David Bordwell's website on cinema   click for CV




Perplexing Plots: Popular Storytelling and the Poetics of Murder

On the History of Film Style pdf online

Reinventing Hollywood: How 1940s Filmmakers Changed Movie Storytelling

Film Art: An Introduction

Christopher Nolan: A Labyrinth of Linkages pdf online

Pandora’s Digital Box: Films, Files, and the Future of Movies pdf online

Planet Hong Kong, second edition pdf online

The Way Hollywood Tells It pdf online

Poetics of Cinema pdf online

Figures Traced In Light

Ozu and the Poetics of Cinema pdf online

Exporting Entertainment: America in the World Film Market 1907–1934 pdf online


Hou Hsiao-hsien: A new video lecture!

CinemaScope: The Modern Miracle You See Without Glasses

How Motion Pictures Became the Movies

Constructive editing in Pickpocket: A video essay


Rex Stout: Logomachizing

Lessons with Bazin: Six Paths to a Poetics

A Celestial Cinémathèque? or, Film Archives and Me: A Semi-Personal History

Shklovsky and His “Monument to a Scientific Error”

Murder Culture: Adventures in 1940s Suspense

The Viewer’s Share: Models of Mind in Explaining Film

Common Sense + Film Theory = Common-Sense Film Theory?

Mad Detective: Doubling Down

The Classical Hollywood Cinema Twenty-Five Years Along

Nordisk and the Tableau Aesthetic

William Cameron Menzies: One Forceful, Impressive Idea

Another Shaw Production: Anamorphic Adventures in Hong Kong

Paolo Gioli’s Vertical Cinema

(Re)Discovering Charles Dekeukeleire

Doing Film History

The Hook: Scene Transitions in Classical Cinema

Anatomy of the Action Picture

Hearing Voices

Preface, Croatian edition, On the History of Film Style

Slavoj Žižek: Say Anything

Film and the Historical Return

Studying Cinema


Book Reports

Observations on film art

Lucky ’13

Friday | August 29, 2008   open printable version open printable version

The Mysterious X (1913).

Each film is interlocked with so many other films. You can’t get away. Whatever you do now that you think is new was already done in 1913.

Martin Scorsese, quoted in Scorsese by Ebert (University of Chicago Press, 2008), 219.

DB here:

Most historical events don’t abide by clocks and calendars. Seldom does a trend begin neatly on one date and end, full stop, on another. Changes have vague origins and diffuse destinies. When Kristin and I, along with others, argued for 1917 as the best point to date the consolidation of the Hollywood style of storytelling, we realized that it’s a useful approximation but not as exact as a Tokyo subway timetable.

It’s just as hard to argue that a year constitutes a meaningful unit in itself. Who expects anything but tax laws to change drastically at midnight on 31 December? Yet evidently our minds need benchmarks. Film historians, while being aware that trends are slippery and dating is approximate, have long spotlighted certain years as particularly significant.

Take 1939, which has become a sort of emblem of the peak achievements of Hollywood’s Golden Age. We had Gone with the Wind, The Wizard of Oz, Only Angels Have Wings, Stagecoach, Gunga Din, Wuthering Heights, Dark Victory, Young Mr. Lincoln, Beau Geste, Mr. Smith Goes to Washington, Ninotchka, The Roaring ‘20s, and Destry Rides Again. I’d watch any of those, except Gone with the Wind, right now—something I find it hard to say about most Hollywood movies I’ve seen in 2008.

Another strong year is 1960, with La Dolce Vita, L’Avventura, Rocco and His Brothers, The Apartment, Elmer Gantry, Spartacus, Psycho, Exodus, The Magnificent Seven, Shadows, Late Autumn (Ozu), and The Bad Sleep Well (Kurosawa). Arguably, 1960 was owned by the French, who gave us Breathless, Shoot the Piano Player, Paris nous appartient, Les Bonnes femmes, Le Trou (Becker), Moi un noir (Rouch), and Letter from Siberia (Marker). (See Postscript.)

Let’s go back still further. Researchers sometimes split the silent-film period in two, with the first stretch, usually called “early cinema,” running up to 1915 or so. (1) The second phase then runs roughly from 1915 to 1928. (2) So for many historians the year 1915 functions as a tacit pivot-point, and it is remembered not only for The Birth of a Nation but also for Regeneration, The Tramp, Kindling, The Cheat, Les Vampires, Daydreams (Yevgenii Bauer), and several William S. Hart films. But another year holds a special place in the minds of silent film aficionados.

Over a decade ago, the annual Days of the Silent Cinema festival (Il Giornate del cinema muto), took 1913 as its focus. (3) It was an extraordinary year. Denmark produced Atlantis (August Blom) and The Mysterious X (Benjamin Christensen). From France we had L’Enfant de Paris (Leonce Perret), Germinal (Albert Capellani), and Louis Feuillade’s Fantomas series. Germany gave us Urban Gad’s Engelein and Filmprimadonna and Franz Hofer’s obsessively symmetrical The Black Ball. The staggering set of Italy’s Love Everlasting (Ma l’amor mio non muore!, Mario Caserini) was matched by the breadth of Enrico Guazzoni’s Quo vadis? In Russia Bauer released Twilight of a Woman’s Soul. American audiences saw Traffic in Souls (George Loane Tucker) and Death’s Marathon and The Mothering Heart, from a guy named Griffith. Several historians would argue that 1913 marked the first major achievements of film as an artform.

Two outstanding films of that annus mirabilis have recently been issued on US DVD. One is a striking accomplishment, the other a flat-out masterpiece. Both discs belong in the collections of everyone who’s serious about cinema.

Your call is important to us

A wife and her baby are alone in an isolated house when a tramp breaks in. As the wife tries to keep the invader at bay, her husband happens to telephone and learn what’s happening. He scrambles to return home. He steals an idle car, and its owner, accompanied by police, race after him. We cut rapidly between the besieged mother and the husband’s frantic drive, as he is in turn pursued. Just as the tramp is about to attack the wife, the husband bursts in, followed by the police. The family is saved.

This is the story of the 1913 one-reel film, Suspense, co-directed by Lois Weber and Phillips Smalley. If the plot sounds familiar, it’s probably because you know that one of D. W. Griffith’s most famous films, The Lonely Villa (1909) tells the same basic tale.There are still earlier versions, including one, The Physician of the Castle (Le Médécin du chateau, 1908), which may have inspired Griffith. The ultimate source seems to be a 1902 play by André de Lord, Au téléphone (translated here).

So Weber and Smalley are reviving an old idea. Their task is to make it fresh. How they do so has been studied in depth by Charlie Keil in his book Early American Cinema in Transition: Story, Style, and Filmmaking, 1907-1913.I can’t match Keil’s subtlety, and it’s better that you see the film first, so I’ll drop only some hints, pointers, and comments.

We’re inclined to say that The Lonely Villa influenced Suspense. But maybe we can capture the situation in a more illuminating way. The art historian E. H. Gombrich has suggested that we can often trace the relationship between artworks in terms of schema and revision. (4) A schema is a pattern that we find in an artwork, one that a later artist can borrow. Most often, later artists copy the schema straightforwardly. This is the usual way we think of influence. But instead of replicating the schema, the next artist can revise it. She can elaborate on it, strip it to its essence, drop parts and add others, whatever—in order to achieve new purposes or evoke fresh responses.

In The Lonely Villa, Griffith uses crosscutting to build suspense. He cuts among the thuggish vagrants trying to break in, the wife and daughters trying to hold them off, and the father learning by phone of the situation and then plunging after them with a policeman. The obvious pattern here is the principle of alternation between different lines of action, all taking place at the same time and converging in a last-minute rescue.

So Smalley and Weber inherit the crosscutting schema, but they go beyond simply copying it. They find ways to revise it, some quite surprising. These revisions aim to create more tension and to dynamize the situation.

The obvious option, at least to us today, would be to use more shots than Griffith does; we think that increasing the cutting pace builds up excitement. Interestingly, however, Suspense uses only a couple of more shots than The Lonely Villa within a comparable running time. (5) We usually expect that American films become more rapidly cut as the 1910s go on, but this isn’t the case here. Shortly, I’ll suggest why.

Smalley and Weber recast Griffith’s parallel editing in several ways. For instance, The Lonely Villa prolongs the phone conversation between husband and wife, building suspense through the husband’s instruction to use his revolver on the thugs. Suspense, by contrast, doesn’t dwell on the telephone conversation but devotes more time (and shots) to the chase along the highway. That’s because Weber and Smalley have complicated the chase by having the husband pursued by the irate motorist and the police, something that doesn’t happen in the Griffith film.

Just as important, Smalley and Weber revise the crosscutting schema through framings that are quite bold for 1913. For example, Griffith’s tramps break into the house in long shot, and they move laterally across the frame.

But Weber and Smalley’s tramp sneaks steadily up the stairs, into a menacing extreme close-up.

Elsewhere, Suspense gives us close views of the wife and of the door as the tramp breaks in. There are oblique angles on the back door of the house, and virtually Hitchcockian point-of-view shots when the wife sees the tramp breaking in and he looks straight up at her.

What struck me most forcibly on watching the film again was the way in which Weber and Smalley’s daring framings serve as equivalents for parallel editing. In effect, they revise the crosscutting schema by putting several actions into a single frame. The most evident, and the most famous, instances are the triangulated split-screen shots. They cram together three lines of action: the wife on the phone, the husband on the phone, and the tramp’s efforts to break into the house (here, finding the key under the mat). (6)

Split-screen effects like this were common enough in early cinema, especially for rendering telephone conversations. Eileen Bowser points out that the three-frame division was one variant, with a landscape separating the two callers. (7) Her example is from College Chums (1907).

But my sense is that in early cinema the split-screen effect was used principally for exposition or comedy, not for suspense. Smalley and Weber have made this framing substitute for crosscutting: instead of giving us three shots, we get one, showing the plot advancing along different lines of action. These splintered frames function much like Brian De Palma’s multiple-frame imagery in Sisters, Blow-Out, and other films. There’s also the nice touch of the conical lampshade over the husband’s head, providing a fulcrum for the composition.

Earlier in the film, instead of crosscutting between the tramp outside and the wife indoors, Suspense gives us both in the same shot, with the tramp peeking in behind her.

A more ingenious revision of the crosscutting schema comes during the shots on the road. Instead of cutting between the father in the stolen car and the police pursuing him, Suspense packs them into the same frame. This is done not only in long shot but also in striking depth compositions.

Flashiest of all are the shots showing the pursuers reflected in the rear-view mirror of the father’s car as he races ahead of them.

Again, a single framing has done duty for two shots, one of the father looking back and another showing the cops coming closer to him. By compressing several lines of action into a single frame, our 1913 film doesn’t need to use significantly more shots than the 1909 one.

These are just a few of the imaginative ways in which Weber and Smalley have recast their standard situation. I could have considered as well the unobtrusive use of the knife as a multi-purpose prop, the echoed shots of mirrors, and the shrewd employment of repetitions in the intertitles.

It would take an early-cinema specialist like Keil to trace out all the connections between Suspense and other films of its era. One among many would be the fact that Griffith wasn’t exactly standing still between 1909, the year of The Lonely Villa, and 1913. For instance, his Battle at Elderbush Gulch, made about the same time as Suspense (though released in 1914), displays far more rapid cutting than Weber and Smalley attempt. Griffith also employed a swelling advance to the foreground, like that of the tramp shot I showed above, in The Musketeers of Pig Alley (1912).

Here, Smalley and Weber seem to have replicated a schema that was believed to ratchet up tension, the so-called looming effect.

The very title of the film exhibits self-consciousness about its artistic purpose. By 1913, it seems, American filmmakers were confident enough in their skills to announce their aims. We want, the title seems to say, to arouse you, to make you wait, hanging there, for the resolution. We know how to tell a twelve-minute story cinematically. The film seems uncannily to anticipate all those one-word titles that Hitchcock invoked to unsettle us–Suspicion, Spellbound, Psycho, Frenzy. As in a Hitchcock movie, the fact that we are pretty certain how Suspense will turn out doesn’t seem to dissipate our anxiety. (For more on this paradox of suspense, see this entry.)

Tunnel vision

It was during the Pordenone 1913 season that the full brilliance of Victor Sjöström’s Ingeborg Holm hit me. I had seen the film a couple of times before and found it deeply moving in its restrained treatment of a poignant situation. The film traces the dissolution of a family. Sven Holm is doing a brisk retail business, but his health problems, along with a thieving clerk, plunge the family into poverty. When Holm dies, his wife Ingeborg must take the children into the poorhouse, and from there they are boarded out to foster families. Her plight worsens over the years, and its sorrowful depths are revealed when her son, now grown, returns to visit her.

Ingeborg Holm has long been recognized as a milestone in European film, not least for its effect on improving Sweden’s treatment of the poor. (8) The acting style is muted and delicate, with no mugging or arm-waving. For long periods, the main actors turn from the audience. (9) Just as impressive are the poised compositions, sustained in unhurried long takes. These give what is essentially a bourgeois tragedy a kind of majestic relentlessness. Watching, I remembered what Dreyer had admired in Sjöström’s Ingmar’s Sons (1918):

The film people here at home shook their heads because Sjöström had really a boldness to let his farmers walk heavily and soberly as farmers do. Yes, they used up an eternity to come from one end of the room to the other. (10)

But sitting in the Cinema Verdi in 1993, I spotted yet another level of artistry. Knowing the story of Ingeborg Holm, I was able to watch the shots unfold. I could study how Sjöström was unobtrusively moving characters so that they became shifting centers of interest. Although he didn’t cut in to close-ups, he harmonized his actors’ movements so that at one point you noticed Ingeborg, at another her husband. Performers spread themselves across the frame, arrayed themselves in depth, turned from or toward the camera. Most subtly of all, one actor might mask another one, driving our attention to other parts of the frame. Sjöström could sustain this intricate play of blocking and revealing for minutes on end.

My favorite example of this tactic is the three-minute passage showing Ingeborg’s daughter and son taken away by new mothers. You can read the whole discussion on pp. 191-195 in On the History of Film Style (1997), but here is an excerpt, with stills intercalated. (These stills, grabbed from the DVD, lack a little on the left. The film has been printed and reprinted so many times, and now fitted to the TV monitor, that it has lost some information along that edge.)

Ingeborg’s entry with her children from the rear doorway establishes the trajectory that will be followed during the scene, as foster mothers come in and take away the children. (Again, the scene is built around movements toward and away from the camera.) In a brilliant stroke, Sjöström immediately plants the young son in the foreground, back to us. The boy will stand there immobile for this first phase of the scene, occasionally serving to block the superintendent.

Ingeborg buries her face in her daughter’s shoulder at the precise moment the foster mother enters from the rear left. She passes behind Ingeborg, and as she is momentarily blocked, the superintendent twitches into visibility, handing the woman a document to sign.

During the signing, when the woman is briefly obscured, the superintendent shifts position again and Ingeborg lifts her face once more. Then Ingeborg and her daughter move slightly leftward as the foster mother comes forward.

This phase of the scene concludes with the departure of the daughter and the embrace of Ingeborg and her son in the foreground, once more concealing the superintendent.

Granted, I picked one of the film’s most exactingly staged scenes, but there are several other examples. Consider the climax, when the adult son comes to visit Ingeborg: the camera position and staging ask us to recall the scene I’ve just mentioned. Or take a look at the handling of the space behind the counter in the various scenes in the Holms’ shop. One instance surmounts this section.

This choreography is hard to catch on the fly: by the time you notice a change, what prepared for it has gone. To understand the overall dynamic, we must reverse-engineer the effects, moving backward from the result to the conditions. As I studied the film, it became clear that creating shifting centers of interest was basic to the scenes’ effects. All the finesse of acting, both solo and ensemble, would come to little if we weren’t primed to watch the most important area of the shot. Directors of the 1910s became supremely skilful in guiding our eye from one major story point to another within the fixed frame.

Some of the tactics of guiding our attention could be borrowed from other arts. Painting and theatre supplied certain schemas, like placing an item in the center of the picture format and turning faces toward the viewer. But some tactics for directing our eye relied on capacities that were as “specifically cinematic” as cutting was argued to be. Theatre is staged for many sightlines, but cinema is staged for a single one, that of the camera lens. That fact allowed directors to organize the unfolding action in depth, prompting the viewer to notice things at many distances from the camera.

It also allowed a precise blocking and revealing of information that would not work on the stage. In a live theatre performance, the slight shifts we detect in the Ingeborg Holm scene would be apparent to only very few viewers; people sitting in other vantage points wouldn’t see them. For another example of this phenomenon, go elsewhere on this blogsite.

There’s a more basic explanation of what’s going on. Cinematic space is a result of optical perspective, like the space of classic painting. The actors may seem to be standing in a cubical space, but in fact they move within a tipped-over pyramid, with the tip resting at the camera lens. They work inside a ground area that’s the shape of a slice of pie. Here’s a reminder from The Black Ball.

When there are figures close to the camera, as here, they fill up more of the frame, indicating that the field of view has narrowed. Filmmakers of the 1910s had discussed this property of their “cinematic stage,” so as researchers we can confirm that the control of position and timing we observe on screen results from the firm intentions of the makers. They might have echoed Uccello, the Renaissance artist who bent over his drawings late at night and exclaimed: “How sweet a thing is this perspective!”

Suddenly the tableau tradition made sense to me. Directors had to organize both the two-dimensional composition and the tapering three-dimensional playing space in front of the camera. The purpose was to create ever-changing centers of interest, laterally and in depth, flowing along with the key moments in the unfolding action.

By studying the same tactics in Feuillade, Bauer, and others, I found that these principles offered a fruitful way to understand much of what was happening in the films of the tableau era. The dynamic of schema and repetition/revision was at work as well, since I found earlier, more rudimentary uses of these principles; these were then refined during the 1910s. Moreover, the idea helped me generalize beyond that era and analyze later directors, including Mizoguchi Kenji and Hou Hsiao-hsien, who also used staging to shape the viewer’s experience. The results can be found in On the History of Film Style and Figures Traced in Light. Those books largely owe their existence to that flash of awareness kindled by Ingeborg Holm. (11)

We can’t reconstruct Sjöström’s stylistic development with much confidence. (12) His first directed film, The Gardener (Trädgärdsmästaren, 1912) survives, but I saw it before I knew how to watch 1910s movies in this way, so I can’t comment on it. Of the twenty-six films he made from 1912 to 1917, nearly all have been lost. Apart from Ingeborg, I have managed to see Havsagamar (The Sea Vultures, 1916). This retains aspects of the tableau tradition, but virtually no scenes display the intricate staging of his 1913 masterpiece.

In the late 1910s, judging by the films we have, Sjöström seems to have moved quickly toward continuity editing in the American vein. (13) The Girl from Stormycroft (Tösen frän Stormyrtorpet, 1917) contains extended passages of analytical editing, and the scenes display a freedom of camera setup one seldom sees in European cinema of the period. Sjöström clearly understands the principles of continuity down to the smallest detail. For instance, he not only uses the frame-edge cut (the cut that lets a character exit one shot and enter another, as the body crosses the frameline); he accelerates it. Our heroine leaves the kitchen, and in the shot’s final frame she hits the right edge. Another director would have had her exit the frame entirely and held the kitchen shot a little longer, to give her time to get to the next room. But Sjöström simply cuts to her already in another room, completing her trajectory.

Sjöström presumes that the constant pace of her approach and the vector of her movement will smooth the shot-change. It does, but few directors of that day would had such confidence that our mind would create continuity of motion.

During this summer’s archive viewing, I spent some time studying Sjöström’s cutting in The Girl from Stormycroft and subsequent films. I may offer some more thoughts later. For now I’ll just say that in these early years we seldom encounter such a versatile director, one who mastered the tableau tradition and then moved, with apparently little effort, to nuanced continuity editing. More generally, examining his technique isn’t a dry exercise. We never lose by studying craftsmanship. Sjöström’s films succeed because his style carefully guides our moment-by-moment apprehension of the heart-rending stories he has chosen to tell.

Have I convinced you to surrender your credit-card number? Suspense is available in a dazzling Flicker Alley collection, Saved from the Flames, compiled from the French series Retour de flame. Ingeborg Holm comes with a tinted print of the estimable Terje Vigen (A Man There Was) on a DVD from Kino. The latter disc includes vibrant scores from Donald Sosin and David Drazin.

(1) For the purposes of the outstanding Encyclopedia of Early Cinema (New York: Routledge, 2005), the editor Richard Abel considers that the early cinema constitutes the period from film’s invention ca. 1894 to the mid-1910s.

(2) Silent films were still made into the 1930s, notably in Russia and Japan—a situation that shows the rough-and-ready quality of period divisions.

(3) For an informative series of articles around the 1913 theme, see Griffithiana no. 50 (May 1994).

(4) Gombrich proposes these ideas at various points in Art and Illusion: A Study in the Psychology of Pictorial Representation, new ed. (Princeton: Princeton University Press, 2000). The book is a milestone in twentieth-century humanistic inquiry.

(5) The Lonely Villa (1909) has, in the version I’ve seen, 54 shots in a 750-foot reel (that is, about twelve and a half minutes at 16 frames per second). Suspense has nearly the same number of shots, 56, in about the same running time.

(6) Sharp-eyed readers will recognize one of these shots as Fig. 5.82 in Film Art: An Introduction, 8th ed. (New York: McGraw-Hill, 2008), 187.

(7) Eileen Bowser, The Transformation of Cinema 1907-1915, vol. 2 of History of the American Cinema (New York: Scribners, 1990), 65.

(8) For more background on the film, see Jan Olsson, “‘Classical’ vs. ‘Pre-Classical’: Ingeborg Holm and Swedish Cinema in 1913,” Griffithiana no. 50 (May 1994), 113-123.

(9) Kristin wrote about this technique in “The International Exploration of Cinematic Expressivity,” in Film and the First World War, ed. Karel Dibbets and Bert Hogenkamp (Amsterdam: Amsterdam University Press, 1995), 65-85. She also discusses it in our Film History: An Introduction, 2d ed. (New York: McGraw-Hill, 2002), 67.

(10) Carl Dreyer, “A Little on Film Style,” Dreyer in Double Reflection, ed. and trans. Donald Skoller (New York: Dutton, 1973), 133.

(11) I was probably primed for this by lectures presented by Yuri Tsivian, who has long been studying mirrors in 1910s cinema and calculating how they revealed offscreen space.

(12) Detailed information on Sjöström’s relation to the film industry in these years can be found in John Fullerton’s epic Ph. D. thesis, The Development of a System of Representation in Swedish Film, 1912-1920 (University of East Anglia, 1994). See also Fullerton, “Contextualising the Innovation of Deep Staging in Swedish Film,” Film and the First World War, ed. Dibbets and Hogenkamp, 86-96.

(13) Several people have analyzed Sjöström’s editing. Ben Brewster and Lea Jacobs’ Theatre to Cinema: Stage Pictorialism and the Early Feature Film (Oxford: Oxford University Press, 1997), 133-136 shows how cutting supports the acting in a crucial scene of Ingmar’s Sons. Tom Gunning’s essay “‘A Dangerous Pledge’: Victor Sjöström’s Unknown Masterpiece, Mästerman,” in John Fullerton and Jan Olsson’s anthology Nordic Explorations: Film before 1930(Sydney: John Libbey, 1999), 204-231, argues that Sjöström’s cutting gives his characters a degree of psychological opacity. Most recently, in an unpublished paper Jonah Horwitz discusses patterns of performance, composition, and editing in Terje Vigen (1917).

Twilight of a Woman’s Soul (1913).

PS 31 August: Some corrections. David Cairns writes to point out that the triangle at the top of the frame in the Suspense split-screen isn’t a lamp but simply the shade behind the father’s head, cropped by the masking. Wishful thinking on my part, I’m afraid. By the way, David has an intriguing movie giveaway going on his site.

Roland-François Lack corrects dates on two of my Greatest French Hits from 1960: “The great Marker from that year is Description d’un combat, not Lettre de Sibérie (1958), and likewise Rouch’s Moi un noir is from 1958.” Thanks to him and David. To my original list, I should probably have added Cocteau’s Testament d’Orphée, released in 1960.

Comments are closed.

David Bordwell
top of page

have comments about the state of this website? go here