in Vilno, she elects to stay by his side-even though he has a price on his head. Kem- pelen comes up with a stratagem to hide him in a new chess-player mannequin which he sends to the Polish court at Warsaw; but a Major Nicolaieff (Camille Bert), who recognizes Boleslas's chess moves, diverts the mannequin to St. Petersburg. There, Catherine II (Mme. Charles Dullin) challenges the man- nequin, is accused of cheating, and promptly orders it shot. With Oblonoffs assistance (for he now realizes Boleslas's greater love for Sophie), Kempelen substitutes himself in his own mechanical creation and dies in Boleslas's place. Struck by the young couple's love, Catherine allows them to return to Vilno, where they are pursued by Nicolaieff. When he steals into Kempelen's house at night, suddenly the doors lock, and he is methodically dispatched by a small army of sword-bearing mannequins. The prophecy of an old gypsy thus comes true-Poland's freedom will come in the shape of a woman. Sophie and Boleslas unfurl the flag of the Polish eagle and (according to an intertitle) "go off bearing high once more the colors of Polish history."109 Since only certain parts of Joueur d'ichecs could be shot on location in Poland (using the Polish army for the Vilno insurrection), the scenario demanded that Bernard's set designers, Mallet-Stevens and Jean Perrier, and his chief cam- eraman J.-P. Mundviller construct a number of mammoth decors at Joinville: the gallery of the palace of Catherine II . . . the streets and squares of Vilno and Warsaw, and the grandiose court of the Winter Palace [at St. Petersburg], which alone covered an area of 8,800 square meters. . .1.0 There were thirty-five decors in all, and the Winter Palace set was described at the time, by G.-Michel Coissac, as one of the three largest ever constructed for the cinema worldwide.1 Perhaps the most interesting feature of Perrier's work, according to Barsacq, was that he "developed a rational concept of film 190 HISTORICAL RECONSTRUCTIONS 96. The Russian patrol in Vilno at the opening )f foIetn d'6ches (1927) set design as a function of the position of the camera and the lenses. . . . The movements of the camera and actors [could be] determined in advance. . . . 12 The result was a "complicated labyrinth of multiple sets, with passageways behind the decors, in which one set sometimes served as a frame for the previous one, with intersections of all sorts that formed a welter of de- tails. ...""..Consequently, Joueur d'echecs is lull of spectacular effects: "the uprising in the streets of Vilno, the battle of encirclement (where in one LS, scores of cannon camouflaged on a hillside fire simultaneously), the masked ball in the court of Catherine II, the night celebration at Vorowski's."i" But, as in Le Miracle des loups, these are integrated smoothly into the narrative. Despite all the spectacle, Bernard's film is marked by an extremely concise form of editing. The opening sequence, for instance, establishes the conflict between the Russians and Poles in a tense, economical pattern of intercutting. While Russian horsemen patrol the night streets of Vilno, members of the Polish resistance meet in a nearby chateau. The Russians are introduced as huge shadows gliding over the city walls, then in MCUs of the horses' hooves, and in dolly shots describing their slow methodical progress. One horseman nonchalantly raises a whip, and an old woman crouched by the wall spins around and collapses from its lash. Inside the chateau, the Poles pause in their singing, waiting for the patrol to pass. Boleslas reaches across the piano to take Sophie's hand, but she looks at him askance. Later, in a much-admired 191 COMMERCIAL NARRATIVE FILM Charles Dullin and 1 chessplayer mannequin in Jfoieur d'i ers (1927) sequence, she imagines the insurrection is victorious as she sings the national anthem at the same piano. m5 As the camera dollies in on her, a superimposi- tion dissolves in of horsemen charging off to the right. Again the camera dollies in on her, and the horsemen seem to sweep out of her hands across the piano top toward the left background. As her excitement fades, Sophie gazes at the necklace Oblonoff had recently bestowed on her, and a subjective image of him is cut in momentarily. She looks up and seems to see the next shot of Boleslas falling wounded in the rocks of the battlefield. Abruptly she rips the necklace from her neck and, in MCU, lets it fall from her hand. Cognizant now of the insurrection's failure, her choice is made-her love will be sacri- ficed to her duty to her country and its freedom. The film's celebration of Polish independence from the Russian monarchy in a period just prior to the French Revolution tempts one to read it as a displacement of the one onto the other. With a few exceptions-e.g., Gance's Napolion-the French cinema generally avoided its country's own revolution, just as the American cinema has avoided its period of revolution. Both stand in marked contrast to the Soviet cinema of the 1920s, which functioned as a major collective ritual in celebrating (sometimes also by displacement) the construction of a new society. Joueur d'checs thus seems to affirm the spirit of independence and revolution (and sacrifice) by safely packing it away to an- other land. The figure of the Baron de Kempelen, however, introduces a cu- rious disruption. He is really half-mad-this teacher, this substitute father- 192 98. The mannequin attack at the end of Joueur d'hecs (1927) a genius who lords it over his own fictional empire. But history intervenes to give his mannequins a double purpose. There is a rather fascinating meta- phorical schema operating in the transformation of this man into a mechanical chess-player, after which the rest of his mannequins change briefly into men. The master takes on the role of one of his subjects and sacrifices himself so other subjects can live. He loses one game so they can win another. The creator/teacher dies so his children can be born. But when they come to life, they take revenge; they destroy a master (albeit a different one) all over again. Does this revenge of the mannequin men substitute for the victorious revolu- tion of the children? And can this be a revolution-this bizarre, half-Chris- tian, half-Oedipal fable of sacrifice and revenge? Bernard's last silent film epic, Tarakanova (1930), returned to the world of Catherine II's court to narrate the tragic end of Elizabeth Tarakanova (Edith Jehanne), a young pretender to the throne, and her lover Prince Orloff (Olaf Fjord), Catherine's court favorite. Wishing only to punish Orloff by separating him from Tarakanova, the Empress inadvertently sends him off to die with his love. "c Shooting on the film (the cameramen were Kruger and Lucas) was 193 99) ( j(-f t) A jprotiii i to photo, worh m1a(//a'11 /Idal/6/11 for murir sIiut prialw- ii t 1$ maiqiiil 1)b fra1li completed in February, 1 929, bUt the edliting was (delayed so it could be releasedl in a sonojrized version dIuring the summer of WO9. Thme delay pushed the cost of the film to almost seven million francs.' Bernard once told Kevin Brownlow that he h imoself comisiderecd 'arik(novel his hest film by far, but no print heas been red isc overedl to con firm or deny his judgment.' The success of these big films by Tourjansky, Volkoff, and Bernard prompted several other producers to try to capitalize on these genre subjects, but with less satisfying results. Plenty Roussel, for instance, took a less active French- POlIShI subject, and in a rather different mode, in Lei Vedse deC /c/wi ( 1928). Subtitled "A page in the life of Frederic Chopin,"' this was a biography, wrote one critic, "more spiritual than chronological. Roussel tried to make Cho- pin into a sN'mbol ic foigure, wrote another: 'In the desperate yiouth and sublime death of a Imusi'-ian twasi the story ot a misunderstood genius destroyed on the calvary of a love that was both patssion and Passion. I2The result, most agrecd, wais aponderous, scntimcnt il film wxhosc only. saxving grace xvas the actin or1 Pir: B- m lncha itnd MIo hell B i a1n Sahenc trota an the r marc Jt il ta k I 11rn Hunri Kr rem.'.ck~r t> cni fl an - 'r1 i scena,-1 -. i1tn IS . .V. ''t I %!.'t r yrac..'':o:, r< ] oral- ti1', ._-( 1:2>.ia.V i ~~---it: dear, Ar~e1'La L:. L): . r.lo orhr C' t CI A:;D FLo.- r:.. . Pierre Batcheff (Albert de Morcerf), Jean Toulout (Villefort), Henri Debain (Cade- HISTORICAL rousse), Mary Glory (Valentine), and Gaston Modot (Fernand Mondego).123 RECONSTRUCTIONS His production team was headed by an incongruous trio: the sober, story- oriented Henri Fescourt as director, the spectacle-minded Boris Bilinsky as set designer, and Daniau-Johnson as technical director.124 The combination ap- parently gelled, for most French film historians agree that Monte-Cristo (1929) was one of the best historical reconstructions of the decade. According to Barsacq, In the winter palace set . . . Bilinsky re-created a vision of the palace as it might have been conceived by Italian opera designers of the eighteenth century, by Bibiena or Piranesi, with a monumental stairway, colonnades, arcades, and balconies; but it was a constructed set, at least up to a certain height: the upper part consisted of a scale model and the base was painted in trompe-d'oel. 12 Fescourt himself remembers with pleasure how he and Bilinsky used this same process to film an 1815 sailing ship entering "the old port of Marseille on a busy day in 1929 without the public's noticing the least bit of contempora- neous movement on the quay or in the harbor. . . .126 Wrote Cinmagazine, In the set pieces such as the evening at the Opera or the Count's celebration, [Fescourt produced] a charming evocation of the aristocratic life in 1845; the crinolines, the tasteless men's fashions, the decorum that reigned at these ceremonies gave to the images a sort of vapid grace. . . .127 The dominant characteristic of this Monte-Cristo, however, seems to have been a sense of movement that, even more than in Casanova, evoked the ballet. Unfortunately for Nalpas, the film was released in October, 1929, just as the first French talkies were hitting the cinemas. 128 The biggest of the French historical reconstruction films sponsored by the Russian emigre money, of course, was Abel Gance's Napolon (1927). Beside it, Henry Roussel's much heralded Destine (1926), an opportunistic but com- paratively tepid, lifeless look at Napoleon during the Revolution, has paled into oblivion. 129 Gance's project, initially, was colossal in scope-six separate films on the life of Napoleon Bonaparte. Yet only the first film of the project actually was completed; and its enormous expense (15 to 19 million francs) and epic length (six hours) quickly dispelled the commercial hopes of Gance's fnanciers.10 Moreover, the film turned out to be profoundly personal and highly experimental, and the filmmaker had the audacity to title it Napolon. vu par Abel Gance. The version that premiered at the Opera, on 7 April 1927, was a condensed film of close to 5,000 meters (including several tryptich sequences).Y' The following October, the complete version of 12,000 meters (but without the tryptichs) opened en exclusivit6 at the Salle Marivaux and ran for nearly three months.' 2 This full-length film was not widely shown, how- ever, and, according to Kevin Brownlow, the three-hour condensed version was generally released throughout France and the rest of Europe.I"3 Whatever its variations and excesses, Napolon fulfills all the conventions of the historical reconstruction film. With Alexandre Benois (Diaghilev's chief set designer) and the Russian-born architect Schildknecht,134 Gance meticu- lously recreated the period of Napoleon's boyhood at Brienne and his young manhood in Corsica, the period of the Revolution in Paris and of his courting 195 COMMERCIAL NARRATIVE FILM A 4 . 100. Vladimir ROUcienko and Albert Diciadonn in Nam/c4';i rii far Abel Gamec ( 19 7) and marriage to Josephine de Beauharnais, and the period of the Italian cam- paign. Much of the film was shot on location (Corsica, Brianqon, Toulon, Nice), but most (even the sea journey from Corsica and the night battle in the rain at Toulon) was done in the huge new Billancourt studio which Dia- mant-Berger, with Mallet-Stevens, had built in 1923. Gance's team of assist- ant directors (including Volkoff, Tourjansky, and Henry Krauss) and camera- men (Jules Kruger, L.-H. Burel, J.-P. Mundviller, Roger Hubert, Emile Pierre, and Lucas) produced enough spectacle for a half-dozen films: the snow- ball and dormitory fights at Brienne, the escape from the army in Corsica, the Marseillaise sequence in the Club des Cordeliers, the long night battle for Toulon, the Victims' Ball, Napoleon's confrontation with the dead heroes of the Revolution. And, of course, there were the famous tryptich sequences of the Double Tempest (at sea and in the Convention in Paris), the military descent over the mountains into Italy, and the climactic victory at Monte- notte. Besides this, Napoleon had some forty major roles for what seemed to be half the number of available French actors. They included Albert Dieu- donne (Bonaparte), Vladimir Roudenko (Bonaparte enfant), Antonin Artaud (Marat), Edmond Van Daele (Robespierre), Alexandre Koubitsky (Danton), Pierre Batcheff (Hoche), Maxudian (Barras), Nicholas Koline (Tristan Fleuri), Armand Bernard (Jean-Jean), Philippe Heriat (Salicetti), Chakatouny (Pozzo de Borgo), Gina Manes (Josephine de Beauharnais), Annabella (Violine Fleuri), Eugenie Buffet (Letitia Bonaparte), Suzanne Bianchetti (Marie-Antoinette), Marguerite Gance (Charlotte Corday), Suzy Vernon (Madame Recamier), Damia (La "Marseillaise"), and Gance himself (Saint-Just). " The complete version of Napoleon was a grandiose paradox. Gance once said that he made the film because Napoleon was "a paroxysm in a period which was itself a paroxysm in time." "" Conceived in the ideological ferment of the early 1920s, his rhapsodic celebration of a single, powerful leader or mover of history (a sort of Gallicized Hegelian ideal) seemed anachronistic, chauvinistic, and even dangerous. Rewriting history, the film depicted Bonaparte as the legendary fulfillment of the French Revolution-"the soul of the Revolution." For Gance, the French hero (much like himself) must have seemed the Ro- mantic artist in apotheosis, a towering figure who made the real world, not an imaginary one, his own province of action. For others, such as Leon Mous- sinac, this figure was the embodiment 'of military dictatorship," frighteningly close to the image of the emperor then held by political groups of the extreme right."- Moreover, the film's conclusion in political and military triumph, in victory (admittedly, a result of the project's incompleteness), went against the French pattern of suffering and sacrifice that so marked even the historical reconstruction genre. Yet Gance's technical innovations alone-e.g., the range of camera movements and the multiple screen formats-influenced French filmmakers for the rest of the decade. Time and time again, as we shall see later, his imaginative handling of both the spectacular and the intimate tran- scended the conventions of the genre. The difference between period spectacle films produced by the two rival groups in the French film industry are probably no more blatant than in the two films that mounted the story of Jeanne d'Arc in the late 1920s. The maid of Orleans had been the subject of several books after the declaration of her sainthood in 1920, and her life seemed a logical "property" for a French film, 196 10 1. The Convention Hall in Napoleon vu par Abel Game (1927) especially after the success of Le Miracle des loups, with its celebration of Jeanne Hachette. The more conventional and decidedly more commercial of the two was produced by Bernard Natan-Marco de Gastyne's La Merveilleuse Vie de Jeanne d'Arc. This ambitious project took nearly two years to shoot (at a cost of between eight and nine million francs), finally premiered at the Opera late in 1928, and was showcased at the Paramount-Palace in April, 1929. "' Jeanne was played by a well-known tomboy actress, Simone Genevois, supported by expert villains such as Philippe Heriat as Gilles de Rais and Gaston Modot as Glasdell. The scenario, by Jean-Jose Frappa, traced Jeanne's story from 1428, when she first had her visions in Domrfmy (visualized in a circle of white- robed women dancing in slow motion, superimposed over a tolling tower bell), to 1430, when she was burned at the stake in Rouen. Gastyne's interest lay primarily in moments of grand spectacle-the coronation of Charles VII at Notre Dame, the capture of Jeanne at Compiegne, and especially her suc- cessful siege of Orleans. The latter was filmed almost entirely on location at 197 COMMERCIAL NARRATIVE FILM 102. (left) Simone Genevois in La Aerveleuse Vie de Jeanne D'Arc (1929) 103. (right) Falconetti in La Passion deJeanne DArc (1928) Carcassonne and was staged on a scale that rivaled the similar siege in Le Miracle des loups. The action is brisk and clear, punctuated by LA shots of horsemen crashing over a line of barricades and by a LAFS of Jeanne halfway up a scaling ladder as she is struck by an arrow. But Gastyne did not ignore other more poignant moments of Jeanne's solitude-a MCU of her praying, after the Orleans siege, between a dead soldier's head and a wooden post from which is suspended a heavy iron chain; a HALS of her alone, head bowed, on a small stool in the empty space of the trial chamber. Although Simone Ge- nevois's Jeanne is angry enough to accuse her interrogators in Rouen's public square, she is also plainly afraid to suffer burning at the stake. The other film, of course, was Carl Dreyer's La Passion de Jeanne d'Arc, which opened in Paris at the Salle Marivaux on 25 October 1928. 11 Appar- ently, the Societe Generale des Films engaged the young Danish filmmaker to make a vast fresco that would complement its production of Napoleon. If Dreyer is to be believed, the choice of Jeanne d'Arc as the film's subject was decided in a match draw ironically, because he picked the headless match. 140 As early as May, 1927, Cinemagazine quoted one of the project's financial backers, the duc d'Ayen, as saying, "We want to reproduce the most accurate and the most poignant events in the life of Jeanne d'Arc, the heroic emblem and mirror of the people."' Despite an extraordinary budget of over seven million francs, two mammoth set constructions (one of great white interiors near the Billan- court studio, the other a reconstruction of the medieval village of Rouen), 142 and major actors (Rene Jeanne Falconetti as Jeanne, Eugene Silvain as Cau- chon, Maurice Schutz as Loyseleur, Antonin Artaud as Massieu), La Passion de Jeanne d'Arc was also a deeply personal, experimental film, some of whose uniqueness comes from its deliberate inversion of the genre conventions. It is clearly an anti-historical reconstruction. Dreyer's scenario focuses neither on the pageantry of the times nor on Jeanne's military and political successes, but on her spiritual journey during the last day of her life (the time of her trial and execution is condensed into what seems a single day). Based closely on 198 HISTORICAL RECONSTRUCTIONS 104. The jousting tournament in Le Tournoi (1929) the records of the trial at Rouen, the film is simultaneously a documentary of Falconetti's ordeal as Jeanne and a symbolic progression of faces within an unusually disjunctive space-time continuum. As we shall see later, La Passion dejeanne d'Arc was one of the crowning achievements of the French narrative avant-garde. Finally, two period films fall somewhat outside these two camps of produc- tion, but both more or less counter the conventions of the genre, although quite differently than does Dreyer's film. The more conventional of the two is another production of the Societ6 des Films Historiques, Jean Renoir's Le Tournoi, which opened at the Salle Marivaux in February, 1929." Set in the court of Charles IX and Catherine de Medici, the scenario by Dupuy-Mazuel is predicated on the Protestant/Catholic conflict of the period, which is re- solved here in a tournament. The exteriors were shot by Marcel Lucien and Maurice Desfassiaux in Carcassonne (again) during "the Bimillennial celebra- tions," and the tournament was staged specially for the occasion. " Mallet- Stevens's interiors and Georges Bargier's costumes were noted for their au- thenticity; and the presence of a former world fencing champion, Aldo Nadi, in the leading role corroborated this impression. In fact, despite the intrigue between the Protestant Francois de Baynes (Nadi) and Catholic Henri de Ro- gier (Enrique de Rivero) over the favor of Isabelle Ginori (Jackie Monnier) and the consequent jousting combat that deteriorates into swordplay and mace- swings, Le Tournoi seems to have been marked by an unusual degree of realism. 199 COMMERCIAL NARRATIVE FILM As one of the few film critics recently to have seen the film, Raymond Durgnat was struck by the weight and feel of the weapons and banquet ware, by the detailed cumbersomeness of the knights on horseback, and by the presence of a blacksmith watching over the lovers-a shot that "locates their idyll in an everyday society."145 Andre Bazin remembered the violence and cruelty of the duels-"the blood from the sword wiped on Lucrece's hair [de Baynes's mis- tress]."146 Perhaps most disturbing was the ambiguity of de Baynes's charac- ter-highly skilled in combat and thoroughly profligate, perfidious and yet finally noble in death. As in Le Bled, the hero seems to mock the genre from within. Although Renoir himself considered the project little more than an exercise in technique (e.g., the construction of an overhead dolly shot down the length of a long banquet table),147 Le Tournoi can be seen as one more instance of a general pattern of genre subversion in his work. That subversion is even more apparent in Renoir's earlier Nana, an ambi- tious film that got out of hand. Aubert released the film in the summer of 1926 (it premiered at the Aubert-Palace) to good reviews and large audiences and then shelved it to make way for its fall releases. 148 When the company tried to redistribute a slightly shortened version in December, interest in the film had passed. 149 That sabotaging of Nana's commercial success contrasts with a different kind of sabotaging within the film itself.150 Before making Nana (with his wife, Catherine Hessling, in the starring role), Renoir had discovered Stroheim's Foolish Wives and had engaged in a study of French gestures in the paintings of his father and others of his gen- eration.151 In adapting Zola's novel, he determined to produce his own study of a society through a combination of strategies. First, Renoir eliminated most of the first half of the novel and concentrated on Nana's life in the theater and her affairs with three men: Count Muffat (Werner Krauss), Count Vandeuvres (Jean Angelo), and Georges Hugon (Raymond Guerin).152 The narrative opens comically, with Nana's last successful theatrical performances, gradually dark- ens, and ends with her death from smallpox. Renoir also conceived of a unique juxtaposition of the decors and the characters. Much of the film's million-franc budget, in fact, went for the construction of elaborate sets that resurrected the final years of the Second Empire: the giant staircase, drawing room, bed- chamber, and dressing room in Count Muffat's townhouse; the racetrack stands at Longchamps; the Variety Theater; the Mabille dance hall (all designed by Claude Autant-Lara at the Gaumont studio in Paris and then reconstructed in Berlin). 153 Yet Renoir deliberately consigned these decors to secondary impor- tance in the film itself. And he underlined their "demode" quality by having Autant-Lara design rather Impressionist costumes according to the later fash- ions of 1871 and after. 154 Moreover he emphasized the acting of his three main characters (in contrast to what he considered a weakness in the French cinema of the time) to the point of "constantly shooting them in American-style two- shots"-a strategy that effectively eliminated the decors.m155 His most important strategy, however, was to have the actors employ quite divergent acting styles. 156 Werner Krauss (the star of Caligari) plays Count Muffat in a humorless, ponderous, pathetic German manner-almost an aris- tocratic variation on Emil Janning's hotel doorman in Murnau's The Last Laugh (1924). By contrast, Jean Angelo plays Count Vandeuvres in a suave detached manner, as if he has just stepped out of an Oscar Wilde comedy. And Ray- 200 i4, 105. (left) The central staircase in Count MUffet's townhonse in Nana ( 1926) 106. (right) A publicity photo of Catherine Hesslin mond Guerin plays Georges Hugon with such a self-effacing, drooping timid- ity that his character scarcely exists. As Nana, Catherine Hessling clearly dominates the film. No sensuous blonde seductress (as in Zola), she plays Nana as a sulky, cajoling, irrational child-woman who reverses the class- and male- dominated patterns of her time. Hessling developed a strange style of acting, derived from pantomime, for the role: syncopated rhythms and crisp, mechan- ical gestures, like those of an automaton or an animated doll. 117 She fascinates and upstages her men just as all of them, seconded by a couple of silent voyeuristic servants (Harbacher and Valeska Gert), upstage the period decors. In this stylized character study, a whole society is revealed in the process of self-destruction. Some years ago Noel Burch called Nana "a key film in the development of a cinematic language" because its "entire visual construction depends on the existence . . . of a {fuctuating] off-screen space." "' Nana was the first film, according to Burch, to exploit the structural use of that off-screen space sys- tematically. Along with Alexander Sesonske, I am afraid I cannot agree. Ac- tually, Burch is less interested in proclaiming Nana's originality than in using the film to exemplify the various ways that off-screen space can be defined and be said to function. There are many earlier instances of off-screen space func- tioning significantly in the diegetic process of a film. Sesonske cites D. W. Griffith's films, specifically, The Lonedale Operator (1911); but one could go back as early as The Lonely Villa (1909) as well. ' Furthermore, as Sesonske argues, off-screen space is not all that influential in Nana. And Burch tacitly admits this in his later writings on the French cinema of the 1920s by avoid- ing any mention of the film." Sesonske, in fact, suggests something of much 201 COMMERCIAL NARRATIVE more importance which operates structurally in Nana-the enactment of the- FILM atrical performances at crucial moments of the narrative. Specifically, there is Nana's comically sensual performance of Venus in the opening sequence, the opening night flop of Muffat's production of La Petite Duchesse, her star turn at Longchamps, and, near the end, her hysterical abandon and collapse at the Bal Mabille. As the fiction that has been her life fails, writes Sesonske, Nana "turns to art, a spectacle, a performance, as a refuge from, and . . . a solution to, life's problems."161 Within this satirical inversion of the historical recon- struction film lies the germ of an idea for Renoir-performance as an act of mediation and resolution. There is one other subgenre of the historical reconstruction film that bears mentioning-the war film, or the reconstruction of World War I. After Abel Gance's J'Accuse (1919)-which will be discussed in Part IV-the Great War vanished from French films except for brief references (e.g., La Croisade, Koe- nigsmark), as if there were a tacit taboo on the subject. But in 1927, several films appeared to break the silence. Their immediate inspiration probably was the extraordinary success of King Vidor's The Big Parade (1925), released in France early in 1927. But also approaching was the tenth anniversary of the Armistice (11 November 1928). In advance of the anniversary, the association of war veterans put together two documentaries from the war footage stored at the Archives d'Art et l'Histoire: Pour la paix du monde and Verdun. 162 Several fictional films with the war as their setting were initiated; of these, the most important was Maurice Tourneur's adaptation of a Joseph Kessel novel, L'E- quipage (1928), which premiered at the Imp6rial Cinema in April, 1928.163 Even though produced by the Societe G6n6rale des Films and released by A.C.E., L'Equipage was called a Franco-American production because Tour- neur and his leading actress, Claire de Lorez, were considered American while the rest of the cast was French (Jean Dax, Daniel Mendaille, Georges Charlia, Camille Bert, Pierre de Guingand).164 Its story revolved around the friendship of two pilots near the end of the war, the one unaware that his mistress is married to, but separated from, the other. When he is killed in combat, the friend forgives his wife, and they resume their life together. This tragedy of the war has none of the habitual faults of the films that evoke mem- ories of those terrible years and contemplate the struggle of millions of combatants. Here is a truly cinematic production where the image is master and where all the effects are visual. 165 Interestingly, Tourneur's film focused on the air war, just as did William Wellman's Wings (1927), which was released almost simultaneously in the United States. Both undoubtedly drew on the interest generated by Charles Lindbergh's solo flight across the Atlantic. Although the French film press considered L'Equipage one of the more beautiful films of the 1927-1928 season, the French public was less impressed. They remembered that Tourneur had not returned to France for the actual war, and (according to Kevin Brownlow) his return to re-enact it created such a hullabaloo that he beat a retreat to Germany. 166 The major historical reconstruction film on the war, however, was Leon Poirier's Verdun, visions d'histoire, whose premiere at the Opera, 10-18 Novem- 202 HISTORICAL RECONSTRUCTIONS 107. A symboli rableau in 4Ve ain. Vins D'Ilit.re bet 1928, was timed to commemorate the tenth anniversary of the Armi- stice. l6 The following summer, it was released for an exclusive run at the Imperial Cinema. " At first, Poirier had the patronage of the French govern- ment to reconstruct the strategic hut costly defense of Verdun in the spring of 1916. Although that patronage was eventually withdrawn either for finan- cial or political reasons (the 1928 elections), the government gave its com- plete, though unofficial, support by allowing Poirier to film on the actual battlefields and to use documentary footage from the Archives d'Art et I'His- toire. With an eerie concern for authenticity, Poirier recalled soldiers who had survived the war to re-create some of the major battles. For instance, the survivors of the Driant Hunrsmen themselves participated in the reconstruction of those terrible days from 21 to 25 February, where, after having been pounded by a twenty-four-hour bombardment, they were assaulted by a mass of soldiers ten times their number who finally gained the position after four days of battle and the death of Colonel Driant, and penetrated the Verdun defenses like a wedge as far as the fort of Douaumont. Their almost incomprehensible willingness to undergo the horror of that ex- perience, again and again, testifies to the importance for the French of that historic moment at Verdun. Even German soldiers volunteered for the re- enactment. Wrote one, "Your film will do much to promote peace by bring- ing together France and Germany because we both need to better understand the spectacle of our common suffering."A Poirier's scenario integrated a chronological progression of documentary footage 203 COMMERCIAL NARRATIVE FILM with re-enacted sequences of battles and of various individuals engaged in or affected by the war. Instead of following a single continuous storyline in the re-enacted sequences, he chose to focus on "symbolic characters"-the French soldier (Albert Prejean), the German soldier (Hans Brausewetter), the peasant (Jose Davert), the son (Pierre Nay), the husband (Daniel Mendaille), the young man (Jean Dehelly), the intellectual (Antonin Artaud), the chaplain (Andre Nox), the German officer (Tommy Bourdelle), the old marshal (Maurice Schutz), the wife (Suzanne Bianchetti), the mother (Jeanne-Marie Laurent)-who, ac- cording to Rene Jeanne and Charles Ford, "would enlarge the scope of the battle, transform it into tragedy. ..171 "Here," Poirier added, "there would be no story, no rousing or extravagant adventure, no sentimental or easily edifying intrigue."172 The film's three parts--"La Force, L'Enfer, Le Destin"- would be a tribute to the collective suffering of the French people. 173 Although critics and historians generally praise the smoothness of this in- tegration of the dramatic and the documentary as well as the carefully re- strained use of symbolic figures, the 1932 sonorized and re-edited version of the film (which is the only print I have seen) looks rather rough and uneven (Poirier himself preferred the silent version). 174 The changes from re-enactment footage to authentic documentary footage (or vice versa) are sometimes ob- vious; some sequences are unusually brief and others verge on incoherence; the overall rhythm of physical movement and narrative seems lax and uncertain. Still, there are stunning and poignant moments: the realistic detail and concise cutting in the initial German attack, a grisly night encounter in "no man's land," the final image of seed-sowing in the devastated earth. Much more technically conventional than Napoleon, say, Verdun, visions d'histoire was also more ideologically suited in subject to the French public of 1928-and it was no less popular. The French spirit was defined in terms of suffering, sacrifice, respect for one's enemies, and a desire to return peacefully to the past (where it still might be possible to cultivate one's garden). As Close Up suggested in 1929, taken as a work of propaganda or ideological statement, if not as an achievement, Verdun clearly bears comparison with Eisenstein's October (1927) and Pudovkin's The End of St. Petersburg (1927).175 For the French film industry, the genre of period spectacle films or historical reconstructions was clearly the most prestigious of the decade. Year after year, its films were among the finest commercial productions: Les Trois Mousquetaires (1921-1922), Les Opprimis (1923), Koenigsmark (1923), Violettes impiriales (1924), Kean (1924), Le Miracle des loups (1924), Michel Strogoff (1926), Joueur d'dchecs (1927), Casanova (1927), Verdun, vision d'histoire (1928), Monte-Cristo (1929). And, until Madame Sans-Gene (which was more American than French), Napo- leon, and La Passion de Jeanne d'Arc (which both deviated sharply from the norm), the genre was also consistently profitable, justifying, at least finan- cially, its high percentage of capital investment. Moreover, except for the earlier films, almost all were produced and/or financed by Russian emigres. The significance of the Russian contribution to the French cinema of the 1920s is perhaps nowhere more evident than here. These 1920s films set the basic conventions for the genre and defined its standards. In the 1930s, the genre's prestige declined slightly as many earlier films were re-made as talkies, though not as well: Diamant-Berger's Les Trois Mousquetaires (1932), Rich6b's L'A- 204 gonie des aigles (1933), Gance's Napoleon Bonaparte (1934) and J'Accuse (1937), Tourneur's Koenigsmark (1935), Baroncelli's Michel Strogoff (1937). But there were at least two clear successes within the conventions: Bernard's reconstruc- tion of the Great War in Croix des bois (1932) and Feyder's reconstruction of seventeenth-century Flanders (with a distinctly collaborationist plot) in La Kermesse hiroique (1935). Besides, in Grande Illusion (1937) and in La Marseillaise (1938), Jean Renoir continued to subvert the genre, to turn it to other ends. Two decades later, a survey of the French public's preferences in film would show that the popularity of the historical reconstruction genre had continued unabated. 176 During the latter half of the 1920s, the French film industry discovered a new genre to invest in that soon challenged the priority of the historical reconstruction film. Perhaps more than previous genres, the modern studio spectacular had quite definite political, economic, and cultural origins. The modern studio spectacular was a product of the cultural internationalism which by then characterized urban life in most of the industrialized countries of Europe. Its development in France coincided with the shift in historical re- constructions after 1924 and also can be attributed in part to the French film industry's increasing involvement in international co-productions. In order to reach as many exhibition outlets as possible, especially the American market, the industry sought formats and subjects beyond those of the period films, that would appeal not just to French audiences but also to Americans and other Europeans. The American craze that was sweeping France by the mid- twenties offered a kind of model. While American tourists flocked to the new casinos, dance halls, and beach resorts, the French stood in line to see Amer- ican jazz musicians, singers, and dance troupes.' At La Revue Negre in the Thatre des Champs-Elys6es, for instance, they gaped at Josephine Baker, "The Nefertiti of Now," performing a "stomach dance" before "pink drops with cornucopias of hams and watermelons."2 The modern studio spectacular was designed to exploit such "cross-cultural" exchanges. And it was encour- aged by the success of the American "flapper" films-e.g., Flaming Youth (1923), The Masked Bride (1925), and Dancing Mothers (1926)-and that of the studio spectaculars produced by Paramount and UFA.3 The parameters of the genre were clearly fixed by 1926-1927. The milieu was the key ingredient for this genre of the Parisian nouveau riche, especially younger men and women, who sported and played at the newest nightclubs, restaurants, dance halls, resorts, or at their own art deco mansions. According to Gerard Talon, these films pictured the good life of a new generation and helped define what was modern and "a la mode" in fashions, sports, dancing, and manners generally.4 Instead of looking to the past and to venerated tra- ditions, or to some other land or culture, the studio spectacular looked to the present and the future. But this present or future was not particularly French. For the genre tended to produce either a picturesque image of contemporary France according to American stereotypes or a kind of generalized milieu that was socially and aesthetically neutral.5 Stylish set designs and costumes (some- times as elaborate as those in the period films) were showcased in sensational set pieces. International casts were common, but they were natural to this MODERN STUDIO SPECTACULARS Modern Studio Spectaculars 205 108. Messenier's Montmartre restaurant set for Konigsmark (1923) kind of neutral decor where actors could easily replace one another as types. The overall effect of the genre was a fantasy of internationalism that denied the specificity of French culture and acceded to the hegemony of the American cinema and the new ideology of consumer capitalism or conspicuous consump- tion.6 The modern studio spectacular, therefore, was probably the least French of all the genres deployed by the French film industry. Isolated elements of the genre can be found in the French cinema at least as early as Perret's Koenigstnark (1923) and Mosjoukine's Le Brasier ardent (1923). Perhaps the major set piece of Koenigsinark was the wedding-night celebration in a lavish Montmartre restaurant constructed by Henri Menessier at the new Levinsky studio in Joinville-le-pont. Menessier's work in the United States for Maurice Tourneur and then Metro Pictures-he had just returned to France with Pearl White for Terreur (1924)-led him to design the Montmartre set specifically as "an amusing attraction with a little color for America!" Its central pool-enclosed dance floor and art nouveau/deco walls provided the pro- totype of an essential setting for the modern studio spectacular. Besides, sev- 206 eral of Huguette Duflos's costumes, designed by Bou6 Soeurs, were advertised as the latest in fashionable elegance.8 A shorter but similar sequence occurs in Le Brasier ardent. Designed by Lochakoff on a smaller scale at Albatros's Mon- treuil studio, its setting is an underground Parisian caf6 where Mosjoukine incites a group of chorus girls to a marathon contest of faster and faster danc- ing.9 The genre almost gelled in Robert Boudrioz's L'Epirvier (1924) and, more importantly, in Marcel L'Herbier's L'Inhumaine (1924). A Paramount produc- tion based on a popular melodrama by Frangois de Croisset, L'Epirvier told the "tragic" yarn of a card-sharping Hungarian couple in Rome, who separate when the wife falls in love with a diplomat. 10 Unfortunately, Boudrioz's cast of Sylvio de Pedrelli and Nilda du Piessy was almost unknown (in contrast to L'Herbier's 1933 version, starring Charles Boyer), and the film had little im- pact. L'Inhumaine, which premiered at the Madeleine-Cinema in late Novem- ber, 1924, on the other hand, caused a sensation, but an unfavorable one." Produced independently by Cinegraphic and financed almost entirely by American money through its star, Georgette Leblanc, L'Herbier's film was deliberately designed to showcase the most modern French painting, sculpture, architec- ture, costume design, and music (as well as cinema) to the American public. 2 The milieu was contemporary Paris-more specifically, art deco mansions (ex- teriors by Mallet-Stevens, interiors by Autant-Lara and Cavalcanti), the Th- atre des Champs-Elys6es (with the Ballet sudois), and a scientific laboratory (designed by Fernand Leger). All these sets were constructed at the new Le- vinsky studio, with an airy lightness that contrasted sharply with the dark decors of conventional films. 13 The characters included the famous singer, Claire Lescot (Georgette Leblanc), and a cosmopolitan group of men attracted to her-Einar Norsen (Jaque Catelain), Maharajah Djorah (Philippe Hriat), Frank Mahler (Kellerman), and Tarsky (L. V. Terval)-whose offers of love and wealth she accepts with glacial, impartial indifference. The first part of the film narrates Einar's apparent death (by suicide) and "resurrection," which almost breaks Claire's composure. The second part narrates her actual death (from a viper hidden by the jealous Djorah) and her miraculous resurrection in Einar's laboratory, which causes her to admit her love for him at last. Most of the genre elements come together in L'Inhumaine: the milieu, the ultramodern decors (although they are quite specifically French), the cosmo- politan characters (and cast), the narrative of a modern woman who is cele- brated, defeated, and transformed. However, there was both too little and too much in the film for 1924. The "d6mode' nature of the narrative-it is not much different from some of Feuillade's and Pearl White's wartime serials- blatantly contradicts the ultramodern settings. 14 And the decors themselves seem to exist independently of one another (almost in separate fictions), never quite producing contingent spaces (one could argue that they were never meant to). Besides, just as in L'Atlantide, the aging figure of Georgette Leblanc makes a rather unconvincing Circe-like seductress, a very problematic object of desire, even for the spectator. 15 Yet L'Herbier also used the project to deploy and experiment with all the technical devices of film discourse then available to him-that is, various forms of montage, including accelerated montage; various forms of masking; superimpositions; brief flashes of solid- color film stock; the fragmentation of objects. The result is a film that, some- MODERN STUDIO SPECTACULARS 207 139. (ieft) Georgerte Leblanc in I In/ mi ne (1924) S11)) (rigzht) Bilinsky's poster for Le Lmon deis MXogo/ (1l92-1) what like Gance's La Roue, was as outdated as it was avant-garde. And the combination was not commercially advantageous. Another film which opened, at the Mogodor-Palace, about the same time as L'Inh/maine combined fewer elements of the genre somewhat more conven- tionally and with a rather cool facility. That was Albatros's Le Lion des AMogols (1924), directed by Jean Epstein from a Mosjoukine scenario, with sets by Lochakoff and costumes by Bilinsky. 1 The scenario juxtaposes a fantastical Arabian Nights palace and modern Jazz Age Paris, neatly mediated by the presence of a film production company. As a consequence, Mosjoukine gets to play the rebellious son of the Grand Khan, a mysterious naive celebrity in Paris, and a movie actor in various heroic roles. More than half of the film takes place in Paris, and two major set pieces there develop the prototypical sequences from Koenigsmark and Le Brasier ardent. These are a night of drunk- enness at the popular Jockey Bar (also reminiscent of the Coaly Hole sequence in Kean-the chief cameraman again is Mundviller) and a climactic masked ball at the fashionable Hotel Olympic. The film's attitude toward this final sequence is not exactly fashionable: the masked bail itself masks a murder in the rooms above, and it finally serves to unmask the real nature of the hero 208 and heroine (brother and sister), who return to the country of their origin and to the fantasy palace. Henry Roussel's La Terre promise (1925), which opened at the Cine Max- Linder in January 1925, was much more serious in its juxtaposition of old and new worlds.17 It had a seriousness that was acknowledged by most film critics and even honored by official bodies such as the Comite franqais du cin6ma. And it became the subject of the first issue of La Petite Illustration (4 April 1925), a deluxe variant of the popular magazine novelizations, published by the prestigious biweekly, Illustration. 18 For the film's subject was unique: the conflicts within a Jewish family, with one branch in London and the other in a ghetto village in Poland. The story begins in the Polish village of Scaravaloff on the eve of the war, as Moise Sigoulim (Maxudian), a London financier, makes his annual pilgrim- age to celebrate Passover with his brother, Rabbi Samuel (Bras), and his two daughters, Lia and Ester. A business deal ensues-an arrangement to drill oil on the land of the provincial governor, Count Orlinsky-and Moise invites Lia and Ester to London where they can receive a suitable education. Ten years later, Moise has become a millionaire oil baron; Lia (Raquel Meller) has taken a degree in engineering; and Ester (Tina de Yzarduy) has become a vain co- quette. On a visit to London, the count's son, Andre Orlinsky (Andre Roanne), finds himself attracted to Lia over Ester, despite his antipathy, as a Christian, toward the Jews. In Scaravaloff, meanwhile, Count Orlinsky, who now man- ages Moise's oil holdings, has decided to lay off many of the Jewish workers; and Samuel travels to London to appeal directly to Moise on their behalf. When they all return to Scaravaloff, the family conflicts intensify: Ester vows to avenge Andre's rejection of her love, and Moise asks to marry Lia, who reluctantly agrees, even though she loves Andre. On the day of the wedding, however, the Christian workers threaten to attack the oil wells, fearing that the Jews are about to be hired in their place (a rumor started by Ester). Lia and Andr6 find themselves jointly trying to calm the workers and, then, with the help of Samuel's assistant David (Pierre Blanchar), saving the oil wells from a fiery destruction. In the end, Moise and Ester admit their errors and ask for forgiveness, while Lia and Andre recognize their mutual love and plan to marry. At least two points could be made about this audacious film. Although the set decors for the London townhouse sequences, especially Moise's Friday eve- ning reception dinners, were as lavish as any that had yet been constructed for a French film, critics were particularly astonished by the detailed evocation of the Jewish ghetto milieu.9 And the film's attitude toward this juxtaposition of milieux was intriguingly ambiguous. Moise's London world is roundly con- demned-through Ester's addiction to it as well as Samuel's disapproval of the risque costumes, jazzband music, and dancing couples. But the Scavaraloff ghetto scarcely draws unequivocal praise-Samuel is willing to accept an in- cestuous marriage, and he apparently refuses to bless the marriage of Lia and Andre at the end. This ambiguity raises questions that are central to the film, for it seems unusually concerned with the representation of religious, ethnic, and class differences and their resolution. Moise's seeming denial of his ethnic and religious heritage pits him against his brother, while Samuel's rigid ad- MODERN STUDIO SPECTACULARS 209 COMMERCIAL NARRATIVE FILM herence to that heritage brings him into conflict with his daughter. At the same time, the two brothers are separated by class tensions that threaten to destroy their family as well as the larger community. The distribution of these conflicts reaches a point where the contradictions seem irresolvable, except through the displacement common to melodrama-a climax of near catastro- phe and individual heroic action. Created by the vengeful jealousy of one woman, this catastrophe is dispelled by the love and self-sacrifice of her sister, abetted by the man who loves her. In Lia and Andr6, then, all conflicts and contradictions seem to be mediated and resolved, much as they will be two years later in Fritz Lang's Metropolis. But here the mediation remains momen- tary, even suspect, as Samuel's recalcitrance attests. Despite its critical and commercial success, La Terre promise did not put in play all of the elements that were coming together to constitute the modern studio spectacular. That honor probably goes to Cin6-France-Westi's Ame d'artiste (1925), which premiered at the Salle Marivaux in July, 1925.20 Directed by Germaine Dulac, it was adapted by the director and Alexandre Volkoff from a contemporary Danish play by Christian Molbeck. The cosmopolitan nature of the production-its financing was French, German, and Russian emigre- also extended to the cast, which included English, French, and Russian emigr6 actors. Dulac's way of handling this diverse group was quite opposite Renoir's in Nana and may well have set a precedent for the genre: she imposed a uniform style of acting which effectively neutralized any cultural differences among them. Ame d'artiste was set in modern London, in the milieu of the theater, probably, as Charles Ford suggests, in order to capitalize on the suc- cess of Kean from the previous year.21 However, this image of London was created almost entirely in the studio with Lochakoff designing the sets and Jules Kruger doing the camerawork. There was a complete theater set (as in Kean), a huge hotel set for a masked ball (as in Le Lion des Mogols), a luxurious townhouse (as in La Terre promise), a simple cottage on the city's outskirts (again as in Kean), and a cheap hotel and bar. And there were elegant costumes for the two principal actresses and for nearly everyone at the masked ball. Yet, except for the names and exterior shots designating London, all these settings could just as easily have been French or American. The melodramatic narrative of Ame d'artiste has some affinity to L'Inhumaine but also has the advantage of being more plausible and more emotionally authentic.22 The film opens rather cleverly in a series of FSs and CUs: a do- mestic quarrel between a husband and wife (with two children in the back- ground) climaxes with her picking up a knife to stab him. Suddenly, an ELS reveals the previous action to be a play, a play whose dramatic conflict ironi- cally sets up expectations quite opposite to what actually happens in the film. For a while thereafter, the narrative develops rather conventionally through contrasts between the frivolous life of a famous actress, Helen Taylor (Mabel Poulton), and her lenient father (Nicolas Koline) and the rather sedate life of the Campbells-Herbert (Ivan Petrovich), an aspiring dramatist, his wife Edith (Yvette Andreyor), and her mother (who owns the cottage they live in); between the power of the rich theater-owner, Lord Stamford (Henry Houry), and the naivete of Herbert, who falls in love with Helen. However, the masked ball sequence, in which Stamford discovers Herbert and Helen embracing, has several nice touches: Herbert is disguised as a clown, and Stamford's falling 210 MODERN STUDIO SPECTACULARS 1 1 1. Ivan Petrovich in the bar sequence in Arne d'artste (1925) cigar ash is intercut with the confetti that descends over the lovers. The se- quence in which Herbert asks Edith for a divorce is unusually poignant: it ends with a LS of Herbert going out the front gate into an empty street and then a MCU of Edith closing the door to simply stand there waiting. The sequence between Helen and Edith has a similar poignance, but its narrative function is stronger: Edith brings her Herbert's supposedly lost play, and they consider who loves him most. The climax of the film comes in a melodramatic but quite effective alter- nation between Helen's performance of Herbert's play, Ame d'artiste (with Edith in the audience) and Herbert's attempted suicide in a cheap hotel room (he believes Helen has gone back to Stamford). The alternation is complicated and intensified by subjective shots of Herbert's delirium, and his rescue by Edith is delayed until the final shots of the film. Thus, the subject ultimately is sacrifice: that of Edith for her husband's career as a playwright and that of Helen, who, transformed by Edith's action, yields Herbert to her while also advancing her career. In the end, Helen has her career, Edith has her love, and Herbert has both love and career-through the sacrifice of the two women. The "artist's spirit" transcends her milieu to articulate a typically French theme, one in which Dulac must have taken little pleasure. In one major aspect, however, Ame d'artiste differed from most of the mod- ern studio spectaculars that followed-beyond the fact that it is still quite watchable even today. Its milieu depends on the theater rather than on the 211 112. A production photo of L'Herbier (right of camera) and his crew in Mallet- Stevens's restaurant set for Le Vertige ( 1926) 1920s phenomenon of nightclub restaurants, dancing, and automobile tour- ing. In the next two years, through several big films released by three of the top French film production companies, the genre anchored itself securely in these modish settings and activities. The first of these was Cineromans' Films de France production of Le Vertige (1926), adapted by L'Herbier from a celebrated boulevard melodrama by Charles Mer6.' The film's popularity so unnerved L'Herbier, who had given much more of himself to L'Inhu;aine and Feu Mathias Pascal, that he avoided further film projects for several months.' According to Jaque Catelain, Le Vertige was a rather simplistic drama in which Countess Svirski (Emmy Lynn) escapes her tyrannical and-just for good measure--criminal husband (Roger Karl) through the intervention of the hero (Catelain himself).5 In record time, L'Herbier shot the exteriors at the fashionable resorts of Eden-Roc and Eze on the C6te d'Azur.6 More time was taken for the interiors-especially an ultramodern house for Catelain and a late night restaurant-designed by Mallet-Stevens and furnished by Jacques Manuel." L'Herbier even got Robert and Sonya 212 Delaunay to donate several of their large canvases to decorate the house, and the set was so striking that Cin6romans used it as a set piece in one of their current serials, Rene Le Somptier's chic, sporty Le P'tit Parigot (1926).28 Now that the 1925 Paris Exposition des Arts Decoratifs had legitimized the art deco style, these interiors were "a la mode." Despite L'Herbier's own reser- vations, several critics at the time accepted Le Vertige as "a prodigious synthe- sis" of the avant-garde and the commercial.29 Even Cinia-Cine-pour-tous's read- ers voted it fourth on the 1926-1927 list of best films shown in Paris.30 When L'Herbier had recovered sufficiently from the experience, Cineromans naturally asked him to repeat Le Vertige's success, this time in a joint produc- tion with Gaumont-British. The result was Le Diable au coeur (1928), starring Betty Balfour and Jaque Catelain in another "marvelous decor for dancing."3 L'Herbier himself remembers it only as his first exposure to panchromatic film stock-which, in the shift from interior to exterior scenes, led to peculiar changes in his star's eyes and costumes.32 Despite the relative failure of Boudrioz's L'Epirvier (1924), Paramount also did its part in establishing the genre by backing its chief French director, Leonce Perret, in an adaptation of one of Henry Bataille's best sex plays, La Femme nue (1926). Updated to the twenties, the film told the story of a young painter who, once having achieved success, abandons his wife and model who, in turn, tries to commit suicide and is rescued by a friend. The subject allowed Perret to pose his stars, Louise Lagrange and Ivan Petrovich, against a series of chic Parisian tableaux, such as Montmartre caf6s and Sacre-Coeur. Despite his reservations about its theatricality, Jean Dreville considered La Femme nue Perret's best film; and Cinia-Cine-pour-tous devoted nearly half an issue to the film when it opened at the Madeleine-Cinema, in December, 1926.33 Later the magazine's readers agreed with all that attention by voting it ninth on their list of 1926-1927 top films.34 Just over a year later, Perret repeated the combination in La Danseuse orchide (1928) for Franco-Film."3 Both films were instrumental in elevating Louise Lagrange to the position of "Princess of the French cinema" in 1929.36 Another co-production (by SOFAR) very similar in format to these two films was Augusto Genina's L'Esclave blanche (1927)-in the words of Dr6ville, "one of the better commercial productions from which the artistic sense has not been completely excluded."37 The third company to invest heavily in the genre was Aubert, with Julien Duvivier's adaptation of a popular Pierre Frondaie novel, L'Homme a l'Hispano, which opened at the Salle Marivaux, in January, 1927.38 Raymond Chirat sums up the film's narrative quite aptly: It is a common story where the old theme of the handsome knight, pure and invincible, who frees the captive princess, is adapted to the decors of luxury hotels, chateaux set in large gardens, and vast estates. A young and charming woman, badly married to an English lord, is attracted to an emotional young man, the owner of a new sportscar, and, after a good number of incidents of middling interest, finds herself passionately, happily, in love.39 According to Aubert's publicity, Duvivier took advantage of the new resorts in "the loveliest spots along the Cte d'Argent and in the Basque country" to shoot his exteriors, and Aubert himself bragged about spending a million francs on the interiors of the Oswill villa.40 In fact, Aubert was so impressed MODERN STUDIO SPECTACULARS 213 a & Fe 'IIv 113. (left) Jaque Catelain in a Mallet-Stevens's interior in Le Vertige (1926) 114. (right) Louise Lagrange and Ivan Petrovich in La Femme nue (1926) with the film-or worried about its costs-that he told an interviewer, "Here's the first French film which makes me want to sail to America just to show the Americans what we are capable of doing. " Julien Bayart was less ecstatic: it was "a good narrative film" which at least included a dandy race between a train and the famous Hispano, with some exciting, nicely edited tracking shots.-' Along with Bernard Natan, Aubert also helped Diamant- Berger dabble in the genre in one of his first films after returning from two years of filmmaking in the United States.-" Education de Prince (1927) was apparently an awkward synthesis of sentimental melodrama, American-style adventure, and studio spectacular. Only the presence of Edna Purviance in the starring role kept it from falling apart completely.' By 1927, the modern studio spectacular was so popular that, when several filmmakers associated with the narrative avant-garde wanted or needed to make a film with supposedly commercial appeal, they turned to the new genre. Jean Epstein, for instance, placed his doubled story of lost love, 6'2 X 11 (1927), in the milieu of the Theatre des Champs-Elysees (singers, chorus girls, and clowns) as well as the spacious winter residences on the C6te d'Azur. Yet today, the sequences in the theater seem almost irrelevant to the film's fasci- nating rhetorical and narrative experimentation. Similarly, Jean Gr6millon situated much of the second half of Maldone (1928) in the plush hotels and casinos of the C6te d'Azur. Yet his purpose was to convey the sterile conven- tions of this noveau riche milieu in contrast to the natural simplicity of the vagabond barge life that his central character (Charles Dullin) had previously enjoyed and to which he returns. Maldone, much like 6'2 X 11, was marked by unusual rhetorical and narrative strategies; and neither film did all that well commercially. Apparently the only film in this group to be successful 214 with the public was Jean Renoir's Marquitta (1927), which premiered at the Aubert-Palace, in July, 1927.45 Taking its title from a currently popular song, Marquitta's scenario by Pierre Lestringuez conceived a Pygmalion story with suitably French and American twists. Among the reviews and reminiscences on this lost film, there is a curious lack of agreement about what exactly happened in it; but the main lines of the narrative seem clear enough.46 Marquitta (Marie-Louise Iribe) is a young street singer in the working-class district of Paris. One day she is discovered by a foreign prince, Vlasco (Jean Angelo), whom she christens Coco; they become lovers, and she scandalizes the casino society of Cannes with her vulgar behavior. Her father, however, steals an enormous sapphire from Vlasco; Marquitta is accused and thrown out-with a bit of money. She goes on to become a star in the luxurious new nightclubs performing the same tunes she once sang on street corners, while Vlasco loses his money and prop- erty in a revolution and degenerates into a Russian dancer in a Nice cabaret.47 There she finds him in a sequence with a neat comic reversal: "She invites him to dine, and the audiences cracked up when they saw the prince throw himself like a dog on a leg of chicken . . . while Marquitta, very properly, delicately, peeled a pear. "48 When Vlasco tries to commit suicide in a taxi driven by his former valet-secretary (Henri Debain), Marquitta and her 22CV Renault save him after a long chase along the cliff roads above the Mediterranean. The sapphire is recovered; the couple restored. More amicable than tongue-in- cheek or caustic in tone, Marquitta was probably just the kind of "high life daydream fairy tale," to quote Raymond Durgnat, that epitomized the genre.49 Renoir himself best remembers the opening sequence, where he had a large- scale model of the Barbes metro station reflected in a huge mirror with small unsilvered areas behind which several actors were deployed (a technique that was similar to the Shufftan process).50 The film's success, however, seems to have spurred Marie-Louise Iribe to direct her own exotic melodrama variation on the genre, the recently rediscovered Hara-Kiri (1928).1 About the same time, one of Renoir's close friends, Alberto Cavalcanti, made a very different studio spectacular, with a correspondingly opposite au- dience response. Produced by Pierre Braunberger, who was using the genre's popularity to shift from short film to feature film production, Yvette (1928) was a contemporary adaptation of a Maupassant story enacted by a half-French, half-British cast.52 The narrative is unusually slight. Yvette (Catherine Hess- ling), a young woman known for her skill as a dancer, falls in love with a shallow young man of her own class, Jean de Servigny (Walter Butler). When she realizes that he does not mean to marry her and is just as interested in her mother (Ica de Lenkeffy), Yvette tries to commit suicide. She is found in time; Jean seems to pledge his love; and her mother nonchalantly returns to her lover, Jean's friend Saval (Clifford McLaglen). In one sense, Yvette is a vehicle for Catherine Hessling, who gets to exhibit several different dance numbers (including one in a Black African mask for a West Indian Night party, which prefigures Marlene Dietrich in Blonde Venus), do a half-dozen imitations, and die in a lovely full-length white gown in a luxurious bedroom. However, she is far more subdued here than in her pre- vious films, perhaps because just as much attention seems to be given to Eric Ads's incomparably suave set constructions: a large sunken-floor room for danc- MODERN STUDIO SPECTACULARS 215 115. Catherine Hessling in Yvette ( 1928) ing and dinners, an art deco house exterior, and a spacious bar with an elevated dance floor.53 The opening sequence juxtaposes rather too obviously the old and the new: an afternoon tea for the queen of England and then a modern evening party of club sandwiches, drinks, and dancing. Thereafter, each se- quence seems to document another aspect of the modish good life, from swim- ming at an exclusive pool to trying out every new form of dancing. The film's attitude toward all this is anything but sympathetic and goes directly against the genre's expectations. Yet the overall effect is not really disturbing; in fact, it is rather distasteful and numbing. The pathos of the central character and the sometimes acute social and psychological observations are gradually sub- merged and neutralized in an impeccably American mise-en-scene and mon- tage whose unvaried rhythm finally proves wearyingly monotonous. By imi- tating the style of American studio spectaculars to a fault, Cavalcanti seems to have found a consistent means of satirizing the boredom and emptiness of this milieu; but that very consistency enervates the film itself. Audiences seem to have responded to Yvette's cool, humorless cynicism with a reciprocal cool- ness. Today, its air of elegant sterility reminds one a little of Feyder's Gribiche. Full-scale exploitation of the modern studio spectacular was underway just before the French film industry changed to sound films. After a couple of mediocre period spectacle films, Paramount committed its whole French film production to the genre. In 1928, it released Maurice Gleize's adaptation of a Dekorba novel, La Madone des sleepings (with Claude France and Olaf Fjord)." In 1929 came contemporary adaptations of two more Henry Bataille plays Luitz-Morat's La Vierge folle (starring Emmy Lynn, Jean Angelo, and Suzy Vernon) and Andre Hugon's La Alarche nuptiale (starring Louise Lagrange and Pierre Blanchar)and Roger Lion's short Une Heure au cocktail bar, which sim- 216 ply used a modern bar decor, writes Gerard Talon, to intermingle "a complete MODERN STUDIO set of stereotyped characters."55The same year, SOFAR, an Italian-German-French SPECTACULARS company, mounted two big productions by Augusto Genina: Quartier Latin from another Dekorba scenario (starring Carmen Boni, Ivan Petrovich, and Gina Manes) and Prix de beaut6 from a Rene Clair scenario (starring Louise Brooks).56 At the same time, Cineromans devoted a good portion of its pro- duction budget to Henry Roussel's Paris-Girls (with Suzy Vernon) and, espe- cially, to L'Herbier's L'Argent. Aubert invested perhaps more heavily in the genre than anyone else. In 1928, the company released Rene Hervil's Minuit . . . Place Pigalle (starring Nicolas Rimsky and Ren6e H6ribel) and H. Eti6- vant's La Sirene des tropiques, from another Dekorba scenario (Josephine Baker in debut, with Pierre Batcheff).57 It also helped produce two major films by Julien Duvivier: Le Tourbillon de Paris in 1928 (with Lil Dagover) and Au bonheur des dames in 1929. At least two films from this extensive "catalogue" survive and offer quite different transformations of Zola novels into the modern studio spectacular. Duvivier's Au bonheur des dames (1929) is much the less costly and more conventional of the two films.58 Its narrative is predicated on the inexorable movement of progress and the unequal conflict between an old-fashioned cloth- ing shop and a new department store (which Zola had based on the Bon March6, then the world's largest department store),59 conveniently located on the same street opposite one another. From the provinces, Denise (Dita Parlo) comes to Paris to work for her uncle Baudu (Armand Bour), only to find his once prosperous shop nearly abandoned. Baudu's competitor, Octave Mouret (Pierre de Guingand), hires her for his department store across the street, and they fall in love. After several secondary intrigues involving Mouret's mistress, Mme. Desforges (Germaine Rouer), Denise's cousin Jouve (Albert Bras), who runs off with a model (Ginette Maddie), and another cousin Genevieve (Nadia Sibirskaia), who dies abandoned by her fiance-after all this, Baudu goes ber- serk and charges into the department store with a gun. After shooting at Mouret, the store manager, and several women customers as well, he tries to return to his shop and is promptly run over (accidentally) by one of Mouret's delivery trucks. Mouret has to sell his business, and both he and Denise end up in the now deserted old shop. Perhaps because of the Zola source, Au bonheur des dames is much more critical of the modern milieu than are either of Duvivier's earlier films in the genre. But the overall tone is less antagonistic than pathetic and nostalgic- the tragic deaths, the desolation of the shop (increased by the demolition of buildings on either side), the disappearance not only of the family business but of the family itself. Despite itself, the film also cannot control its admiring attention to the department store (designed by Christian Jacque, later to be- come an important commercial filmmaker), the modeling and shopping that are its chief activities, and the vacation resort of L'Isle-Adam (with a bathing- suit contest) in the then fashionable Seine-et-Oise district north of Paris. Bau- du's ineffectual attack and demise become an apt metaphor for the film's own compromised critical position. Distinctly German, American, and French ele- ments combine here in a way quite characteristic of the genre. "The street scenes," writes Raymond Chirat, "nicely echo the German films of the era. "60 The attention to crowd scenes in the street and in the department store, with 217 COMMERCIAL NARRATIVE FILM 1 16. Pierre Alcover and Brigitte Helm in a Lazare Meerson interior in L Argent (1929) one serving as the context for the violent climax (as in a misguided western), imitated American films whose crowd scenes the French took as standards of veracity and authenticity. Peculiarly French, however, were the numerous and often elaborate tracking shots, employed usually in long takes rather than in short clusters of shots or in isolated moments. Denise's movement into Paris and through the streets to her uncle's boutique is marked with tracking shots, as is Baudu's assault on the department store and his withdrawal. The film thus exhibits quite well the synthetic nature of the modern studio spec- tacular. The Cin6romans-Cin6mondial production of L'Argent (1929), with its budget of nearly five million francs, was one of the biggest French films of the 1928- 1929 season." It was immediately embroiled in controversy when, at the start of the production, Andr6 Antoine reported that L'Herbier's transposition of the narrative from 1868 to 1928 would betray and disfigure Zola's novel. 6 Antoine's argument hinged on locating the real, the authentic, in the source text, which would have produced a period atmosphere film such as Therese Raquin. But L'Herbier was interested in the overwhelming power of money or capital in contemporary society and so used Zola's novel basically as a pre- 218 text.64 In one sense, however, the question of verisimilitude or authenticity was germane, for L'Argent was much more an international production than was Au bonheur des dames. Although its narrative was situated clearly in Paris, the characters were generalized types (played by a French-German-English cast) that did not really represent French society. Nicolas Saccard (Pierre Al- cover) looks like a stolid, ambitious German who has acquired an American sense of mental and physical quickness. His rival in financial speculation is a coldly calculating reclusive German, Gunderman (Alfred Abel), and his part- ner-betrayer is the equally cool blonde Baroness Sandorf (Brigitte Helm). The aviator hero, Jacques Hamelin (Henry Victor) and his threatened wife, Line Hamelin (Mary Glory), act like innocent foreigners (American or English) caught up mistakenly in Saccard's schemes. This form of character typing makes L'Argent an exemplary film in the genre of modern studio spectaculars. But it also raises the question, first expressed by Jean Lenauer, about whether the dramatic conflict between Saccard and Gunderman distracts an audience from the insidious power of money at the base of that conflict.65 There were other exemplary features as well. During the three days of Pen- tecost in 1928, L'Herbier was allowed into the Paris Bourse (the first film- maker accorded that privilege) to shoot several spectacular set pieces for the film.66 For one night, he also electrified the Place de l'Opera in order to shoot crowd scenes that would later be intercut with Hamelin's record-breaking solo flight from Paris to Cayenne, French Guyana. Lazare Meerson and Andre Bar- sacq constructed several magnificent interiors, including an enormous bank interior, several banquet and party halls, and an unusual circular room deco- rated with a wall-size world map. Apparently, L'Herbier intended L'Argent to be critical of all these sumptuous fashionable images, more critical than was Marquitta or even Yvette. However, there is some question as to which is more pronounced-the critique of the milieux or the celebration of the set decors and what they represent. The film seems much more effective in creating patterns of disruption in the genre by means of several highly original formal strategies. Perhaps no other commercial narrative film seems so conspicuously marked by camera movement-of all kinds. And that insistent movement is paralleled by unconventional editing patterns that are no less striking.67 The result is a peculiarly uncommercial, experimental work that, very much like La Passion deJeanne d'Arc, presents itself as a major text of the French narrative avant-garde. From his analysis of the French cinema in 1928, and especially the genre of the modern studio spectacular, Gerard Talon concludes that "the ambition to make French films no longer existed in France. "68 Ideologically and tech- nically, the French film industry was bent on reproducing the studio spectac- ular of the United States and Germany. By presenting on the screen, "under the guise of a universal spectacle, a reality that was no longer specifically theirs," writes Talon, the French were admitting implicitly to failure-an economic failure to challenge the American cinema's hegemony and an ideo- logical failure to reconstitute a sense of national identity.69 The argument is harsh but generally persuasive. The attitude of many French studio spectacu- lars toward the modern good life they showed may have been ambiguous, but the alternative they usually offered was nostalgia for the past. There were MODERN STUDIO SPECTACULARS 219 COMMERCIAL NARRATIVE FILM Comics and Comedies exceptions, to be sure, almost exclusively in the films of Epstein, Cavalcanti, Gr6millon, and L'Herbier; but none of these (with perhaps the exception of L'Argent) were commercially successful. The coming of sound-and French voices, French speech-tended to reverse this pattern and gradually returned the genre to a more distinctly French base, but with its own stereotypes. During the first years of the sound film period, the genre sustained its prom- inence in films such as L'Herbier's L'Enfant de l'amour (1930) and Le Mystre de la chambre jaune (1930), Pierre Colombier's Le Roi des resquilleurs (1930), Duvivier's Allo Berlin . . . ici Paris (1931), and Rene Guissart's Tu seras du- chesse (1931).70 Together with the boulevard melodrama, and often in con- junction with it, the modern studio spectacular soon came to dominate French film production. Before the Great War, when the French film industry was not turning its national literary heritage into solemn celluloid tableaux, it was usually falling all over itself, anticipating Arthur Freed's credo, "Make 'em laugh, make 'em laugh, make 'em laugh!" As most film historians agree, it was the French who almost singlehandedly created film comedy. From 1906 to 1914, according to Francis Lacassin, more than fifty series of short comic films were produced in France, most of them by Pathe-Freres and Gaumont.1 Their one-reel comiques used teams of vaudeville clowns and circus acrobats often to stage one kind of chase or another which usually ended in a triumphant, destructive furor. But soon each film had at its center a single comedian who, throughout the series, developed a singular comic type. Outside the particular skills of comedians, directors, and scenario writers, the only difference between the two rival com- panies' productions was Gaumont's preference for shooting location films-in order to keep the tricks and gags cheap. The period between 1911 and 1914 was especially productive. At Path, Andr6 Deed (the first major French film comic) had returned from Italy to resume his Boireau series about a comical idiot, with an unruly shock of hair, who resorted to gags based on mechanical effects.2 In the United States, he was dubbed Foolshead. The Variet6s actor, Prince (Charles Seigneur), was appearing in the popular Rigadin series about "a comic simpleton." Rigadin was a plump, hapless Pierrot with a lugubrious clown face and upturned nose who was hopelessly in love and at odds with everything.4 The top-liner at Pathe, however, was the Max series with Max Linder. In contrast to his pred- ecessors and rivals, who were basically clowns, Max was a trimly turned out bon vivant whose nuanced acting made him the first real star of the French cinema.5 "A proper young man, impeccably dressed," writes Georges Sadoul, "Max lived in fine apartments, was served by domestics, frequented salons, but never worked. "6 His adventures often were prompted by the women he was infatuated with or bound to, and thus his actions were psychologically (if simply) motivated and frequently involved self-mockery. But his comedy was predicated chiefly on the contrast between his dapper figure and an unusual or unexpected situation. When Linder's assets all came together in a scenario of cleverly structured gags, his one-reel comic shorts could be the equal of any short Chaplin film. In Max, victime du quinquina (1911), for instance, Max gets high on an overdose 220 of quinine he has taken for an illness.7 In rapid succession, he gets into a quarrel with the Paris police chief, an ambassador, and the war minister, each of whom presents a calling card and challenges him to a duel. With equal speed, a succession of policemen try to arrest him for drunkenness; but when Max blithely produces a card, they respectfully escort him in turn to the homes of the police chief, the ambassador, and finally to bed with the minister's wife. In each case, his puppetlike behavior is misperceived, and he is carried off like an overgrown baby or an expensive piece of luggage. In the end, the minister returns and tosses Max out a bedroom window and onto the street, where he lands on top of all three policemen. Simultaneously, they snap to attention, but, once recognized, he is tossed about like a bundle of laundry. Here, a vaudeville routine, assorted visual gags (Max is pinned to a lamppost when he puts on his topcoat while leaning against it), and a series of obses- sively repeated situations are skillfuly woven into an economical social satire. As with Chaplin, writes Sadoul, everything for Linder depends "on the sce- nario, the play within the scene, the gags and the comic timing. The film technique, which remains secondary, is as simple as possible. "8 At Gaumont, even more filmmakers were caught up in the rage of comic shorts. Leonce Perret was developing a series after a character named Leonce, which he himself played: He was a jolly fat fellow, full of good spirits, less intelligent than Max Linder and possessed of a short temper (whose aptness kept him from being imbecilic), not very forward with women (whom he could not give time to anyway), and who, after walking about a little, concluded that he was better off than most: a sympathetic type of ordinary Frenchman-more bourgeois than Max-who had a good chance of pleas- ing the masses.9 Much like the American comic, John Bunny (Vitagraph Films), Leonce was usually married, so most of the series' comedy grew out of domestic squabbles with his wife. Louis Feuillade was writing and directing two different series for Gaumont. One was Bout-de-Zan, the most popular of the comic shorts centered on the antics of a precocious child. As Bout-de-Zan, Rene Poyen was a real enfant terrible who terrorized his parents, nurses, and anyone else who ventured near. W. C. Fields would have crossed the street to avoid him. The other series was called La Vie drole and starred Marcel Levesque, a new co- median who had made a name for himself in the boulevard comedies of Tristan Bernard. 10 The most interesting Gaumont comedies, however, were those done by Jean Durand. Durand created his best comic shorts in the Onesime series, which starred Ernest Bourbon, an outrageously costumed, acrobatic Auguste. One- sime was usually a hard-pressed buffoon who got his way through hare-brained schemes or luck. But the real fascination of these films, according to Sadoul, was the accelerating rhythms of their gags and "the imperturbably logical development of an absurd situation." 1 Onesime horloger (1912), for example, presents Onesime with the problem of collecting an inheritance his uncle has stipulated cannot be distributed for twenty years. 12 He merely rewires the main clock in the Paris Central Bureau, and undercranking or pixilation takes care of the rest: studies, marriage, children, a new house, a seat at the Opera and at the Cafe de la Paix. By the end of the reel, Onesime has the inheritance COMICS AND COMEDIES 117. Max Linder 221 COMMERCIAL NARRATIVE FILM as well as more than his share of responsibilities and ways to spend it. As Bardeche and Brasillach first noted, Onisime horloger may be the germ of Rene Clair's Paris qui dort (1924).13 The war nearly killed off French film comedy. Several of the comics, such as Deed and Linder, volunteered for military service. Perret and then Linder accepted offers to make films in the United States, and Deed went back to Italy. Alone, Prince continued the Rigadin series for Pathe, but his character was becoming tiresome. While Feuillade regularly ground out Bout-de-Zans until his actor outgrew the role, Durand turned to literary adaptations and patriotic films. By the close of 1916, Gaumont's only regular comic short was Feuillade's retitled series, Pour remonter le moral de l'arriere, with Marcel Le- vesque.14 The flood of imported American comic films starring Charlie Chaplin or "Charlot," Fatty Arbuckle, Mabel Normand, and Harold Lloyd quickly filled the vacuum. Outside of Levesque and the aging Prince, there were no French comics who could even begin to challenge them-and there was pre- cious little money to do it with. Besides, the war made it difficult for the French to laugh at themselves, at least in the ways they used to. It was easier for someone else to provide the slapstick and the embarrassing situations; and the American films had a level of zest, inventiveness, and cinematic construc- tion that created a new standard in film comedy. Perhaps the French comic short would have been superseded by Charlot and company anyway, but the war eliminated any chance that the genre could have been sustained and re- vitalized. After the war, there were a few feeble attempts to resurrect the one- and two-reel comic films. At Gaumont, Durand directed a short-lived series with Marcel Levesque called Serpentin (1919-1920). This series found favor with the critics (even Louis Delluc); but, denied distribution beyond the small French market, it could not generate enough profits. 11 After Serpentin was dropped, Feuillade built a series called Belle Humeur (1920-1922) around another vau- deville actor, Georges Biscot. However, in the words of Lacassin, "replacing the calm immobility of Levesque with the tremblings [of Biscot produced [occasional] guffaws rather than [sustained laughter."16 At Path6, the Rigadin series wound down slowly until it was finally halted in the company's reor- ganization in 1920. Max Linder returned briefly from the United States (by way of a Swiss sanitarium, after convalescing from double pleurisy) and al- lowed Diamant-Berger to re-make one of his earlier one-reelers as Le Feu sacr6 (1920).17 Nothing seemed to work. After 1922, the French comic short simply disappeared except for an occasional sortie by one of the narrative avant-garde filmmakers, none of which had more than limited distribution: Gance's Au secours! (1923) with Max Linder, Clair's Entr'acte (1924), Renoir's Charleston (1926), Cavalcanti's La P'tite Lily (1927) and Le Petit Chaperon rouge (1929). The death of the comic short, however, did not mean the end of French film comedy. For two things happened during the decade after 1915 to re- position it. One was the integration of the comics into the police serial, the Arabian Nights film, and the historical reconstruction. Louis Feuillade and Marcel Levesque were instrumental in this transformation. For both Judex (1917) and La Nouvelle Mission deJudex (1918), Levesque developed a character named Cocantin who served as a comic foil and befuddled assistant to the hero. In Tih-Minh (1919) and Barrabas (1920), Feuillade gave a similar role 222 to Biscot who, in one, was even called, ironically, Placide. It was Levesque who carried the comic role into feature films in Louis Nalpas's production of La Sultane de /'amour (1919). And it was Armand Bernard who made the role a viable part of the historical reconstruction film in his reluctant performance of Planchet, d'Artagnan's valet, in Diamant-Berger's Les Trois Mousquetaires (1921-1922).18 Thereafter, the comic confidant, usually a male sidekick, be- came a prominent fixture in the French cinema of the early 1920s. Along with Levesque and Bernard, certain actors became identified with the type: Lucien Tramel, Henri Debain, Nicolas Koline, Alex Allin.19 The other strategy to transform the comic short was the attempt to produce a feature-length film comedy. Apparently, the idea originated with Diamant- Berger who, in 1919, had just returned from studying, at Pathe's insistence, production and distribution techniques in the United States.20 Caught up in the dream of restoring the French cinema to international status, Diamant- Berger got Path6's consent to film a sure-fire property with an international star. He chose Tristan Bernard's most popular boulevard comedy, Le Petit Cafe (which had played over a year at the Palais-Royal), and asked Max Linder (whom the French now considered an American) to head its cast.21 For a director, he took a chance on Raymond Bernard, the playwright's son, who had been Feyder's assistant at Gaumont and who had directed a number of short films based on his father's scenarios.22 Max Linder, writes Sadoul, was perfectly cast in the role of the cafe waiter who, on suddenly becoming a million- aire, was still bound by contract to the bistro and so passed his evenings as a pleasure- seeker and his days as a mere garqon.23 Henri Debain also made an auspicious debut as a cafe dishwasher in the kind of secondary comic role that would become his trademark.24 As for Raymond Bernard, he knew full well the spirit of his father's work, according to Ric- ciotto Canudo, and "the slight but quite funny comedy was translated intact onto the screen, with its characters and situations, its ironies and gaiety."25 Le Petit Cafe seemed to have just the right blend of physical, social, and psychological comedy. Everything went smoothly until the completed film was shown to a group of distributors-and no one wanted it. Diamaat-Berger complains, perhaps disingenuously, the exhibitors were unanimous! A Max Linder without fast chases, without slapstick, a comedy that lasts as long as a drama . . . how could one hope to spark the audience's laughter and sustain it for such an unusual length of time?26 Only four cinemas agreed to open Le Petit Cafe, in December, 1919: the Omnia-Path6, Tivoli, Colis&e, and Cine Max-Linder, which the star was then in the process of selling.27 Against all odds, the film was a smash and soon was playing everywhere in Paris, in the provinces, and across Europe. In Ger- many, it was the first French film to receive acclaim after the war. Costing just 160,000 francs, it accumulated over a million in grosses.28 Astonishingly, nothing came of this success. All of the principals went on to other things-Diamant-Berger began preparations for Les Trois Mousque- taires; Linder returned to the United States to do a parody of Fairbanks in The Three Must-Get-Theres; Bernard became engaged in a series of psychological COMICS AND COMEDIES 223 COMMERCIAL NARRATIVE FILM "chamber music" films scripted by his father-and no one else in the industry seems to have had the courage or desire to make another Petit Cafe. This is all the more puzzling since the French theater at the time was in the midst of a popular revival of boulevard comedy.29 Why did this happen? Was there a rivalry between the theater and the cinema which restricted the chances of presenting a comedy simultaneously on stage and on film? Were the theatrical comedies so verbally oriented that it was difficult to adapt them to the visual film medium? Was the industry so overwhelmed by the American film comics and so despairing of its own comic actors? Whatever the exact combination of reasons, the feature-length film comedy developed fitfully over the next several years. The most consistent formula was also the most retrograde: Robert Saidreau's series of vaudeville comedies, a dozen of which were released from 1920 to 1925. The least interesting were Feuillade's uninspired adaptations of boulevard comedies such as Bayard and Vanderbuch's Le Gamin de Paris (1923) and Labiche's La Fille bien gardie (1924), both of which starred a grown-up Bout-de-Zan.30 Much more promising were two sets of films that were closed off by untimely deaths. Pathe-Consortium seemed to have given the go-ahead to a series of comedies, starring Lucien Tramel, based on the Bouif character of de la Fouchardiere: Le Crime du Bouif (1922) and La Resurrection du Bouif (1922).1 But the director of the series, Henri Pouctal, died suddenly early in 1922. Several years later, Aubert launched Max Linder's last film, Le Roi du cirque (1925), a French-Austrian production which was shot by E. Violet at the Vita-Film studios in Vienna.32 The subject, a succession of amusing episodes loosely organized around the circus arena, foreshadowed and may have influenced Chaplin's more tightly structured film, The Circus (1927). But Linder's suicide, in late October, 1925, unfortunately closed off what promised to be a series of fine comedies and a new stage in his career.33 The most important effort of all, as it turned out, had little to do with either the boulevard comedy, the vaudeville, or the circus. Instead, it represented an original synthesis of American film comedy and earlier French fantasy films, interpolated with parodic elements from other film genres. Two writer-directors were principally involved in this strategy: Ivan Mosjoukine and Rene Clair. Ironically, it was the young Albatros company formed of Russian emigres that initiated this renewal of French film comedy. Apparently, their model of comedy construction was the story of a naive provincial fellow come to the sophisticated city, which suggests a parallel to the company members' own transposition from Russia to France and Paris. This juxtaposition was the basis for linking together all sorts of mocking episodes in different milieux. For instance, in Tourjansky's Ce Cochon de Morin (1924), adapted from a Maupas- sant story, Morin (Nicolas Rimsky), a lawyer from La Rochelle, has a series of drunken adventures in Paris that carry him from a jazzband dance hall to court. Perhaps overstating the case, Jean Pascal saw the film as "high comedy" in contrast to "the heavy-handed burlesque too often employed in American comedies."3 An even better film was Volkoffs Les Ombres qui passent (1924), from a scenario by Mosjoukine.35 Louis Barclay (Mosjoukine) and his young wife Alice (Andr6e Brabant) live simply and sanely in Happyland, the country estate of Louis's father, Professor Barclay (Henry Krauss). In order to collect an inheritance of several million francs, the young innocents travel to Paris, 224 where Louis is quickly seduced by the vamp Jacqueline (Natalie Lissenko) and becomes the victim of numerous misadventures. Les Ombres qui passent seems to have been a fantastic tale, a pastiche parodying various genres, from Grif- fith's sentimental country melodramas to the French bourgeois melodrama. Mosjoukine's most interesting contribution, however, was Le Brasier ardent (1923), which he wrote, directed, and performed in as the leading character. Although not all that successful commercially, Le Brasier ardent was unques- tionably one of the most bizarre films of the decade. Cindmagazine called the film a labyrinth whose every turn opened onto a different genre.36 Some of the comedy comes from Lochakoffs strange set designs-e.g.,, the detective agency, which the husband literally falls into, or the woman's bedroom, which oper- ates much like a Mack Sennett set. But most of it is produced by Mosjoukine's conception of the detective, who rapidly regresses from his initial heroic image through a shifting parody of disguises to a mama's boy who lives with his adoring mother and her pet bulldog. Mosjoukine was never again to be more mockingly exuberant than in this film. Within a year, stardom had tamed him into respectability. Just as Le Brasier ardent was being previewed for the critics and exhibitors, Rene Clair was engaged in a similar attempt to revive French film comedy. His first project, based on an original scenario, Paris qui dort (1924), was an unexpected commercial success, despite a certain roughness in its mise-en- scene and editing.37 Here Clair quite consciously wanted "to revert to the prewar tradition, that is to say, the tradition of the first French comic shorts . . . films [that address directly the greatest number of spectators. "38 His narrative was predicated on the inverse of Durand's Onisime horloger. An angry scientist invents an invisible-ray machine which he uses to stop time and immobilize all Paris. The only people not affected are the scientist him- self, his daughter, a planeload of five people from Marseille, and the caretaker atop the Eiffel Tower. Paris becomes simultaneously an immense waxworks museum and a rich, ripe world for the six characters to wander through-and plunder. Their wanderings are both comic and magical. A whole society of manners, character types, public and domestic situations lies exposed to their scrutiny and ridicule. They themselves, however, repeat those gestures and actions by taking whatever they desire and by cavorting and then fighting among themselves while suspended over the city on the Eiffel Tower. The exuberant, mocking spirit of playfulness in Paris qui dort, along with that in Le Brasier ardent and Clair's second film, Entr'acte (1924), carry all three films well beyond the conventions of the commercial cinema. Although Clair's next few films do not really renounce this comic renewal, they tend more and more to experiments in fantasy. Le Voyage imaginaire (1926) probably most exhibits his continued interest in comedy. Much of the comedy here evokes Le Brasier ardent and Clair's first films. It opens, for instance, with the hero being chased by a bulldog who in turn is being chased by its master; and its long fantasy section ends with the hero transformed into a toy bulldog before the eyes of the disappointed heroine. Early on, the film also includes some effective satire of French bureaucracy-especially in the way a bouquet of flowers passes round and round a small town bank office, in order to define all of the characters and their relations with one another. However, Le Voyage imaginaire was no more a commercial success than was Le Brasier ardent.39 The COMICS AND COMEDIES 225 COMMERCIAL NARRATIVE FILM Mosjoukine-Clair renewal seemed little more than a minuscule eddy in the stream of profitable period films and emerging modern studio spectaculars. After Mosjoukine moved on to lavish spectacle films, Albatros turned to Nicolas Rimsky to sustain its film comedy projects. Rimsky was not a major comic actor; in fact, he looked and acted like a milquetoast version of Marcel Levesque. His films, done in collaboration with Pierre Colombier and Roger Lion, depended heavily on comic situations and on wholesale borrowings from other films, both French and American. One of the first of these films, Co- lombier's Paris en cinq jours (1926), revives the story of provincial characters in the city, but in a new format.40 Here Rimsky is teamed with Dolly Davis as a New York couple (he loves The Three Musketeers; she is a swimmer) who win $10,000 in a contest, get engaged, and set sail for Paris. This sets the stage for five days of comic adventures. In later films, such as Le Chasseur de Chez Maxim's (1927)--which originally was to have starred Max Linder- Cinia-Cine-pour-tous argued that Rimsky was able to transpose completely "American methods-gags, touches of imaginative detail, comic gestures drawn from Chaplin and Lloyd-into an atmosphere of French gaiety."41 Jim la Houlette, roi des voleurs (1926), the only one of these films that I have seen, illustrates how well, and not so well, this actually worked.42 Rimsky plays Jacques Morton, the lowly secretary to a popular novelist, Brettoneau, whose daughter Pauline (Gaby Morlay) he loves madly. Much like a latter-day Orlando, he writes anonymous poems to her and attaches them to lamp shades around the Brettoneau's chic ultramodern villa. Brettoneau's agent, an Amer- ican-style advertising maniac, conceives a plan to reverse the slumping sales of his books-throw a party and have Jim la Houlette, a Fant6mas-like char- acter who has terrorized Europe for ten years, steal his latest manuscript. Morton is bamboozled into disguising himself as Jim (Pauline says she could love such a man), but the real thief appears and leaves him to take the rap. In the court tria (a kind of parody of Crainquebille), Brettoneau and his agent testify against their own duped accomplice-the publicity campaign has worked; the new novel is a best-seller. But when Pauline confesses that she loves him, Morton proudly accepts his criminal name. At the last moment, his lawyer (who turns out to be the real Jim) allows him to escape from prison. On a train bound for the lawyer's country estate, Morton disguises himself as a priest and performs a few routines imitating Chaplin in The Pilgrim. The lawyer invites Brettoneau to his estate, and Morton pretends to haunt him (once more disguised, as a ten-foot-tall figure perched invisibly on a servant's shoulders). Pauline shows up unexpectedly, just as her father (threatened with disclosure) has made out a blank check for 500,000 francs. Suddenly, the police arrive. The lawyer and his wife go off in one underground tunnel; Morton and Pauline go off (with the check) in another. There the film ends, as improbably as a serial, as she sums up her love: "The things I loved in Jim, I love in you-bravery and daring; but besides you are kind and honest." So it goes, indeed. After the success of Max Linder's Le Roi du cirque (1925), Aubert decided to revive the Bouif character in a series of films starring Lucien Tramel. Ac- cording to Cinda-Cine-pour-tous, the most important of these was the six-part serial, Le Bouif errant (1926): "Sly and naturally optimistic, without overzeal- ous outbursts, without theatrical effects, Tramel provokes laughter through 226 the aptness of his gestures and expressions to the situation of the scenario. " His production team was also quite supportive: an original scenario from de la Fouchardiere, competent direction by Rene Hervil, and solid comic acting by Albert Prejean and Jim Gerald. Aubert's attempt to exploit Tramel in a double role in Duvivier's Le Mystere de la Tour Eiffel (1927), however, frittered away a chance to sustain momentum.44 Clearly, one problem in most of these films was an enervating tendency to exploit sumptuous modern decors, as if comic appeal were not enough to generate public interest. That tendency was so marked in Aubert's Les Transatlantiques (1928), directed by Diamant-Berger and Colombier, that a satire of Yankee tourists became little more than a puff of charming pleasantries in the chic milieux of Paris and Deauville.45 Much like the modern studio spectacular, the French film comedy seemed on the verge of losing even the remotest connection with its own national identity, its own social structures.46 In the last two years of the decade, the genre suddenly flowered in a quick succession of films, most of them produced by Albatros. Unlike the previous films, they were deft adaptations of classic or current boulevard comedies. These films were grounded solidly in French society, and they consistently subordinated fantasy and psychological responses to it, integrating them into the representation of a specific milieu. They also announced a new generation of comic actors: Albert Prjean, Pierre Batcheff, Michel Simon. The first two films to begin the cycle were directed by the filmmaking brothers, Henri Chomette and Rene Clair. Le Chauffeur de Mademoiselle (1928), which opened at the Omnia-Palace in April, 1928, seems to have been a carefully calculated project written for Dolly Davis and Albert Prjean.47 Chomette himself freely admitted this, but distinguished this, his first feature film, from the then current "league of nations" productions: In the Chauffeur de Mademoiselle, a film that I made in full knowledge of the commer- cial concessions I decided were acceptable, I have tried to achieve a homogeneity by using French elements only.48 Among others, Jean Dreville remarked on the comedy's singularity: Ah yes, a charming film, full of good taste and balance, and according to the French formula. For once, the formula seems to me excellent. Le Chauffeur de Mademoiselle is indeed a French film; the publicity is not lying.49 Although the film seems to have been modestly profitable, most film historians have accepted the critic's disappointment at Chomette's failure to carry over the array of techniques manifested in his few short experimental films, Jeux des reflets et de la vitesse (1925) and Cinq Minutes du cinema pur (1926). Ren Jeanne and Charles Ford, for instance, conclude that, in Le Chauffeur de Ma- demoiselle, "Chomette was no more than a good craftsman."50 Drville, how- ever, valued the film precisely for this restraint and notes several images which suggest that the film was more than hack work: "It is definitely interesting to study; the design produces some striking dissolves between disparate images, a sky that progressively 'eats' half a tree in the background, etc."51 The dcou- page of one sequence published in Cinegraphie also reveals a rather effective use of multiple exposures or split-screen framing for comic effect: a couple's tele- COMICS AND COMEDIES 227 118. (left) Albert Prejean, Olga Tschekowa, and Vital Geymond in Un Chapeau de Paille d'Italie (1928) 119. (right) Jim Gerald and Albert Prdjean phone conversation ("Don't come yet. I haven't dared break the news of our marriage to my aunt") is gradually overheard by a half-dozen people listening in on the party line.2 The moment has a Lubitsch-like touch that combines the technical facility of Gance's Napoleon and a typical gag from the prewar French comic shorts. Unfortunately, no print of the film seems to have sur- vived so that we could resolve these critical contradictions. Ren6 Clair's Un Chapeau de paille dItalie (1928) presents no such problem. It was Alexandre Kamenka who bought the rights of this clever Labiche play from Marcel L'Herbier and then asked Clair to adapt and direct it for Alba- tros.' The film pleased the critics more than it did French audiences (an exclusive run at the Omnia-Palace, early in 1928, lasted only three weeks), but it was just successful enough-and without a single major star.' Clair's adaptation was brilliant in at least two crucial ways. The play's early Second Empire setting (1851) was transposed to that of the Belle Epoque bourgeoisie (1895), specifically, to the Charenton suburb of Paris bordering on the Bois de Vincennes. Keeping the Second Empire setting, reasoned Clair, would have created a period film of nostalgic beauty; to produce properly ridiculous cos- tumes and decors for a 1928 audience, the fin-de-sicle period was perfect. Sepia film stock provided an appropriate period color-tone (the chief cameraman was Maurice Desfassiaux); and Lazare Meerson's interiors, with their ornamental Henri II funishings, caricatured early Pathf film sets.5" Thus the society which enjoyed Labiche's play is itself satirized in the film. Furthermore, Clair per- ceived that Labiche's verbal comedy was based on a comedy of situations, a burlesque of movement. His task was to drop the verbal comedy and graft 228 onto this "burlesque intrigue . . . , transposed cinematically and treated like a ballet, . . . a series of striking visual observations."56 The result, although perhaps indebted to Lubitsch and Chaplin, was a truly French film comedy- without a single comic caption. At its summer preview, Edmond Epardaud dubbed it a "comedy of observation."17 The narrative of Un Chapeau de paille d'ltalie intermeshes two simple plots in a delightful web of complications. On the afternoon of his marriage to Helene Nonancourt (Marise Maia), Fadinard (Albert Pr6jean) crosses the Bois de Vincennes to inspect his new apartment and nuptial chamber. On the way, his horse nonchalantly munches on a lady's hat in a bush; from behind the bush quickly come Anais de Beauperthuis (Olga Tschekowa) and her lover, Lieutenant Tavernier (Vital Geymond). In a rage, Tavernier orders Fadinard to find another Italian straw hat exactly like the half-eaten one or he will destroy his apartment. For the lady is married and must return home "intact." Throughout the day, as he anxiously goes through the series of social cere- monies, Fadinard and his servant Felix (Alex Allin) pursue the elusive dupli- cate hat. Tipped off by a saleswoman, he finally goes to the home of Beau- perthuis (Jim Gerald) and explains his predicament, only to discover the man is Anais's husband. When the hat eventually shows up, aptly enough, it has been there all along-the first of the bride's wedding presents. After a final ballet of circulating objects, running figures, and desperate glances, Anais returns safely to her own bedroom and husband, while Fadinard sends Taver- nier off and goes to Helene in the nuptial chamber. The marvelous play of visual observation which this narrative accrues makes for an unrelenting attack on the society of the Belle Epoque bourgeoisie. Clair's method is clearly an extension of the bureaucratic satire in Le Voyage imaginaire. The characters are all recognizable types with different objects to define them, and they behave like marionettes going through the motions of social interchange. Tavernier is all strutting uniform, with a ridiculously ob- solete sense of decorum and gallantry. Nonancourt, the bride's squat father (Yvonneck), is stuffed into a new shirt front and boots that pinch. Tall, thin cousin Bobin (Pr6 fils) is at loose ends over a lost glove. Uncle Vesinet (Paul Olivier) has a blocked earhorn and cannot hear or see anything that is going on. Felix, on the other hand, sees entirely too much. At one point, every time he peeks into Fadinard's sitting room, the lovely but powerless Anais is in a different man's arms. When the huge blustering Beauperthuis is finally intro- duced, his feet are soaking in a pan of hot water, which at last allows Non- ancourt to pilfer some comfortable footwear. Each social ceremony is undermined by a parody of conventions. The may- or's speech to the wedding party, for instance, is interrupted by a loose tie- everyone straightens his own, including the mayor, until finally an uncle dis- covers that it was his tie that initiated the ricocheting comedy.58 Then the mayor is cut off in the middle when the increasingly nervous Fadinard inter- prets his lifted arms as a sign of conclusion and leads the whole family off past the astonished officials. Later, while a drunken Nonancourt delivers his homily on marriage to H6lene and the family in what he thinks to be the new apart- ment, Fadinard is in the next room unwittingly revealing to Beauperthuis Anais's affair with Tavernier. Even the objects gain an importance of their own as possessions, an importance that is ridiculed in several wedding gift COMICS AND COMEDIES 229 / 120. (left) The final recovery of the hat in Un Chapeau de paille dItalie (1928) 121. (right) Pierre Batcheff as the defense lawyer fantasizing in Les Deux Timides (1929) exchanges. Momentarily mystified, Tavernier accepts the present of a clock, which turns out to be a replica of one in the bedroom; just as the prim couple who offered it prepare to drive off in their carriage, the clock sails through the second-floor window and smashes on the sidewalk beside them. When the exasperated Nonancourt tries to annul the marriage at the end, the relatives immediately attack Felix and grab their gifts back-inadvertently exposing the hat, the source and resolution of all the troubles. Un Chapeau de paille d'Italie neatly sets in opposition the reality and illusion, the emptiness and hypocrisy, of the French bourgeois marriage. Although initially juxtaposed to the newlyweds, the adulterous couple soon literally replaces them. It is they who occupy the new apartment and who finally hole up in the nuptial chamber (although nothing untoward happens). While the relatives come bearing gifts to adorn the apartment, Tavernier methodically destroys some of the things that are already there. In a brief sequence (using wild camera movement, slow motion, and fast motion), Fadinard even imag- ines a whole set of furniture disappearing. But Tavernier and Anais only pre- pare for the pattern of doubling that succinctly closes the film. While Taver- nier runs off to rescue the hat, first from the police and then from an inconveniently placed lamppost (its last resting place), Helene hides Anais behind her full wedding dress and Fadinard blocks Beauperthuis's vision with a raised umbrella. In their attempt to preserve a marriage, the newlyweds provide an innocent front for deception. The implication is clear-it is only a matter of time before that deception is embodied in them. The final images 230 confirm it. While Anhis smiles in her sleep, Beauperthuis sits bewildered in his bedroom chair, gazing at her hat (actually Helene's). Because the hat seems intact, is she really in his possession? This illusory emblem of "the proper marriage" (fitting woman to man) is then matched by its corollary-the floral wedding wreath placed on the bedroom mantel and preserved by a glass cover. How soon will Helene find her Tavernier? After Un Chapeau de paille d'ltalie, Clair had wanted to direct a realist film based on a recent criminal case; but early in 1928, Chiappe, the new prefet de police in Paris, began to concern himself with cinema exhibition (e.g., Les Amis de Spartacus), and Clair abandoned the idea.59 Since Albatros had a contract with Maurice de Feraudy, he decided to use him in another adaptation of a Labiche play. Les Deux Timides (1929) was less successful than its prede- cessor, perhaps because its release coincided with the first sound films late in 1928 and early in 1929.60 But the film itself was rather different, and critics have been divided on its merits ever since.6' La Revue du cinma, for instance, found it deftly charming in its caricature, but not really ferocious enough.62 The two timid characters are Frmission (Pierre Batcheff), a young pro- vincial lawyer, and an old landowner, Thibaudier (Maurice de Fraudy), whose daughter Cecile (Vera Flory) is the object of Fremission's inept courting.63 His rival, Garadoux (Jim Gerald), is a former wife-beater (which no one knows, since he has resettled) whom Fr6mission had defended in his first court case several years before. After a multiplication of misunderstandings that culmi- nate in a near siege of the Thibaudier estate, plus another court trial, Fr6- mission and Cecile are married. Several things distinguish Les Deux Timides from Clair's earlier films. This is the first film in which Clair shot exteriors outside Paris, and several se- quences-for example, the young lovers meeting in an open pasture-have a charm and freshness usually associated with the provincial realist fi1m.64 The interiors, again by Meerson, were designed to complement the authenticity of the location shooting and emphasized exact details-such as the old sewing machine in Garadoux's first apartment, the mixture of comfortable and uncom- fortable chairs in Thibaudier's drawing room. A comedy of situations struc- tures this film as well, but the comedy is less socially determined and more interior, that is, focused on the psychology of character types-harmonizing with the more realistic decors.65 In the opening trial sequence, for instance, a mouse interrupts the proceedings much as did the loose tie at the wedding ceremony in Un Chapeau; but here the effect is to so unhinge Frmission that he becomes a blathering fool who condemns his own client, Garadoux. When the young suitor calls on Thibaudier to broach the subject of marriage, the two men are so afraid to speak that three hours later, when Cecile enters the room to turn on a light, she startles them both out of a sound sleep. Batcheffs stylized but finely nuanced performance as Frmission (his shy courting and frantic avoidance of danger are clearly modeled on Chaplin) is just right for this psychological comedy, but the other principal characters either shade off toward naturalism (Thibaudier) or broad caricature (Garadoux and his family who come to visit Thibaudier).66 In fact, as Jean Mitry argues, the caricature and physical comedy-which are strongest in the final sequences, especially in the siege and second trial-do tend to throw the film slightly out of balance.67 Finally, Les Deux Timides contains several inventive gags that ridicule the COMICS AND COMEDIES 231 122. The final triptych sequence in Les Deux Timides (1929) mental state of its characters, using optical effects and the split-screen concept Henri Chomette had recently exploited in Le Chauffeur de Mademoiselle. The film opens with shots of Garadoux cruelly beating his wife, which are suddenly revealed to be the representation of the prosecutor's speech at the beginning of the trial (this seems to cleverly incorporate Clair's original plan for a realist crime film). In contrast, Fremission's defense envisions a gentle, generous Garadoux and culminates in five overwhelmingly sentimental images within one shot. After the mouse wreaks havoc in the courtroom, however, Fremis- sion becomes flustered. And his lapses are conveyed in shots of Garadoux and his wife that slow and freeze, then speed up and freeze again, then reverse; finally, when the lawyer's mind fails, Harry Alan Potamkin writes, "the do- mestic scene explodes [in a spiral] from the screen leaving it blank. "68 In another sequence, after they have recognized one another outside Thibaudier's house, the two rivals go through a mock battle in separate rooms-each vic- torious within his half of a split screen. The final shot is a delightful parody of Gance's triple-screen in Napoleon. Fremission and Cecile are in bed in the center panel, flanked by Garadoux and Thibaudier, separately in theirs. The two lovers embrace and look shyly off to the left and right. Thibaudier duti- fully turns off his light; they embrace again and turn off their overhead light; Garadoux looks startled and angrily throws down his lamp. Simultaneously, the center light pops on, and a much vexed Fremission bolts upright and pulls down a shade in front of the camera. The last silent film comedy produced by Albatros was Jacques Feyder's Les Nouveaux Messieurs (1929), adapted from a current play by Robert de Flers and Francis de Coisset.69 Both the play and the Charles Spaak/Feyder scenario were 232 COMICS AND COMEDIES n 123. Gaby Morlay and Albert Pr6jean in Les Nouveaux Messieurs (1929) overtly political, which made the film rather unique in the French cinema of the 1920s. Jean Gaillac (Albert Pr jean), chief electrician at the Paris Opera and an important labor union official, is interested in Suzanne Verrier (Gaby Morlay), a prima ballerina at the Opera and mistress of Comte de Montoire- Grandpre (Henry Roussel). Jean is elected to run for the Chamber of Deputies and, in a left landslide, defeats his opponent, Montoire-Grandpre. Suzanne becomes his mistress, and soon he has accepted all the trappings of authority and power. When the government is overturned in a crisis, Jean accepts a diplomatic post in the colonies (arranged by his rival), and Suzanne reluctantly returns to Montoire-Grandpr6. The French government responded to this sub- ject by refusing Albatros a license to distribute the film. Their objection was not to the narrative, which seemed to satirize the labor union movement (a topical subject after the left lost the elections of 1928), but to the disrespectful depiction of the Chamber of Deputies and its members." After months of haggling and several brief cuts, the film was exhibited at the Paramount- Palace, in April, 1929, with its basic satirical structure intact.1 Perhaps be- cause of its notoriety, Feyder's film was quite successful. The readers of Cinea- Cine-pour-tous even voted it second to The Jazz Singer as the best film of 1929.7 Les Nouveaux Messieurs can be described as a sentimental romance or melo- drama in a comic mode that synthesizes many previous film comedies. On the one hand, it continues the pattern of most Albatros and Aubert comedies of 1926-1927. There are several large decors (again by Meerson) straight out of the modern studio spectacular: a detailed set reconstruction of the Chamber of Deputies and a sumptuous modern townhouse for the Comte de Montoire- Grandpr6. Moreover, the heroine is a star and has the role of a dancer at the 233 COMMERCIAL NARRATIVE FILM Opera, where the film's long opening sequence of a rehearsal takes place. However, the light comic observations are closely modeled on Clair. The count is first defined by his lapel pin, his monocle, his coat and cane, and the overall image reflected in his chauffeured car. Jean becomes what he opposes by using the same poster design as his opponent and by blithely donning a top hat and tails. And, like Clair, Feyder and his cameramen, Georges Prinal and Maurice Desfassiaux, also employ technical devices for comic effect. The dedication of a workers' city ends in a Sennett-like fast-motion parade and speech so Jean can return to the crisis-wracked capital.73 In the Chamber itself, the Minister of Art falls asleep and dreams his colleagues have been transformed into bal- lerinas in tutti. On the crucial vote of confidence for the government, a soft focus area keeps moving back and forth over the chamber, signaling the ca- pricious shifts in voting. The comedy, as one can see, is not quite the equal of Clair's, and the film does drag in some sequences.74 Perhaps it is most effective in the way the overall mode shifts from romantic comedy to ironic, or even tragic, satire.75 The ultimate subject of the satire is not the blindness and corruption of the working class or labor union movement but rather the social and political system that remains unchanged and easily co-opts them. Jean and Suzanne enjoy a brief idyll on the morning of the leftist victory-driving along the Seine (with the Billancourt factories behind them and the Eiffel Tower off to the side) and impulsively going for a swim-but the rest of the film reveals the moment to be an illusion. Perhaps the real victim of the film and its central character is Suzanne. She only accepts Montoire-Grandpre's favors be- cause it is the "right thing" to do. Although she genuinely likes Jean, he pays little attention to her once he is launched on a career. It is she who suffers on his politicking trips: struggling with the baggage while he is greeted cere- moniously at the station, sleeping away the afternoon in a cheap hotel room, finding dirt in her luncheon coffee cup. In order to enjoy even a bit of the good life, Suzanne seems to have no other choice than to go back to Montoire- Grandpr6. For her, too, "the new masters" are no different from the others. That Moussinac and his Socialist-Communist friends defended Feyder's film against the government's censorship made perfect sense. There was one more French film comedy of a very different sort-Jean Renoir's most underrated silent film, Tire au flanc (1928).76 Although an in- dependent production, financed by Pierre Braunberger, Tire au flanc was defi- nitely a commercial project, predicated on the appeal of Mouezy-Eon and Sylvanie's vaudeville comedy about army barracks life, which had played at the same Paris theater for twenty years.77 According to Alexander Sesonske, Renoir added a new character to the play, the big-hearted, bumbling servant set off against a blithely assured, self-centered master.78 As in Nana, the con- trast between the two central characters was established by divergent acting styles: the smooth grace of dancer Georges Pomies versus the awkward, gro- tesque fumblings of Michel Simon. And here, in a comedy, it worked perhaps even more effectively. Jean Dubois d'Ombelles (Pomies), a naive, pacifist poet, is about to be conscripted into the army, along with his servant Joseph (Simon), who, alleg- edly, is going to protect him. A dinner party given by Jean's aunt, Mme. Blandin (Maryanne)---to pull a few strings for him-goes awry; and the guests, 234 Colonel Brochard (Felix Oudart) and Lieutenant Daumel (Jean Storm), leave convinced that he is an idiot. Once at Casserone, Joseph makes himself at home while Jean quickly becomes the "Sad Sack" butt of the barracks pranks led by Muflot (Zellas). Both men's fiances visit Casserone, with predictable results. Mme. Blandin's maid, Georgette (Fridette Fatton), becomes the whole barrack's sweetheart and causes a near-riot that sends Jean and Muflot to prison. Her daughter, Solange (Jeanne Helbling), to Jean's chagrin, falls in love with Daumel. But he soon discovers Solange's sister Lily (Kitty Dorlay), who has loved him all along. Promising to reform, Jean is released from prison and performs with Joseph at the colonel's annual barracks party. When a fire breaks out, he proves himself a hero at last, by capturing the arsonist, Muflot. In the end, there is a triple wedding celebration--Mme. Blandin and Colonel Bro- chard toast Jean and Lily as well as Daumel and Solange in the dining room, while several enlisted men revel with Joseph and Georgette in the kitchen. Tire au flanc was shot rapidly in the early summer of 1928, just before the opening of the Carcassonne festival that would provide the locations for Le Tournoi (1929).19 Even more than in La Fille de /'eau (1925), this was a co- operative enterprise that established the model of community filmmaking for much of Renoir's work in the 1930s. Renoir stuck to simple, realistic sets designed by Eric Ass, and he and his cast and crew apparently improvised a great deal. Following the tendency of several French films made in the wake of Gance's Napolion, some of this improvisation took the form of extensive and elaborate camera movements. Their peculiarly rough, erratic nature may be explained by a curious story Renoir once told Kevin Brownlow: I had an electrician [Louis Nee). He used to tell me what to do. He had some very good ideas. So I gave him a handheld camera and I let him stand by the main camera Jean Bachelet) and shoot the scene however he liked. And I used a lot of his material. He became a famous cameraman.80 Complementing this camera movement (which included all manner of tracks, dollies, and pans) is a pattern of foreground/background juxtapositions (of characters and parts of the decor), both of which recur prominently in Renoir's 1930s films-e.g., Boudu sauvd des eaux (1932).81 The episodic narrative, the carefree acting, the "wild" camerawork, the seemingly slapdash editing-all gave Tire au flanc an unruly look and an unbridled, off-balance rhythm that set it apart from Clair and Feyder's smoothly crafted films.82 At the time of its premiere at the Electric Cinema, in December, 1928, the vaudeville com- edy source did indeed guarantee the film's success; and later its playful exu- berance much impressed the French New Wave filmmakers.83 There are several levels of comedy operating in Tire au flanc. Physical com- edy dominates the satire on barracks life, in short mock-documentary sketches that made Francois Truffaut compare Renoir's film to Jean Vigo's satire on boarding school life in Zero de conduite (1933).84 Arriving at Casserone, for instance, Jean walks from his chauffeured car with a briefcase and a respect- fully raised hat only to be immediately pummeled by his fellow soldiers-not once, but twice. At bayonet practice, he gently pokes the enemy dummy until, prodded suddenly from behind, he hysterically attacks everything and everybody in sight. Then, on maneuvers one day, the recruits don gas masks in response to an imaginary attack. While their leader struts off confidently COMICS AND COMEDIES 235 124. (left) The "gas attack" in Tire au fanc (1928) 125. (right) The opening shot of Michel Simon and Fridette Fatton in one direction, they stagger off in the other, stumbling about the woods like blind men using their rifles as canes. Finally, they all roll down a long hill and, like frightening (and frightened) prehistoric monsters, scatter a class of schoolchildren on a nature study outing. All this is a bit daring because Renoir dressed his soldiers, not in the usual silly Belle Epoque uniforms, but in World War I blue.85 The sexual comedy involving both masters and servants, which is even more important, depends on ironies of situation and juxtaposition. Perhaps its high point comes during Jean's incarceration, in several variations on a "midsum- mer night's dream." Outside the barracks one night, Joseph and Colonel Bro- chard circle in and out of a clump of trees-in a clockwork pattern, each ignorant of the other-awaiting their lovers. When Georgette and Mme. Fle- chois appear, they mesh with the wrong men. While the Colonel presents a poem to the startled maid, Joseph realizes the error, although Mme. Flechois seems not to mind. Georgette simply pockets the poem, finds Joseph, and pulls him off. Another master has become a poet, ergo, an idiot. The night before, however, had seen a more bittersweet encounter. Daumel and Solange are strolling through a "forest" of drying men's shirts just outside the barracks prison. From a bush beside Jean's cell window, he plucks a rose and they kiss. Noticing the hand mysteriously appear in the moonlight, Jean is drawn to the high window, where he watches in despair, his body spread-eagled against the wall (in Andre Bazin's phrase) "like a great nailed bird. "86 As Sesonske points out, this supposedly Stroheim-like moment is quickly reversed four short sequences later.8I Smuggled into the barracks, Lily tosses pebbles at Jean's prison window, which he ignores (he is reading a book on how to be audacious!). Finally, she stands on a cart to get his attention, and Jean again pulls himself up to the bars, but joyfully, to a very different apparition. When the guard comes to release him at that moment, Jean reaches for a rose; and, while Lily watches, he dances merrily out of the cell, presenting the rose to 236 COMICS AND COMEDIES 126. Michel Simon as an A ha angel in Tire au flanc (1928) his captor. Jean's character has finally undergone the change necessary for both his survival and happiness. For all its roughness and exuberance, Tire au flanc exhibits a remarkable symmetry in its structure." The end circles back to the beginning in a series of clever reversals on master-servant relations. In the initial dinner sequence, the characters of master and servant are neatly juxtaposed. The first shot opens on a Daumier print depicting a fat gentleman kneeling eagerly before a young buxom woman whose hand is daintily touching his balding head. The camera dollies out and over the dinner table as Joseph and Georgette arrange the linen (a toss of the cloth momentarily whitens the frame completely) and pause to stretch over it and kiss. After the camera dollies back into the Daumier print, the second shot repeats the movement of the first, only now Jean is standing at the head of the table, practicing his speech for Solange, who seems less than interested-the image will be echoed and transformed in the cell window encounter between Lily and Jean. A third shot extends the contrast by having Joseph and Georgette reenter the dining room and kiss, shocking Jean, who orders them out. The formal parallels set master against servant, movement against immobility, overt affection against repression, action against words.8" The arranged dinner quickly deteriorates into chaos when Joseph drops the meat off the platter, drips sauce on the Colonel's precious uniform, and then casually tosses a glass of Benzine (brought out to clean the uniform, but which the Colonel mistakenly drinks instead) into the fireplace. The resulting roar of flames drives everyone into the hall, where the guests try respectfully to take their leave. At the barracks party, master and servant play ironically reversed roles. Jean now is an exuberant satyr, scampering about in a short fur piece and a crown of laurel, piping on a flute. Joseph, however, is a large, gangling, bemused angel, dressed in a full white dress with flimsy paper wings. He is dropped, wriggling, from the flies just as Nana was in Renoir's earlier film. When this 237 COMMERCIAL NARRATIVE FILM ballet is disrupted by a fire set deliberately backstage, it is Jean who manfully douses it with a water hose (some satyr) and who now pummels the arsonist, his prankster nemesis, Muflot. Unlike Nana, Jean both proves and discovers himself in this performance. An epilogue of just a half-dozen shots carries the reversal one step further. Joseph leaves his wedding banquet in one room to serve at the banquet of his superiors in the next. Yet both tables are decorated with the same castle cake. And the final tracking shot comes to rest on Jean and Lily kissing below the table (he has dropped something again)-echoing Joseph and Georgette in the beginning of the film. Tire au flanc is more than a tuneup for such films as Boudu sauvi des eaux and La Regle du jeu (1939); it is a first-rate social satire on a favorite Renoir theme-the comic interaction of masters and servants. By the end of the decade, the French film industry evidenced little concern for producing French films, except in the genre of film comedy. The period spectacle film generally was reconstructing historical eras elsewhere; the mod- ern studio spectacular was producing a luxurious "no man's land" milieu; while the realist film and the fantasy film had been abandoned to the narrative avant-garde. Only the comedy presented the French as they saw themselves- through mockery. With the development of the sound film, the genre gained added prominence. Some films-such as Jean Choux's Jean de la lune (1932), starring Michel Simon and Rene Lefevre; Marc Allegret's Mam'zelle Nitouche (1931), with Raimu; and Renoir's On purge Bibi (1931), with Simon and Fer- nandel-merely imitated the conventions established by the Aubert and Al- batros films of the middle 1920s. Others, however-such as Rene Clair's Le Million (1931), with Lefevre, and A nous la liberti6(1931); Renoir's Boudu sauvi des eaux (1932), with Simon; and Jean Vigo's Zero de conduite (1933)-all extended the biting and caustic mockery of the late 1920s comedies. It was through these films, like it or not, that the film industry began to regain its Frenchness. 238 111The Alternate Cinema Network .I.  r We are witnessing the birth of an extraordinary art. The only truly modern art perhaps, assured already of its place and one day of astonishing glory, because it is simultaneously . . . the child of technology and of human ideals. -Louis Delluc (1918) People are only barely beginning to realize that an unforeseen art has come into being. One that is absolutely new. We must understand what this means. -Jean Epstein (1921) The dominant faction of the French film industry, from 1915 to 1929, defined the cinema as a spectacle entertainment-a commercial product in the system of economic exchange. Certain kinds of films or genres, particular stars, certain methods of production and exhibition were developed to gain profits within that system of exchange. Another smaller, but quite articulate, seg- ment of the industry conceived of cinema differently. For it, the cinema was a cultural product, an art-or something that could become an art-and the individuals and organizations who believed so considered themselves engaged in the avant-garde of its creation. Some thought of the avant-garde chiefly within the context of a narrative cinema; others thought of it as a "pure cinema"; still others thought of it in terms of documentary. Yet whatever the mode or theoretical base, their efforts depended on an alternate means of exchange. As Jean Tedesco succinctly put it: The actual exhibition market of the film industry . . . is almost completely closed to one category of films. We have called them avant-garde films only for the purpose of better distinguishing them from the current production and not because of any preconceived idea of a chapel or school.. . . We must extend the marketplace of the intellectual cinema in Paris for such films.1 That marketplace depended on a specific material base-an interrelated net- work of film critics, cinema journals, cine-clubs, and specialized cinemas, all somewhat independent of the major companies in the industry. It was this alternate, essentially cooperative system of cultural exchange, often diametri- cally opposed to that of the dominant industry, which provided much of the impetus for the alternate cinema that was the French avant-garde. The first area of struggle for an alternate cinema was located in the press-- in the newspapers, literary magazines, and specialized film journals. The key figure in conceiving and promoting such a cinema was Louis Delluc. The major journals that provided professional information and publicity about the French film industry during this period were launched either before or during the war, at the moment when the industry was being forced to adapt to the interests of the American cinema. They included Cin6-Journal and Filma (both of which published the earliest annual summaries of cinema activity in France), Le Courrier cinimatographique, Cinopse (whose editor, G.-M. Coissac, had founded the first popular industry organ, Le Fascinateur, in 1903), and, finally, the most important of all, La Cinimatographie franfaise.1 Among this group was a small but influential weekly journal published by Henri Diamant-Berger, beginning in 1916, Le Film.2 In its pages the concept of an alternate cinema seems to have had its beginnings. The significance of Le Film lies in its nurturing of an autonomous film criticism, a criticism that assumed film could be a form of art and that began the attempt to isolate its specific features in order to analyze and evaluate individual works. Most film historians attribute this new practice initially to Louis Delluc, who had been a poet, novelist, dramatist, and drama critic for Comoedia Illustri before becoming editor-in-chief of Le Film in June, 1917.3 But Delluc had an important precursor in another young French writer, Co- lette. Introduction The Beginnings of a Film Criticism 241 ALTERNATE CINEMA NETWORK It was Colette who started the "Critique des films" column for Le Film, on 28 May 1917. Diamant-Berger had asked her to write for his journal, not only as a friend but because she had a keen interest in the cinema and had connections in the industry.4 Colette was a close associate of Musidora, the famous villainess of Les Vampires (1915), and the two women had just collab- orated on a film shot in Rome early in 1917, La Vagabonde. Besides, she had already written several film reviews for the Paris daily newspapers. One of these, in Excelsior (7 August 1916), was crucial, according to Delluc, in pro- claiming the "artistic merits" of Cecil B. De Mille's Forfaiture [The Cheat (1915), starring Sessue Hayakawa.5 Within eight days of its opening, Forfai- ture was elevated to exclusivitl status at the Omnia-Pathe cinema.6 That one film persuaded Delluc and Marcel L'Herbier, respectively, to become a film critic and a filmmaker; and it heavily influenced Abel Gance's first artistic success, Mater Dolorosa (1917). In Colette's own words, The Cheat was a veri- table art school to which writers, painters and dramatists came nightly like students. It offered "the profound, if less than crystal clear, pleasure of seeing the crude 'cin6' groping toward perfection, the pleasure of divining exactly what the future of the cinema must be when its makers would want that future. . . ."' Seven of Colette's columns appeared regularly in Le Film, from 28 May to 21 July 1917, and most were devoted to American films, especially the Tri- angle films of Thomas Ince. When she quit the journal-"because there was no money in it"8-Louis Delluc was already acting as editor-in-chief and was publishing, concurrently with her column, his own articles on particular films. From his first column (25 June 1917),9 Delluc perpetuated Colette's interest in American films at the expense of French serials and literary adaptations, and he began to proselytize for a truly French cinema art. His witty, acerbic, trenchant pieces won a small but enthusiastic following among young writers and artists suddenly awakened to the cinema's potential. Within a year or so, the members of this "circle" had expanded and had begun to function on their own. In the summer of 1918, Delluc ceased publishing his own critical work in Le Film (with a couple of exceptions) but continued functioning as editor-in- chief, a post he held until 1919 when Diamant-Berger sold the journal to Georges Quellian and went into film production. Throughout this period, Le Film became a forum for several other young writers to take up Delluc's chal- lenge. In April, 1918, he published the full text of L'Herbier's "Hermes et le Silence," a philosophical rhapsody on the cinema's paradoxical ability to doc- ument life with acuity and exactitude (e.g., the war newsreels) and, at the same time, to create a "new symphony" of landscape, gesture, light and shadow. t In September, he introduced Louis Aragon's provocative essay, "Du Decor," which celebrated the cinema's transformation of reality through magnification/ isolation (e.g., the close-up) and through the subordination of set/landscape to character (e.g., Chaplin)." "It is indispensable," wrote Aragon, "that film take a place in the preoccupations of the artistic avant-gardes."12 In December, Delluc persuaded his friend Leon Moussinac (they had spent schooldays to- gether at the Lycee Charlemagne), to begin writing on the cinema. 13 All this was possible in part because Delluc had been able to satisfy his own desire for an even larger reading audience. 242 He succeeded in convincing Leon Parsons that the time had come for a major newspaper to provide its readers with a serious cinematographic criticism alongside its dramatic criticism and, on 1 June 1918, he gave Paris-Midi his first article devoted to Douglas Fairbanks." Paris-Midi was not the first Paris newspaper to publicize or report on the cinema. Beginning in 1908, Comoedia and, later, Le journal once a week had listed cinema programs in the capital and occasionally carried brief reviews and articles (as did others, such as Excelsior). Then, late in 1916, Le Temps had begun to publish a biweekly film review column by its music critic, Emile Vuillermoz.1' In fact, before it accepted Delluc's work, Paris-Midi had been printing, since January, 1918, a sometimes weekly, sometimes biweekly col- umn, "Les Spectacles," by Jan de Merry. But Delluc's column, "Cinema et cie," was the first critical appraisal of the cinema as art to appear weekly, on a regular basis. After January, 1919, it became a daily event. For a full year or more, Delluc was the single most important voice in the cinema market- place. And he consolidated that position by publishing two collections of his film criticism (the earliest in France): Cinema et cie (1919) and Photogenie (1920). Delluc's early film criticism was certainly not systematic or highly theoret- ical, but his columns and books were more than "mere bouquets of impres- sions," as he ironically described them. 16 The significance of his writings be- comes evident when compared to Henri Diamant-Berger's concurrent survey of the cinema as a craft in Le Cinema (1919). " As an admirer of American films as well as a proteg6 of Charles Pathe, Diamant-Berger advocated a film practice in France based on the production methods and aesthetic conventions of the American cinema. No less an enthusiast of American films, Delluc, however, sketched out, through repeated insights and ideas, a framework for an alternate French cinema. " For one, he singled out certain French films and filmmakers who were developing a form of film discourse in parallel with, but differing from, the Americans: Jacques de Baroncelli's Rarnuntcho (released Feb- ruary, 1919), Abel Gance'sJ'Accuse (released April, 1919), Marcel L'Herbier's Rose-France (released July, 1919), and Germaine Dulac's La Fete espagnole (re- leased May, 1920). For another, he advocated a different concept of filmmak- ing. The filmmaker should be the auteur of the ideas and stories he films rather than, as Diamant-Berger and Andre Antoine would have it, the metteur-en-scene of a scenario developed by a recognized writer. 19 And those ideas and stories should originate in the real world of contemporary life rather than be adapted from the theater or from fiction.1'0 Finally, he suggested the possibility of alternate methods of film distribution and exhibition. Attacking the then cur- rent "salad" of cinema programs (which he compared to an uncoordinated jumble of skits, excerpts, and one-act plays in a theater), Delluc asked for programs of separate individual screenings, at definite times.2' More impor- tant, in opposition to the industry's initial reluctance to re-release films, he began to consider the idea of a repertory of significant films that could be collected for repeated screening. Within three years of Delluc's pioneering efforts in the pages of Paris-Midi, the concept of a regular film review column was adopted by all of the major Paris daily newspapers. Andre Antoine's clarion call-"it is necessary to create a veritable, independent screen criticism, as now exists for the theater"- FILM CRITICISM 127. Louis Delluc by Beqan 243 ALTERNATE CINEMA NETWORK 128. Ricciotto Canudo seemed to announce the capitulation." Following Le Temps (Vuillermoz) and Comoedia (J.-L. Croze), most of the Paris newspapers had inaugurated a weekly film column or page by the autumn of 1921: L'Information (Lucien Wahl, editor of Eclair-Journal newsreels), Le Matin (Jean Gallois), Le Journal (Jean Chastaigner), Le Petit Parisien (J.-L. Croze), L'Intransigeant (Boisyvon), and L'Humanite (Leon Moussinac).A However, the most important of these, be- cause it brought together a circle of writers and filmmakers, probably was Le Petit Journal. In addition to his own column there, Ren6 Jeanne printed a series of articles by Diamant-Berger, Gance, Delluc, Ricciotto Canudo, Leon Poirier, and the art critic, Charles Leger. By 1922, the cinema was accepted as a permanent form of popular spectacle by the press, and certain critics (Vuillermoz, Wahl, Jeanne, Moussinac) were writing as if it were an autono- mous art. Delluc's example also had an impact on the literary magazines of Paris. Before the war, Guillaume Apollinaire's avant-garde monthly, Les Soires de Paris, had expanded its provocations with a regular film review column by Maurice Raynal-which culminated in the whimsical "Soci6t6 des amis du Fant6mas," based on the famous Louis Feuillade serial. Two small journals founded during the war-Pierre Albert-Birot's SIC and Pierre Reverdy's Nord- Sud-tried to sustain and extend Apollinaire's interest with a few notes, es- says, and interviews.1 In 1919, however, three literary magazines that would have some influence in the 1920s instituted regular essays and/or review col- umns on the cinema. One was Litterature, an irreverent new magazine edited by the young poets, Andre Breton, Louis Aragon, and Philippe Soupault. Their interest coincided with Delluc's, but it led to a series of playful "cine- matographic poems," based especially on Chaplin and William S. Hart films. Another highly polemical, but more long-lived magazine was Le Crapouillot, edited by the robust eccentric, Jean Galtier-Boissiere.21 As early as 1919, Galtier-Boissiere published a seminal essay on film by the painter, Marcel Gromaire; and, in March of 1920 and 1923, he devoted special issues exclu- sively to the cinema.2' Between those dates, Le Crapouillot emerged as an important forum for film criticism, especially when Leon Moussinac became its regular film reviewer in September, 1921. Before writing for Le Crapouillot, as well as for L'Humaniti (the Communist Party newspaper), however, Mous- sinac had joined the staff of Le Mercure de France. For six years, beginning in May, 1920, this prestigious literary magazine published Moussinac's trimonthly film review column along with early articles by young writers such as Jean Epstein and Alexandre Arnoux.2 In one of his first reviews, Moussinac pro- phesied: A new art is born, develops, discovers one by one its own laws, progresses toward perfection, an art which will be the expression-bold, powerful, original-of the ideal of a new age. The most flamboyant prophet of the new art was Ricciotto Canudo. In Paris before the war, this polemical Italian writer had published Montjoie/, a small review devoted to modern tendencies in the arts, and had actively encouraged filmmakers to join his circle of artist contributors." To that end, he had written one of the first credos on the cinema-"Manifesto of the Sixth Art" (1911). After being demobilized from the French army, Canudo founded a 244 second journal, La Gazette des sept arts, consecrated more specifically to the FILM JOURNALS aesthetics of the cinema.31 The purpose of this journal was twofold: -the conquest of intellectual and artistic milieux, until now recalcitrant to the cin- ema;-the amelioration of the quality of film production.32 Canudo's connections in the arts world were even more extensive than Del- luc's, and La Gazette des sept arts (which appeared irregularly until its founder's death in late 1923) included a host of important articles by writers (e.g., Alexandre Arnoux, Jean Cocteau), painters (e.g., Fernand Leger, Robert Mal-. let-Stevens, Marcel Gromaire), musicians (e.g., Arthur Honegger on his score for selections from Gance's La Roue), filmmakers (e.g., Jean Epstein on shoot- ing his first film, Pasteur), and critics (e.g., Moussinac). To advance his ideas on a more regular basis, beginning in October, 1922, Canudo also initiated a film review column in the new literary weekly, Les Nouvelles littiraires. This "missionary of poetry in the cinema," as Jean Epstein later called Canudo, like Delluc, proselytized the idea of the filmmaker as an auteur or 6cranist.33 And, more consistently than any other writer, he articulated an expressive or Symbolist theory of the cinema, emphasizing the film image's evocation of the filmmaker's as well as of a character's feeling, imagination, or state of mind.34 Soon other magazines began opening their pages to the cinema-some, appropriately, to Delluc himself. One of Le Mercure de France's chief rivals, La Nouvelle Revue Franfaise, initiated the practice of publishing film scenarios as literary works, the most famous of which was Jules Romains's "Donogoo- Tonka" (November and December, 1919).35 Another avant-garde journal, Le Corbusier and Ozenfant's L'Esprit nouveau, from 1920 to 1922, published three Delluc pieces championing the cinema (one of them devoted to Chaplin).36 During the same period, the journal also accepted an early essay on the aes- thetics of cinema by B. Tokine, a second piece on Chaplin by the art historian Elie Faure, and one of Jean Epstein's first essays on current French films.37 Les Choses de theatre (first issue) and Le Monde nouveau published important essays by Delluc which were to form the basis for a book tentatively titled "Les Cineastes," but which he left unfinished at his death.38 Finally, two monthly review journals established short-lived, but influential columns. One was Paris- Journal which, in 1923, began printing the reviews of Georges Charensol and the young Surrealist poet, Robert Desnos.3 The other was the deluxe Thatre et Comoedia Illustre, now published by Rolf de Mare, who also managed the Theatre des Champs-Elysees where the famous Ballet suedois performed.40 For over a year and a half (from December, 1922), Theatre et Comoedia Illustr included a special film supplement in which Ren6 Clair wrote a comprehensive review of the month's films in Paris while twenty-year-old Jean Mitry (in between assignments as a publicity poster artist) transcribed some half-dozen interviews with major French filmmakers.4' The Paris newspapers were now providing information and increasingly so- Film Journals phisticated film reviews to the public, and certain literary magazines were educating their elite audiences to the value of the cinema. Could there be a publication format, a form of organization, to bridge the gap between these two-to create a popular movement which could put pressure on the French 245 ~i'~~ji iK I le journal du o ce a A n 1 Hebdomadaire Cirngraphiqte - LE JOURNAL OU CINII-CLUB publiant les programmes des CINEMAS DE PARIS PEARL WHITE 129. (left) The cover of the first issue of Cine-pour-tois 130. (right) The cover of Le Journal d CinI-Club film industry and influence the direction of its operations, or which (failing that) could establish an alternative to it? To this task, Louis Delluc and several other writers now turned their efforts. On 15 June 1919, a new film journal appeared with a masthead which proposed all Paris for its audience. Edited by Pierre Henry, Cine-pour-tous ran eight tabloid-sized pages and cost but twenty centimes (old currency). Each fortnightly issue included a list of French films in production, notes on current releases, a half page of credits and reviews on a selected film (usually French), a two- or three-page essay on a film star (usually American), information on a particular area of the film industry, letters from readers, and an editorial by Henry. In Paris-Midi, Delluc drew attention to the journal by devoting a full column to it. The "grand public"-which loves the cinema and which gives it life-wants someone to address it. I have always thought that would happen one day. So it begins with Cine-pour-tous, a biweekly review which is not content to address the public occa- sionally but wants to address it exclusively. This ambition pleases me. Everything in it has intelligence, taste, precision, force. Sadly, the pages are limited. But it will grow, and you will find again, by waiting, all that you want to know about films, about actors, about the present and sometimes the future of the cinema.1 According to Jean Mitry, who began his cinema education by surreptitiously 246 reading Cine-pour-tous at school and then became one of its writers, words cannot do justice to "the role that it played, the influence that it had on the formation of young minds and the first cinphiles."2 Cin6-pour-tous undoubtedly spurred Delluc's own interest in publishing a magazine for the growing number of film enthusiasts. With the help of friends such as Charles de Vesme, Georges Denola, and Leon Moussinac, he prepared a weekly film magazine whose first issue appeared on 14 January 1920.3 Its title, Le Journal du Cine-Club, was more prophetic than accurate. Actually there was no cine-club or formal organization of cinphiles. LeJournal du Cine- Club was essentially a magazine whose purpose was "[to help build relations between the public and the 'cinematographistes,' [to support the work of all the young filmmakers" and "[to organize] lectures accompanied by projected film clips, dealing with the history of the cinema, its achievements, its artistic nature, [and its social and educational ends. . . ."4 Through its readers, wrote Charles de Vesme, "in an epoch in which the masses played such a large role and exerted on all things such a great influence, [Delluc hoped to mobilize a cadre of the elite and the professional, together with an army constituted of the vast public who were passionate about the cinema. "5 Published twice as often as Cine-pour-tous, Le Journal du Cin6-Club also included more pages (twelve) and more information: complete listings of cinema programs in Paris, the suburbs, and even some of the provinces; brief reviews of all the new Paris releases of the week; and articles on filmmakers as well as film stars. Despite their intentions, neither Cine-pour-tous nor Le Journal du Cin-Club ever appealed to more than a small elite of the French film audience, much like the elite readers of the literary and art magazines. A truly popular film magazine did not emerge until January, 1921, when Jean Pascal and Adrien Maitre launched Cinimagazine. This one-franc weekly journal of some fifty pages aligned itself quite closely with the commercial film industry and ca- tered to the tastes of the mass cinema audience.6 Cinmagazine served as a major publicity outlet for the French and American film producers and distributors by reviewing the week's new releases and previews, doing several articles in each issue on popular stars, publishing episodes from a filmed serial or novel, printing a great number of publicity stills and advertisements, and providing a gossipy information column. So quickly did it become popular that, within eight months of its founding, Cinemagazine had its own correspondent in the United States, a former assistant to Louis Feuillade, Robert Florey. But Pascal and Maitre did make some effort to bridge the gap between the masses and the elite, between the dominant industry and the alternate cinema advocates. Important critics such as Moussinac, Jeanne, Wahl, and Vuillermoz were printed frequently enough in the magazine; and independent French filmmakers often reported on their production or distribution activities and problems. Cinma- gazine also did tend to single out independent narrative films for their artistic value. Cinimagazine's success quickly generated some rivals. Within two years, at least two more popular weekly film magazines became available in Paris: Mon- Cine and Cin-Miroir (published by Le Petit Parisien).' Both of these proto-fan magazines concentrated on stories about stars, filmmakers, and particular film productions, as well as novelizations of current films, at the expense of actual film reviews and cinema program listings.8 The simultaneous release of a film FILM JOURNALS 247 CE NULUCUNTIENT ULUX PLACE-" : S im:. I .3 O CINEMA A TAIM ULOUIT " r magazine S T PI ___-ABEL DANCE ____ 131. (left) The cover of Cinema gazie i 132. (right) The cover of Cinia: Rio Jim poster by Beqan and its novelization in a newspaper or popular film magazine had become so profitable by 1922 that several publications-e.g., the weekly Le Film Complet and the biweekly Les Grands Films-did nothing but package recits of current films.9 The most expensive of these-Cinema-Bibliotheque (published by Jules Tallandier -appeared actually as a small booklet, with inserted specially printed sheets of production and frame stills."' All of these magazines now constitute an important, if approximate, record of lost and incomplete films; at the time, however, they served as little more than organs of the industry. The most influential alternate film journal had its origins in Louis Delluc's last critical venture. In April, 1920, Delluc turned over the editorship of Le Journal du Cine-Club to Georges Denola when he himself became involved in directing films. When the journal was forced to cease publication one year after its inception, Delluc persuaded Arkady Roumanoff (a Russian emigre collector and amateur artist) to finance a deluxe weekly review, Cinea, which would be for the cinema what Comoedia Illustr( was for the theater." Director- ship of the new journal was first shared by Delluc and Roumanoff and then handled by Delluc alone; but there were many collaborators: critics (Moussi- nac, Wahl, Vuillermoz, and especially Lionel Landry), poets (Canudo, Jean 248 Epstein, and Ivan Goll), filmmakers (I'Herbier, Baroncelli, Poirier, Louis Nalpas, Henry Roussel, and Alberto Cavalcanti), actors (Philippe Heriat and Andre Daven, and graphic artists (Serge and Beqan)."2 For nearly two years after its first issue (6 May 1921), Cinea would be the most consistent, outspoken de- nigrator of the commercial French cinema and advocate of a national cinema art. In Cinea, Delluc and his colleagues began to sketch out a loose set of criteria for determining film art. Chief among them was the elusive term, photoginie, which Delluc used to point to the sometimes artless, sometimes artful, trans- forming power of film in relation to reality, but which. became a kind of "floating signifier" that recurred frequently throughout the 1920s.13 Delluc himself came to focus on the filmmaker's "composition" of a central idea or theme, described sometimes in terms of an analogy then current, that of mu- sical orchestration, which he used to define "cinegraphic rhythm.""' Although the musical analogy was quite prominent in early French film theory (e.g., Canudo, Gance, Dulac, and even Epstein)," another formulation probably would be more precise-"poetic composition." For the process was similar to that of rhetorical and rhythmic patterning in poetry, a kind of poeticization of the process of representation. On a more practical level, Delluc's analyses of individual films, along with those of Moussinac, Vuillermoz, and Clair, became models of an informed film criticism. 16 Delluc also became more in- sistent in promoting the re-release of older films and the exhibition of inno- vative new films, especially those from Sweden and Germany. He was now sketching the outlines of a history of cinema art that was both national and international. A library or repertory of significant films was needed, he argued, not only to preserve but to promulgate the idea of cinema art and to educate cinema audiences in order to support further innovations and the cinema's eventual achievements. Of the contributors to Cinea, the one who developed the implications of Delluc's work most thoroughly and imaginatively was the young poet and essayist, Jean Epstein, whom Jean Mitry recently admitted was "the first real theoretician of the cinema."7 Epstein's ideas emerged full-blown in a peculiar blend of poetic and scientific language, in the last chapter of his first book, La Poesie d'aujourd'hui: un nouvel etat d'intelligence (1921), and especially in Bonjour Cinema (1921). In its original design format, Bonjour Cinema was a witty parody of a film program. It included poster photographs of film stars, adulatory "fan" poems, a serial episode, and several "features"-the essays, "Le Sens 1 bis," "Grossissement," and "Cind-Mystique."18 More important was the book's correlation to the work done by Apollinaire, Blaise Cendrars, and even the Dadaists during the war and the postwar period. It combined a variety of different materials-three major essays, a half-dozen poems printed in free typography, photographs and posters of several American stars, and Cubist-style paintings and drawings. In a sense, Epstein was already playing with the idea of "editing" together the diverse strands of modern life into something analogous to a film. And in his essays was the germ-or quantum theory, if you will-of a cinema of discontinuity or, rather, of a continuity quite different from that developed by the American cinema. Delluc's own film practice during these years-especially Fievre (1921) and La Femme de nulle part (1922)-cut short his work as a film critic and editor. FILM JOURNALS JEAN E FTEI EDITIONS DE L A SIEsENE 29OULEVARDMALESHERBESPARIS921 _ILF-cIiomI"TRACTIS 133. The cover of Bonjour Cinema (1921) 134. Jean Epstein 249 ALTERNATE CINEMA NETWORK 135. The cover of Cinea-Cin- pour-tous: Arlette Marchal Nile H5ri. N 47. 4s OCTOBRE 1925 Prix # I rRANC 50 After the commercial failure of La Femme de nulle part, in October, 1922, Delluc sold his interest in Cinia to Jean Tedesco (a young editor of women's magazines) and gradually yielded the editorship to him. A year later, Tedesco bought out Cine-pour-tous, which was in financial trouble due to the prolifer- ation of film magazines, and the two journals merged. In November, 1923, Cinia-Cin-pour-tous appeared in a glossy, beautifully laid out, thirty-six-to- forty-four-page format that boasted a special section of tastefully mounted photographs. Under Tedesco's editorial control, the new journal included re- views and articles by Edmond Epardaud, Pierre Porte, Dr. Paul Ramain, Pierre Henry, Juan Arroy, Rene Jeanne, and Leon Moussinac as well as state- ments by such filmmakers as L'Herbier, Epstein, Henri Chomette, Fritz Lang, and Lulu Pick. It had an immediate and lasting impact. Throughout 1924, Tedesco and others, such as Jeanne and Moussinac, agitated for Delluc's idea of a repertory cinema; and both Cinea-Cine-pour-tous and Cinemagazine gave extensive coverage to the special lectures and exhibitions such filmmakers as 250 L'Herbier, Dulac, Epstein, and Clair were beginning to give. In a short time, CINE-CLUBS especially through the writings of Epstein, Porte, Ramain, and Tedesco him- self, Cinia-Cine-pour-tous became the principal forum for the theories and ar- guments that engaged the French avant-garde.1 This broadly sustained development of independent cinema journals and an Cine-Clubs embryonic film theory/criticism were unique to France. Nothing much like this existed in the United States or Germany, let alone Great Britain, Italy, or the Scandinavian countries. Only the revolutionary society of the Soviet Union had a somewhat comparable phenomenon. But the Soviet writings came slightly later and developed quite differently. There, filmmakers such as Lev Kuleshov, Dziga Vertov, and Sergei Eisenstein were the ones to initiate discussions of film theory and practice in Kino-fot, Kino-Gazeta, and especially LEF (1923- 1925).1 Within a couple of years, in the pages of Novy LEF (1927) and in several books, most notably The Poetics of Cinema (1927), the Russian Formalist literary theorists and critics (Boris Eikhenbaum, Viktor Shklovsky, S. Tre- tyankov, and Yuri Tynyanov) then took up the aesthetics of the cinema, the practice of film criticism, and even scriptwriting.2 In France, the movement had a broadly based constituency (crossing several classes and professions), but it was more exclusively cultural than political (as the critics saw it) and more oriented (at least initially) toward a conception of cinema based on the film- viewing process rather than on filmmaking. Its purpose was to encourage, first of all, an informed audience that would support advances in film art and, only secondarily, real social change. Whereas Soviet film theory and practice were stimulated by an exhibition system that included mobile agit-trains and steamers,3 the French development was supported by the unprecedented rise of cine-clubs and special film lectures and exhibitions in Paris. Again Delluc, along with Canudo, Moussinac, and a few others, were crucial instigators. The idea of a cine-club or an organization of cinephiles dedicated to the advancement of cinema art seems to have occurred simultaneously to Delluc and Canudo. Although Delluc coined the term cin-club in 1919 (from the groups called Touring-Clubs, according to Sadoul)4 and published Le Journal du Cine-Club, he never really organized any club or circle that would meet regularly and engage in concerted activity. Instead, he tried out several ideas which would later become standard practice in the cin6-club movement. The first was a conference or special program that would provide a retrospective look at the work of a particular filmmaker. Under the rubric of LeJournal du Cin- Club, on 12 June 1920, at the P6piniere-Cin6ma, Delluc organized a special program devoted to the animator, Emile Cohl (most of whose short French films had been made between 1907 and 1912), and to the theater director turned filmmaker, Andr6 Antoine, whose lecture on "The cinema of yesterday, today, and tomorrow" was illustrated with film clips.5 Soon after, Antoine broached the idea of redirecting the programming of one particular cinema in Paris for the clientele of cinphiles-to have a kind of "Vieux-Colombier of the cinema. "6 Delluc's second idea was to present an unreleased film (the work of a cineaste) to a limited audience. Now under the rubric of Cinia, on 14 November 1921, at the Colisee cinema, Delluc organized a charity matinee screening for "Tout-Paris." The film shown was a shock-The Cabinet of Doctor 251 ALTERNATE CINEMA NETWORK Caligari (1919).' It was the first German film to be seen in France since before the war, and the director of the small Cin6-Opera cinema was interested enough by it to arrange an exhibition of German films the following year. Two months later (22 January 1922), again at the Colis6e, Delluc hosted a matinee session for an art circle, "Ideal et Realit6." The program brought together the twin interests of the young cinephiles into sharp juxtaposition: while he himself spoke on "Cinema, the popular art," L'Herbier screened his short experimental film, Promethe . . . banquier, as a kind of entr'acte.8 Once having tested his ideas, however, Delluc seems to have pursued them no further-either out of a deep commitment to his film practice or, more likely, because of financial and emotional difficulties (e.g., his marriage to Eve Francis).9 But his friends would not let those ideas drop. The first actual cine-club seems to have grown out of the informal gather- ings of artists, writers, and professional filmmakers at the Caf6 Napolitain in 1920. At their head was Ricciotto Canudo who, by April, 1921, had dubbed the group C.A.S.A., "Club des amis du septieme art."10 Its members included some of the most prominent of the avant-garde in the arts: filmmakers (Delluc, Dulac, Poirier, L'Herbier, Cavalcanti, Epstein), critics (Moussinac, Wahl, Landry, Jeanne, Pierre Scize), writers (Faure, Arnoux, Cendrars, Cocteau), artists (Gromaire, Mallet-Stevens, Leger), musicians (Honegger, Ravel, Ro- land-Manuel, Maurice Jaubert), and actors (Eve Francis, Jaque Catelain, Jean Toulout, Harry Bauer, Gaston Modot).11 This was a cine-club that fulfilled Delluc's expectations. The principal activity of C.A.S.A., at first, involved little more than dinner speeches and discussions, at the Poccardi restaurant or at Canudo's spacious apartment, 12 rue du Quatre-Septembre, much in the manner of the famous Belle Epoque banquet celebrations before the war. 12 However, C.A.S.A. mem- bers were the major contributors to Canudo's La Gazette des sept arts; and several filmmakers associated with the group (Gance, Dulac, Nalpas, Rene Le Somptier) took part in a series of lectures on the cinema given early in 1921.13 Canudo himself published an important manifesto for C.A.S.A. in the second issue of Cinda (13 May 1921). Several of the principles laid down there were harbingers of the future: b. To raise the intellectual standard of French cinematic productions, for aesthetic as well as commercial ends. . . . d. To consider as urgent the establishment of a "hierarchy" of cinemas such as exists in the theater: popular cinemas and elite cinemas. . . . f. To agitate, by every propagandistic means, so that equitable laws and reasonable supports be provided by the State to the "Art of the Screen," in the same measure at least as they are accorded the "Art of the Stage." g. To attract public attention to the origins and evolution of the Cinema in France, through the organization of a first French Cinematic Festival. 14 Apparently, Canudo organized several private film screenings for C.A.S.A. in the hall of the "Syndicate de la Grange-sur-Belles" (about the same time Delluc was presenting Doctor Caligari at the Colisee). 15 But his real coup, achieved with the aid of Moussinac, was to persuade the president of the prestigious Salon d'Automne, Franz Jourdain, to hold a special exposition of 252 film screenings in 1921 and again in 1922 and 1923.6 At the first exposition, CINE-CLUBS following an address by Jacques de Baroncelli (read by the actor, Signoret), Canudo arranged the screening of "a selection of films by Baroncelli, Feyder, Roger Lion, Henry Roussel, (and Edouardj Violet . . . as well as several fragments from La Roue. . . ."" Two years later, he selected excerpts from the best films of the year which drew attention to what he considered to be the major cinematographic styles: "1. Realism, 2. Expressionism, 3. 'Essays' in cinematic rhythm, 4. Pictorial Cinema."8 Canudo's film programs were particularly significant because they represented the first attempt, in Jean Ep- stein's words, "to present to the public specifically chosen film sequences, to constitute an anthology of cinema."" These expositions thus marked the ear- liest semi-official recognition of the cinema as art. To cap this work, just before his death in November, 1923, Canudo also arranged for a similar pro- gram of screenings in Lyon (the birthplace of the Lumiere brothers).210 In the words of Henri Fescourt, Ricciotto Canudo's influence was decisive for the recruitment and education of cu- rious, newly fascinated spectators everywhere. His propagandizing effort had a special, courageous character since it tried no less than to build a bridge of communication and sympathy between the traditional arts and the young, unruly, but promising savage that was the cinema. That effort bore fruit.' 136. Loon Moussinac Alongside the dinners and private screenings of C.A.S.A., which continued into 1922 and 1923, there appeared a number of other clubs and special events. In imitation of Canudo's club, in December, 1921, the editors of Cinemagazine created an informal group called simply "Amis du cinema."-- Throughout the next two years, the group held monthly lunches and occa- sional lectures, including sessions with Bernard-Deschamps and Diamant-Ber- ger.23 In June, 1923, Marcel L'Herbier was invited by Robert Aron to speak to the Students' International Circle at the College de France.4 His speech, "Le Cinematographe contre lart," was in such demand that he repeated it during the next few months in Geneva, La Haye, and Lausanne.25 Inspired by this mushrooming activity, young Georges Sadoul invited Jean Epstein to give a lecture at Paris-Nancy, in December that same year, a lecture which Epstein then repeated a month later at Montpelier. 26 The most important new organization to emerge from this wave of interest was started by Leon Moussinac, either late in 1922 or early in 1923.2' The "Club franais du cinema" was a semi-professional organization which seems to have formalized and recharged the loose circle of people Delluc had brought together around Le journal du Cine-Club and Cinia. Directed by Leon Poirier, who along with L'Herbier had just been fired at Gaumont, its agenda was set up explicitly to defend filmmakers as artists (or cin6astes, to use Delluc's term) and to attack the restrictions of the commercial industry. They demanded 1) that film writers and directors enjoy the benefits of artistic copyright; 2) that they cease being at the mercy of production companies who too often ignore the subjects of their films or, at least, distort them; 3) that film writers and directors, like their actors, have contact with public opinion, by means of an independent criticism, such as exists for the theater, literature, or painting. . . ." Toward these ends the club organized a number of evening screenings of both 253 ALTERNATE CINEMA NETWORK French and other European films, in order to "unleash new initiatives, fervent convictions, creative energies-elans even-which, by breaking the restrictions of mercantilism and routine, will propel the art of silence to the heights of its predecessors." 29These screenings included Delluc's La Femme de nulle part (1922), Epstein's La Coeurfidele (1923), Lulu Pick's Sylvester or New Year's Eve (1923), the earliest Soviet feature film, Polikouchka (1919), and a new version of Gance's La Roue.30 The drive for an alternate cinema was reaching a new stage of militancy. That drive seemed to accelerate rather than diminish after both Canudo and Delluc died suddenly within five months of one another (November, 1923, and March, 1924). For one, the Amis du cinema group expanded its activity. Affiliated clubs were organized in several provincial French cities.31 In May, at the Artistic cinema, the Paris club sponsored a special preview of Jaque Catelain's second film, La Galerie des monstres (1924); and in November, at the Colis6e cinema, it sponsored major lectures by Germaine Dulac and Ren6 Clair.32 For another, C.A.S.A. itself underwent a transformation. Early in March, 1924, at the club's first meeting since its founder's death, Ren6 Blum proposed the revival of the weekly members-only dinners and, more impor- tant, the sponsorship of regular public film screenings and lectures.33 With the help of Moussinac, Tedesco, and others, Blum organized a biweekly series of evening film programs at Raymond Duncan's dance studio on the rue Co- lisee.34 The first program, on 11 April, was dedicated to Delluc. Blum read an homage (prepared by Dulac) introducing a rescreening of Fievre, and then Jean Epstein projected his short montage film, Photognies, edited specially for the occasion.35 Two weeks later, C.A.S.A rescreened Epstein's controversial film, Coeur fidele, along with what Tedesco called "S6lections Symbolistes"- selected fragments from La Roue and Mosjoukine's Le Brasier ardent.36 For its third program, the club took the risk of presenting Dmitri Kirsanoffs first film, L'Ironie du destin (1924), which had received scant commercial distribu- tion.37 Although this series of special C.A.S.A. screenings was something less than an unmitigated success with the public, it seems to have encouraged several different lines of activity. Tedesco's selection of film excerpts and Epstein's Photogenies most likely helped stimulate the avant-garde production of short non-narrative films--e.g., Lager's Ballet micanique (1924), Clair's Entr'acte (1924), Autant-Lara's Fait-Divers (1924), and Chomette's Jeux de reflets et de vitesse (1925). More certainly, the weekly exhibition strategy convinced Tedesco to begin looking around for a permanent cinema to provide continuous programming in the fall. Meanwhile, Blum and Moussinac turned their attention to an upcoming exposition. In the winter of 1923-1924, Moussinac had pulled off a major coup at the Musee Galliera. The Galliera's director, Henri Clouzot, had become a member of the Club fran ais du cinema, and Moussinac persuaded him to form a commission in order to mount a full-scale exhibition devoted to the cinema-- in effect, realizing one of Canudo's dreams.38 On that commission were Louis Lumiere, Gaumont, Path6, Gance, L'Herbier, Mallet-Stevens, Jean Benoit- Levy, Vuillermoz, and Moussinac himself. The resultant "L'Exposition de lart dans le cinema franqais" extended from May through October and included displays of film stills and scripts from particular filmmakers, animation draw- 254 ings, examples of film titles and credits, set design models and sketches, CINE-CLUBS costumes, posters, film books and periodicals. It also boasted two series of lectures, one lasting a month (May through June), the other in October. The first amounted to a veritable seminar of ten lecture-presentations on the cinema as art. Tuesday May 27 M. G.-Michel Coissac: The history of the cinema-prophets, precursors, filmmakers. Friday May 30 M. Leon Moussinac: The characteristics of this new art-modern, popular, interna- tional. The styles of cinema. Descriptive cinema and the cinegraphic poem. Genres. Relations between cinema and the other arts. Tuesday June 3 M. Leon Moussinac: The visual idea, its specific character, its development in the scenario (adaptations). The importance of scriptwriting and mise-en-scene-elemen- tary techniques. Cinegraphic rhythm and balance. Friday June 6 M. Marcel L'Herbier: Photoginie. The important role of light, interior and exterior lighting methods. Tuesday June 10 M. Jaque Catelain: The interior rhythm of images. Acting-expression, movement, makeup. Friday June 13 M. Mallet-Stevens: Decors. Costumes. Props. The cinema-the means of popularizing the modern forms. Intertitles and posters. Tuesday June 17 M. Jean Epstein: The expressive techniques of the cinema-the function of different shot distances and angles. Dissolves. Lap dissolves. Superimpositions. Soft focus. Deformations. Animation. Friday June 20 M. Lionel Landry: French cinema; its characteristics, its aesthetic development. For- eign influences. The function of the critic. Tuesday June 24 M. Ren6 Blum: Cinema and music. Composers. Orchestra conductors. Synchronizing machines. Friday June 27 M. Luchaire: French film industry-its impact on French ideas worldwide.39 Substituting for Epstein, Germaine Dulac wrote one of her most persuasive essays of the decade on the possibilities of a subjective cinema.40 If the cin- club movement still exerted less influence on the industry and on the mass French audience than it would like, it was well on the way to establishing film as an equal in the world of the arts. The consequences of the Galliera exhibition were twofold. For one, early in 1925, it led to the merger of C.A.S.A. and Moussinac's Club franqais du cinema into one body, the "Cin6-Club de France."41 At the core of the new group were Moussinac, Blum, Poirier, Feyder, and Henri Clouzot. According to its charter, the Cine-Club de France sought . . . to advocate the study, development, and defense of the cinema as Art. . . . to coordinate all the intellectual, artistic, technical, and economic structures capable of enriching the international landscape of the cinema with films. . . . to encourage the sincere effort of artists from all countries, whatever their 255 ALTERNATE CINEMA tendency or style, and to further, by any means possible, that which provides special NETWORK publicity to the manifestations of its activity. With Moussinac providing much of the impetus, the new club embarked on an extensive schedule of monthly public film screenings-first at the Colis e and then at the Artistic cinemas-especially with the aim of reviving older films unjustly ignored (e.g., Stiller's The Legend of Gosta Berling) and preview- ing completed films (e.g., Feyder's L'Itnage, Lager's Ballet ;necanique), frag- ments (e.g., L'Herbier's Resurrection), and films rejected by the commercial distributors or by the censors (e.g., Eisenstein's Potemkin). ' Under Dulac's guidance (with support from Epstein and L'Herbier), in the winter of 1925- 1926, the club also organized another major series of conferences at Jean Tedes- co's new Theatre du Vieux-Colombier.,4 Entitled "The Creation of a World through the Cinema," the announced twelve lecture-presentations, from 28 November to 20 February, complemented those at the Mus e Galliera the year before: 1. The meaning of the cinema Leon Pierre-Quint 2. Photoginie of the machine world Pierre Hamp 3. The psychological value of the image Docteur Allendy 4. The external world as revealed by sunlight Jean Epstein 5. Human emotion Charles Dullin 6. Fantasy and humor Pierre Mac Orlan 7. The comic Andre Beucler 8. Cinema and time Jean Tedesco 9. The formation of sensibility Lionel Landry 10. The shackles of the cinema Germaine Dulac 1. Photogenie of the animal world Colette 12. The cinema in modern life Andr6 Maurois' Although there were changes in this program (e.g., L'Herbier seems to have 137. Charles _ger substituted for Colette), the series was so successful that the Cine-Club de France added several Friday evening lectures which lasted at least into March.6 The Cin6-Club de France's emergence out of the Galliera experience im- pelled a second group to form around Charles Leger-"Le Tribune libre du cin6ma." Here was a younger generation of cinephiles, including Marcel Carne, Jean Dreville, Edmond T. Greville, Bernard Brunius, Robert de Jarville, Jean Mitry, Jean Arroy, and Jean-Georges Auriol.-' Jean Dreville tells me that the Tribune libre had its origins in a number of meetings and discussions in a hall on the Avenue Rapp, late in 1924." Soon after, the group established a biweekly series of public film screenings." One of Charles Leger's first and most difficult accomplishments was to organize (with Moussinac's assistance) an even larger cinema exhibition for the famous Paris Exposition des Arts D6coratifs et Industries Modernes in the summer and autumn months of 1925.5" The highlight of the exhibition, at "La Maison des artistes," was a series of screenings of selected French films with each filmmaker present for individual discussion sessions. So Mme. Germaine Dulac's last film project, La Folie des vaillants, Rene Clair's Paris qui dort. Entr'acte, and Le Fantome du Moulin-Rouge, Lulu Pick's Shattered, Louis Del- luc's Fiere, Jean Epstein's L'Affiche, etc. . . . received an enthusiastic response from a mass cosmopolitan public. 256 The film programs changed every Thursday, and for the inauguration of each new SPECIALIZED CINEMAS film, at a gala sance on Thursday evening, the filmmaker himself spoke about his film-his intentions and ideas; after the screening, the filmmaker solicited the opinion of opponents, and debates ensued, directed and animated by Charles Leger, the de- lighted program organizer. There one could finally see drawn together, discussing the same subject, the artisans of the cinema, men of letters, painters, sculptors, archi- tects. . . .51 After this astonishing success, Tribune libre immediately set up a second season of biweekly public film screenings at the Salle des Ingenieurs civils and . then at the Salle Adyar, Square Rapp.52 Following the format developed at the Exposition, Lger applied the combination of short lecture, projection (of excerpts or a complete film), and public discussions on a systematic basis.53 This strategy quickly opened up the cin6-club movement to a wider audience. This upsurge of cin6-club activity coincided with the institution of the first Specialized Cinemas specialized cinema in Paris. The man to implement Delluc's original idea of a repertory of films, perhaps naturally enough, was Jean Tedesco. In his search for some nook or cranny to project films on a permanent basis, late in Septem- ber, 1924, Tedesco discovered to his surprise (did he remember Antoine's suggestion of three years earlier?) that Jacques Copeau's famous but recently abandoned Th6tre du Vieux-Colombier was available for rent. Initially, ac- cording to Andr6 Brunelin, he and an entrepreneur by the name of Simon Gantillon concocted a scheme to use the Vieux-Colombier for alternate eve- nings of theater performances and avant-garde film screenings. 1 Gantillon soon pulled out of the venture, but Tedesco persisted. Although Copeau himself disdained the cinema, he was sufficiently interested (apparently Gantillon smoothed the way) to give the young madman a year's lease on the 500-seat auditorium.2 In less than a month, Tedesco and his friends transformed the theater-installing a projection booth and screen, repainting the fa ade and interior, printing posters and invitations. Since Copeau, in his retreat to Bour- gogne, had taken along the original pair of Florentine doves which marked the theater's entrance, Tedesco was in need of a new emblem. One day by chance he found just what he wanted-the simple circular design of the ap- erture shutter in a film projector.3 On 14 November 1924, the Vieux-Co- lombier finally opened its doors as the beacon of cinema art. On the posters and advertisements announcing his new cinema, Tedesco set forth a few simple principles. The Vieux-Colombier's programs would be com- prised of avant-garde works and a film repertory. In that repertory would be included "quality films that the commercial industry had not allowed the majority of the public to see" as well as "films of such value that they merit a second screening under the title of Cinema Classics."4 The first week's pro- gram established a format for those that followed: Andre Sauvage's La Travers du Crepon (a documentary on mountain-climbing), Marcel Silver's L'Horloge (an experimental film without intertitles), and Chaplin's short feature, Sunnyside (1921).5 The films were projected nightly and ran for one week. And they were accompanied by a secluded chamber orchestra of just three or four in- struments whose musicians prepared the music specially for each program.6 For seven months, until the first week of June, 1925, the Vieux-Colombier 257 ALTERNATE CINEMA NETWORK fulfilled its mission of a full season of film screenings. Most of the films shown were revivals-from the United States (Chaplin's shorts, Griffith's Way Down East, Fairbanks's The Mark of Zorro), Sweden (Sjdstrdm's The Phantom Coach [in a new print, Stiller's Trdsor d'Arne), Germany (Wiene's Doctor Caligari, Lulu Pick's Shattered, Lang's Destiny), and France (Feyder's Crainquebille, Ep- stein's Coeurfidele, and Roussel's Les Opprims).' But there were also several new films-Carl Grune's The Street (1923) and the Soviet feature, Polikouchka-and excerpts projected as set pieces of the narrative avant-garde film practice: "S6- lections de rythmes" from La Roue, "Etude d'Expressions" from L'Affiche, and fragments of accelerating montage (from La Roue, Coeur fiddle, and Kean).8 Although Tedesco sometimes organized a double bill of feature films or a selection of excerpts, his programs almost always included a short film, usually a documentary.9 At first, "Tout-Paris" came to the little cinema out of curi- osity, but the public interest waned, particularly after Tedesco tried to screen Arthur Robinson's Warning Shadows. 10 Perhaps a more serious problem for him was a lack of publicity. Except for his own announcements in Cinia-Cine-pour- tous, a diatribe from Moussinac in defense of the Vieux-Colombier in L'Hu- manit6, and a brief reference by Emile Vuillermoz in Le Temps, the Paris news- papers and film magazines either ignored or took little interest in Tedesco's operation.11 By the end of the first season, the Vieux-Colombier was in danger of closing down once again. Yet Tedesco continued to believe in his exhibition strategy, and he con- vinced Copeau to extend his lease on the Vieux-Colombier for one more year. Several factors were now in his favor. The Exposition des Arts Decoratifs had excited, both among the masses and the elite, a wider interest in cinema art. Following the Exposition, with the support of the Cin6-Club de France and the literary magazine, Les Cahiers du mois, Tedesco hosted a series of lectures and screenings on the "Creation of a world through the cinema"-each Sat- urday afternoon, at 4:30, from 28 November 1925 to 20 February 1926.12 The speakers included writers Pierre Mac Orlan and Pierre Hamp, critics Lionel Landry and Tedesco himself, filmmakers Dulac and Epstein, and the actor Charles Dullin. For Pierre Hamp's lecture, Jean Grmillon even pro- jected a short film, Photoginie mcanique, comprised of excerpts from his docu- mentaries, apparently edited in the manner of Epstein's Photognies and L6ger's Ballet mcanique.13 This lecture series provided the Vieux-Colombier with a good deal of prestige. Futhermore, the opening film in Tedesco's second season of programs played upon the interest generated by the series and was an im- mediate sell-out. The film was Dmitri Kirsanoffs Menilmontant (originally en- titled Les Cent Pas), which the commercial distributors had offered to him because they did not know what to do with it.14 Its success was assured, in part, by Vuillermoz's enthusiastic one-column review in Le Temps and by its co-billing with Chaplin's The Pilgrim (which Cinia-Cine-pour-tous readers had voted the second best film of 1925).15 Finally, Tedesco added to his schedule a Monday matinee screening for students in the Latin Quarter. 16 This, too, proved a success, and the students became passionate defenders of the Vieux- Col. Inspired by Tedesco but independent of him, another cin6phile nurtured the dream of opening a specialized cinema to show films unlike the others.' Armand Tallier was a well-known actor who had worked with Copeau at the 258 THCATRE DU VIEUX-COLOMBERA s Ii recio :JAN I!FDESCO 1 S~ S~t5 ANNI C rxu e -_IRllit I Tous lea soirs i 21 heurea Le Dimanche en matinee a 15 heurea Du Vendredi 26 Mars au Jeudi 8 Avril inclus LES CHASSES AUX GRANDS FAUVES Fl documenaire de MARTIN JOHNSON Repreeationa excp--nne16 du Cycle d&, NIBELUNGEN la Vengearice de Kriembild (Les Nibelungen: 2° panic), - Sure de LA MORT DE SI EGF RI ED Du Vendredi 9 au Jeudi 14 Avr lit FILNS A TIOVES 4 6F A - S , - L . o i. r , r 138. (left) Advertisement for the Th6ratre du Vieux- Colormbier 139. (right) Armand Tallier in Jocelyn (1922) Vieux-Colombier and who, with Myrga, his constant companion, had starred in several Gaumont films, the most important being Poirier's Jocelyn (1922). In the summer of 1925, the two found a small (300-seat) cinema for rent behind the Pantheon and prepared to finance a season of film programs with what they called "a light truck, . . . some money left over from an African tour, certain trinkets which would fetch a little money on the flea market, and . . . for the rest, admission tickets. . . ."N On 21 January 1926, the Studio des Ursulines opened its doors with this profession of faith: We propose to enlist our audience from the elite of writers, artists, and intellectuals in the Latin Quarter and from those, ever increasing in number, whom the poverty of commercial film production has driven from the cinemas. Our programs will be composed of diverse tastes, styles, and schools: anything which represents originality, value, effort will find a place on our screen." Like the Theitre du Vieux-Colombier, the Studio des Ursulines had a se- cluded chamber orchestra of just a few instruments.11 But the new cinema was quite different in its programming: "twenty minutes of prewar cinema, twenty minutes of avant-garde cinema, and an unreleased film of a more accessible 259 ALTERNATE CINEMA NETWORK Proliferation and Crisis character and aesthetic."21 Tallier and Myrga's first program included L6once Perret's comic Mimosa la derniere grisette (1906), Rene Clair's reedited version of Entr'acte (1924), and G. W. Pabst's The Joyless Street (with Asta Nielsen and Greta Garbo). The latter film was an immediate sensation, especially after Tallier provoked the audience with an inserted intertitle-"The images that you should see here have been cut by order of the censors"-which forced the attending minister of education to beat a hasty retreat from the cinema.22 Two months later, Tallier's second program revived Clair's Le Voyage imaginaire (1926) and then a third presented the first Japanese film in France, Musum6, along with Claude Autant-Lara's short, Fait-Divers (1924).23 In just a few months the Studio des Ursulines was astonishingly " la mode," and the cin6- philes were rubbing elbows with the young nouveaux riches. By 1925 or 1926, a network of film critics, cinema journals, cine-club lectures and exhibitions, and specialized cinemas was well established in Paris. This loose cooperative system of cultural exchange and persuasion had staked out several areas of concern. It was promoting the production of formally radical, modern films by its own members (Delluc, Gance, L'Herbier, Dulac, Epstein, Feyder, Clair), both within the French film industry and on its mar- gins. As a corollary, it was educating and expanding the audience for such films. It was also articulating key issues in the attempt to define an aesthetics of film and produce a sophisticated critical practice. And it was taking the first steps to preserve older films and establish a tradition of cinema art. The last five years of the decade were marked by a proliferation of these efforts and a number of significant crises. In 1925, several books were published in Paris which clearly revealed the divisions that had developed between the dominant industry and the various advocates of an alternate cinema. Against G.-M. Coissac's Histoire du cinma- tographie, which surveyed the technological and industrial evolution of the film industry, Lon Moussinac's Naissance du cinma contended that a new narrative art had been created.' In a polemical but clearly reasoned analysis, Moussinac summed up the argument for a film aesthetics based on a conjunction of plastic and rhythmic elements (the mise-en-scene, framing, and editing of a cinematic discourse) and then sketched the stages of development in each of the major national cinemas (American, Swedish, German, and French). And like Delluc before him, he reiterated the belief that "the cinema will be popular or it will not be at all."2 Against the polemical essays advocating a "pure cinema," published principally in Cinda-Cin6-pour-tous, Henri Fescourt and Jean-Louis Bouquet's L'Ide et l'dcran: opinions sur le cinma (1925-1926) defended an up- dated and more sophisticated concept of the commercial narrative film.3 Al- though admitting the value of lighting and rhythm, for instance, Fescourt and Bouquet contended that the principal basis of film art was its subject, meaning the logical development of a story.4 Along with the publicity gen- erated by the film exhibitions at the Musee Galliera and the Exposition des Arts Dcoratifs, these books stimulated the publication of other works over the next several years. In 1926, Les Ecrivains reunis alone published two books on the cinema. In one, a slim volume entitled Le Cinematographe vu d'Etna, Jean Epstein collected 260 several of his lectures together with a new theoretical essay. In the title essay, Epstein developed a concept of the cinema apparatus as a machine of the Imagination, producing its own epiphanies of revelation-"one of the most powerful forces of the cinema is its animism."5 In another, however, he called for "a new avant-garde" and a mode of filmic construction that would direct the polysemic play of associations in a chain of images. That play, as rhetorical figuring, could be controlled through patterns of simultaneity and alternation (or, more generally, through paradigmatic and syntagmatic relations), in what Stuart Liebman has called a "thematic montage of associations. "6 In the second book, L'ABC du cinma, Blaise Cendrars, the idiosyncratic poet who had helped Abel Gance shoot and edit La Roue, finally published the full text of his essay on the modernity of the cinema.7 In his elliptical, telegraphic prose style, Cendrars argued that film's fragmentation of reality gave the viewer an inten- sified experience of the simultaneous flux of life and that the worldwide ex- hibition of a film created a kind of "global village" of simultaneous audience participation. The following year, a Geneva publisher collected many of Ricciotto Canu- do's published and unpublished essays into L'Usine aux images (1927). The most important of these, "R6flexions sur la septieme art," which articulated Canudo's expressive theory of film, seems to have been written in 1921, while the reviews of film genres and individual films that complete the book date from 1922-1923. As the decade drew to a close and the silent cinema to an end, several books attempted to sum up the historical development of the era: Moussinac's own Panoramique du cinma (1929), Alexandre Arnoux's Cinma (1929), Georges Charensol's Panorama du cinma (1929), and Henri Fescourt's Le Cinema des origines d nos jours (1932). There was even more activity among the film journals and literary maga- zines. Several literary magazines, such as La Revue nouvelle, followed Le Mercure de France in establishing film review columns; but the most important phe- nomenon was the increase in special issues, modeled after Le Crapouillot. Les Cahiers du mois put out the most significant of these. In 1925, it devoted two special book-length issues to the cinema: #12 on scenarios, #16/17 on film aesthetics and criticism. The latter included essays-some from the Cin6-Club de France lectures at the Vieux-Colombier-by most of the critics, filmmak- ers, artists, and writers associated with the cin6-club movement. There were major statements by Epstein, L'Herbier, Clair, and Dulac, as well as summary critical pieces by Tedesco, Vuillermoz, Charensol, Henry, and Moussinac. Two other literary magazines came out with important special issues-La Revue fidraliste (November, 1927) and Le Rouge et le noir (July, 1928)-both of which focused more on the cinema itself as an art than on its relation to literature.8 In 1926, Rene Jeanne convinced the Librairie Flix Alcan to pub- lish and market a special collection of essays on the cinema in a serial format, much like the issues of a magazine.9 L'Art cinmatographique (1926-1929) ran to six book-length volumes and collected most of the lectures given at the Vieux-Colombier in 1925-1926, along with additional pieces by Gance, L'Herbier, Vuillermoz, and others. Several are notable as summary or original statements: Dulac's "Les Esth6tiques, les entraves, la cingraphie int6grale," Landry's "Formation de la sensibilite," Moussinac's "Cinema: expression so- ciale," and Andr6 Levinson's "Pour une potique du film." PROLIFERATION AND CRISIS 261 ALTERNATE CINEMA NETWORK 140. The cover of Cinographiue This phenomenon of special issues was paralleled by the emergence of sev- eral specialized, rather deluxe film journals, all of them edited by filmmakers and cin -club leaders. Although none lasted more than a few years, they were important organs for the exchange of ideas on an alternate cinema. Germaine Dulac's Schemas (1927), which appeared but once, was envisioned as a theo- retical journal modeled after Les Cahiers du mois (#16/17). That one issue contained a cross-section of critical debate (and no photographs): Hans Richter and Dulac herself advocating a "pure cinema" analogous to music, Henri Fes- court and Jean-Louis Bouquet again defending the narrative cinema, Dr. Paul Ramain celebrating the "oneiric incoherence" of film, Dr. Commandon ex- plaining the value of scientific films, and Jules Romains rhapsodizing in a poetic epilogue. Jean Dreville edited no less than three film journals over a two-year period: Photo-Cini (1927), Cinegraphie (1927-1928), and On Tourne (1928). Of these, Cinegraphie, which ran only five issues, had the most impact. Printed in a folio-sized format, it was the most deluxe film journal of the decade, reproducing large-scale stills from major films along with Dr ville's own exquisite landscape and portrait photographs. It was also the only film journal to publish excerpts from actual shooting scripts: Dulac's La Folie des vaillants (1925) and Epstein's 6' X 11 (1927).( In Switzerland, Kenneth McPherson edited a monthly English-language film journal, Close Up (1927- 1930), which provided continuing coverage of the avant-garde French cinema through Jean Lenauer. Finally, there was the monthly La Revue du cinema (1928-1931), edited by Jean-Georges Auriol and addressed to the elite clien- tele of Gallimard and the Nouvelle Revue Frangaise.'" Besides the comprehensive film criticism by Paul Gilson, Louis Chavance, Bernard Brunius, and himself, Auriol published several of Robert Desnos's scenarios as well as Bunuel and Dali's scenario for Un Chien andalou (1929)."1 Although large (forty to eighty pages) and rather expensive, La Revue du cinema took consistently leftist polit- ical positions, especially after Janine Bouissounouse and Robert Aron joined the editorial staff. It was the only film journal, for instance, which, following Moussinac, envisioned an alternate cinema in political as well as aesthetic terms. Coincident with the first issue of La Revue du cinema, late in 1928, the field of popular film journals expanded with the appearance of two new rivals to Cinemagazine and Cinea-Cin-pour-tous. The first was Cinemonde, edited by Gas- ton Thierry, a slim twenty-page folio-format weekly. Cinemonde's attention was focused on the major commercial films then in release or in production, and it was filled with publicity photographs, sometimes artfully arranged accord- ing to the whims of an imaginative graphics designer. Although it defended the work of several filmmakers associated with the narrative avant-garde, their work and that of the avant-garde generally received even more attention in the other new magazine, Pour Vous. Pour Vous was edited by Alexandre Ar- noux, who hoped to make it an inexpensive, independent film journal: "In this magazine, our readers will find not a single line of film industry publicity, whether blatant or disguised (as is too often the case)." ' To that end, Pour Vous used a newspaper-size format of only sixteen pages (much like Cin(- miroir). And Arnoux and his writers-e.g., Wahl, Charensol, Roger Regent, Nino Frank, Jean Lenauer, and Lucie Derain-devoted a good deal of space not only to industry matters and the latest commercial films but to the broad 262 range of avant-garde film practice as well as to the exhibition activities of the cine-clubs and specialized cinemas. Although it could not sustain the level of independence that Arnoux initially desired, Pour Vous survived, along with Cinemagazine and Cinemonde, to become the major weekly film journals of the 1930s. For all that Delluc, Canudo, Moussinac, and others had done to create a specialized/popular press devoted to the cinema and, in the words of Harry Alan Potamkin, "a body of critics, as authentic and authoritative as the critics of the other arts,"14 their legal status was less than clearly defined. To the industry, the press served as a kind of publicity department. extension; to the critics, it served as an educational forum, an arena of exchange on aesthetic values and on the social function of the cinema. These conflicting attitudes came to a head in March, 1928, when a state court finally handed down a decision on a suit that Jean Sapene and Cineromans had brought against Mous- sinac for one of his columns in L'Humanit6 (26 September 1926). The charge was that he had maligned a film they were then distributing, Jim le Harponneur [The Sea Beast, starring John Barrymorel, as "the perfect example of a bad American film or, simply, a bad film."15 Incredibly (or perhaps not-given the results of the recent national elections), the court decided against Mous- sinac and accepted Cin6romans's demand of 100,000 francs in damages.16 Moussinac and L'Humanit6 appealed the verdict, but it was not until 12 De- cember 1930 that the Court of Appeals overturned the lower court's decision and, in effect, granted the film critics freedom of expression. 17 After more than a decade of struggle, film critics could finally enjoy the same legal rights as their colleagues in literary, theater, and art criticism. Moussinac found himself at the center of an even more important struggle in the cin6-club movement in 1928. The success of the Cin6-Club de France and the Tribune libre had been phenomenal. Early in 1927, alongside its regular public film screenings, the Cine-Club de France organized, for the College libre des sciences sociales, a series of lectures called "The Cinema in modern life and thought": PROLIFERATION AND CRISIS January 7: January 14: January 21: January 28: February 4: February 11: February 18: February 25: M. Charensol-The state of the cinema M. A. Berge-Literature and cinema M. A. Obey-Music and cinema M. Moussinac-The social expression of the cinema M. Levinson M. Jean Laran-The documentary and instructional cinema M. Rene Clair-The cinematographe vesus intelligence M. M. L'Herbier-The cinematographe as a cosmic medium. 18 A month later, it sponsored a second "Exposition du cinema" for the Galerie d'Art de la Grande Maison de Blanc. 19 And during the commercial exhibition of Gance's Napoleon at the Madeleine-Cin6ma, the club held a special screening of the triptych sequences that had been cut from the film.20 Meanwhile, the Tribune libre opened its third season of weekly public film screenings with Kirsanoff's Menilmontant.21 In April, it sponsored Robert de Jarville's series of avant-garde film screenings and lectures, held on Saturday afternoons at the Theatre de Chateau d'Eau.22 Later that same year, for the Salon d'Automne, 263 ALTERNATE CINEMA NETWORK Jarville also delivered a lecture on "Images du monde" and screened several undistributed Soviet films.23 Finally, for its fourth season, the Tribune libre began with what it called "the most characteristic specimens of each country's production": Feyder's L'Image (France), Lulu Pick's Sylvester (Germany), Sj6- str6m's L'Epreuve du feu (Sweden), and Kuleshov's Dura Lux (Soviet Union).24 All of this activity soon had repercussions throughout France and the rest of Europe. In Paris itself, a host of small cin6-clubs grew up-with names like Club de l'Ecran (Pierre Ramelot), Phare tournant (Raymond Villette), Regards, Le Lanterne magique, L'Effort, Les Spectateurs d'Avant-Garde.23 Outside the capital, as early as 1925, Dr. Paul Ramain had organized the first provincial cine-club at Montpelier.26 Over the next few years, others were started in Agen, Lyon, Reims, Strasbourg, Bordeaux, Chalons, Lille, Tours, Grenoble, and Marseille-Jean Vigo's Amis du cinema, Cine-Club d'Avant- Garde.27 In London, in 1925, there was the Film Society organized by Ivor Montagu and Hugh Miller and run by Iris Barry; in Lausanne and Geneva, in 1926, the "Cin6 d'Art," partly sponsored by the Vieux-Colombier.28 The most active cine-club outside France, yet still associated with the French groups, was in Brussels. Organized by Albert Valentin in 1925, as "Cin6graphie Fri- days" at the Cabinet Maldoror, the Brussels cine-club soon changed its name to "Visions fortis" and held an extensive series of lecture-presentations late in 1926 at the Salle de L'Union Coloniale.29 The list of speakers reads like a directory of the French avant-garde: Dulac, L'Herbier, Epstein, Gance, Poi- rier, Clair, Feyder, Chomette, Kirsanoff, Dr. Comandon, Moussinac, Vuil- lermoz. By 1927, the film critic Carl Vincent, who took over from Valentin, had transformed the group into the "Club du cinema," which held weekly screenings at Sever House both Friday and Saturday evenings.30 By the end of the decade, the cin6-club movement had proliferated to such an extent that it climaxed in two major international events. In September, 1929, Robert Aron and Janine Bouissounouse organized a ten-day Interna- tional Congress on the Independent Cinema at the Chateau de La Sarraz in Switzerland.31 The congress gathered together filmmakers and critics from France (Aron, Bouissounouse, Auriol, Cavalcanti, and Moussinac), Germany (Hans Richter and Walter Ruttman), England (Ivor Montagu and Isaacs), Austria (Bela Balizs and Fritz Rosenfeld), Holland (H. K. Franken), Italy (Enrico Prampolini and Alberto Sartoris), Spain (Gimenez Caballero), Switz- erland (Robert Guye, Arnold Kohler, Georg Schmidt, Alfred Masset, and Jean Lenauer), the United States (Montgomery Evans), and the Soviet Union (Sergei Eisenstein, Edward Tiss6, and Gregory Alexandrov). Two months later, in Paris, Germaine Dulac, Charles Leger, and Robert de Jarville organized a Congress of Cin6-Clubs, out of which emerged the first international Fdera- tion des Cin6-Clubs.31 Despite this expansion, the cine-club movement remained basically elitist, appealing to a restricted number of artists, intellectuals, cinephiles, and (to use an unflattering label from the period) "boisterous snobs."3 But there was one exception-"Les Amis de Spartacus." Two events in 1926 and 1927 finally led Moussinac and several of his Communist friends to wed their interest in an alternate cinema with their political ideology. On 13 November 1926, at the Artistic-Cin6ma, Moussinac and Dulac arranged a Cin6-Club de France screening of the banned Eisenstein film, The Battleship Potemkin (1925).34 It 264 was one of the most successful of all their programs and aroused a keen interest among some Parisians to see more Soviet films. Then, in 1927, a workers' cooperative took over the Bellevilloise, a popular cinema in the 20th arron- dissement, to screen "exceptional films considered 'uncommercial' by the major- ity of cinema owners."3 By the early 1930s, the Bellevilloise would become an important outlet for workers' newsreels.36 Its early success convinced Mous- sinac to attempt a similar venture in the form of a cine-club. In July, 1927, along with Jean Lods, Francis Jourdain, Paul Vaillant-Cour- turier, and Georges Marrane, Moussinac organized the "Club des Amis de Spartacus" with the purpose of creating a mass cinema movement.37 Their first step was to acquire the Cinema du Casino de Grenelle, the largest cinema in the 15th arrondissement.38 The second step was to arrange for the rights to project a series of provocative films. The third was to publicize their princi- ples: To the public that loves and understands the cinema, that foresees its destiny, there remains only a single means of battling this dictatorship of money: to band together. Henceforth, it is the purpose of "Amis de Spartacus," through the organizaiton of restricted screenings, to assure the distribution of major works of the French, German, American, and Soviet cinema. The future of the cinmatographe is entirely in the hands of the public. It is indispensable that every film advocate work against commercial publicity, against French protectionism, against American colonization.39 The cinema would be conceived, not as an end in itself, but as a means of combat and of social liberation. Finally, on 15 March 1928, the Casino de Grenelle cinema opened with the first of six months of weekly programs. Four thousand people reportedly showed up to vie for the 2,000 available seats.40 "We couldn't believe our eyes," said Jean Lods, "when we found ourselves with a few thousand members after several weeks; we didn't know what to do."41 One strategy they tried was scheduling simultaneous screenings at the Bellevilloise as well as at the Casino de Grenelle. Within three months, membership in Les Amis de Spar- tacus had swelled to at least ten thousand.42 Soon the club was organizing efforts in the suburbs and provinces until, within another couple of months, the membership had quadrupled.43 Just what films was the group exhibiting that could cause this phenomenon? Older film classics, such as Delluc's Fievre and Stiller's Trisor d'Arne, recent popular films, such as Flaherty's Moana and the Swedish film, Charles XII, and, most important of all, three banned Soviet films-Eisenstein's Potemkin and Pudovkin's Mother (1926) and The End of St. Petersburg (1927).44 According to Jean Lods, as many as twenty-five to thirty thousand people saw Potemkin alone.45 What was the group planning for its fall season in 1928? Series of French, American, and German films, and, in honor of the tenth anniversary of the Soviet revolution, Eisenstein's new film, October (1927).6 Les Amis de Spartacus was the one overtly political organization in the French cin6-club movement of the 1920s. And it was the only organization to come close to making that movement a mass movement, to lay the ground- work for a truly popular alternate cinema which could confront the dominant industry with any degree of force. As long as the cin6-clubs and specialized PROLIFERATION AND CRISIS 265 ALTERNATE CINEMA NETWORK cinemas were restricted to small audiences, the industry could ignore them or even offer mild encouragement. But when a club could compete on a par with the largest commercial cinemas in Paris, that was another matter. Les Amis de Spartacus was also making a mockery of the government practice of per- mitting officially censored films to be projected privately in the cine-clubs but forbidding them in the commercial cinemas.47 That most of these films were Soviet in origin made them a highly visible target of the political attacks on French Communists that reached a peak in 1928. In September, Jean Chiappe, the dapper, devious Paris chief of police, called Jean Lods into his office and told him the Spartacus group was on a collision course with his duty to protect the public order. He even threatened to have his agents (who had infiltrated the group) disrupt any further Soviet film screenings at either the Casino de Grenelle or the Bellevilloise.48 Given Chiappes power, Moussinac and Lods had no legal recourse. By October, Les Amis de Spartacus was disbanded, and the energy that had gone into the organization dispersed among the other film groups and into other political activities. Although the Spartacus group never reached the point of engaging in film production, the Soviet films it exhibited had a decided impact on a number of French avant-garde filmmakers. Another crisis of sorts occurred among the specialized cinemas as they ex- panded in number and scope to the point of threatening the apparent unity of the alternate cinema network. Jean Tedesco steadily continued to build up a film repertory at the Vieux-Colombier and promoted his effort in the pages of Cinia-Cind-pour-tous.49 For the most part, his programming consisted of re- screening films, whether successful (e.g., L'Herbier's Feu Mathias Pascal, Lang's Siegfried) or unsuccessful (e.g., Murnau's The Last Laugh, initially booed at the Aubert-Palace),50 which had contributed to the development of film art. His choice of films struck a fairly equal balance among the American, German, and French cinemas. In the second and third seasons, however, Tedesco began to introduce new films as well (e.g., Menilmontant, Cavalcanti's En rade).51 Most of these were documentaries-by Sauvage, Boge, Chaumel, and Jean Grmillon-a preference that soon distinguished the Vieux-Colombier from the other specialized cinemas in Paris.52 That preference seems to have coin- cided with, and possibly fed, a growing public interest in documentary films- e.g., Poirier's La Croisiere noire at the Salle Marivaux (April, 1926), Cooper and Schoedsack's Grass at the Madeleine (June, 1926), and Flaherty's Moana at the Electric (August, 1926).53 In November, 1926, Tedesco offered a spe- cial screening of Moana, which had to be rescheduled twice to meet audience demands.54 In early 1927, he presented Andre Gide and Marc Allegret's film on central Africa, Voyage au Congo, and then Gr6millon's Tour au large, a lyrical celebration of trawler fishing off the Brittany coast.55 For the latter, Gremillon himself constructed an experimental sound synchronization system using Ple- yela player piano rolls. The popularity of the Vieux-Colombier persuaded Te- desco, early in 1927, to lease the first floor of a building on the Right Bank of Paris, the Pavillon de Hanovre. There, at the Pavillon du cin6ma, until the summer of 1928, he could screen films that had been programmed at the Vieux-Colombier one or two weeks before.56 Tedesco's example and the expanding activity of the Tribune libre led to the transformation of a small cinema near the Mouffetard area in the same year 266 the Pavillon opened. This was the 300-seat Cine-Latin, owned by Jose Miguel Duran, whose model for programming strategy was Charles LUger.57 Much like the Vieux-Colombier, the Cin6-Latin specialized in repertory programs, but its emphasis was on the French cinema. In February, 1928, for instance, following the exclusive run of Napoleon, Duran rescreened Gance's early war epic, J'Accuse.58 In March, he devoted several weeks to a retrospective of films by Rene Clair; in June, he persuaded Marcel L'Herbier to project, along with L'Inhumaine, his incomplete version of Rdsurrection.59 The Cine-Latin is even credited with rediscovering, and so preserving, Louis Delluc's La Femme de nulle part and L'Inondation as well as Max Linder's L'Etroit Mousquetaire.60 Just as did the Tribune libre, Duran scheduled discussions, every Thursday eve- ning, for the spectators to debate the films shown during the past week.61 The audience of the Cin6-Latin guaranteed a lively dialogue, for it was comprised equally of students and working-class people from the neighborhood. Paul de la Borie reports on one of the summer programs: Seated side by side, intellectuals and workers fraternize in front of the screen. I cannot say that the films they are watching produce the same impression on one and the other. They are not, evidently, in the same state of receptivity. The comic film pre- ceding the piece de resistance provokes among the common people of the audience a hilarious outburst in which the university students do not share. Instead, the students amuse themselves over the naive delight of their neighbors. But the second part of the program begins, and soon there is a silence that is absolute. The entire audience follows the unfolding of the images and the development of ideas with unfailing attention. But, in general, the common people applaud very little. They are trying too hard to understand. And the fact is that they do understand for they don't get annoyed. . . . And they come back for more.61 Much like the Bellevilloise cinema across Paris, then, the Cin6-Latin served as a precedent for Moussinac's Les Amis de Spartacus. The Studio des Ursulines flourished even more than did its two companion cinemas. By late 1927, Tallier told the Cinimatographie franfaise, his cinema had drawn 23,000 to 25,000 people to its programs.63 His programming format of prewar films, avant-garde films, and unreleased films distinguished it from both the Vieux-Colombier and the Cine-Latin. And Tallier's choice of newly released German and American films, along with French experimental shorts, catered to the snobs of his audience. Ironically, that snobism affirmed the very status quo the alternate cinema movement was trying to combat: that the French cinema merely existed on the margins of the American and German cinema. An unspoken rivalry seems to have arisen between Tallier and Te- desco, perhaps because the latter saw himself as a founder or father confronted by a usurper.64 The rivalry is implicit in Cina-Cine-pour-tous's relative neglect of the Studio des Ursulines in its pages and perhaps explicit in Tedesco's abandonment of a joint exhibition of G. W. Pabst's Secrets of a Soul (1926), planned for May, 1927.65 In its second and, especially, its third season, the Studio des Ursulines shifted its programming to include more and more new French films-but with no letup in the rivalry. In the fall of 1926, Tailler had countered Tedesco's screening of Moana by being the first in Paris to exhibit Cavalcanti's semi-documentary, Rien que les heures (1926).66 Late in 1927, he countered Le Voyage au Congo and Tour au large at the Vieux-Co- lombier with Leon Poirier's new documentary on Africa, Amours exotiques. That PROLIFERATION AND CRISIS 267 ALTERNATE CINEMA NETWORK same year, Tallier scored another coup by offering en exclusivit6 two of Jean Epstein's independently produced films, 6 2X 11 and La Glace d trois faces. The latter had an exceptionally successful run, for a French film, of nearly three months.67 What really made the Ursulines distinctive, however, was short avant-garde films. It was the single Paris cinema to consistently offer an outlet for young independent filmmakers. By exhibiting Clair's Entr'acte, Autant-Lara's Fait- Divers, and LUger's Ballet mdcanique on its earliest programs, the Ursulines built up a loose corpus of short experimental films which were first projected pub- licly on its screen. Besides those mentioned, they included Chomette's Jeux des reflets et de la vitesse (1925) and Cinq Minutes du cinema pur (1926), Du- champ's Anemic cinema (1926), Cavalcanti's La P'tite Lily (1927), Clair's La Tour (1928), and Man Ray's Emak Bakia (1927) and L'Etoile de mer (1928).68 Tallier even made room in this corpus for several short documentaries: Georges Lacombe's striking study of the ragpickers of the Clignancourt flea market, La Zone (1928), Marcel Carn6's Nogent, Eldorado du dimanche (1930) and Jean Vigo's A propos de Nice (1930).69 At least one of his selections, however, em- broiled the Ursulines in considerable controversy. This was La Coquille et le clergyman, which Cin6-Club de France had screened privately as early as Oc- tober, 1927.70 Thinking that Antonin Artaud (the author of the scenario) had denounced Germaine Dulac's production of the film, the Surrealists provoked a near-riot when the film was shown in February, 1928.1 The so-called scan- dal demeaned La Coquille et le clergyman at the time, but so marked it histor- ically as to create continued interest in it. During the last two years of the decade, at least four more specialized cinemas sprang up in Paris. The popularity of the Ursulines's programming convinced a M. Querel, late in 1928, to open another little cinema, the 550- seat Salle des Agriculteurs.72 According to Jean Dreville, Autour de l'argent, his short documentary (and montage study) on the making of L'Herbier's L'Argent (1929), was the main attraction of the inaugural program-even more than Ralph Ince's Shanghaied (1927).73 For its first half year, the Agriculteurs had a shaky existence. La Revue du cinma alternately praised and damned its programming-they especially supported an early festival of Cavalcanti films.74 In May, 1929, Tallier and Myrga announced that they were going to take over management of the cinema, and for the next six months or so they seem to have run the Agriculteurs successfully as a repertory cinema.75 Even more successful than the Agriculteurs was Jean Vallke's L'Oeil de Paris, located near the Place de l'Etoile. Although less luxurious than some of the other cinemas, L'Oeil de Paris had a chic Argentinian decor and was conveniently inexpen- sive.76 Its inaugural program, in May, 1929, was modeled on the Ursulines- Dulac's Arabesques, Robert Florey's Life and Death of a Hollywood Extra, ex- cerpts from the old serial Les Mysteres de New-York, and Jean Epstein's new film, Finis Terrae.77 Within months, L'Oeil de Paris had become the meeting place for the biweekly-sessions of the Club de l'Ecran.78 The only specialized cinema to fail during this period was the Studio Diamant. Ever the opportun- ist, Henri Diamant-Berger conceived this cinema as part of an ambitious art center for the avant-garde.79 All he completed, however, was the 160-seat cinema and an adjacent bar. Perhaps because of his former position in the commercial industry, his rather unfocused programming included very few 268 French films.80 Although Diamant-Berger says that his mismanagement of the bar caused him to close the Studio Diamant, Harry Alan Potamkin suggests another reason (besides the programming): it was so designed (with steel stalls, a leveled or stairlike ceiling, and a distant screen) as to distract and trouble one's vision.81 The last major specialized cinema was Jean Mauclaire's Studio 28. Mauclaire opened his 337-seat cinema in January, 1928, not on the Left Bank, but halfway up the hill of Montmartre.82 His programming format aligned the Studio 28 closely with the Ursulines: unreleased or newly released features plus short prewar and avant-garde films. The inaugural program he put to- gether-a documentary on the making of Gance's Napolion and a new Soviet film, Abram Room's Bed and Sofa (1927)-was so successful that it ran for five weeks.83 That summer, Studio 28 featured Epstein's new film, La Chute de la maison Usher, en excusiviti, with similar success.84 Like the L'Oeil de Paris, it, too, became a favorite meeting place for several cin6-clubs.85 From the beginning, Mauclaire's ambition was to equal or surpass the prestige of the Vieux-Colombier and the Ursulines. To that end, according to Potamkin, before each program he projected color slides on the cinema's silver-toned walls and then accompanied each film with "an arranged selection of music from a mechanical piano. "86 Within a short time, Studio 28 was acknowledged as the most attractive and "a la mode" small cinema in Paris. On one occasion, however, Maucliare's programming actually endangered his own cinema. In April, 1929, at the Ursulines, the Surrealists had arranged a private screening of the unknown Un Chien andalou, along with Man Ray's Les Mysteres du chdteau du Di.87 Initially skeptical, they came away enthusiastic advocates of the Bunuel-Dali film. When Mauclaire agreed to premiere Un Chien andalou at Studio 28 the following autumn, Andre Breton decided to make it a gala event, with a foyer exhibition of paintings by Masson, Mir6, Ernst, Man Ray, and others.88 Rumor and the usual polemical publicity led to an ugly confrontation at the early November premiere. Members of the right-wing Jeunesse patriote and Camelots du Roi stormed into the cinema, attacked the film, and retreated, slashing some of the paintings. This scandal assured Un Chien andalou's success, and for nine months at Studio 28 it played off and on to capacity crowds whose responses oscillated wildly from outrage to enthusiasm and hysterical laughter.89 Cyril Connolly, for instance, reports that The picture was received with shouts and boos and when a pale young man tried to make a speech, hats and sticks were flung at the screen. In one corner a woman was chanting, "Salopes, salopes, salopes!" and soon the audience began to join in.90 As the cine-clubs and specialized cinemas multiplied in the second half of the decade, there was a reciprocal increase in the demand for avant-garde films. The problem was how could the production and distribution of more and more French films be financed? The production of a short silent film was not all that expensive--Carn's Nogent, Eldorado du dimanche (1929) reportedly cost but 5,000 francs-but that was beyond the means of most avant-garde film- makers.91 One source of capital was a wealthy patron whose interest in the cinema made him susceptible to dabbling in the new art. Rolf de Mare, for instance, financed Clair's Entr'acte as an interlude for one of the 1924 Ballet PROLIFERATION AND CRISIS 269 ALTERNATE CINEMA NETWORK suddois programs at his Theatre de Champs-Elys~es. To compete with de Mare, the following year, the Comte de Beaumont asked Henri Chomette to make several short films for his famous ballet productions, "Soires de Paris."92 Perhaps through his indirect involvement in L'Herbier's Cingraphic com- pany, the Vicomte de Noailles became enamored of the cinema and hit upon the idea of producing one film each year as a gift to his wife: Man Ray's Les Mysteres du chteau du De (1929), Luis Bufiuel's L'Age d'or (1930), and Jean Cocteau's Le Sang d'un pote (1930-1932).93 But the whole French production of avant-garde films could hardly depend on the whim and largess of a couple of latter-day Maecenases. In contrast to Germany, where both UFA and the state helped finance short avant-garde films,94 capital was not forthcoming from the French commercial film industry, especially after the tight money conditions of 1925. Perhaps inevitably, the specialized cinemas began to venture into film production. As early as 1926, Jean Tedesco turned the loft above his cinema into the "Labo- ratoire du Vieux-Colombier" for the production of short scientific films-e.g., Etoiles et fleurs de mer, Papillons et chrysalides.95 A year later, Jean Renoir's pro- duction team joined Tedesco to expand the laboratory's meager facilities for their fantasy film, based on a Hans Christian Andersen fable, La Petite Mar- chande d'allumettes (1928).96 The atelier now had its own electrical generator, specially built reflector lamps (for panchromatic film stock), wooden tanks and an old Paths camera (for processing the negative and making the initial posi- tive prints).97 After more than six months of work, La Petite Marchande d'al- lumettes was projected in the very place of its creation, in June, 1928.98 Un- fortunately, the film's exhibition and distribution was halted almost immediately by Maurice Rostand (the author of a 1914 operatic adaptation of the same fable), who charged Renoir and Tedesco with plagiarism.99 Although the film- makers won their case, it was in litigation for more than a year; and Tedesco was unable to develop the Vieux-Colombier laboratory into the major produc- tion facility he had envisioned.l100 Still, it was used occasionally, at least until 1930, when the Hungarian animator, Berthold Bartosch, worked there on his film version of a book of woodcuts, Une Ide (1934).101 Besides these few studio films, Tedesco also seems to have helped finance Kirsanoffs short experimental sound film, Brumes d'automne (1928), and Epstein's short documentary, Le Pas de la mule (1930).102 The Ursulines and L'Oeil de Paris also sponsored several short films, notably the abstract experiments of Germaine Dulac-Themes et variations (1928) and Arabesques (1929). But the most assiduous practicioner of independent film production financing was Jean Mauclaire of Studio 28. Even before he opened his cinema on Montmartre, Mauclaire had signed a contract with Abel Gance that gave him exclusive exhibition rights to several short triptych films. 103 In May, 1928, Dr6ville identified three of these works (which have since disap- peared) as Galop, Danses and Marine-the last of which, he said, was "with Tour au large, the most beautiful poem that has been created about the sea."104 During the next two years, Mauclaire provided financing to several young filmmakers for their first films: Henri d'Arche and Georges Hugnet's La Perle (1929), Lucie Derain's Harmonies de Paris (1929), Eugene Deslaw's Marche des machines (1928) and Nuits 6lectriques (1929), and A. Sandy's Lumieres et ombres 270 (1929) and Pritexte (1929). According to one source, he also partially under- wrote Epstein's La Chute de la maison Usher as well as his last silent film, Sa Tete (1929).105 Just as they were forced into film production, the Vieux-Colombier and Ursulines both found themselves involved very quickly in film distribution. By 1926, Tedesco was shipping films off to Switzerland and Belgium to pro- vide current programs for outpost cinemas and cin6-clubs. Tallier soon fol- lowed his example, and, by the summer of 1928, he was advertising that the Ursulines could supply a program of films, a projector, and a projectionist to any provincial cinema on payment of costs and a moderate fee. 106 The two primary distribution agencies for young avant-garde filmmakers, however, were set up independently of Tedesco and Tallier. One was the "Cooperative du Film," directed by Robert Aron, a newly appointed editor for La Revue du cinima.107 The other was run by Pierre Braunberger. Braunberger's production company, Neo-Film, had already financed short films by Renoir and Caval- canti-e.g., Charleston (1926) and La P'tite Lily (1927)-as well as the latter's feature film, En rade (1927).108 In May, 1929, he announced the formation of Studio-Film to distribute and sell "all the films of artistic quality (experimen- tal films, documentary films, films called 'Avant-Garde')."109 In just a few months, Braunberger had compiled a list of available films that summed up much of the current French and foreign avant-garde film practice: A. Cavalcanti-En rade, Rien que les heures, La P'tite Lily. Man Ray-L'Etoile de mer, Emak Bakia, Les Mystires du chteau du D. E. Deslav-La Nuit ilectrique, La marche des machines, Montparnasse. D. Kirsanoff-Brumes d'automne. E. Deslav-La Nuit lectrique, La marche des Machines, Montparnasse. L. Bufuel-Le Chien andalou. G. Lacombe-La Zone. Rene Clair-Le Voyage imaginaire. Henri d'Arche-La Perle. J. Renoir-La Fille de l'eau. C. Lambert-Voici Paris, Voici Marseille. J. de Casembroot-Ernest et Amilie. R. Landau-Rhythmes d'une cathdrale. S. Silka-Le Mdle mort de Canart. Duhamel-Paris express. 0. Blakeston-I Do Love To Be beside the Seaside. P. Sichel-Bithulite. C. Heymann-Vie heureuse. Gide et Allegret-Voyage au Congo. Allegret-Les Troglodytes. V. Blum-Wasser. A. Strasser-Partie de campagne. Robert Florey-Symphonie des gratte-ciels. Kenneth Macpherson-Monkeys, MoQn. Michel Gorel-Bateaux parisiens. Pierre Chenal-Un Coup de d. C. Autant-Lara--Construire un feu. Jean Lods et Boris Kaufmann-Aujourd'hui."0 PROLIFERATION AND CRISIS 271 ALTERNATE CINEMA NETWORK Preservation Even Path6-Cin6ma inadvertently became part of this alternate cinema distri- bution circuit through its sale of 9.5mm copies of films to collectors and the smaller cine-clubs."' Unfortunately, this distribution effort was quickly un- dermined and diverted by the sound revolution sweeping the film industry. The change to sound films in France, from 1929 to 1930, had a disrupting effect on two crucial areas of the alternate cinema network-production and exhibition. The proliferation of independent avant-garde film production was halted and reversed by the expense and inaccessibility of material and equip- ment. Only a few filmmakers were able to work briefly-and independently- in sound: Bunuel's L'Age d'or, screened at Studio 28 for several weeks before it was banned on 11 December 1930; Jean Cocteau's Le Sang d'un pote, made in 1930, but not screened until 20 January 1932, at the Vieux-Colombier; and Jean Vigo's banned Zero de conduite (1933), which was screened only once publicly at the Artistic cinema. Epstein tried to maintain his independence with a little-known three-reel semi-documentary on Brittany fisherman, Mor- Vran (1930), but he had no control over its musical soundtrack. Most film- makers were soon forced to join the dominant industry to work on commercial feature films (L'Herbier and Gremillon at Natan-Path6, Clair and Chomette at Tobis-Klangfilm, Cavalcanti at Paramount, Renoir at Braunberger-Richeb6) or newsreels (Dulac for Gaumont) or singing interlude films (Epstein for Syn- chro-Cine). Of the specialized cinemas, only two survived beyond 1930. Te- desco sonorized the Vieux-Colombier with an eccentric system, Equipment Synchronista, and kept the cinema open until 1932 with his repertory films plus a few new foreign releases.2 Tallier also sonorized the Ursulines in 1930 and was lucky enough to premiere Sternberg's The Blue Angel (with Marlene Dietrich), which played for a record fourteen months.3 When his staple of German films dried up by 1934, the Ursulines reverted to a repertory program that alternated with the usual commercial releases. The cin6-clubs also changed their orientation. Where before they had fo- cused on educating film viewers and supporting independent filmmakers, they now became sanctuaries for the masterpieces of the Belle Epoque of silence. For some, a period of nostalgia set in. For others, more serious, the impulse to establish a heritage of cinema art and to preserve a body of films from the economic cycle of production and destruction (negative film stock melted down for chemical extractions) or from blind neglect (negative and positive prints thrown away or stored haphazardly and carelessly) suddenly became para- mount. A precedent of sorts had been set by the educator, Victor Perret, in establishing an educational cinema library, the Cinematheque de la ville de Paris, in 1925.4 The repertory film concept of Tedesco, Lger, Clouzot, and others had yielded a small collection of films of widely varying conditions. Cin6philes such as Jean Mitry and Jean Mauclaire agitated among filmmakers and critics for a national cin matheque, not for educational, technical, or military purposes, but to honor the cinema itself.5 As evidence of his concern, Mauclaire managed to organize a retrospective of Louis Delluc's films (April, 1929) and, with Jean-Georges Auriol of La Revue du cinma, sponsored a special program devoted to Georges M6lies, with eight newly reconstructed, tinted prints (16 December 1929).6 In his book, Panoramique de cinma (1929), Mous- 272 sinac joined in to argue at length for an international bibliothque and cinma- PRESERVATION thque.' According to Raymond Borde, however, it was not until 1933 that a Cine- matheque nationale was established and installed in the old Trocadero palace.8 The impetus came not only from cin6-club lobbying (Maurice Bessy, for in- stance) but from the public outcry in the press over the loss of films. Mous- sinac recounts the story well. A number of cinema owners, in effect, faced with the lamentable mediocrity of the talkies available to them, still resist reequipping their cinemas and, in order to offer "silent" programs, specifically seek out quality flms released during the past five or six years. Their interest leads them to compose a kind of program sharply different from that of their competitors and thus serves to maintain an existing clientele. Certain other cinema owners, even though equipped for sound, continue to screen silent films from time to time, preferring to schedule an interesting nontalkie rather than an imbecilic talkie. Such actions cannot agree with the interests of the large companies which need to circulate their production schedules through the greatest possible number of cinemas in order to write off, at least, the cost of the films. Hence, the recent offensive which translates itself into the sabotage of the good silent films that still exist. We have already mentioned that, for a great number of these films, the distributors are resisting any rentals and that, if a rental is allowed, the prints provided are so scratched, spliced, and deteriorated, that a public screening provokes the ire of spec- tators. In that way, the masters of the international film industry hope they can force all the "rebellious" cinema owners-if they want to avoid closing down-to screen "their" programs of sound films, however mediocre they may be. They are going so far-and even the bourgeois press is scandalized, for appearance's sake-to destroy certain works which truly deserve preservation for the history of the cinema.9 Unfortunately, the Cinmatheque nationale proved little more than a name since it lacked any funding. To rectify its inaction, two years later, Mitry, Georges Franju, and Henri Langlois organized the Cinematheque fran aise under the nominal direction of Paul-Auguste Harle, a top administrator of La Cinematographie franfaise.10 Thus did the cin6-clubs and the few cinephiles who collected films, along with key industry figures such as Alexandre Kamenka and Germaine Dulac as well as second-hand film distributors, become the primary sources for what has been preserved of the French silent cinema art. The French movement to establish an alternate cinema in the 1920s seems to have gone through three fairly distinct stages. The first can be marked off by Louis Delluc's assumption of the editorship of Le Film in 1917 and by the deaths of Canudo and Delluc, respectively, in 1923 and 1924. Those six or seven years saw the founding of independent film journals, regular newspaper film review columns, and the earliest cine-club organizations. From the writ- ings and conversations of Delluc, Canudo, Moussinac, Vuillermoz, and others emerged the theory and praxis of an embryonic film criticism and the concep- tualization of an alternate cinema. The second period can be marked off from 1924 to 1926 or 1927. For two or three years, a loose network of cin-clubs, specialized cinemas, and film journals seemed to present a united front in actively promoting an alternate cinema. Popular exhibitions at the Musee Gal- 273 ALTERNATE CINEMA NETWORK hera and the Exposition des Arts Decoratifs, lectures at the Vieux-Colombier, and regular public screenings by the Cine-Club de France and the Tribune libre-all these activities encouraged independent filmmakers in their work, built up a permanent audience, and began to articulate a history of cinema art. The third and final period can be marked off from 1927 to 1930. In these three or four years, the proliferation of organizations and the intensification of activity had two major consequences. With the founding of Les Amis de Spartacus, the cooperative system supporting an alternate cinema verged on becoming a mass political/socioeconomic movement. The threat was grave enough to the dominant industry and the state that the Paris police reacted quickly to dissolve and disperse its force. Simultaneously, the alternate exhi- bition system established by the cin6-clubs and specialized cinemas expanded to include film production and distribution. This, too, was swiftly curtailed by the industry's transformation to sound films. Stymied by political and economic conditions, what was left of the alternate cinema system redirected its energies into the cinematheque movement and into limited subversion from within the industry itself-e.g., the films of Clair, Renoir, Dulac, Vigo, and Jacques Pr6vert. This history of an alternate cinema structure in France also reveals certain internal contradictions or oppositions which the movement failed to resolve. What was conceived as a mass movement, a popular cinema, consistently functioned as an elitist operation. Despite Delluc and Canudo's expressed wishes, their journals and clubs (and Delluc's films, for that matter) attracted only a small cultured audience. L6ger's Tribune libre and Moussinac's Les Amis de Spartacus came closest to fulfilling those initial dreams, but their work was outweighed by the elite-oriented programs of the Ursulines, Studio 28, and even the Vieux-Colombier. The " la mode" audiences of the Ursulines and Studio 28, especially, led to repeated charges that the French avant-garde merely provided excitement and entertainment for snobs. The elitism of the alternate cinema structure also depended in part on its overriding interest in aesthetic and even epistemological questions raised by the new medium of film. The debates that animated the pages of Cinda-Cine-pour-tous, Les Cahiers dA mois, Cinigraphie, and others often revolved around the possibility of a "pure cinema" or a non-representational cinema, the definition of cinematic specific- ity and cinematic rhythm, the relationship between cinema and the other arts, especially music. At their best, they explored the ways that film could expand human perception and knowledge. Only Moussinac, and Clair to a lesser ex- tent, consistently argued in writing that an alternate cinema had to be con- ceived politically and economically. It would depend on a strong alternative to or a transformation of the dominant industry. That did not, and perhaps could not, happen in the French society of the late 1920s. Moussinac's bitter conclusion has some validity: "because it envisioned the problem according to the point of view of aesthetics only, because it wanted to ignore the economic laws which determined it, the avant-garde is dead."11 Yet the avant-garde did not really die. If it failed to sustain a lasting cultural revolution on the scale it had hoped for, its counterpart in the Soviet Union was hardly more successful in its mission. One should not forget that, even with all of its contradictions, the French alternate cineha movement produced a good deal of valuable work, in theory and practice, that has not been without 274 descendants. Its heritage is visible in the British documentary movement of PRESERVATION the 1930s, in the film theory and criticism of Andre Bazin and Jean Mitry (though unacknowledged and somewhat obscured), in the American avant- garde that finally developed after World War II, and in the French film activ- ity in the 1950s that eventually became known as the New Wave. In this current period of rediscovery and reinterpretation, may that heritage once more make a useful intervention. 275  i1