20th Century Fox Film Corp./Everett Collection

Dana Andrews and Gene Tierney in Laura, 1944

One night not long ago I found myself once again drawn into a movie from the tail end of the 1940s. This one, with the thoroughly generic title Backfire (not to be confused with Crossfire or Criss Cross or Backlash), did not come with a high pedigree. It had sat on the shelf for two years after it was filmed in 1948, and afterward seems to have faded quickly from recollection. But movies of that time, when they emerge decades later, have devious ways of holding the attention: beguiling hooks and feints lead deeper into a maze whose inner reaches remain tantalizing no matter how many times those well-worn pathways have been explored, and no matter how many times the interior of the maze has led only to an empty space.

In a state of suspended fascination only just distinguishable from the preliminary stages of dreaming, I remained absorbed as documentary-style scenes of wounded war veterans recuperating at a hospital in Van Nuys, California, gave way to progressively more disjointed and often barely comprehensible episodes: an unidentified woman creeping in the dark into a patient’s room with a message about a missing war buddy, a murder investigation, a cleaning woman peering through a keyhole in a fleabag Los Angeles hotel, a winding path from mortuary to boxing arena to gambling den to the office of a particularly corrupt doctor, a plaintive piano theme playing over and over accompanied by reminiscences of its origin in a far-off Austrian village. The story splintered into flashbacks, spiraling into deepening confusions of identity and chronology, punctuated, as if to keep the spectator awake, by a series of point-blank shootings.

All the while, the dialogue was evoking the trauma of wartime injury, questioning the difference between memory and hallucination, talking about nightclub rackets, tax evasion, gambling debts, obsessive love, as the narrative coiled in its final swerve toward a strangling, silhouetted on a bedroom wall while Christmas carols were sung in the background. And then, abruptly, the nightmare was over, we were back in the veterans’ hospital after a second and successful round of rehabilitation, and the three surviving principal characters were even managing to laugh as they took off in their jeep for Happy Valley Ranch. By then I could hardly have said what the movie was about or even if it was especially good—few viewers have ever thought so—yet could not deny that something had caught me in its grip and stirred up troubling associations, like partially retrieved memories of a parallel life.

Backfire is one of the hundreds of movies that come up in the course of David Bordwell’s magisterial Reinventing Hollywood, cited as a particularly extreme example of the use of multiple flashbacks. (Bordwell even provides a rundown of the chronological order in which Backfire’s backstory is laid out: 3-5-2-4-1-6-7.) At the risk of making it sound less entertaining than it turns out to be, Reinventing Hollywood might be described as a taxonomy of narrative devices and strategies (“dead narrators and multiple flashbacks and surprise voice-overs and bizarre dream sequences”) in 1940s Hollywood: a decade that for Bordwell’s purposes stretches from 1939 to 1952, and to which he attributes a singular richness of experiment and invention still being picked over and recycled by contemporary filmmakers.

The scientific ring of such a description is apt for Bordwell, who in his many books and articles has consistently avoided impressionistic, anecdotal, or theoretically grandiose approaches to film history. He prefers the direct examination of palpable and measurable evidence. In Poetics of Cinema (2008), while rejecting the label of science for what he does, he modestly declares that his aim “is to produce reliable knowledge, both factual and conceptual, about film as an art form.” He has generally preferred to explore how films are actually made and precisely how their makers intended them to function, bringing to bear a remarkable command of technical and economic detail and an equally remarkable curiosity not just about the sweep of film history but about the peripheral circumstances that have impinged on every aspect of that history.

These investigations have at times become densely detailed. Bordwell has never been one for cutting corners when it comes to measuring shot lengths, parsing the intricacies of eyeline matches and aspect ratios, tracing the industrial development of film stock and lighting techniques, or the evolution of visual style in pre-1945 Japanese cinema, but his exposition of these matters is ever clear and to the point. However staggering a range of information he sifts through, a sense of interrelatedness is at work, no matter how arcane a particular byway might appear. In Reinventing Hollywood he discourses in great detail, for instance, on the various ways in which a telephone conversation can be represented cinematically or on the intricate inconsistencies of the voice-over narration in Laura.


What holds it all together is a sense of form, fittingly enough since form is his overriding preoccupation—narrative form above all. Irresistibly drawn to the comparison of examples taken from as wide a field as possible, Bordwell collects specimens of filmmaking and lays them side by side with the alert eye of a natural historian. His sensitivity to the ways in which filmmaking has changed over time is informed by a feeling for the human element at both ends of the process: as he puts it baldly in Poetics of Cinema, “films are made by human beings to provide other people with experiences.”

In this perspective, films—even the products of the heyday of studio filmmaking—do not roll off an assembly line as ready-made commodities demonstrating social and economic pressure points, any more than they emanate mysteriously from the zeitgeist or the collective unconscious. Bordwell quotes James Agee kicking back against the fashionable determinism of some 1940s film critics: “A movie does not grow out of The People; it is imposed on the people—as careful as possible a guess as to what they want.”

Reinventing Hollywood hews closely to the point of view of the filmmakers. Bordwell’s juxtaposition early on of epigraphs by Darryl F. Zanuck and Henry James serves not to contrast them but to show each of them engaged pragmatically with the problems of a job. Zanuck suggests that how the story is told is more important than the story itself, James that the best results can come from being part of a group: “Every man works better when he has companions working in the same line.” In tirelessly cataloguing the tools developed by 1940s filmmakers, Bordwell summons up the atmosphere of a workshop in which “forms were generated, repeated, swiped, tweaked, parodied, pounded flat, turned inside out.”

For the most part he is not concerned with the motives of the creative workers, beyond their intention to satisfactorily meet the challenge of the task at hand. The mood becomes that of a particularly open-ended and wide-ranging story conference, at which an array of storytelling elements new and old are hauled onto the table and sifted and recombined with a sense of infinite possibility. We come to feel something of how it was to be always looking for some fresh variant on a familiar narrative trope—the elusive “switcheroo”—or at least some fresh way to disguise that one had not quite been discovered. Was there a new gimmick to be found in a story involving twins (the basis for at least thirty features during the period) or amnesia (“rare in real life but common in movies”)? Could the old Grand Hotel set-up of a fixed location serving as the crossroads of destiny be given one more refurbishing, as in the desert motel of One Crowded Night or the nightclub of Club Havana? Was there any way to top the restricted space of Alfred Hitchcock’s Lifeboat or the restricted seventy-two-minute time frame of Robert Wise’s The Set-Up?

Reinventing Hollywood conjures an exciting, nerve-wracking atmosphere in which everyone is looking over everyone else’s shoulder in search of a new ingredient or a usable twist: producers keeping an eye on other producers’ releases, writers scavenging among novels and Broadway hits and overviews of Freudian psychology in search of some untried element. Bordwell makes his way through the back pages of 1940s popular culture, ranging beyond the world of film into the places where filmmakers foraged for ideas: slick magazine stories, radio plays, news items, the psychological thrillers that were beginning to supersede the old-fashioned whodunit. He spends a good deal of time parsing what he calls Murder Culture, the explosive popularity in the 1940s of crime fiction in every style and every medium. He reads contemporary manuals on screenwriting as a window into a culture invested in the technicalities of the storytelling for which it had such an appetite—in which anyone who could write could easily imagine that he too could make millions as a Hollywood scenarist or a big-time radio writer.

Bordwell’s curiosity about forgotten source material is impressive. In showing how Hollywood came up with its own version of modernism, he traces a multitude of obscure connections. The radical formal innovations of literary high modernism provided a powerful model of just how differently things could be done, but Hollywood did not respond by adapting Joyce or Woolf to the screen. Filmmakers turned instead to intermediary experimental works, instances of what Bordwell likes to call “moderate modernism” (a term he uses to describe, not disparage): popular novels or plays or radio dramas that in their own way and for their own audience fooled around with unorthodox chronology and fragmented point of view and interior monologue. He brings up the bifurcating storyline of J.B. Priestley’s play Dangerous Corner (1932); the multiple viewpoints of Gerald Bullett’s novel The Jury (1935); Archibald MacLeish’s choral verse play for radio The Fall of the City (1937); Peter Bowman’s Beach Red (1945), a war novel in verse told by an anonymous narrator; a Rex Stout novel (How Like a God, 1929) whose whole action is framed by a man walking up a flight of stairs; a Rumer Godden novel (Take Three Tenses: A Fugue in Time, 1945) in which a multigenerational story densely interweaves past and present. These were not extraordinarily exceptional works but instances of a widespread innovative tinkering going on at many levels.


Everett Collection

Gloria Swanson in Sunset Boulevard, 1950

All this ransacking and reshaping of material, this search for new tricks and twists, was a response to what seemed like limitless demand. Hollywood in the 1940s was riding an expansive wave that began in the late 1930s, crested in the war years, and persisted creatively if not commercially in the years just after the war. Many of the most splendidly inventive films of the decade (the film noir cycle, Sunset Boulevard, A Letter to Three Wives) were being made even as the studio system began to reel from the toll of anti-monopolistic legal rulings and competition from television. Movie attendance was never higher than during the war, and from 1939 to 1952 the eight major studios released over four thousand films.

That commercial boom in itself can’t account for the formal experimentation of 1940s films. Having come to Bordwell’s book after a long immersion in the films of an earlier period—the pre-Code talkies between 1929 and 1934, with their freewheeling mixing up of every available farcical, melodramatic, romantic, hardboiled, and muckraking plotline—I can imagine the dilemma of the studios was what to do after all the stories have been rehashed to the point of dissolving into burlesque. In Bordwell’s account, the solution was to tell them again but differently, and to amplify narrative innovation with technological improvements that could elicit fresh responses to anything from a shadow on a wall to the sound of a dripping faucet.

Pre-Code movies often had the flash and raucous directness of a live performance or a tabloid front page; the 1940s perfected a hypnotic absorption. Thus many of the devices on which Bordwell focuses, whether voice-over narrations or dream sequences or flashbacks, serve to undermine any sense of fixed location. The present dissolves into the past, the most solid dwelling recedes into a shadowy subdomain, and voices situated in some undefined in-between state of being navigate among shifting scenes. A self-conscious artificial world blossoms in the dark. Precisely delineated dream worlds and alternate realities and ghostly romances flourish as never before: Portrait of Jennie, The Curse of the Cat People, The Unsuspected, It’s a Wonderful Life, The Ghost and Mrs. Muir.

By the same token, as urban location filming made its inroads late in the decade, the constructed artifice dissolved, and by a different kind of magic the cities themselves, their el trains and police precincts and creaking tenement stairs, emerged in full counter-reality, the world outside the theater offered for your inspection without having to leave your seat. Yet this documentary realism itself partook of conscious artifice. In Jules Dassin’s The Naked City (1948), shifting narrators commented on the scenes as they unfolded, giving a double layer to otherwise banal scenes of everyday life. The effect was of waking up to the real world while remaining within the movie dream.

Earlier, in Orson Welles’s Citizen Kane (1941), counterfeit cinematic documents were made to seem real by being clearly marked as damaged or archaic, spurring an unprecedented awareness of celluloid as the medium of what was being perceived: “For perhaps the first time in film history, newly made shots were given scratches, jumps, and speeded-up motion to masquerade as archival footage.” By using such a device, Welles relied on an audience capable of picking up in an instant on the visual point being made—or at least capable of being instantly educated by him. He had after all for years been making use of analogously fine-tuned distinctions in the sonic sphere as the impresario of Mercury Theatre on the Air, where he had succeeded in creating out of sound alone a space virtually cinematic in its clarity and depth.

Radio was a primary laboratory, something that tends to be overlooked, because while people continue to listen to old music and watch old movies, few take much trouble to track down old radio plays. How many hours did Americans spend in the 1940s listening to voices in the dark, as Welles or Arch Oboler created dense dramatic soundscapes for an audience that had learned to pay attention to every audible clue? The fluidity of radio drama—its ability to shift from place to place, to switch narrators, to vividly describe events through sound effects alone—was a source for much of what seems most characteristic of 1940s movies. The potent emotional possibilities of voice-over narration had no cinematic precedent: the offscreen voice of Irving Pichel breaking into the narrative of How Green Was My Valley and at the same time announcing its conclusion with “Men like my father cannot die,” or the voice of Welles accompanying the shattered Tim Holt through the streets and into the deserted family mansion in The Magnificent Ambersons as he tells us that “George Amberson Minafer had got his comeuppance,” or Robert Mitchum describing the long days of waiting for Jane Greer to walk into a Mexican cantina in Out of the Past.

Radio was a more intimate medium than film. It whispered suggestions, and you completed its pictures within yourself. Improvements in sound technology made it more intimate yet, as the faintest gasp of breath or rustling curtain became distinctly audible. Those same improvements made their way into movie theaters, until soundtracks were as detailed and descriptive as the visuals:

A B picture like One Crowded Night (1940) can capture sleeves brushing a diner’s counter. A medium shot of two men drinking coffee can include the tiny sounds of their swallowing (Fallen Angel, 1945). In merely twenty seconds, Lady on a Train (1945) gives us the noises of footsteps on carpets, stairs, and wood flooring, a bag tossed from a window onto a canvas car top, the swish of a door, and telltale creak that reveals the heroine hiding behind it—all standing out against a fluctuating orchestral score.

The malleability of recorded sound lent indispensable support to the tensions and terrors so often depicted: the heavy breathing of a threatening phone call, a haunting remembered phrase echoing in the mind, the approach of a distant police car siren. On the screen, the noises that on the radio were shorthand for unseen events became an overloading of the visible, in which the sound of the smallest thing that happened was isolated and amplified, often, as Bordwell suggests, in distinct counterpoint with a symphonic environment created by Max Steiner or David Raksin.

Hollywood innovated in order to compete. Modernist stylization, Freudian plot lines, documentary realism, the otherworldly abstraction of theremin music: all these were no more than plausible ways of giving the product line a fresh sheen. But whether in the hands of artists within the system with their own preoccupations or simply by the expressive force of the forms themselves, new ways of constructing stories could end by generating new stories. The form itself became the story. The splintered personality of the troubled Laraine Day in The Locket is rendered through a complicated series of contradictory flashbacks, a Hollywood-style Cubism that literally embodies her madness to far more effect than any glib verbal diagnosis. The triangular romance of Otto Preminger’s Daisy Kenyon takes on an unexpected ambiguity not through any particular originality in its situations or plot points—even if it did manage to nudge an inch or so beyond Production Code conventions—but through a deliberately elliptical presentation that leaves its characters’ intentions unexplained: “The obscurity of their motives and purposes is enhanced by the utterly objective narration that rules nearly the entire film.…They may be dissembling or simply assuming a certain attitude by habit.”

At times—as Bordwell demonstrates amusingly in a detailed analysis of Arthur Ripley’s The Chase —the structural devices may have exercised an aesthetic power beyond the intentions of the filmmakers. This adaptation of Cornell Woolrich’s The Black Path of Fear has a deserved reputation as one of the most hallucinatory of B movies, but the film’s dream-within-a-dream loopiness resulted in large part from the producers’ pragmatic desire to refashion the plot so that the female lead would not get killed halfway through the picture. The old “it was all a dream” device was brought into play, but in such half-logical fashion as to throw the whole story into a Borgesian mode of infinite regression.

The book has many accounts of similarly bizarre examples: a movie in which virtually the entire running time is devoted to a flashback that turns out to be a lie (The Guilty), or a mystery constructed within a psychoanalytic framework that unaccountably culminates in a Sid Caesar monologue satirizing psychoanalysis (The Guilt of Janet Ames). Bordwell keeps the book in focus by fixing his eye essentially on the narrative devices themselves rather than the ambitions of those who used them or the ultimate merit of the films in which they did so, and thus Citizen Kane is juxtaposed with One Crowded Night, The Guilt of Janet Ames with Double Indemnity, all these hundreds of films flung together as in the world in which they came into being.

A few sidebars are reserved, however, for further consideration of those who most consciously transformed the available means: Joseph Mankiewicz (A Letter to Three Wives and All About Eve each receive a meticulous structural appraisal); Preston Sturges (with a warm appreciation of Unfaithfully Yours and its multiple fantasy sequences); Alfred Hitchcock (Vertigo, made in 1958, is accurately described as “one of the most typical forties movies”); and, of course, Orson Welles, because it is impossible to imagine what 1940s movies would have been like if Citizen Kane and The Magnificent Ambersons had not existed. Describing how a character in the latter half of Ambersons reminisces about things that happened in the first half, Bordwell remarks: “The early stretches of the film become part of the characters’ past, and our memory is folded into the reverie.” It provides an appropriate coda for this tour of a culture given over with something like religious intensity to the technology of storytelling: a quick reverse shot of the spectator, with motives as ambiguous as any character in a 1940s movie, as his memory is folded into a reverie.