The centenary of movies as a public spectacle is nearly upon us. Only specialists will care whether the benchmark date ought to be May 9, 1893, when members of the Brooklyn Institute of Arts and Sciences lined up to peep into Thomas Edison’s kinetoscope and watch a twenty-second film of three men hammering on an anvil and sharing a bottle of beer; or December 28, 1895, when at the Grand Café in Paris the Lumière brothers inaugurated a program featuring The Arrival of a Train at the La Ciotat Station and other movies made with their camera-projector-printer the cinematographe; or any of the other candidates. What matters is that about a hundred years ago a new species of language began to alter human life irrevocably.

A toy, a peepshow, a scientific novelty: astonishingly small-scale points of origin for such a phenomenon. Neither Edison nor the Lumière brothers nor most of the other hard-headed capitalists and dogged tinkerers who got movies off the ground could have fully imagined the depths of the fascination their machinery would provoke, or that a century later a whole planet would remain transfixed by its simulation of vital movement. The medium has evolved, becoming a multitude of different media in the process; but like one-celled organisms, Blacksmith Scene or Workers Leaving the Lumière Factory already contains the essential. What followed was just a matter of accretion and technological fine-tuning.

The story is never done being told, because the process isn’t finished yet. Every new plot development leads back to the beginning, forcing a revision of all earlier accounts. If an earlier model of film history went from the twenty-second one-shot movie to Intolerance, the latest goes from Cleopatra to the music video and the twenty-second TV commercial. The story of movies has been, at various times and places, a story of technological progress; of a sensation in the vaudeville business; of society overrun by seductive images of vice; of art and science collaborating harmoniously to produce a universal medium; of solitary film geniuses purifying a language of gesture and visual rhythm; of radical visionaries seizing the means of production in order to deprogram the brainwashed masses; of multinational corporations conspiring to reassert control over all forms of electronic communication.

From today’s perspective it could be a story of unforeseeable virus-like proliferation: how a small gadget capable of showing a single flickering scene became the global environment. The metaphor is not too far-fetched, in light of the close similarities that contemporary information theorists find between patterns of viral infection and the way information diffuses and mutates. Told this way, the story leads into science fiction, into a decentered hyperspace swarming with random and infinitely reproducible images.

In 1905 people stepped into nickelodeons to encounter the novel excitement of a disordered, constantly shifting world of pratfalls, state funerals, heroic firemen, and scenic tours. The “nickel madness” of that time was the first great wave of enthusiasm for the new medium. Since then we have moved ever closer toward perceiving the world as an immense nickelodeon, an enclosed area delimited by walls of screens—movie screens, television screens, computer screens—on which the action never stops. Indeed, with the promise (or threat) of untold interactive modes and virtual realities, we may be on the verge of taking the next step beyond spectatorship and entering the screen itself, like Buster Keaton in Sherlock Junior.

These books, then—only the latest in an unending succession—can like their predecessors at best retrieve fragments from the ceaseless flow of images. More than many of their predecessors they attempt to marshal their data in a systematic, even scientific fashion, while racing to collect an immense and scattered body of evidence before it disappears altogether. As Kevin Brownlow notes of the film stock on which most pre-1950 movies were printed, “1990 inaugurates the last decade of nitrate…. It is not expected to last into the new century.”

Film historians and critics confront more fundamental problems, however, than the profusion and frequent inaccessibility of materials. Even after a century, written language is not yet at ease with the language of film. The most basic terminology remains an occasion for acrimonious contention. In order to talk about it at all, language has to radically transform what happens on the screen: slow it down, simplify it, break it up into abstract units. The written texts seem to exist quite apart from the films they discuss, in a separate world where words are still sovereign.

The tack taken in the first volumes (three out of a projected ten) of History of the American Cinema, under Charles Harpole’s general editorship, is often to avoid the problem by studying primarily not what is on the screen, but what is around, in front, or in back of it. This approach reflects a current scholarly preoccupation with questions of production, exhibition, and spectatorship, with the context rather than the contents of the spectacle; and in the expert hands of Charles Musser, Eileen Bowser, and Richard Koszarski it yields immensely valuable information. Margins and backgrounds become momentarily central as the authors survey the role of illustrated slide lectures with their “beautiful dissolving views,” an extravagant range of early machines (choreutoscope, phasmatrope, eidolscope, cineograph), the evolution of fan mail and movie magazines, the vagaries of musical accompaniment. In addition the series functions as a small-scale archive, with its splendid miscellany of production stills, frame enlargements, ads, memos, and cartoons: a magic lantern version of The Raven, scenes from Kicking Willie’s Hat, What Demoralized the Barbershop, and The Gay Shoe Clerk, an array of rooftop studios and store-front theaters, and hundreds of other items.1


Such spreading out of materials is what film history does best, in keeping with the fundamental randomness with which film captures some but not other pieces of the world, leaving us a collection of “living photographs” (as they called them in the 1890s) which could just as easily all be different. These are simply the movies that happened to get made and that happened to survive; only retroactively do they acquire an air of inevitability. Chance and improvisation punctuate film making and provide many of its most enduring pleasures.

Where current film writing seems most at odds with its subject is in its attempt to impose a systematic determinism. The essays in Thomas Elsaesser’s anthology Early Cinema: Space, Frame, Narrative and Noël Burch’s Life to Those Shadows embody some of the best-informed scholarship going, but one needs a pair of tweezers to extract the useful observations from the relentless and fruitless mongering of abstractions, the frequently circular methods of argumentation, and (in Burch’s case especially) the almost caricatural political simplifications. (Burch likes to insert, into his otherwise intensely theoretical and sometimes brilliant discussions of “diegesis” and “syntagma,” side comments such as: “Informing has traditionally been encouraged in Britain in the name of social consensus.”) Little of the enthusiasm felt by either spectators or film makers finds its way into these writings: if film viewing actually resembled the tortuous processes formulated here, the movies could hardly have become the most well-loved of modern entertainment forms.

Of the prose itself the less said the better, although a few examples may indicate the stylistic tenor. Raymond Bellour, discussing the structure of Griffith’s The Lonedale Operator, speaks of

micro-condensation of the textual system, which integrates into one of its units components of several earlier units, and thus constructs itself, by means of displaced similarities which constitute its repeated difference,

while Noël Burch lingers over

the “morphological moment” at which this originary fissiparity arose, i.e., at which a distinction began to be made between two meanings attributable to the transition between two biunivocally concatenated shots.

Certain tics of vocabulary recur often enough to serve as signatures: situations are never revealed or clarified, they are articulated; events are not caused, they are determined. It is peculiar that while there are writers who find it possible to discuss cosmology or quantum mechanics in quite lucid terms, so many film critics must deploy an arsenal of neologisms and technical arcana to describe, for instance, a scene of a dog stealing a sausage in a one-shot movie.2

Most of these writers represent a reaction to the “auteur theory” which flourished in the 1950s and 1960s. That misnamed, much-maligned tendency is easy enough to criticize, especially along its fringes, yet it does have the virtue of acknowledging that movies are actually made by people, although those people obviously include many others aside from the directors whom the auteur-minded critics concentrated on. What the auteurists at their best conveyed was their relish of filmic pleasures, an emotional engagement and capacity for joyful discovery seemingly rare in film studies today.

The theoretical and political approaches that became institutionalized in the universities in the 1970s—summarized by Burch as “structuralism, dialectical and historical materialism…and left-wing feminism”—have shifted attention from the concrete process of movie making as a collaboration among individuals toward a more generalized analysis of how basic film practices embody (or “inscribe”) dominant social and political motives. Film as commodity, narrative as the end-product of ideological and economic determinants, the spectator as “constructed subjectivity”: these rather mechanical conceptions tend to create the impression of an almost Augustinian absence of personal initiative. It’s as if films were generated automatically by a gigantic impersonal industrial-ideological apparatus, something like the machinery of Metropolis. Hence the abstraction, the reductiveness, the relentless sorting of filmic elements into “progressive” and “bourgeois,” the denial of any role to the creative imagination or to the sense of play which is imagination’s conduit.


It isn’t a question of denying the reality of political and economic pressures and controls in the film business—or in life. What has to be questioned is the theorists’ patent lack of interest in the directors, artists, writers, and others who somehow functioned and sometimes fluorished within those pressures and controls. To deny the specificity of temperaments and talents is to deny precisely what makes movies more interesting than the manufacture of socks. (Some theorists would undoubtedly reply that they are not more interesting.) Deterministic film theory ends up burying the individual in an analogous fashion to the industrial juggernaut it decries, and implying that particular films—the presumable object of study—are of interest only as embodiments of ideological and economic tendencies.

More often than not, an ostensible concern with social contexts and struggles leads us away from the flavor of actual situations and eras. The bourgeois ideologues whose heavy hand is read into every permutation of film form or content are a rather faceless crew. Social history without empathy or imagination becomes a parade of empty abstractions such as “the site at which product and spectator interact with and construct each other.” The great pleasure of Kevin Brownlow’s decidedly untheoretical Behind the Mask of Innocence—a survey of the social problem films which flourished before the studio system took full control in the 1920s—is to get back to specifics such as corrupt political bosses, exploited prostitutes, the drug trade. By concentrating on individual anomalies rather than hypothetical collectivities—Mary MacLane, Margaret Sanger, Carry Nation, Jack London, Texas Guinan, Evelyn Nesbit, Harry Thaw, Governor William Sulzer of New York, the elusive Russian monk Iliodor—Brownlow imparts a much more tangible sense of a period when accused murderers, judicial reformers, and impeached politicians might be recruited to reenact their deeds for the camera, when movies were weapons in debates over prohibition, birth control, child labor, and “social hygiene,” when events like the Triangle Fire of 1911 and the Ludlow Massacre of 1914 were rushed to the screen in fictionalized form. (The text’s magnificent photographic accompaniment certainly helps; it is rare and exciting to encounter so many completely unfamiliar images in a single book.)

Behind the Mask of Innocence would make a good starting point for arguing the superiority of anecdote over theory. Anecdote is of course next door to fiction, and movie lore regularly slides over the line, with the help of frequent infusions of self-serving or malicious gossip, public relations legends, and a strain of tall tale bravado. Of the director Raoul Walsh (himself an authentic cowboy), his wife noted: “Raoul never bored you with the truth.” An aura of half-truth—intoxicating at times—is the native air of movie history.

Christopher Rawlence turns this fact of life into a formal game in The Missing Reel, his investigation into the life of Augustin Le Prince, who may or may not have invented the first motion picture camera, and who disappeared mysteriously after boarding a train for Paris in 1890. The facts are skimpy, the speculations hazardous—Le Prince’s family persisted in believing that he had been murdered by agents of Thomas Edison—and so Rawlence adopts the methods of “docudrama,” freely fictionalizing dialogue and hypothesizing inner monologues. The lack of a clear line between evidence and improvisation can be disconcerting. Yet The Missing Reel does introduce, into its world of patent wars and hard-pressed small-time entrepreneurs, a note of real human sadness in what would otherwise be an assortment of exhibits as lost to memory as Charles Foster Kane’s sled.

People make up a fiction, a myth, about the history of movies for the simplest of reasons: it happened to them, it really mattered, there’s a personal compulsion to make some sense of it. Movies are not “out there”; they have long since been internalized by just about everyone on the planet. They were internalized, in a sense, long before the medium was invented. The germ of the idea of movies goes back centuries, or even millennia if you care to retrace it to the imagery of living shadows or deceptive visions, to the reflexive trap of Narcissus or Orpheus’ backward glance.

Charles Musser, in The Emergence of Cinema, begins his history of “screen practice” (of which movies are only a part) in the seventeenth century, with the magic lantern and with the German Jesuit Athanasius Kircher’s treatise on The Great Art of Lights and Shadows. Musser credits Kircher with demystifying those optical tricks of reflection and enlargement, the magician’s repertoire of mirrors and silhouettes, that had long been used to simulate the supernatural. As to how long such practices had been going on, Kircher ventured a reference to the time of King Solomon. Musser doesn’t choose to probe any deeper into this prehistory of projection (understandable given the length of his book), but it provokes fantastic speculations about a subculture of projectionists carefully guarding its technical secrets: medieval movies, Graeco-Roman movies, ancient Egyptian movies, or who knows what mystifying play of flame and shadow cast against a cave wall to induce Cro-Magnon wonderment.

In any event the secret got out, and magic-lantern techniques passed from the realm of superstition to that of science; but as Musser’s account makes clear, it could not shake the aura of the fantastic. In the late eighteenth and early nineteenth centuries, showmen staged elaborate lantern shows which offered to immerse the spectator in supernatural visions which were at the same time debunked. “This is a spectacle,” declared one impresario, “which man can use to instruct himself in the bizarre effects of the imagination, when it combines vigor and derangement: I speak of the terror inspired by the shadows, spirits, spells and occult work of the magician.” New York saw its first phantasmagoria exhibition in 1803, a Wonderful display of Optical Illusions which paraded “the Phantoms, or Apparitions of the Dead and Absent,” supposedly in order “to expose the practices of artful imposters and exorcists, and to open the eyes of those who still foster an absurd belief in Ghosts or Disembodied Spirits.”

These details are worth lingering over because they point up, even at this embyronic stage, a certain duality of intent: to create illusions and at the same time to lay bare the real. The audience will first of all be terrified by apparitions, and afterward instructed in how the illusion was produced. The past two centuries have witnessed the gradual erosion of the supernatural through the combined effects of electric lights, photography, movies, and modern transportation. There were to be no more unlighted corners or inaccessible wilderness areas where ghosts and demons could congregate: what registered on the lens of the camera was an absolutely limited, physical world. Before the emergence of the story film, actualities—street scenes, views from railroad carriages, parades, coronations, lifelike re-enactments of crimes and battles—seemed to embody the medium’s future.

Such a prognosis reckoned without that appetite for the fantastic that has persistently diverted movies from the straightforward task of recording what is there. Upon the motion picture—the most alluring mechanism of the age of mechanical reproduction—would devolve the task of reconstructing the imaginary worlds that it had helped to dismantle. The editor of History of the American Cinema divides the world of cinema into three domains, the technological, the socio-economic, and the aesthetic. He could well have added a fourth: the religious. No medium having so vividly intimated the disappearance of God—there are sacred books but no sacred movies—it stands to reason that film would overcompensate by the systematic cultivation of visions, icons, exorcisms, martyrdoms, paradisiacal landscapes, and sacred rituals.

Sacred, or magical. This last inheritance of the nineteenth century could be one of those dangerous legacies that are regularly passed on in tales of the supernatural: keys to rooms never before opened, sealed vials of mysteriously potent liquids, ancient books containing necromantic inscriptions. For all the stolid workmanlike attitude of its inventors, movie technology fulfilled the most Faustian of ambitions. At least two fin-de-siècle novels, The Future Eve (1880) by Villiers de l’Isle Adam and Jules Verne’s Castle in the Carpathians (1893), conjure up a technology capable of generating simulacra indistinguishable from living humans. In Verne’s novel, a reclusive inventor copes with the death of his beloved by endlessly contemplating a hologram-like duplication of her, captured while she was still alive.

In the most interesting pages of Life to Those Shadows, Noël Burch explores this desire “to extend the ‘conquest of nature’ by triumphing over death,” uncovering some revelatory first reactions to the new technology. “When these cameras are made available to the public,” writes a journalist in 1895, “when everyone can photograph their dear ones, no longer in a motionless form but in their movements,…death will have ceased to be absolute.” Or, in the even more hyperbolic words of the scientist Georges Demenÿ: “The future will see the replacement of motionless photographs, frozen in their frames, with animated portraits that can be brought to life at the turn of a handle…. We will do more than analyse, we will bring back to life.” One can hardly help thinking of the frenzied painter in Poe’s The Oval Portrait, raving as he completes his lifelike picture of the wife who has died while sitting for him: “This is indeed Life itself!”

There is in fact a peculiar awe attendant upon watching the earliest movies. We are looking not only at people who are dead but at a world that has disappeared: the world in which movies were unknown. Dai Vaughan, in a meditative essay on the Lumière films included in Early Cinema, writes that viewing them is “like pondering what happened…in the first few microseconds after the big bang.” We might be extraterrestrials surveying the delayed signals from an already vanished planet. Yet the people move about casually and do not know they are being transformed into a picture for the eyes of spectators yet unborn. They are being filmed but have not yet grasped what that implies. To intrude on that innocence can seem as much a violation as opening a mummy case.

In the carnival tents, fairgrounds, music halls, and shop fronts where films were first exhibited, it is doubtful that such metaphysical perplexities were uppermost in the minds of the viewers. For those who stared at Annabelle Serpentine Dance or The Boxing Cats or Chinese Laundry Scene, through peepholes or at the first flickering screen projections, the novelty of movement was enough. Clearly the first movies had nothing like the sustained hypnotic power of the narrative films that became dominant within ten years: what held the audience was not emotional identification but amazement and curiosity.

The spectator was still consciously comparing the images on the screen to what they were supposed to represent, so that filmed waves are described as behaving “in the most natural manner.” Such astonishment—at the simple fact of something, anything, being filmed—could not last very long. Once the initial shock was past, the movement of a wave no longer sufficed. Entrancement by mere optical illusion became the subject of joke films, as in the 1902 Uncle Josh at the Moving-Picture Show, where a yokel tries to communicate with the woman on the screen. (Godard would reinvent the joke decades later in Les Carabiniers.) In the absence of strong narrative, subject matter was paramount.

This was what the historian Tom Gunning has called a “cinema of attractions,” a variety show of disconnected views with the exhibitor in creative control, “editing” in deejay fashion through his mix of one-shot subjects: the cinema that was supplanted by story films like Griffith’s but that seems now to be re-emerging in the fragmented formats of television and video. Such programs—interspersed with live entertainment—went after contrast and diversity rather than coherent themes or moods. It must have been something like a visual party.

And what did the spectators see? They saw bricks and crockery, babies and dogs, chattering women and destructive children, windows, sausages, a robber on stilts, a wife with a rolling pin, banks, horse-drawn carriages, tramps, policemen, wharves, ladders, grief-stricken women and stolen children, taverns, bottles of poison, telephones, railroad tunnels, escaped lunatics, armed anarchists, collapsing walls, acrobats, flim-flam artists, washerwomen, incomprehensible foreigners, nursemaids with criminal tendencies, lecherous young men in straw hats, women smiling at the camera as their skirts were blown up by the wind, comical bumpkins making a spectacle of themselves on their visit to the big city.

And mixed indiscriminately with all that they saw pieces of real life. A fire breaking out in a warehouse; a cavalry troop charging; a battleship being launched; President McKinley speaking (without intertitles, a sublimely empty sight); the coffin of President McKinley being lifted from a train; the assassin of President McKinley being led from his cell, strapped into the electric chair, and executed. This last was a re-enactment, whether or not the audience knew it was, any more than they knew that many of the “war-graphs” of the Spanish-American War were in varying degrees misrepresentations. The battleship shown was not really the Maine; the battleship shown was not really a battleship at all but a miniature; the battle shown “through the dense smoke” was a fake, with the smoke there to hide the traces of deception. Such frauds may well have advanced the range of techniques and textures, as film makers learned to imitate the look of documentary footage.

The film business was a study in chaos in those days, and the historians take up much of their space just in unraveling the legal and commercial battles that disrupted the movies’ first decades. The War of the Patents over one kind of camera or another is actually a good deal less interesting than the censorship skirmishes that erupted as film makers poked around for new subjects to point their cameras at: kooch dancers, boxing matches, cock fights, passion plays, representations of police brutality. These primal battles took place, it should be noted, under a Supreme Court ruling that movies were not protected by the First Amendment, a judgment not rescinded until the 1950s.

The censorship fights often raised, quite unintentionally, some curious philosophical problems. It was argued, for instance, that while boxing itself was illegal, a film of a boxing match, as a representation rather than the thing itself, was exempt from the law. Similarly, for an actor to play Jesus on stage was viewed as sacrilege. To make a film of an actor playing Jesus, however—and there was a wave of such films from 1897 on—could be construed as not only acceptable but inspirational.

The sacrilege of the stage Jesus lies in the fact that he was of flesh and blood; you might reach out and touch him; you might smell him. The figure on the screen belonged to a different order, odorless and beyond touch: he was a phantasm, a trick of light. “There will be no ‘real’ actors,” wrote the Boston Herald in 1898, “no living personages in the presentation of this most sacred and sublime of the world’s tragedies.” A clergyman testified that while “the performance of this play…by living actors and actresses was prohibited…to the rendition of it by these pictures there can be no objection.”

So much for one theological take on film ontology. On the other hand, Pope Leo XIII consented to be filmed in Rome in 1898, and in the process bestowed a blessing directly on the camera filming him, as if the spiritual energy thus set loose could be canned in the apparatus, projected on the screen, and thereby transmitted to an audience on the other side of the Atlantic. The gesture implied a sort of cinematic transubstantiation.

The range of permissible subjects for an American movie in 1991 is in many ways narrower than at any point before the 1920s. Before Hollywood got a stranglehold on distribution, before the Hays Code cut back on controversy, people would and did try out anything. Ibsen, Schiller, companionate marriage, the Yellow Peril, capital punishment, female detectives: they all tumbled out so fast that there was no time or concern to sort them. Independents, even downright amateurs, still had a shot at getting in the game. Movies had gone irrevocably narrative and feature-length, but standardization of plot and genre still had a long way to go.

Nothing evokes this free-for-all of form and content better than Brownlow’s Behind the Mask of Innocence. The book is all the more valuable because most of the movies it discusses are either lost—Raoul Walsh’s prison picture, The Honor System, Allan Dwan’s street gang saga, Big Brother, Lewis Milestone’s political corruption exposé, The Racket—or inaccessible to the ordinary viewer. The period on which Brownlow concentrates (roughly 1912 to 1919) calls to mind Samuel Fuller’s definition: “A movie is a battleground.” Polemic fury rages on all sides, with cops and clergymen moving to stamp out vice while on the other hand a progressive-minded club woman can declare that “motion pictures are going to save our civilization from the destruction which has successively overwhelmed every civilization in the past” by providing “a means of relief, happiness and mental inspiration to the people at the bottom.”

The resulting cacophany—a traffic jam of moral uplift, unhinged propaganda, and enthusiastic exploitation in dozens of often cheaply made films—would be edited out in subsequent decades, largely through the mechanisms of self-censorship that took hold between the Fatty Arbuckle scandal of 1921 and the final imposition of the Production Code in 1934. In mainstream movies there would be no more revelations of drug addiction (Human Wreckage), birth control (Where Are My Children?), venereal disease (Damaged Goods), or prostitution (Traffic in Souls, The House of Bondage, and other instances of what Variety called “patchouli and kimono pictures”). By the time of the Red Scare of 1919, things took a more repressive turn, with Adolph Zukor and the secretary of the interior sharing notes on how to counteract “Bolshevism, radicalism and discontent against the U.S.” and the editor of Photo-Play Journal declaring that “there is much freedom of thought that should be imprisoned, especially at this critical time in our national history.”

Still, as late as 1926 Cecil B. De Mille could produce a (mildly) pro-Bolshevik film like The Volga Boatman; and curiously enough, De Mille’s 1928 melodrama of reform school sadism, The Godless Girl, enjoyed a hugely successful run in the Soviet Union as an exposé of American brutality. Even more curiously, The Godless Girl’s star, Lina Basquette, acquired a devoted Austrian fan: Adolf Hitler, who may have been entranced by the film’s prolonged episodes of punishment and torture. The ambiguity of silent film imagery cut across all sorts of national and ideological lines.

It wasn’t only subject matter that changed. Someone who fell asleep in 1907, when D. W. Griffith was beginning his career at Biograph, and woke up in the 1920s would encounter a drastically streamlined world of complex feature-length stories, seamless editing, carefully crafted close-ups, fluid compositions and camera movements, subtly modulated lighting, and (sometimes) carefully synchronized live music, all framed by the glittering decor and precision-drill ushers of the dream palaces. As the opening title of von Sternberg’s The Last Command declared: “Hollywood—1928! The Magic Empire of the Twentieth Century! The Mecca of the World!”

Above all the emotional relationship of the spectator to the screen had changed. There were movie stars, for one thing, a phenomenon driven not by producers but by audiences, who (as soon as close-ups showed the actors’ faces clearly enough) recognized their favorites and made stars of them before they even knew their names: “the Biograph girl,” “the Indian.” It was not the first time that film makers had miscalculated the effects of their productions. As Miriam Hansen notes in Babel and Babylon (a study of spectatorship, especially female spectatorship, as it evolved from the earliest silents to the more complex cases of Intolerance and the films of Rudolph Valentino), the promoters of the first boxing movies were surprised by their popularity among women. Although movie producers learned soon enough to study and exploit the desires of their diverse audiences—turning the star system, for instance, into an institutionalized method of control—they could still be caught off guard. In the case of Valentino—“a cult object, the center of a collective ritual”—it could still threaten to escape from conventional limits into explosions of riot, suicide, and erotic obsession.

But the stars could not exist apart from the narratives they inhabited. People cared about them in large part because of the complex things that happened to them. Early story films—The Great Train Robbery (1903), The Escaped Lunatic (1903), Rescued by Rover (1905), The Automobile Thieves (1906)—had evolved out of physical actions that were easy to track: chases, abductions, and recoveries, criminal attacks and police counterattacks. It took a decade of formal experiment (traditionally associated with Griffith, although today’s scholars are more skeptical) to come up with ways of handling more complicated or more subtle plots.

That juncture is for many contemporary theorists the crossroads where film fatally enters on the path of “classical narrative,” “Hollywood continuity,” or what Noël Burch calls “the institutional mode of representation.” The heterogeneous world of film becomes “the movies,” a standardized industrial product designed (again Burch) “to misinform and anaesthetise the masses of the people,” a cinema of evasion centered in America but ultimately achieving global dominance.

Escape was of course precisely what was intended, and presumably precisely what was sought. A commentator in the 1920s, the director of a training school for movie theater managers, spelled it out in perfumed prose:

People come to the motion picture theater to live an hour or two in the land of romance…. For a small charge they can be picked up on a magic carpet and set down in a dream city amidst palatial surroundings where worry and care can never enter, where pleasure hides in every colored shadow and music scents the air.

Paramount touted its motion pictures in similar terms: “A magician somewhere waves his wand, and we’re off on our travels into the realms of laughter and tears…. We are youthful romancers living in another world.”

The pleasure of that escape became the most commonly shared pleasure of the century. Everybody could go into the same dark room—no matter where it happened to be located—and zero in on precisely the same dream. None of the noises and smells of circus or live theater; it was a more futuristic experience, the vision of the individual space traveler strapped in and watching the universe roll by outside his porthole. Narrative only offered a way in, a way to stay in that intimate and trance-like space as long as possible.

Perhaps it is time to defend, again, the aesthetic that accommodated itself to the movie palaces’ dream voyage. Whatever tribute it may have paid to the needs of backers and bookers and renters, whatever it may have sacrificed to the imperatives of bourgeois respectability, the film art that triumphed in the 1920s owed most to the audiences who—as children or adolescents—watched the movies of the nickelodeon era and wondered. Some of those spectators were named Charlie Chaplin, F. W. Murnau, Fritz Lang, Ernst Lubitsch, Raoul Walsh, Josef von Sternberg, King Vidor, John Ford, and Buster Keaton. The movies they made may indeed be read as an outgrowth of historical circumstances, but they were first of all a response to other movies: a dream replying to a dream.

In looking at old movies, historicism is indispensable, but it has its limits. By watching Foolish Wives do we come to understand anything about 1922 and the mentality of its spectators, or do we construct a fiction from the haunting shapes and moods of Foolish Wives, and call it 1922? Must we, in fact, linger unduly over the fact that Foolish Wives was made in 1922? To an archivist the answer to the latter question clearly is yes, and films when shown in academies and museums are always locked securely into their historical provenance. Viewers, however, have other options. The availability on video of an increasing number of remastered, often lovingly restored silent movies makes possible a refreshingly direct access to the likes of Pastrone’s Cabiria, Griffith’s Intolerance, Sjostrom’s The Outlaw and His Wife, Lang’s The Spiders, Stroheim’s Greed, Vidor’s The Crowd, or Keaton’s Steamboat Bill Junior.3 After prolonged immersion in contemporary film theory and historiography, it is an inexpressible delight to let words rest and begin to see.

It takes a bit of adjustment to allow the effect to take hold. With no dialogue to listen for, the eyes assume complete control, gradually learning to navigate the depths and rhythms of a universe of illuminated gestures. It’s all utterly concrete: only the visible exists, whether we are watching (in Intolerance) Constance Talmadge playfully biting a goat’s ear or a series of razors poised to cut the rope that will hang a man. But no matter how tightly the images are linked to a narrative, the silence allows extraordinary leeway for each spectator’s fantasy. The silent movie must be met halfway, the viewer’s imagination filling in all that is left unexpressed. The movie becomes an active dream, a waking trance cunningly orchestrated by visual musicians: a ghost opera, perfected just at the moment when synchronized sound was about to change everything yet again.

This Issue

May 30, 1991