The Important Umbrella

screenshot_2474Not long ago a journalist asked me to name what I considered the most important film of the last 15 years and I somewhat perversely suggested Pixar’s short The Blue Umbrella (2012) directed by Saschka Unsled. I’m not really a fan of “important” films because the term too often implies overt social relevance rather than artistic/cinematic/film-historical value. Surprising no one, my vote didn’t make the journalist’s cut — romance among umbrellas is not yet a burning issue, apparently.  So I will make the claim here, in blog form. I’m elevating these seven minutes of cinema to the pantheon of “importance” because I think they develop the medium in a fundamental way. To be precise, a fleeting moment of graphic change near the end of the film, about four seconds long, might have implications for how we understand cinema.


screenshot_2454I only know about The Blue Umbrella because my former student (and blogger) Lilly Holman brought it to my attention during her senior thesis research . The short tells an unassuming story of two umbrellas in love: Blue and Red exchange glances on the street, a gust of wind pulls them apart, and they are reunited. I think Blue Umbrella ranks with the best of the Pixar shorts. Like most of them (don’t judge the form by Lava) it is elegantly crafted, touching, and powerfully visual (not a line of dialogue). It has the clean and simple storytelling of Luxo Junior, the tactile world building of Tin Toy, and the emotional flow of Partly Cloudy. Holman can fill you in on why all of these films deserve our attention on her Manic Pix(ar) Dream Girl vlog series.

The movie is also a bravura technical experiment in photorealist animation. The opening shots mislead viewers, using framing, pacing, and a slight camera jiggle to convincingly emulate location shooting. A mailbox appears to be filmed through a telephoto lens from across the busy street as passing cars make blurry intrusions in the foreground.

screenshot_2455 screenshot_2456 screenshot_2457

The closer shot is zoomed in and unsteady (in the illustration below, the mailbox moves slightly down and up in the frame) as though the cinematographer struggled to keep the composition from a distance.

screenshot_2458 screenshot_2459 screenshot_2460

The painstakingly detailed world is indistinguishable from live-action. Until the objects start moving in impossibly expressive ways (the mailbox smiles), most spectators assume that Pixar’s artists have abandoned their digital workstations for an afternoon on a Manhattan street.  Only the umbrellas’ minimally drawn and faces belong to what we might conventionally call a “cartoon.” Limited to eyes and a mouth, the schematic designs have just enough detail to suggest gender: Red’s eyelashes strike me as feminine (as does her conventional role as the desired object, but that’s a different blog).

screenshot_2461In its realism, The Blue Umbrella erases the boundary between art and representation; it turns technology that we usually associate with high-end special effects into a completely synthetic and verisimilar world. The look lends gravity to the love story and the love story lends poetry to the everyday. Storefronts, sidewalks, mailboxes, and sewer grates gather feeling.

This is all clever and moving film making, but the moment of importance occurs at the end. Beaten and disheveled, the Blue umbrella rises from the pavement and a red sheen of light passes over it. He looks up and discovers that the Red umbrella has entered the frame, her color reflecting off his surface. The subtle change in color underscores their reunion; warmth and love suffuse the frame just before Red comes into view.

screenshot_2462 screenshot_2463 screenshot_2464  screenshot_2466 screenshot_2467 screenshot_2468

It is an emotionally charged play of color temperature that harkens back to Walt Disney’s experiments with Technicolor in the 1930s, as when the warm glow of Mother Mouse’s hearth beckons her wayward son back home in The Flying Mouse. The Blue Umbrella’s widescreen sheen more closely resembles Robert Elswit’s cinematography in Punch Drunk Love (2002), but it participates in a long history of color control in animation.

screenshot_2472 screenshot_2473

The “cartoon” can now take on a photoreal cinematographic skin. This is beautiful, but not obviously important.

The truly groundbreaking thing is that the animators did not plan this color effect, they discovered it. The Blue Umbrella employs Pixar’s “Global Illumination” system, a program that emulates the physical characteristics of light in an artificial space. The color effect takes place because, essentially, a real red umbrella held over a real blue umbrella under street lamps would create a warm glow. As Unsled explains to The Verge: “We had the moment when the red umbrella is being held over the blue, and there’s this reflection on blue which is the bounce of her light. That was never planned for. That wasn’t boarded like this, it wasn’t something we gave direction to for the lighter…. We got something and were like, ‘Oh, let’s use that because it’s beautiful.’ ”

This strikes me as more than an impressive technical achievement–it might be a historical turning point. One quality that separates photography from drawing, or cinema from animation, is film’s openness to contingency, to capturing an unexpected or unintended event. Film historian George Sadoul famously observed that the audience for the Lumiere Brothers’ first public projection of motion pictures in 1895 were captivated by “the trembling of the leaves through the action of the wind… of nature caught in the act” as much as they were by the baby eating his breakfast in the foreground.

feedingTheBaby-imageCameras capture the contingent, the transient details of a world in flux. Ephemeral, unnoticed texture seeps into film, sometimes despite the filmmaker’s intentions. This quality was the cornerstone of André Bazin’s aesthetic theory. He proclaimed, “All the arts are based on the presence of man, only photography derives advantage from his absence.” In photography’s power to “lay bare the realities,” to record an event rather than construct an image, Bazin saw the prospect of a unique art form. His favorite filmmakers, particularly the Italian neorealists, coaxed meaning from the world rather than imposing it; the camera, in Bazin’s view, was like a magnet passing over and collecting the iron filings of reality. Vittorio De Sica, for instance, created the drama of The Bicycle Thieves in collaboration with the living texture of the Roman street.

Bicycle Thieves

François Truffaut, Bazin’s student and a vocal proponent of “faith in reality,” similarly celebrated accidental details caught by the camera. He found “the height of truth” in Debbie Reynolds’ unconscious gesture when she adjusts her skirt to cover her leg at the end of the “Good Morning” number in Singing in the Rain. Though her character wouldn’t have cared, the actress revealed her sense of propriety before the camera. Such interweaving of reality and illusion would seem impossible in a cartoon. If Betty Boop or Jessica Rabbit adjust their skirts, they do so only at the will of the animator.

Singing in the Rain

This is not the only rationale for seeing film as an art (though some of Bazin’s followers would claim otherwise), but since its inception only live-action cinema could boast “happy accidents” from the wind in the trees on a location to the gestures of an actor on a sound stage. Thousands of unanticipated micro-details give real life to the moving image.

When Red passed over Blue, Unseld stumbled across an unanticipated revelation. He observed a play of light that he did not place there. These few milliseconds of The Blue Umbrella ease this line of demarcation between live action and animation. In a small way, perhaps the digital world crafted by Pixar has taken a step toward passing a cinematic Turing test. A transitory and fleeting fact of the world emerged on film (or in the digital render), and the artist found a way to make it serve the story. In a small way, perhaps the world has changed.


Posted in Uncategorized | Comments Off on The Important Umbrella

The Great Severity of Howard Hawks

Poster - Air Force_02

Bemoaning the state of Soviet film in the early 1930s, Sergei Eisenstein asked if the term “sound-film” meant that “what you see while you’re listening does not deserve attention?” He demanded a return to silent cinema’s “great severity of form” in which each shot was admitted into a sequence “with as much care as a line of poetry is admitted into a poem, or each musical atom is admitted into the movement of a fugue.”

On the face of it, we wouldn’t expect Eisenstein’s call to be heeded by Howard Hawks, Hollywood’s master of rhythmic speech. This severity of artistic form might seem even less likely in a propaganda effort like Air Force (1943), but the film evinces the deliberate precision of a carefully crafted poem. Hawks achieves this visual grace without announcing it, exemplifying what David Bordwell calls the “quiet virtuosity” of Hollywood cinema.

Consider the brief sequence in which the crew of the Mary Anne (a B-17 bomber caught up in the aftermath of Pearl Harbor) listens to Roosevelt’s declaration of war over their radio headsets. [You can find the scene here.] In 15 shots Hawks unites the group, signals their personal stakes in the mission, and connects this crew to the war as a whole. The scene cuts between the crewmembers grouped in their assigned parts of the plane. Captain Quincannon (John Ridgely) and his co-pilot Williams (Gig Young) share the cockpit;


bombardier Lt. McMartin (Arthur Kennedy), navigator Lieutenant Hauser (Charles Drake), and pursuit pilot/passenger Lieutenant Rader (James Brown) gather around the navigator’s desk at the nose;


Corporal Peterson (Ward Wood) and Private Chester (Ray Montgomery) listen at their radio post in the mid-section;


and mechanic Sergeant White (Harry Carey) joins gunners Corporal Weinberg (George Tobias) and Sergeant Winocki (John Garfield) near the tail.


By this point, just under an hour into the film, we are comfortably familiar with the characters and the layout of the Mary Anne. It is a testament to Hawks’ handling of a multiple-protagonist plot that we recognize each crewmember on sight, know his job on the ship, and understand his personal stakes. We have learned the ropes, joined the unit.

Having laid this physical and psychological groundwork, Hawks reduces his means of expression. He cycles through the character groupings three times during the speech as they react silently, save for a single line of dialogue. During the first pass, the men settle in, take up their headsets, and begin to listen.


When applause interrupts Roosevelt’s address, White remarks: “I hope he tells us something about the Philippines, my son’s at Clark Field.” Hawks prefers medium double and triple shots but cuts twice to a medium single of Winocki, the rebellious sergeant who plans to quit the military played with cool ambivalence by Garfield. An understated exchange activates the scene’s dramatic, personal thread. Winocki and White are at odds: the young man sees no future for an enlisted man, while the weathered mechanic remains devoted to advancing his family name in the military by way of his pilot son. In our first single of Winocki he glances from White to Weinberg, checking their reactions as we watch his.

screenshot_2004 screenshot_2002

screenshot_2003  screenshot_2004

White mentions his son in a two-shot that groups him with Weinberg who shoots him a look of concern, and then we return to Winocki.


He glances up, holds his gaze for a moment on the off-screen White, and glances down.


With this Hawks establishes a parent’s concern for his family, the group’s concern for White, and Winocki’s awakening to the stakes of the fight. All this with 13 spoken words, a glance up, then down.

Having broadened the sequence’s formal repertoire just enough to sound these notes, Hawks narrows his means. From here on, no character speaks to or elicits a reaction from another. Instead, glances around and out of the frame, small gestures, and gentle camera movements tell the story. Hawks’ second rotation groups actors in wider shots. In the cockpit Quincannon glances in Williams’ direction when the copilot adjusts his posture.

screenshot_2007 screenshot_2008

At the navigator’s desk Rader exhales a puff of smoke and glances back toward McMartin who straightens his back a bit.

screenshot_2009 screenshot_2010

At the radio, Peterson adjusts the controls while Chester leans forward and glances off down the plane’s corridor.

screenshot_2011Where the other characters are occupied with the speech and with their immediate surroundings, Chester’s look reaches out, connects space, and registers concern for the crew. It is a minimal hint of group empathy that builds on Weinberg’s glance to White at the start. We know from the layout of the plane that Chester is looking toward the tail, but rather than complete the circuit with a point-of-view shot or eye-line match, Hawks returns us to the cockpit. Sealing the glance to its object would seem too firmly sentimental, perhaps too personally focused, for this sequence. The point is that the men care for one another, not that Chester is worried about White. Hawks asks viewers to lean in and watch carefully for the emotion beneath the procedure.

But distance isn’t indifference. At the same time he teases and denies the easy empathy, Hawks amplifies our investment. The final round through the plane begins with characters in close-up. Back in the cockpit, Gig Young scales his performance as Williams to the tighter frame. He moves his hand from his ear and looks rightward. Hawks’ camera follows this gaze, panning to discover Quincannon whose reaction is even more minute. Where Williams looks to his leader, the Captain fixes on the horizon. Ridgly limits his acting to a slight wince.

screenshot_2012    screenshot_2013

screenshot_2014   screenshot_2015

The pattern continues when we cut back to the navigation desk where the camera frames McMartin in medium close-up before dollying back to reveal first Hauser and then Rader. At shot’s end McMartin grips his headphone cord and glances down while Hauser tilts his head upward catching an eyelight.





The gesture is in precise coordination with Roosevelt’s speech. As the president intones “So Help Us God,” Hauser glances up. We know, however, that he is also looking in the direction of the picture of his father that he has taped above the desk, just off screen. Hawks’ tactile environments enable spatial precision and subtlety. We know that Hauser lives in the shadow of his father, a WWI ace. He may be looking to God, and certainly the film’s propaganda mission benefits from connecting the soldier to a higher power. On the other hand, Hauser’s gaze activates a meaning within the sealed world of the airplane (if just out of frame). White’s vocalized concern for his son echoes in the navigator’s desire to live up to his father’s image.

These choices build to an unexpectedly powerful four-shot climax. To appreciate his achievement, it helps to imagine alternatives to Hawk’s style. A director might play up the content of Roosevelt’s speech with pregnant cutaways. In this regard, it seems significant that Hawks omits the most famous line of the address, in which Roosevelt refers to a “day that will live in infamy.” But what remains could support more pointed imagery. The president’s line “I regret to tell you that very many American lives have been lost” is ripe for embellishment, from a close-up of the wincing captain to a crosscut back to the burning airstrip we’ve just left in Hawaii. Hauser’s glance off screen might lead to an insert shot of his father’s photo, a reminder that an earlier generation also gave their lives. The characters, lost in somber reflection, might still exchange a few words during Roosevelt’s pauses, helping to articulate their resolve or anger. If Hawks were willing to break from the world of the airplane, the speech could be the occasion for a sweeping montage of Americans from different walks of life, all posed next to their radios in nervous anticipation.

In fact, Hawks does relatively little to align specific images with specific lines from the address. The president’s words float through the plane as a general, shared experience. Roosevelt sets a rhythm for movement and cutting, but doesn’t dictate the content of the shots. This leaves room for emphasis. As Hauser glances upward we get a first indication that word and image are drifting into alignment.

When we return to Winocki, Weinberg, and White in the tail, Hawks lands his punch. The composition holds the characters in medium shot, White in the foreground right, Winocki at midground left, and Weinberg further back and between them.

screenshot_2020Our attention is free to wander between listeners. Light and movement favor Winocki as he polishes his gun, but White is largest in the frame and Weinberg the only figure facing the camera. As Roosevelt nears his conclusion, “I ask, that the Congress declare…”, background movement flags Winocki’s portion of the frame. Private Chester, who has heretofore been leaning unnoticed against the fuselage near the radio, takes a seat directly behind Winocki.

screenshot_2021The moment is a “Hawksian” reminder of continuous space and action; it makes concrete the contiguity implied by Chester’s earlier glance. At the same time, this bit of staging guides the viewer’s eye to Winocki exactly at the moment of Roosevelt’s conclusion: “…a state of war.” The sergeant who began the film declaring that he was quitting the army looks up, but only slightly.

screenshot_2022If we missed it, Weinberg is there in the center to direct us by shifting his eyes leftward and locking them on Winocki.

screenshot_2025More meaningfully, White shoots Winocki a quick glance, checking the impact of Roosevelt’s words on him, and then looks away. The wizened old-timer doesn’t waste much time contemplating the soldier’s awakening, and neither do we.

screenshot_2024All of this transpires in the final second of an unmoving 9-second shot. With more or less static and unspeaking characters, deliberate pacing, and compositional precision, Hawks attunes the viewer to minute change. Drama is nearly microscopic.

After the slow unfolding of our previous compositions, the sequence closes with startling directness. The lengths of the previous three shots decrease from 15 to 9 seconds, but now Hawks keeps them under 5 and trades staging and camera movement for cutting. With applause on the radio, we return to a close-up of Williams in the cockpit. He turns, glancing once again toward Quinncannon and we cut to the captain, gaze still fixed on the horizon.


Without moving his head, Quinncannon shifts his eyes rightward and we cut to the object of his look: the tiny toy aviator charm that his son gave him, dangling from the control panel.


These decisive edits complete the pattern of glances off-screen and provide the scene’s few unambiguous eyeline matches. Williams’ glance to his captain ties off the thread of crewmembers looking to one another, and Quincannon’s eyeline match crowns the sequence’s paternal theme. The military unit’s inward compassion, their care for one another carried by glances and look, will culminate in the cutting around Quincannon’s deathbed late in the film. [You can see that scene here.] Meanwhile, the allusions to sons and fathers signal concerns that reach beyond the plane and reinforce the quasi-familial structure among the crew. If Roosevelt speaks to and for the nation, the images anchor patriotism to the group’s love for family and one another. The toy aviator, the scene’s most emotionally specific shot, dissolves to a transitional exterior long shot of the Mary Anne slicing through clouds and into a storm. Hawks leads us to the individual in order to emotionalize the whole. It is going to hurt when the Mary Anne breaks apart. We are primed for Winocki’s attempt to save it.


The scene’s premise is simple and its means elemental. It works as a cinematic haiku. Untold depths of feeling emerge from austere form. Hawks could easily have given viewers something to watch while they were busy listening, treating the image as an adjunct to sound. Indeed, we might not recognize the filmmaker’s care, attributing the scene’s emotional energy to Roosevelt’s speech or to our personal historical experience. This is Hawks’ (and Hollywood’s) own version of cinema’s great severity, made all the more potent by going unnoticed.




Posted in Uncategorized | Comments Off on The Great Severity of Howard Hawks

“Digital Seriality” — Panel at #SCMS15 in Montreal


At the upcoming conference of the Society for Cinema and Media Studies (March 25-29, 2015 in Montréal), I will be participating in a panel on “Digital Seriality,” co-chaired by Andreas Jahn-Sudmann and Scott Higgins, along with Dominik Maeder and Daniela Wentz.

Here is our panel description, along with links (below) to the abstracts for the various papers:

Digital Seriality

Seriality and the digital are key concepts for an understanding of many current forms, texts, and technologies of media, and they are implicated in much broader media-historical trajectories as well. Beyond the forms and functions of specific cultural artifacts, they are central to our global media ecology. Surprisingly, though, relatively few attempts have been made at thinking the digital and the serial together, as intimately connected perspectives on media. This is precisely the task of the present panel. On the one hand, the papers interrogate the serial conditions, forms, and effects of digital culture; on the other hand, they question the role of the digital as technocultural embodiment, determinant, or matrix for serialized media aesthetics and practices. The panel thus brings together heretofore isolated perspectives from studies of new media culture (cf. Manovich 2001, Jenkins 2006) and emerging scholarship on seriality (cf. Kelleter 2012, Allen and van den Berg 2014).

Seriality and digitality are understood here in terms not only of their narrative/representational manifestations but also their technical-operational impacts on our media environments. Accordingly, Shane Denson and Andreas Jahn-Sudmann’s paper looks to the case of the Xbox One in order to show how computational platforms affect the serial forms and practices emerging within, among, and around digital games (“intra-,” “inter-,” and “para-ludic” serialities; cf. Denson and Jahn-Sudmann 2013), but also how these platforms inscribe themselves – as a serialized factor in their own right – into the parameters of computational expression. Whereas video games serve here to highlight the differences between digital and pre-digital serial forms, Dominik Maeder approaches things from the opposite direction, arguing that the interfaces of Netflix, Hulu, and other digital streaming services embody a form of spatio-temporal serialization that, already anticipated by TV series, is closely related to (pre-digital) televisual seriality. As a complementary perspective, Daniela Wentz’s paper shows how certain TV series anticipate their own digital afterlives in the form of fan-made gifs and memes. Finally, Scott Higgins provides an “archeological” perspective, exploring the ludic dimensions of the operational aesthetic, which anticipates computer games in pre-digital forms, thus offering a fruitful case for rethinking digital seriality from a media-comparative perspective.


Allen, Robert, and Thijs van den Berg, eds. Serialization in Popular Culture. London: Routledge, 2014.

Denson, Shane, and Andreas Jahn-Sudmann. “Digital Seriality: On the Serial Aesthetics and Practice of Digital Games.” Eludamos: Journal for Computer Game Culture 7.1 (2013): 1-32.

Jenkins, Henry. Convergence Culture: Where Old and New Media Collide. New York: New York UP, 2006.

Kelleter, Frank, ed. Populäre Serialität: Narration – Evolution – Distinktion. Zum seriellen Erzählen seit dem 19. Jahrhundert. Bielefeld: Transcript, 2012.

Manovich, Lev. The Language of New Media. Cambridge, MIT, 2001.

Finally, here are links to the individual abstracts:

Shane Denson and Andreas Jahn-Sudmann, “The Xbox One as Serial Hardware: A Technocultural Approach to the Seriality of Computational Platforms”

Dominik Maeder, “Serial Interfaces: Publishing and Programming Television on Digital Platforms”

Daniela Wentz, “The Infinite Gesture: The Serial Culture of the Gif”

Scott Higgins, “Ludic Operations: Play and the Serial Action Sequence”

Posted in Uncategorized | Comments Off on

Digital Seriality

Here’s a panel that I’m participating with during the SCMS meeting next March in Montreal. Shane Denson has posted all the info on his blog:

Posted in Uncategorized | Comments Off on Digital Seriality

Republic Lands the Punch: Art of the Serial Dust-up


Fisticuffs were an obligatory component of almost every serial, but Republic gained distinction for developing the form into a precise articulation of force and agility. Columbia and Universal serials tended to cede pacing and presentation of fights to casually organized stuntmen, covered by a more or less distant camera.

Universal In Action


Ford Beebe and Ray Taylor’s The Green Hornet (1940) offers a good example of Universal’s fighting style. The serial as a whole places less emphasis on fisticuffs as an attraction. Beebe and Ray tend to stage brief scuffles that interrupt standoffs or lead to car chases. When the Green Hornet (Gordon Jones) and Kato (Key Luke) do fight, they tend to struggle with henchmen in close formation in long to medium-long shots. For instance, in chapter 5, Time Bomb, the hero fights three racketeers in a garage as (unbeknownst to the Green Hornet) the clock runs out on a time bomb in a nearby car. screenshot_1172

Green Hornet punches out two assailants when a third, armed with a wrench, attacks. As they struggle, Kato rushes toward the garage and a gun heavy revives enough to take aim at Green Hornet. Kato arrives just in time to blast the heavy with his gas gun and Green Hornet overpowers the wrench heavy, just before the garage clock reaches 11:00 and the car explodes.

Beebe and Ray cover the 50-second fight in 26 shots for an average shot length of just under two seconds. The speed of cutting, though, does not guarantee kinetic engagement. Most of these cuts are away from the fight to intensifying details like the garage clock inching its way to the 11:00 deadline, a heavy preparing to enter the fray, or Kato rushing to the rescue.



The physical struggle between The Hornet and the henchmen receives little cinematic elaboration. Beebe and Ray film the action from three camera positions: one long shot and two medium shots. Shot scale is dictated by practical concerns.When Green Hornet grapples with a single assailant, medium shots prevail. If a third participant should enter the frame, or the action occurs closer to the ground, the editor selects a long shot.


Since the camera constantly reframes during the fight, the dominant impression is that cinematographers are working just to capture the staged event in an efficient and legible way. Only three cuts articulate the fight by lending punctuation and emphasis to a moment. When Green Hornet swings two of his attackers to the ground in medium shot, a match on action to long shot allows us to follow movement to the floor, and the graphic change accentuates the fight’s development. More effective is a cut from medium to long shot that punctuates Green Hornet’s left hook to a heavy’s chin. The edit is timed with an audible snap on the foley track, and it is one of the few affective moments in the fracas.


The third articulating edit is something of a jump cut. When the wrench heavy enters the frame and raises his weapon to strike, we move suddenly from a long to medium shot that brings both figures closer, obscuring his attack but amplifying it nonetheless.


The hand-to-hand combat isn’t wholly uninteresting, but it reflects Universal’s priorities. The editors are more precise when cutting between contiguous spaces. When Kato fires his gas gun and his target collapses we are treated to two distinct bursts of action in two shots, each with their own beat. Action within the frame can be treated as so much filler, quickly staged and easily captured.


The Columbia Method

Columbia serials follow much the same model as Universal’s; cinematography and editing tend to follow the staged event. In an effort to inject dynamism into fairly rote dust-ups, the Columbia unit used more undercranking (running the camera at a slow speed so that the resulting footage appears fast) than other producers. The effect is at the forefront of James Horne’s Captain Midnight (1942).


Horne treated henchman as bumbling comic relief characters, and though his fight scenes were “straight” the fast motion could lend the action a light quality. When Midnight (Dave O’Brien) is set upon by Ivan Shark (James Craven) and three of his henchmen on a country roadside in chapter 9, Horne sticks to long and medium shots, cutting away for comic asides, and shifting shot scale to accommodate the action. Columbia’s stunt team stages the fight more broadly than The Green Hornet, and they pack several combinations into each shot. One shot begins with a henchman clocking Midnight with a right hook, continues as Midnight returns the blow, throwing his assailant out of the frame, then pans left to follow our hero as he grabs a second heavy by the shoulders, tosses him rightward, and spins to face a third henchman who surprises him from the rear.


A cut back to a long shot opens room for Midnight to fall on his back from the punch, and then catch his attacker with his feet, flipping him over his head. Just as Midnight stands up, another heavy enters and pulls him into striking range before finishing his punch in the next, closer, shot.


Each take delivers two or three small stunts and sets up an action that bridges the next edit. Undercranking speeds up the choreography and helps mask the contrived staging which sets the stuntmen up for each gag. The fighters also telegraph the punches they throw by swinging out from their shoulders and exaggerate the impacts they receive with a good deal of tumbling and rolling. They tend to use more space within the frame than their Universal counterparts.  

Later Columbia serials, produced by the notoriously cut-rate Sam Katzman, retain the undercranking for even looser choreography. In Jack Armstrong: All American Boy (Wallace Fox, 1947) and Blackhawk (Spencer Gordon Bennett and Fred. F. Sears, 1952), for instance the stunt teams stage group fights with multiple assailants and defenders within a single frame. Fast motion and constant movement give the fights a “free—for-all” quality that can degenerate into disorganized flailing. The images below from a single take of a single fight in Blackhawk illustrate the late Columbia method. Note that in the first frame, the heavy reacts to a punch before our hero can throw it, while in the fourth, a stuntman has time to adjust his hat before his next engagement. This is the fighting style lampooned by the camp Batman and Robin television series in 1966.


Our Heroes: Witney, Sharpe, and Republic


The fights and stunts at Republic in general, and particularly in Daredevils of the Red Circle (1939), achieve a level of craft polish well beyond the competitors. In his autobiography, Republic Director William Witney (pictured above with his wife Maxine Doyle) takes credit for innovating a new style of shooting the sequences. In a description that well suits the Universal style, Witney writes:

I was never satisfied with the way movie fights were shot. …The stuntmen staged the fights, and they stunk. … The fights always seemed to be okay for the first punch. Then the stuntmen were always out of place for the next punch. By the time three or four minutes had passed, the stuntmen were out of breath, scattered all over the set and seemed to be staggering around waiting for someone to hit them. (135)

Witney was inspired by a visit to the Warner Bros. lot around 1937, where he watched Busby Berkeley at work. Berkeley had broken a long dance into short segments and intensively rehearsed sequences of steps. Once the forty dancers had mastered the steps to the point “you could have shot a bullet down the line and not hit anyone,” they were sent to wardrobe and makeup while Berkeley’s assistant “shot a close-up of one of the leads doing the same dance steps that the other girls had done.” (136) Witney and his parter John English decided to bring the “rehearse, break, close-up” method to Republic’s stages. Hurst suggests that Daredevils of the Red Circle was the first serial for which Witney choreographed his own fights in this manner, though he offers no citation. Witney describes the process thusly:

Each cut might be only fifteen seconds: a punch, cut, a fall over a chair, cut, a charge into someone over a desk, cut. Each time you saw “cut” in the lines above represents a close-up of one of the leads.

The stunt people caught on fast. It made their work easier. A fall over a table could be done with precision and without the chance of being off balance as they hit the table…. And after a few walk-throughs, everyone knew exactly where they should be at all times. There was no more wandering around the set looking for someone to hit them so they could fall down.” 

Witney’s technique roughly resembles “segment shooting” which David Bordwell identifies with Hong Kong martial arts cinema from the 1960s on. Facing small budgets, short schedules, and high demand Hong Kong directors found it more efficient to rehearse and film combat in short segments, setting camera angles and constructing the fight as they went. (Planet Hong Kong 129) As Bordwell notes, this method is labor intensive, requiring long continuous workdays from the crew, but it yielded more “tangible benefits” than spending equal time on story construction and screenwriting. (130). The Republic fights are not as painstakingly precise as those of the Hong Kong masters, and it is almost certain that schedules were at least as tight for serials as for Kung Fu films. Still, one senses a kindred “kinesthetic artistry” in the work of Republic’s stunt team. (199)


The studio’s stunt crew originally formed around the incomparable Yakima Canutt (above), who worked for Mascot when Republic formed. Though he continued to stunt and direct for Republic, by the late 1930s Canutt was in high demand and already considered Hollywood’s premiere stuntman. In 1939 alone he appeared in Stagecoach, Young Mr. Lincoln, Dodge City, and Gone with the Wind, and 16 other productions. His protégé at Republic was former vaudeville acrobat, juggler, and national tumbling champion, David Sharpe (below).


According to William Cline, Sharpe took leadership of the Republic team between 1939 and the start of WWII. (In the Nick of Time, 162) In industry parlance, the lead stuntman was known as the “ramrod,” a crew chief who organized the other performers as instructed by the director. Cast as a lead in Daredevils, and therefore indispensable to the production, Sharpe was himself doubled in the film by Jimmy Fawcett, another vaudeville acrobat who started at Republic that year. (Mathis, Valley of the Cliffhangers 111) Still, Daredevils owes some of its dynamism to the fact that Republic’s ramrod was a major player. Sharpe brings physical panache to routine exposition scenes, as in chapter 3 when he sits down to the breakfast table by effortlessly leaping over the back of his chair.



The other stunt doubles on Daredevils included veteran professionals Cy Slocum and George DeNormand (for Gene), and Ted Mapes (Tiny). Together, they created gag-packed energetic routines in which each hero was fully individuated.


Many of the Daredevils’ fights develop into wide-ranging chases as each of the trio engages a fleeing henchman or two. The dust-up in chapter 6 Thirty Seconds to Live is contained to a single room and so offers a nice comparison to the Universal and Columbia sequences. The trio has been dispatched to investigate Professor Seldon (Stanley Price) who supplies scientific secrets to criminal mastermind 39013 (Charles Middleton). Seldon tricks the three and locks them in a storeroom from which they almost immediately escape. In the meantime, four of 39013’s henchmen interrogate Seldon and threaten to eliminate him. To hold his assailants at bay, Seldon reveals a contact switch bolted to the edge of his desk and announces: “There’s a time bomb planted in this room, and in just 30 seconds after I throw this switch, it will blow up.” One of the henchmen (a gun heavy) shoots Seldon in the stomach and he falls to the ground. The Daredevils have been spying on this confrontation from the doorway, and they immediately burst into action. Twenty-nine shots fly by in 63 seconds and Gene, Burt, and Tiny engage the four henchmen. Unlike the Green Lantern example, only two shots might be called “cutaways,” a medium shot of Seldon writhing on the ground and a quick close-up of a heavy knocking the contact switch closed. All of the remaining shots feature the heroes punching, flipping, and wrestling the heavies, and this action is covered from 13 different camera positions. Clearly, Witney and English have invested more time and creative energy in their brawl than their Universal and Columbia counterparts.

Professor Seldon’s laboratory is a simple three-wall set with his desk near a window at the far left and a doorway on the opposite wall at right. The rear wall is lined with what appears to be radio equipment, and a second desk is placed near the doorway. This wide set design affords plenty of floor space in its center, as well as desks, chairs, and cabinets for the stuntmen to knock into. When our trio burst through the door, they have room to race leftward in a panning long shot, picking up momentum as they near the heavies at Seldon’s desk.



All three leap into the air at precisely the same time. In a flamboyant opening maneuver, Bert uses his running start to dive over the desk and bounce across the back of a thug on the other side. Bert’s (in this shot played by Jimmy Fawcett) altitude is breathtaking; he clears the desk, a tall lamp, and a henchman standing beside the desk in a full shot that rules out any hidden trampolines or other devices. 


As Bert lands at the left, Tiny punches two heavies down to the floor in the background, beyond the desk, while Gene drags a foreground thug off frame right by his leg. In a single fluid movement and less than two seconds, the Daredevils have floored all four thugs. Choreography distributes action across all areas of the. Bert’s leap activates the air above the desk and pushes our attention leftward, at the same time Tiny uses the background to sweep his thugs downward and in the foreground Gene whisks his target in the opposite direction. The timing is immaculate. The camera pans back right as Bert on the left and Gene on the right rise upward in symmetry to reengage their opponents.


 The entire shot runs just under five seconds, and in that time the viewer’s attention courses first to the left, then toward the background and finally it is pulled between to synchronized foreground actions at either edge of the frame. It may be too swift for us to track each individual, but the flurry of action high and low, left and right, fore and back, catches us up in a smooth kinetic movement.

This wide shot works as a master: a camera position that takes in all of the action from a distance by panning and reframing. But it isn’t an all-purpose composition. The framing of the opening leap is too precise for that, and the very set furnishings have been adjusted for maximal clarity and dynamism in this shot. Four shots later, when we cut back to a seemingly identical wide composition, a large worktable loaded with electrical equipment suddenly appears in the middle of the room, between Seldon’s desk and the door.

screenshot_1146 The table materializes just in time for Gene to crash into it, reeling from a thug’s right hook.  Rather than shooting coverage of the entire action sequence and then inserting a series of close-ups, Witney and English have broken things down into discrete units and that allows a finer control of each frame. Even so, their work isn’t flawless. When Bert and Gene each rise on either edge of the composition, the camera operator does not pan far enough right and Gene steps out of the frame where we can miss the punch he throws at his opponent. In the next shot, a heavy falls backward against a radio on the far wall, but since we haven’t seen the strike, his relation to the larger fight isn’t clear.


It is a momentary incoherence that might reflect the editor’s practice of covering gaps in action with closer shots of random thugs falling down (a nearly identical example can be found in the first fight in episode 1).  The two shots would have proper continuity if the first were reframed just a little. The filmmakers seem to have recognized the error, because they eliminated the cut when fight was replayed at the start of the next episode. As a form, serials are invariably marked by the compromises and contingencies of quick production. This must not blind us to the filmmakers’ considerable achievements within those constraints.

The fight is constructed around the individual struggles of our heroes, mostly Bert and Tiny, in tighter shots, followed by wide compositions that allow the combatants to interact. Small stunts can occur within single frames, or cutting can punctuate a blow. Toward the end of the sequence, for example, Bert blocks a thug’s jab and returns a right hook in a medium shot. A ninety-degree cut to long shot reveals new space near the doorway, a desk chair, and another worktable against the wall. In a match-on-action, Bert completes his punch and the thug careens backward knocking the chair and table away.


Cutting doesn’t follow action so much as create it; the sudden widening of the frame and revelation of new space carries the force of Bert’s punch. The new composition has been arranged to accommodate the stuntman’s fall, allowing him to spectacularly (and safely) hit his mark. This simple articulating cut achieves elegance when the shot continues to facilitate a second stunt. While Bert’s thug is still crashing into the table, our hero races back into the frame, picks up another heavy and tosses him leftward as the camera pans with him.


Two distinct beats, foreground and background, in a single take.  Witney and English vary this method in the shot of Gene crashing into the surprise worktable described above. While Gene is socking his thug in the corner, a cutaway shows Bert rolling a heavy across Seldon’s desk and landing him on the floor. When Gene is propelled into the table, Bert has been well placed at the foot of Seldon’s desk. As the thug steps forward to finish Gene off, Bert bounces up from the floor and intercepts him, plowing him back into the frame. Action surges forward and then backward in the frame as the two lines interconnect in a single shot. The rough and tumble clash achieves the grace of a dance.


In this way, phases of action flow into one another and overlap. The filmmakers pack the short sequence with flips, leaps, and falls, and these tend to cohere into larger units of action. The first seven shots follow Bert’s opening leap through his defense of Gene just described. Shots eight through thirteen generally focus on Tiny and his struggle against two heavies. When one of them lifts a desk chair to bash Tiny, Gene appears and disarms him. All three lines again intersect in the fourteenth shot, a wide composition in which Gene is tackled by the chair heavy, Tiny knocks his opponent over Seldon’s desk, and Bert punches down another assailant.



The next seven shots alternate between Tiny battling a tough near Seldon’s desk and Bert fending off his attacker near the door. The editors blend the Tiny and Bert scuffles by carrying the direction of action across cuts. When Bert throws a henchman off toward the left, a cut to Tiny punching his thug leftward against a cabinet creates a continuous motion.


Organizing the larger fight into smaller sets of action lends the whole a sense of progress as we shift among characters encountering and solving physical problems. The emphasis, though, is on process rather than goal. It is standard for an assailant to help his opponent to his feet so that he can deliver a second blow. Both sets of combatants appear endlessly capable of leaping back up to throw another punch, no end is in sight.

The deftly woven series of stunts comes to a sudden halt when Tiny knocks his thug into Seldon’s contact switch (an action depicted in four continuous shots). The thug pulls a gun and everyone freezes when he announces: “Hold it! We gotta scram outta here. I closed the contact!” It is an arbitrary interruption.


The fight could have continued indefinitely or been halted at any moment that the henchman remembered he was carrying a gun. The fight is an intricately choreographed, neatly designed end in itself. The Republic crew has mastered a particular brand of kinetically engaging physical storytelling. In place of an undefined mêlée, viewers are given a flowing series of defined but interconnected physical feats. Frames are filled with speed and motion, so details like suit color (the trio always wears contrasting shades), fighting style (Bert leaps and tumbles, Tiny prefers to fight two men at once), and the distinct cracks and thwacks on the foley track keep things legible. At any moment, the immediate stakes and the status of the combatants are clear. This is a sophisticated form, crafted from space, frame, movement and tempo.  

Posted in Uncategorized | Comments Off on Republic Lands the Punch: Art of the Serial Dust-up

Pi Eyed

The problem with writing an aesthetic history of the present is that the present has a nasty habit of always changing (even more than the past). Last year I published an essay on 3D in a special issue of Film History devoted to digital cinema. There, I focused on Coraline (2009) as a case study of how some filmmakers were exploring the potentials of 3D depth for storytelling. I think that Coraline is a watershed experiment in carefully regulating stereoscopic space to support character psychology and viewer emotion. I suggested that it introduced a way of thinking about 3D that has been developed in films like Hugo (2011). Despite ever-present predictions that 3D will soon vanish, our multiplex and art-house screens have been steadily plunging and thrusting over the past year. My thoughts about 3D are evolving, and this blog seems as good a place as any to sketch out a few more ideas.


The idea of a single “3D aesthetic” seems less and less viable to me. Instead, as the field broadens different filmmakers are carving out distinctive options within the medium. Coraline’s precise experimentation is in a class of “art-house” or “prestige 3D” with films like Hugo (2011), Cave of Forgotten Dreams (2010), and Pina (2011). Tent-pole movies like The Amazing Spiderman (2012) and Avengers (2012) (a post-conversion) exemplify “blockbuster” 3D. Meanwhile lower-budget horror remains the stalwart genre of “exploitation 3D.” Such generalizations raise questions, of course. Do animated family films deserve their own category? What of prestige brands like Pixar and Disney? There are plenty of borderline cases in live-action too. Prometheus (2012) and Oz (2013) straddle the prestige/box-office divide, while Judge Dredd (2012) and Transformers Dark of the Moon (2011) are franchise entries with a heavy dose of interesting experimentation. The potentials of 3D are still being charted and conventions are still forming, which makes it an exasperating and exhilarating topic.


The technology is at that awkward age when its novelty is waning but it hasn’t made the leap to maturity as a norm. Whether it will ever make that leap is an open question, but Life of Pi’s Oscar for best cinematography (the third 3D film to earn the award) grants it some artistic legitimacy.  Like the other 3D Oscar winners Avatar (2009) and Hugo, Pi binds stereoscopic effects to the central character’s journey of self-discovery. As a matter of course, all three films pour on 3D spectacle (Hugo’s train crash, Avatar’s battles and banshees, Pi’s shipwreck). The promise of spectacle loosens purse strings for both producers and spectators. But the Hollywood test of a new technology is whether it can help deliver an emotional experience to popular viewers. To some degree, this means that technique has to get out of the way of tried and true storytelling conventions. James Cameron helped promote Pi (which employed his company’s hardware) by proclaiming: “It does what good 3-D is supposed to do, which is, it allows you to forget you’re watching a 3-D movie.” In other words, the process must reach beyond gimmickry; it must earn the right to go unnoticed.


Ubiquity is the most reliable road to invisibility. Rouben Mamoulian’s early sound film Applause labors mightily to bend the soundtrack to expressive aims, but it remains an awkward affair because in 1929 the technology was inflexible and novelty was king. Just over a decade later, Mark Robson’s low-budget masterpiece Ghost Ship (1943) builds an equally experimental sonic world, but does so upon a foundation that moviegoers had learned to accept. You can “forget you’re watching” a sound film. 3D may never have such advantage. Judging from the past, I’d say that 3-D looks a lot more like color than sound. The transition to sound was astonishingly fast. Major studios adopted their technology in May of 1928 and by September 1929 they were exclusively producing sound films. Full color lingered as a special attraction from the 1930s until the mid 1960s, when NBC began broadcasting its prime-time lineup in color. The only thing that could finally topple black-and-white from the cinematic throne was the conversion of a competitor (and ancillary market). 3-D may well travel that path. It seems to be settling into a sustainable niche of two or three screens at the multiplex. A total conversion is unlikely unless our other screens (TV, Computers, Phones, iPods, DSs) suddenly make the transition, essentially forcing theatrical’s hand.


A film like Pi can never lure contemporary viewers to forget the premium admission price or the glasses they are wearing. It can, however, create a context in which the novelty recedes while story values come forward. Blockbuster fair like The Amazing Spiderman does this by simply alternating more or less “flat” expositional conversation scenes with action sequences of dimensional hyperbole. Pi is more thoroughly integrative. Leaving aside the unexceptional frame story, in which the adult Pi relates his story to a visiting author (and nearly kills all poetic ambiguity in the film), Ang Lee tends not to flatten his image for the sake of unobtrusiveness. Instead, the film nudges stereoscopic depth forward to refresh our awareness of it between spectacles. The film also binds dimensional effects to Pi’s character arc, which gives 3D an emotional resonance. My discussion here is necessarily limited because I’ve only been able to see the film in 3D once, though it was at the Los Angeles ArcLight in former Cinerama Dome, one of America’s best venues. The film’s innovations stand out plainly in a single screening.


From the opening credits, designed by Garrison Yu’s yU+co firm, Pi primes viewers to notice depth. In a series of images of Pi’s family zoo, the film’s titles appear and float at various distances behind the screen surface (positive parallax). Photographic (or at least photo-real) elements sandwich the credits at fore and rear, and even obscure them, as when Avy Kaufman’s credit slides behind a tree, or a lemur extends its nose over Adil Hussain’s name.





The sequence is an exercise in layering planes beyond the frame. Shallow focus renders the bars of a cage as abstract vertical stripes with the associate producers’ credits floating just in front of it. Space is literally shallow. Then an apparent rack focus reveals the leopard’s habitat behind the bars with the co-producer’s name inset above some foliage but still well in front of the big cat. A single shot behaves like a spatial accordion, stretching away from the screen surface. Ang Lee’s director’s credit floats on water, between a pond plant and the pool’s edge.


In each shot, the text defines a finely graded cinematic depth laid out in paper-thin surfaces. The effect is reminiscent of Disney’s multiplane animation in Snow White and the Seven Dwarfs, which represented deep space by photographing cells stacked in front of the camera. As with Disney, Pi’s strategy is to alert the viewer to the illusion of depth by staggering image planes. The title sequence is clever and beautiful, but it is also functional. It primes us to notice depth effects right from the start, reorienting our expectation that 3D means a poke in the eye. It is a gentle and precise way of introducing the film’s world, and especially important because the spectacle-hound will have to wait through the first act before shipwreck promised in the trailers. In the mean time, the credits seem to say, attend to depth.


Pi pushes this concept of planar layering to the peak of abstraction in scene transitions and brief montage sequences, also designed by yU+co. In something like a 3D version of a dissolve, the planes of one image fade in over, or under, elements of another. The first instance is a quick transition from Pi’s memory of his childhood swimming pool to his kitchen. The image of the boy splashing into the water hovers momentarily in the middle of the kitchen, between the two characters. The space beyond the swimming-pool plane is rippled as though it is seen through the water’s surface. The past is sandwiched between elements of the present.



Later, during two elliptical montages at sea, Pi writes in his journal and reads from a survival manual over scenes of life on the raft. In my work on Technicolor, I noted that montage sequences tend to be open to formal experiments. The compression of time already makes narration overt so viewers can more readily accept stylization. Pi takes this motivation and runs with it. The first montage offers images from different times that dissolve and wipe over one another, each layered as a plane that lingers momentarily in the frame. A close-up of Pi sharpening his pencil with a hunting knife hangs in the foreground as a long shot of him fishing emerges in the background. Even more striking, a close-up of Pi writing in his diary moves to the background as a majestic shot of Richard Parker materializes in the fore.

The second montage is simpler. Pi holds the foreground left in a medium-close shot of him reading the manual, while the right two-thirds of the frame depict his methods of survival.  The montage begins with three layers: Pi, pages of the manual close behind him, and shots of life at sea further back and on the right. Eventually the book pages vanish, leaving two layers and two different time frames on screen. Depth helps clarify the sequence. In 2D it seems odd to see two images of the same character at different points in the story superimposed in sharp focus. The stereo separation between the planes in 3D somehow reinforces the temporal distance between events. In 2D, the images appear as flat collages, but 3D provides an elegant coordination of space and time.

The play of layers and planes can feel gimmicky, and it is used only in brief stretches. But it also sets up an emotional payoff at the film’s end. Where Pi’s voice-over narration during his voyage occasioned a complex mix of flat but staggered image elements, the story he tells the Japanese insurance adjusters in the hospital is handled differently. Here, cleanness and directness are the rule. Sitting in medium-close up in a flatly lit, barren white hospital room, Pi stares toward the camera. In place of the planar layering, Lee presents the audience with a frame that bows outward into the theater (negative parallax). Pi’s head emerges from the screen in a shot that exaggerates its roundness.

I suspect that Brian Gardner, Lee’s stereographer, either shifted the interocular distance (essentially the distance between the two camera “eyes”) during the monologue to accentuate dimensional volume, or manipulated a “dynamic floating window” to tilt the top of the frame backward, though I cannot confirm that without an additional 3D viewing.  Scorsese uses a similar effect in Hugo to signal interiority when the main character remembers his father’s death. In both cases, facial emergence accentuates expression and brings the audience closer to the sympathetic character. The gentle use of negative parallax to create a bulge around Pi’s face functions something like a carefully lit close-up might in a studio-era melodrama. It also signals a strong departure from the stacked planes in the earlier montages. Here we might actually “forget” we are watching 3D, in part because the film has guided us to notice layering in depth as a style, and because the scene is a moving climax. I think the viewer actually “feels” the difference in volume as an increase in emotional directness and connectedness with the character. If we don’t recognize the change, Lee’s return to the flat world of the frame story in which the older Pi concludes his interview with the author punctuates the shift. The moment of raw insight has passed and space is deflated, the screen snapped back.


Of course Pi also has its share of spectacular emergence. In her excellent blog on 3D aesthetics, Miriam Ruth Ross notes that underwater scenes in particular motivate protrusion: “There are numerous shots filmed from under the water where bubbles and small pieces of jetsam drift in negative parallax to the viewer and the space between the audience and the screen seems to become thick and heavy. In a couple of these shots, Pi’s floating body hangs into the auditorium space.” Flying and floating objects have long been privileged in 3D because they need not be anchored to a spot behind the frame. Think of Avatar’s helicopters, Hugo’s snow, or even the swimmers in Creature from the Black Lagoon. Likewise, Pi delivers its own version of the “Lion in Your Lap” spectacle promised long ago on posters for Bwana Devil. The first time Richard Parker bolted from beneath the tarpaulin the audience I was with reflexively jerked back from the screen. Later, Pi plunges his grappling hook toward the camera when he spars with the tiger.

Lee’s most notable use of negative parallax comes during the flying-fish sequence. Here, he shifts the film’s aspect ratio over a cut so that black bars appear at the top and bottom of the frame. The animators at Rhythm and Hues supplied fish that dive into and out of the black areas so that they appear to slip below the limits of the screen. The effect is a variation on the ‘dynamic floating window’ developed by Gardner known as “breaking the mask,” and introduced in G Force (2008). (Sara Ross tells me that the opening sequence of Oz does something similar, and Marc Longenecker reminded me of a overt instance in the end-credit sequence of Despicable Me).


Like most 3D features, though, Pi spends most of its “depth budget” behind the screen. This is not a neutral or “flat” option in which dimensionality recedes to move the story along between actions sequences. Instead, like Hugo and Coraline, Life of Pi helps develop 3D’s expressive vocabulary in its quiet moments.  For all the buzz over its CGI animals and elaborate set pieces, Pi’s most powerful 3D moments evince an integration that makes stereoscope a flexible, nuanced story medium. Not to heap too much praise on the film, but it strikes me as closer to Ghost Ship than Applause.

Posted in Uncategorized | Comments Off on Pi Eyed

Chromo-Drama: Innovation and convention in Douglas Sirk’s Color Designs

The Book is OUT!


This is a new anthology based on a terrific conference on Color Cinema held in Bristol England a few years ago. I’m proud to be among the contributors. I’ll say more about my contribution later — but for the moment, here are color illustrations that go with my essay. My chapter really needs about 90 color illustrations (!), or at least I like to think that. This is impossible in academic publishing, so I put references in the text to this blog page. I hope that some readers will seek these images out — since they help make the whole thing make sense. On the other hand, the stills are probably better than the essay — AND they are free! So, enjoy.




Posted in Uncategorized | Comments Off on Chromo-Drama: Innovation and convention in Douglas Sirk’s Color Designs

Emily Nussbaum on Cliffhangers!

Some of the greatest shows on television today owe a debt to the serially told forms of the past, such as novels and soap operas.


The New Yorker has a terrific piece on Cliffhangers – pointing out the form’s long history and currency on television. Nussbaum traces a pretty clear trajectory from serials to daytime TV to prime-time dramas and thrillers. Her point that the cliffhanger crosses taste cultures and market identities seems to me pretty important. I’ve written here on the BBC’s Sherlock, which in the US has a snooty appeal, but Nussbaum’s opening example is the equally venerable Melrose Place. Melodrama is not as class-based as it once was thought to be (e.g. 19th century — though Brewster and Jacobs even dispute that). I am intrigued with the resurgence of the form on TV from the late 1990s onward. I’ve noted elsewhere that situational sound-serial plotting made a vengeful resurgence in shows like Alias and 24 about a decade ago. But the connection between these action mellers and the romantic/family soaps of day and night is certainly worth pondering. Alas, I’m spending my days and nights pondering The Drums of Fu Manchu and Daredevils of the Red Circle, so TV will have to wait for another summer.

Incidentally, I did finally catch up with season 2 of Sherlock and can report that the solution to the problem space was in the room the whole time. It isn’t particularly clever and not half as elaborate as the scenarios I’d built in my mind — but, that is the beauty of a good cliffhanger. For the interim, the viewer becomes the showrunner/filmmaker/storyteller. And like a serial, the show moves on quickly and never looks back.

Posted in Uncategorized | Comments Off on Emily Nussbaum on Cliffhangers!

A Moment of Cinema

Last November, Steve Collins, Jeanine Basinger and I presented a WeSeminar to students, parents, and alums entitled “The Cinematic Moment: A Glimpse into the Wesleyan Film Studies Classroom.” The idea was to give a quick and vivid idea of how we approach cinema: seeking the explain the power of an artwork by LOOKING AT that work and THINKING LIKE the artist. I felt more than a little apologetic for imposing on my colleagues at what turned out to be the tail end of Middletown’s second major power-outage of the season. In the end, though, the event was the highpoint of my year as a Wes Film professor. We each showed a brief excerpt from a favorite film, and then tried to explain why it moved us. Here is my discussion of the “Have Yourself a Merry Little Christmas” number from Meet Me in St. Louis.

For me, many of the medium’s highest achievements come from the Hollywood Studio Era, a time when a mastery of craft met the mandate to entertain and engage, and move audiences. This style reached an apex in 1944, in a simple scene of a young woman comforting her little sister on Christmas eve. She does this just after she’s promised to marry the boy next door. She does it with a song.

It helps that the young woman is Judy Garland, the sister is Margaret O’Brien, and the song is by Hugh Martin. The color is by Technicolor, Art Direction by Cedric Gibbons, and Cinematography by George Folsey. All of this was marshaled, organized, and directed by Vincente Minnelli; also it has monkeys.

See the sequence here:

Why is this a highpoint of cinema history? How does it work? What does it do? And what can it tell us about Art, Cinema, and Film History?

It is deceptively simple, only eleven shots from the moment Garland enters her sister’s bedroom to the end. And the story, such as it is, is also deceptively simple – she comforts her sister with a song. But the depth, the emotional punch, the gripping engagement that Minnelli offers in this scene far outstrips simple summary. In fact, we are hard pressed to articulate in words what happens to us, what transpires on the screen, in these four minutes. Like the best Hollywood films, it has an effortless beauty, one that captures us, but also deflects analysis.

But let me try. I want to draw attention to one moment in particular, a small gesture that takes place in the 10th shot, but that changes everything. In fact, I think it is the heart of the scene.

Judy begins the song in shot number 9, and the whole of the song is covered by only three camera positions, cut back and forth. Shot 9 is a two shot from a medium distance (in itself vital  – we’ll come back to that). Shot ten gives us a close-up: Judy sings to her sister who is now off-screen. And then – as she hits the line “Next year all our troubles will be miles away” she takes off her headscarf and glances forward and a little bit upward. She takes a breath, and begins the song’s bridge “Once Again as in Olden Days, Happy Golden Days of Yore.”

The power of the moment comes from Vincente Minnelli’s meticulous control of the flow of detail, the choices made among the thousands of possibilities and the shaping of emotion out of a hundred little touches. The touch here involves George Folsey’s Technicolor cinematography, and his work within the limits of a fairly inflexible technology. The Technicolor Camera was famously stubborn when it came to light – it required a LOT of light and cinematographers had to fight hard to achieve the precision and balance of highlight and shadow they so valued in black and white. Close-ups were a challenge.

Here we have a relatively dim scene, and one that depends on small and precise shifts in light and dark to tell the story. The fact is, this scene isn’t all that dark, but Folsey has built up and layered highlights to give the impression of contrast. The lighting set up is quite precise, and it requires Garland to perform within a very small area of movement, she finds expression within the space allotted her by the cinematographer.



Folsey arrays shimmering highlights to add emphasis and connect us with the moment. Garland’s headscarf is an important costuming decision. It throws off a blaze of sparkle, of dazzle, that is essentially a-chromatic. This was important for Garland’s awkward love scene just moments before, which Minnelli designed around highlight and shadow, silvers, blacks, and dark blues, rather than strong color: an almost black-and- white scene in a Technicolor film.


Now the sequined scarf gives the impression of contrast, dapple and shimmer, but within a color lighting scheme.



In this moment, as we move in on Garland’s face, she removes the scarf — removes the sparkling mass of highlights, and, with perfect timing, glances forward, catching two pin-points of light near her pupils. This is an eye-light, a small light source set off to one side of the camera with the sole purpose of adding dazzle to the actress’s eyes. As she looks up and to the left, her eyes engage the light; highlight is shifted from costume to face in one fluid gesture.


The look into the eye-light is a pretty powerful way that cinematic form shapes performance, and the late 1930s and 1940s seems to be the heyday of this technique (though you can find much earlier, as in William Daniel’s work on Flesh and the Devil in 1926). It is what gives punch and intensity to Ingrid Bergman’s reaction when Paul Henried tells her he loves her in Casablanca, or Bogart says “here’s looking at you, kid” one last time;



it lets us see Vivian Leigh’s spark of self interest and ambition near the beginning of Gone With the Wind;


and, in a recent favorite of mine, it signals the evil Vultura’s heartless resolve as she prepares a human sacrifice in the Perils of Nyoka, a Republic Serial from 1942.


The glance toward an eye-light is a tool in the Hollywood cinematographer’s box, and the ability to play to the eye-light, a skill of the Hollywood actor.

What Garland, Minnelli, Folsey, and the crew of Meet Me in St. Louis have done, is to match this effect to an emotional beat – or, more accurately, they create an emotional beat out of this formal effect. When Garland looks away from her sister and into those lights, she broadens and deepens the song – she isn’t only singing to another character, in fact she isn’t comforting her little sister, but herself, and I think, the home-front viewers of 1944 for whom family union was held in abeyance by war.

She is, after all, looking toward the window of the boy next door – not her sister. This is the relationship that has been lingering before us, unresolved, since the last scene. Garland’s character seeks reassurance through song, and in doing so she acknowledges the impossibility, the sadness of it all. But too, the close up and the shift of the highlights places us in a different relationship to the character and the star. She is singing to and for herself, and to and for us. As Dr. Sara Ross pointed out to me, the bridge of the song broadens it – she looks ahead for “Faithful friends who were dear to us, will be near to us, once more” — image and sound coordinate to address the audience. To borrow Jeanine Basinger’s terminology, the film slides into a “musical world” where we can be addressed, where we make a connection not just with character but also with star.

The glance into the eye-lights tops off a series of revelations across the scene.  It is part of a texture, a weave of moments and telling impressions. Garland’s trip from the door to the window brings her in and out of shadow, until she steps into a light that edges her from the background, in a nearly monochrome frame. Garland realizes she cannot just rush O’Brien off to bed, that this will take some serious talk; she takes off her pearl grey coat, returning the vibrant red ball gown from the evening’s Christmas dance (emotional beat, it recalls the joyous celebration in a somber scene). Next, the light from her boyfriend’s window shifts color temperature, and Garland’s colors pulse momentarily, before returning her to a colder, gray winter light  (another thread in the film which associates warm lamp and firelight with home – the image keeps reminding us what is about to be lost by the family). As the music begins, Garland makes a small entreating gesture to O’Brien, really a kiss in the air, a maternal touch that individuates her character and marks her empathy. This tiny movement also helps smooth the shift to Garland’s lip-syncing to on-set playback.  All of this preparing the way for Garland to take off the icy scarf, and catch us in her eye. She is opening up for us and for her sister, lowering barriers, and making a connection.

This is the brilliance of Minnelli’s work – a dense weave of textures, colors, movements, lights, sounds, and gestures – not one of which can do the job on its own. They collaborate to sweep over the viewer with an impression of order, plentitude, and feeling.

If that makes any sense, then I think it says something about how cinema can work on viewers, about what I find cinematic. People often think of film as a means of communication — as though there is a message, or a content, that a filmmaker pours into a spoon and delivers to the viewer’s lips. That content might be delivered by another vessel, a play, a novel, a dance, a political manifesto. Film Communicates an Idea To the Viewer.

But this is only part of what cinema can do, and in the Hollywood tradition it is perhaps a rather small part. Film form, the stuff up there on screen, the myriad stimuli that flow through sound and image, doesn’t communicate TO us, it captures us, harnesses and shapes our perception with pin-point accuracy and infinite finesse, Film form entertains, and moves us, and THAT experience is uniquely compelling – no other art does that in quite the same way.



Posted in Uncategorized | Comments Off on A Moment of Cinema

Cliffhanging by the Pool


I’m starting to see my topic everywhere. So, it shouldn’t have surprised me that when I took a break from my chapter on cliffhangers to watch the last episode of the BBC’s Sherlock season 1, I found myself face to face with… a cliffhanger. Sherlock is an updating of the Holmes universe, produced by Doctor Who alums Mark Gatiss and Steven Moffat. I think I am among the very last people to have watched it, but if not, be warned that SPOILERS LIE AHEAD.


In the The Great Game (the third of three episodes), Sherlock is taunted and run around town by Moriarty who has taken to wiring innocents with plastic explosives and threatening to blow them up with a sniper’s bullet should the detective be unable to solve various mysteries before an arbitrary deadline. Moriarty has apparently learned a bit from Dirty Harry’s Scorpio. Why he relies on a sniper’s bullet to set off the explosive, as opposed to the more pedestrian cell phone, is a mystery — though the little pinpoint of red light does build a visual motif, and the episode, refreshingly, avoids any red LED readouts ticking down to the deadline.


But the cliffhanger takes place the night after the last mystery is solved, and Sherlock arranges a meeting with Moriarty at an indoor swimming pool. The episode stages two standoffs, back to back; the second constituting the cliffhanger. In the first, Moriarty has rigged Watson with a vest bomb, and has his unseen sniper trained on him. If Holmes shoots Moriarty, Watson, and perhaps everyone, will die. This is quite a good situation, and it is a dry run for the cliffhanger. Watson, in an attempt to turn the tables, grabs Moriarty and uses him as a human shield. His move fails, though, when the sniper re-focuses his aim on Sherlock. The trap seems pretty well inescapable, and it is only resolved when Moriarty, having said his share, leaves the pool area. Sherlock rips the vest off of Watson and kicks it down the pool deck; both breathe a sigh of relief.



Then it all begins again. Moriarty reappears, claiming to have changed his mind on a whim, and a cluster of laser-sites begin dancing around on both Sherlock and Watson.  In response, Sherlock draws his pistol and aims it squarely at the explosive vest lying on the floor between the enemies. The episode ends, and we wait until next August (in the UK, later in the US) to see the resolution.


I’ve been trying to understand the power of the cliffhanger – a narrative structure that assured viewers would return to see the next installment of a serial in the 1940s, and that spurs them to stick around through (or fast forward past) commercials in any given episode of any given police procedural. At its most basic, the cliffhanger works by interrupting a suspenseful situation, holding the resolution in abeyance. In his essay “The Paradox of Suspense” Noel Carroll offers the cleanest and most concise explanation of fictional suspense that I have read. Suspense occurs when a character the viewer aligns with is caught in a situation with two clearly defined opposed outcomes, a desirable “moral” outcome, and an undesirable “evil” one. For Carroll, the “evil” outcome must be equally or more likely than the “moral” solution. Situations that progressively tip the balance of probability toward the evil outcome are excellent vehicles for suspense. Every moment that the time bomb ticks down, or the bus hurtles toward the unfinished bridge, or the laser beam approaches James Bond’s crotch, means that “time is running out on the good and therefore evil is becoming more likely” (in Suspense: Conceptualizations, Theoretical Analyses, and Empirical Explorations, 1996; 83).  Our hero’s fingers begin to ache, the sand slips out from beneath them, pebble by pebble, as she clings to the edge of the cliff.


Standoffs are particularly good situations because they can be so nicely visualized. As I’ve discussed in my Cinema Journal article “Suspenseful Situations,” the standoff has a venerable history, stretching back at least to the 19th Century melodramatic stage where the villain could hold the hero at knifepoint as the curtain closes before an act break. Melodrama thrives on making the stakes and emotions of scene visual and instantly graspable – a stage picture. Sherlock’s standoff gives suspense a clear geometry. Our hero is in the foreground, our villain at the rear, and between them lays the bomb. The “evil” option seems to have clearly won out.


[ingredients of the cliffhanger: Sherlock (Benedict Cumberbatch); Gun; Bomb; Moriarty (Jared Harris)]

In fact the very diagrammatic simplicity of the standoff reminds me of cognitive psychologist Richard Gerrig’s description of suspense as a “problem space” (Experiencing Narrative Worlds,1993; 82). For Gerrig, suspense is a participatory structure that cues viewers to seek some piece of withheld knowledge. The “problem space” analogy compares problem solving to searching a space for information that will allow one to achieve a goal (82). The contours of the trap, and the heroes aim, are clear, but the means of liberation, the escape hatch, remains hidden: “to make the reader really feel suspense, the author must sufficiently constrain the space of possible solutions so that the situation appears beyond hope” (83). The ingredients of Sherlock’s problem space are, heroes, villain, bomb, sniper(s), and swimming pool. The first standoff serves, in part, to lay out the parameters of the trap and to narrow the options. As Gerrig notes, a successful suspense situation “mimics the process of problem solving… by specifying and then eliminating potential means for escape” (78). We know that Watson’s “human shield” maneuver is off the table, and now doubt that Watson can serve any real purpose. We also know that Moriarty can simply change his mind – he isn’t just a villain he is a mad genius. Mad geniuses make excellent bad guys because they fuse malevolence with creativity and unpredictability. As simple as a standoff is, he remains a wild card.


And so, over the intervening months, the fan is left to search the problem space. I suspect the swimming pool will come into play, partly because its role hasn’t been directly acknowledged in the proceedings. Aside from the nice lighting opportunities it provides, why set this climax poolside in the first place? Atmosphere can step forward to save the day. The cavernous tunnel that climaxes episode 2 works in this way. It stops a villain from firing his gun because of the likelihood of an uncontrolled ricochet. The pool is the elephant in the room – though, search the space as I may, I can’t quite figure out how it might help resolve the situation. There might also be something to Moriarty’s use of snipers, since it seems a gratuitous method for setting off explosives. Are there any snipers at all?


As cliffhangers go, this is a fair one. It differs from its serial predecessors in two ways. First, a serial would end with a bang. We would see the room, the pool, the building explode in a fiery maelstrom before cutting to black. Serials tend to push situations to the point of cataclysm in order to leave a lasting impression. The drawback was that screenwriters had to resort to “cheating” for the cliffhanger’s resolution. So that explosion didn’t really happen, or everyone found a trap door (jumped into a pool?) just before the explosion, etc. The other difference is that Sherlock’s cliffhanger is just a little muddy. For all its clear simplicity, it lacks a ticking deadline. If Sherlock has managed to call Moriarty’s hand by aiming his gun at the bomb, then we are in a deadlock pure and simple. They could all stand that way forever, at least until August. More than that, though, viewers aren’t privy to the entirety of the situation’s dynamics. Why, I ask, does Sherlock aim at the bomb instead at Moriarty’s head? And why does he seem to think this is a cunning maneuver? Our hero already seems to know the way out.  It is in keeping with the show, which leaves the viewer (and Watson) just behind the hero’s problem solving process, but it interferes with the suspense. There is something about the situation that we don’t yet understand; some part of the problem space has been withheld, and we know it. This grain of uncertainty eats away at the set of opposed outcomes that the standoff usually makes us entertain.


One of Gerrig’s most valuable ideas is that suspense situations often rely on “functional fixedness” to throw the viewer off the trail of the solution. Our access to the scene’s secret is effectively blocked because we perceive each ingredient as locked into a single role – the bomb is there to explode, the guns are there to fire, Moriarty is simply evil and genius and mad, the swimming pool is there to create atmosphere. Sherlock’s move tips his hand just a little, he has thought outside of this fixedness and found another use for the bomb. I’m not sure why I find this cliffhanger less than completely satisfying, but perhaps it has something to do with the attention it draws to our own fixedness. It is better when we really don’t recognize the solution that will seem blindingly obvious after its reveal.


The bigger questions raised by cliffhangers point to the paradox of suspense, as Carroll terms it. Why are we fascinated or uncertain at all when we very well know that Sherlock will survive this ordeal. Beyond that, why do we find this kind of suspense pleasurable, since, from Gerrig’s perspective all this cognitive processing takes a good deal of work. Both theorists have their own answers. For my part, I’m beginning to think it has to do with the pleasures of problem solving. Sherlock has given me a puzzle to dwell on for the next several months, a situation to replay and tease out. And just as it was for serial fans of the 1940s, the solution may leave me unsatisfied – but the cliffhanger will have fulfilled its function (I’ve tuned back in). The real challenge for Gattiss and Moffat isn’t to get everyone off of that pool deck alive, but to set up another problem quickly enough to keep us watching.




Posted in Uncategorized | Tagged , , , , , , , , , , , | Leave a comment