David Bordwell's website on cinema   click for CV

Home

Blog

Books

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

Video

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

Essays

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

Articles

Book Reports

Observations on film art

Archive for the 'Narrative strategies' Category

Merrily he rolls along: A belated birthday tribute to Stephen Sondheim

Sweeney Todd: The Demon Barber of Fleet Street (Tim Burton, 2007).

DB here:

Last month Stephen Sondheim celebrated his ninety-first birthday. By coincidence, I just finished a section on him in my current book on popular storytelling.

The argument there is that art at all levels, high and low, mass and elite, depends on novelty. That demanded accelerated in the twentieth century, with the explosion of book publishing, magazines, film, radio, theatre, and television. I see this process as a vast energy of crossover. Popular and “middlebrow” storytelling picked up on avant-garde innovations. But it seems to me that modernism, even the High Modernism of Joyce, Woolf, and Faulkner borrow more than is usually noted from conventions of popular storytelling. I hazard the view that it’s enlightening to look at how experimentation, innovations that open new avenues of artistic expression, emerge in mass-audience art.

Nowadays, most intellectuals have found something to love in mass culture. (“We are all nerds now.”) It wasn’t always so. In England and America, the “battle of the brows” had as one consequence the idea that true art, usually typified by High Modernism or the more radical avant-garde, was being squeezed. On one side was mass culture, manufactured as a commodity designed to sooth the masses. On the other was “middlebrow” art, which mimicked the techniques of modernism but made them simple enough for suburbanites to follow. By the 1960s, the people who believed in this trinity were in the minority, partly because they began to realize that centering on the most difficult side of modernism created too austere a prototype of all ambitious art. Irving Howe called this “The Decline of the New.”

Not that everything is a mashup. But I think now we all realize that crossover is a valid expressive option, probably the most pervasive one. The mass media make room for a huge amount of creativity that can’t simply be dismissed as lowbrow or middlebrow. And one reason it can’t is because we can still be excited by unexpected innovations.

Hence the importance of the nearly seventy-year career of a man who confessed a “taste for experiment in the commercial theatre.”

 

Experiments that are fun

Stephen Sondheim and Leonard Bernstein.

You couldn’t, I think, find a better case for the prospects for experiment in popular culture than his work. His fondness for stage games is part of a theatre tradition we might call “light modernism,” a merging of avant-garde strategies with traditions of popular entertainment. It stretches back to Cocteau’s Parade (1917) and includes the work of Pirandello. Similarly, a trend in British theatre fused aspects of modern theatre (Pinter, Beckett, Ionesco) with the comedy of P. G. Wodehouse and The Goon Show. Tom Stoppard’s Rosencrantz and Guildenstern Are Dead (1966) showed that what was “offstage” in one story could constitute the amusingly doom-laden plot of another.

Alan Ayckbourn dedicated his career to formal experimentation with space (three bedrooms as one in Bedroom Farce, 1975) and time (simultaneous action in How the Other Half Loves, 1969). Intimate Exchanges (1983) spreads forking-path plots across eight separate plays and sixteen possible endings. (Two of the variants are on display in Resnais’ films Smoking/ No Smoking of 1993.) House and Garden (1999) consists of two plays performed simultaneously in two auditoriums, with actors dashing between them. Ayckbourn’s most famous cycle, The Norman Conquests (1973) displays a “stacked” structure; each play gathers all the action taking place in one location and skips over scenes elsewhere, which are assembled in the other plays. The audience must construct the overall story, remembering what has happened just before the scene we see now.

Michael Frayn’s virtuoso Noises Off (1982) probably owes something to Stoppard and Ayckbourn, but it offers its own switcheroo. The opening scene of a sex farce is rehearsed in our first act, is performed skillfully in our second, and collapses in our third view. Crucially, the smooth show of the second act is presented to us in a backstage view displaying the intricate timing involved. Again, an abstract formal concept is mapped onto a conventional scenario, the comedy of a bungled stage production.

Mystery plots make useful targets for this sort of popular experiment. Ayckbourn plotted some plays as quasi-thrillers, and Stoppard’s The Real Inspector Hound (1968) is a parody of the country-house murder. Stoppard undercuts the mystery  by a Pirandellian device: two critics down in front are commenting as the performance unfolds. This gives Stoppard a chance to mock pretentious reviewing language. The standard whodunit device of reenacting the murder transforms into a replay of the opening act, but now with the critics taking the roles of detective and victim.

Also not surprisingly, both Ayckbourn and Stoppard declared themselves influenced by cinema, a reliable marker of crossover in modern media. Ayckbourn plays were adapted, with ingratiating wit, to film, while Stoppard wrote several scripts, most famously Shakespeare in Love (1998), which if the term means anything must count as defiantly middlebrow entertainment.

Sondheim is on the same frequency as these masters, but has, I think, greater bandwidth. He plunged into cinephilia more deeply. His early musical influences were Hollywood scores, notably that for Hangover Square (1945); Sweeney Todd, The Demon Barber of Fleet Street (1979) was his tribute to Bernard Herrmann. A Little Night Music (1973) and Passion (1994) were adapted from films. Many of his songs refer to movies, and he composed music for Stavisky (1974), Dick Tracy (1990), and other projects. He was even a clapper boy for John Huston on Beat the Devil (1953).

Instead of parodying mysteries, he has been deeply committed to them. Detective fiction is his favorite reading, and as a puzzle addict he has spent hours devising murder games. “I have always taken murder mysteries rather seriously.” Both Company (1970) and Follies (1971) were initially planned as mysteries, with the latter concerned not with “whodunit?” but “who’ll do it?” Sweeney Todd is a paradigmatic revenge thriller. Asked by Herbert Ross to write a film, Sondheim and Anthony Perkins (another admirer of “the trick kind” of mystery practiced by John Dickson Carr) came up with The Last of Sheila (1973).With George Furth, Sondheim wrote a play, Getting Away with Murder (1996), and with Perkins he planned the unproduced Chorus Girl Murder Case, an homage to 1940s Bob Hope movies. The clues would be hidden in the songs.

 

Fiddling with formula

Sondheim and producer/director Hal Prince (left) during rehearsals for Merrily We Roll Along.

Sondheim’s zest for film and mystery fiction reflects a deep admiration for popular culture. It’s one thing to enjoy it, as Ayckbourn and Stoppard clearly do, while also poking fun at its silly side. It’s something else to appreciate its artistry in depth and try to master the conventions yourself. Despite learning “tautness” from the austere avant-gardist Milton Babbitt, Sondheim took as his mentor Oscar Hammerstein II. Tin Pan Alley, with its finger-snap rhythms and suave wordplay, pushed him toward a brisk cleverness. The virtuoso rhymes in his bouncy “Comedy Tonight” (A Funny Thing Happened on the Way to the Forum, 1962) inevitably bring a grin: Panderers! Philanderers! Cupidity! Timidity! . . . Tumblers, grumblers, bumblers, fumblers!

While executing these pirouettes, the song provides a Cliffs Notes guide to Roman New Comedy by contrasting it with tragedy. Tonight, weighty affairs will just have to wait. Sondheim similarly lays bare a convention in his script for the unproduced movie Singing Out Loud, where the couple gradually learn that it’s okay to express emotion in song. “They have to learn to overcome the unreality of it and break into song in the conventional manner of all musicals.”

By taking fun seriously, Sondheim ransacks culture high and low for occasions for experimentation. He was crucially influenced by Allegro (1947), an ambitious Rodgers and Hammerstein musical he considered “startlingly experimental in form and style.” Its use of sliding screens to create “cinematic staging” would become standard in later musicals. He calls Hammerstein “the great experimenter” who used the verse sections of songs to explore possibilities of structure, melody, and harmony.

Sondhim’s breakthrough experiment was Company (1970). It consists of flashbacks framed by the protagonist Robert’s thirty-fifth birthday party. Sondheim claimed it offered “a story without a plot.” The flashbacks don’t supply a goal-directed character arc and instead sample Robert’s bachelor lifestyle and its effects on his three girlfriends and the five married couples in his circle. The result is a compare-and-contrast pattern of parallels.  Complicating things further, bits of action are accompanied by a chorus-like commentary from characters not in the scene. Still, there is a certain progression to the whole, indicated by Robert’s disillusioned but faintly hopeful final song, “Being Alive.”

The nonlinearity of Company poimts up Sondheim’s impulse to play with time, viewpoint, and other techniques. Assassins (1990) moves freely back and forth across a hundred years. In Follies characters argue with their former selves. Sunday in the Park with George (1984), built on parallels between two painters, assigns inner monologues to artist and model; characteristically, Sondheim expresses jumbled thoughts by avoiding rhymes. Another duplex structure (Before/ After) shapes Into the Woods (1987), s a virtuoso braiding of classic folktales into a single plot.

Several plays utilize a narrator, who may take a role or converse with the characters. The Narrator of Into the Woods is killed fairly early in the action. Alternatively, Sondheim conceived Passion as an epistolary musical. Characters writing or reading letters operate “somewhere between aria and recitative,” rendering the emotional climaxes as “read rather than acted.” In this excerpt, Giorgio goes to bed with Clara while Fosca is reading his letter breaking up with her. This is a concert performance; in a full production the couple undress on one side of the stage, while Fosca reads the letter. Time floats uncertainly

Working in musical theatre gave Sondheim a layer of implication beyond what Stoppard and Ayckbourn had available with spoken dialogue. A score can evoke earlier scenes through leitmotifs and can enhance characterization. Starting with Anyone Can Whistle (1964) Sondheim created pastiche songs that vary from the show’s overall style. Merrily We Roll Along incorporates cabaret acts (“Bobby and Jackie and Jack”) while Assassins integrates various popular musical traditions, from vaudeville to melodrama. We might think of these pastiches as akin to the “polystylism” of the chapters of Ulysses, which Sondheim strongly admires.

Pacific Overtures (1976), a chronicle of Japan’s early engagements with the west, has the structure of a “portmanteau” film composed of exemplary episodes. It too has a narrator, the Reciter, and its experiments include turning renga linked verse into a passed-along song. The most formally daring scene, “Someone in a Tree,” stages the March 1854 signing of the treaty opening up ports to American ships. We do not see the ceremony, which is held in a secure house.

The Reciter questions an old man who claims that as a boy he watched the negotiations from a tree. During their dialogue, a boy clambers up the tree, and he and his older self collaborate in reporting the event he sees but cannot hear. Then the Reciter discovers a samurai guard hiding beneath the floorboards. He reports what he hears but cannot see. (Beware the YouTube ad at the start.)

As the singers’ accounts intertwine, a moment is assembled through partial perceptions. In a gesture reminiscent of Joseph Conrad’s novels, the event is broken up, made accessible only through partial viewpoints. And only the witnesses attest to the event; without them, we have no access to history. “I’m a fragment of the day./ If I weren’t, who’s to say/ Things would happen here the way/ That they’re happening?” In adapting novelistic techniques for relativistic point of view to the stage, Sondheim, as an experimental storyteller, is ready to plunder any tradition that can yield something fresh.

 

Form, “content,” and everything in between

Sondheim and collaborator James Lapine.

In his invaluable creative memoirs, Finishing the Hat (2010) and Look, I Made a Hat (2011), Sondheim claims as a basic creative principle “Content dictates form.” But his puzzle-addict efforts to seek out difficulties to be overcome makes me think that this is more alibi than axiom.

I’d rather think that like many experimental artists, Sondheim uses “content” (whatever that is: subject matter, theme, bare-bones story) as at best one ingredient and often as a handy excuse. Sometimes audiences need help when faced with radical novelty. People may have been more receptive to Debussy’s daring pieces because of the “atmospheric” connotations of their titles. The stratagem of labeling one movement of “La Mer” as “From Dawn to Noon on the Sea” was pointed out by Erik Satie: “I liked the part at quarter to eleven best.”

Granted that Sondheim tries to suit his words and music to the genre and story he has selected, he often seems to take “content” as a pretext for solving problems he sets himself. Who else would decide to write all the songs for A Little Night Music in waltz tempo, not least for the challenge of avoiding monotony?

I think Merrily We Roll Along is a good example of how, accepting an initial problem, Sondheim complicates matters “unnecessarily.” This largely forgotten 1934 Kaufman and Hart comedy was formally daring: its scenes play out in reverse chronology, so the last scenes we see in the plot are the first events of the story. In adapting the play to a musical, Sondheim created a story of three twentysomethings trying to break into show business.

The purported rationale for telling their story backward is that it poignantly traces the loss of youthful idealism. But that arc could be just as poignant laid out in chronological order–or, if you want advance knowledge of how the struggles will turn out, via chronological flashbacks wrapped in a contemporary time frame. The reverse chronology seems both an experiment in whether audiences can follow the string of events and an effort to end a sad story in an upbeat way, with the characters still vigorous and hopeful–while we know what awaits them.

Sondheim faced an initial task of making the time scheme clear. He came up with transitional choral passages by the whole company that signal the shift to an earlier block of time. Different productions experimented with ways of reinforcing these musical tags, such as a synoptic slide show reminiscent of the “News on the March” sequence of Citizen Kane.

But the inverted chronology of Merrily We Roll Along also justifies experiments in musical texture. Because the play is about friendship, melodic motifs are swapped among the characters in their soliloquy songs. Moreover, in a 1-2-3 plot, Sondheim explains in Finishing the Hat, fully vocalized melodies are given reprises, shorter versions of the original number. Sondheim could have followed this convention. Instead, in relentless adherence to the reverse chronology, he made the reprises come first, as appetizers for songs yet to be fully heard. The puzzle addict will try to fit everything together in surprising ways, whether the audience realizes the fine points or not.

Solving one problem can launch a cascade of further problems. To introduce a 1980s audience to the musical ambience of Broadway’s heyday, Sondheim opted to revive the thirty-two bar song that he and his generation had “stretched out of recognition.” But then his penchant for pastiche posed a new problem. How to use that schmaltzy song form not just to satirize superficial characters like Joe the producer but to sustain connection to the sympathetic characters when they express authentic emotion?

Sometimes the problem is set not by “content” but by genre, tradition, or purely practical contingencies. Sondheim embraces show-biz conventions to discover what he can do with them. Hammerstein told him that the opening number can make or break a show, so Sondheim strives for a grabber when he can.  The eleven o’clock number, a hangover from the days when shows began at eight-thirty, demands a show-stopping vehicle for the stars–e.g. “Anything You Can Do” in Annie, Get Your Gun. For Anyone Can Whistle, Sondheim wrote “There’s Always a Woman,” a rapid-fire comic confrontation between the two stars Angela Lansbury and Lee Remick, dressed identically. He conceived it as “the eleven o’clock number to end all eleven o’clock numbers.” It didn’t achieve what he wanted (“more like a ten-fifteen”) but it indicates his inclination to display virtuosity in response to purely formal demands.

 

Look, he made a show

Finishing the Hat and Look, I Made a Hat take us into the artisan’s kitchen–not only the studio or the stage where a team comes together but also the kitchen that’s the mind of the creator, who sweats out as much as possible beforehand. Sondheim shares with us the “choices, decisions and mistakes in every attempt to make something that wasn’t there before.”

This is a rarer accomplishment than you might think. Relatively few artists in any medium have the wish or ability to probe their creative process in detail, from conception to minutiae of execution. There are plenty of artists who talk of story sources and bursts of inspiration, but they seldom get down to the compromises, workarounds, and failed efforts. The best such books on popular storytelling, Sidney Lumet’s Making Movies and Patricia Highsmith’s Plotting and Writing Suspense Fiction, are illuminating, but nothing compared to the multilayered self-consciousness Sondheim brings to the task. He notes: “The explication of any craft, when articulated by an experienced practitioner, can be not only intriguing but also valuable.”

It helps when the artist works in brief, segmented forms, such as the songs that make up a musical. They can be analyzed line by line, and can illustrate the challenges of word choice, rhythm, and rhyme as they relate to the ongoing story, the portrayal of character, and of course the score. This engineering side of construction, basic to popular art, is attractive to a practitioner who happens to be a puzzle fiend. He makes every choice a challenge to his ingenuity.

Take one example, “A Little Priest” from Sweeney Todd. Sweeney has decided to turn his lust for personal vengeance into random barber-chair homicide. This scheme suits Mrs. Lovett’s need to produce better meat pies. The result is a list song, a Sondheim favorite. Here’s the rendition of the Broadway original with Len Cariou and Angela Lansbury, for most of us the definitive version.

The couple deliriously catalogue what citizens might furnish appropriate  ingredients. Sondheim’s founding choice: the list will consist not of individuals, who would need naming, but of social types whose titles can rhyme more easily. Sweeney role-plays a customer checking Mrs. Lovett’s imaginary wares.

Todd: What is that?
Mrs. Lovett: It’s priest./ Have a little priest.
Todd: Is it really good?
Mrs, Lovett: Sir, it’s too good, at least./ Then again, they don’t commit sins of the flesh./ So it’s pretty fresh.
Todd: Awful lot of fat.
Mrs. Lovett: Only where it sat.
Todd: Haven’t you got poet/ Or something like that?
Mrs. Lovett: No, you see the trouble with poet/ Is how do you know it’s/ Deceased?/ Try the priest.

Second choice: the song must be constructed so that the rhyming emphasis falls on the types, so the lines end on the name of the profession, as above. Third choice, new constraint: Throughout the play, Sondheim tries for triple rhymes, so one- or two-syllable professions (priest, poet) will allow for that. Fourth choice: the need to build the song out of short lines, which will permit “an increasingly intricate rhyme scheme.” That’s on display here, with priest–least–deceased–priest threaded with flesh–fresh and fat–sat–that and poet–know it.

The song surveys a range of social types, while reminding us of Sweeney’s main target Judge Turpin (“I’ll come again/ When you have judge on the menu”).  The climax of the song declares a perverse commitment to equity. “We’ll not discriminate great from small/ No, we’ll serve anyone/ Meaning anyone–/ And to anyone/ At all.” The catalogue culminates in a sprightly celebration of mass murder.

Finicky as ever, Sondheim confesses that he never liked the line “Meaning anyone,” which is a place-filler, but now he says he knows how to improve it.

Why probe the artist’s craft in such detail? Sondheim notes that journalistic reviewers aren’t trained in technique or don’t know the trade secrets; they’re just declaring what they like or dislike. More intellectual and academic critics are inclined to write broad overviews, usually about the cultural sources and effects of the music. A practitioner who explains the cascade of decisions, freely made or imposed from without, can provide something rare. Just as the sports fan enjoys learning the fine points, learning the tricks of the artist’s trade can boost our appreciation. We can learn to enjoy “the spectacle of skill.”

No surprise: I was reminded of the  historical poetics of cinema. This approach to film studies tries to analyze how filmmakers draw on the menus of options normalized in particular times and places. In trying to discover principles of craft practice, we seek out any information we can glean from artisans. For this purpose, we’ll not discriminate great from small.

 

At the movies

Sondheim with Steven Spielberg on the set of the remake of West Side Story.

The Last of Sheila is a brittle, cynical satire on the film industry while being a well-designed classic mystery. Sheila Greene is struck down by a hit-and-run driver after she storms out of a party she and her husband Clinton have been holding. Under the promise of a potential movie deal, Clinton gathers six suspects on a pleasure cruise on the French Riviera. Once they’re are aboard, Clinton announces the entertainment: a mystery game. Each guest receives a card declaring a secret, such as “You are a homosexual” and “You are an alcoholic.” Clinton assures them that each is simply “a pretend piece of gossip,” but it becomes clear that the game is designed to expose someone, not necessarily the recipient, as guilty of the card’s charge. Meanwhile, Clinton tries to discover Sheila’s killer.

Every night, at the port they visit, the guests must follow clues to link one of them to a past transgression. The game is interrupted when one of the travelers is killed, and so the survivors embark on their own investigation. As in an Agatha Christie novel, the characters are stock types. There’s  the vapid star, her hanger-on husband, the over-the-hill director, the rapacious talent agent, the second-tier screenwriter, and his heiress wife. Again, as in Christie novels like Death on the Nile and Murder on the Orient Express, the yacht cruise isolates the suspects and allows them to debate the identity of the culprit but also also to reveal Clinton’s scheme. In response, the killer has conceived a counterplot for baffling the other guests. Instead of these Christie novels, there is no designated Great Detective to take Hercule Poirot’s place. Any of the people hazarding solutions could be guilty.

In the spirit of the “fair play” detective novel, the audience is provided some clues that flit by, but can be checked on a replay. It’s not a spoiler to note the casual early appearance of an icepick, or the glimpses of some–but not all–of the clue cards.

     

By contrast, as John Dickson Carr points out, the really skillful mystery writer rubs your nose in the clues. It’s not a matter of a passing mention of the color of a man’s tie, or the way a character pronounced a word.

The masterpiece of detection is not constructed from “a” clue, or “a” circumstance. . . . It is not at all necessary to mislead the reader. Merely state your evidence, and the reader will mislead himself. Therefore, the craftsman will do more than mention his clues: he will stress them, dangle them like a watch in front of a baby, and turn them over lovingly  in his hands. 

The best example of this tactic in the film is the photograph at the bottom of this entry. The shot lets us dwell on Clinton’s apparently innocuous photo of his guests. Elsewhere the script plays up that old favorite, the discarded cigarette butt.

Sondheim and Perkins’ passion for classic detection is revealed in a long sequence of pure ratiocination. Across twenty-one minutes in the yacht’s lounge, one guest reconstructs the scenario behind Clinton’s game. The layout of space, in both long shots and shot/reverse shot, is quite precise and varied. Of course there are clues in the behavior of certain characters.

Part of the reconstruction turns out to be erroneous, as in the detective-novel convention positing an initial, faulty solution. In all, The Last of Sheila, while memorializing the sort of scavenger hunts and elaborate games Sondheim put his friends through, remains a rare example of an updating of the classic whodunit.

By contrast, the film adaptation of Sweeney Todd is a suspense thriller. The plot doesn’t hinge on an investigation but a pursuit (although that does yield a surprise revelation). Sondheim considers it the only satisfactory film version of one of his shows. “This is not the movie of a stage show. This is a movie based on a stage show.” But then what was the show? Opera? Operetta? No. “What Sweeney Todd really is is a movie for the stage.”

And a grim, brutal one at that. Tim Burton has developed Sondheim’s original through several cinematic strategies. One emphasizes Sweeney’s willed estrangement from humanity. If the stage Sweeney, often played by a large man, has a certain sweeping bravado, Sweeney on film scowls in soft-spoken fury. His puckered brows and pinched lips are set in a face as unearthly pale as a cadaver’s, or a clown’s. This seething little man is presented as virtually locked in his quarters, peering out at the world on which he will wreak vengeance.

Sweeney’s isolation is given a narcissistic cast when he unpacks his razors. He who has no human ties croons to “my faithful friends.”

     

Mrs. Lovett looks on, trying to secure a connection: “I’m your friend, too.” But he ignores her. Burton provides POV shots of Sweeney’s face reflected in his razor blade, a neat way of showing his self-absorption passing into violence.

At one point, he seems to register Mrs. Lovett’s gestures of affection, and Burton neatly shows the POV reflection shifting to her.

     

His contemplation of her as an ally can, it seems, only see her as another reflection of himself and an instrument of his vengeance–a tool, not a lover.

He rejects her affection (“Leave me”) and he twists the razor, turning the reflection back to him. He’s still locked in his solitary obsession.

               

He finishes the song alone, now even more committed to his mission. A final shot shows him gleefully peering out the window at the city he and his faithful friends will ravage.

Mrs. Lovett becomes his genuine accomplice in the song, “A Little Priest.” In the stage directions, Sondheim asks that Sweeney and Mrs. Lovett use pantomime to evoke the meat pies they want to harvest from Sweeney’s customers. In the film, however, the pies she has already made are surrogates for the ones she proposes.

Given the more tangible setting, Burton returns to the window motif; Mrs. Lovett’s shop becomes an extension of Sweeney’s enclosed world. The couple’s list-song takes place with the two of them at the windows, scanning the street for victims. As you’d expect, they spot a priest first.

     

If the barber shop upstairs is Sweeney’s mission control, the pie shop becomes his supply center. Mrs. Lovett realizes that she can get his friendship only by joining his plan for vengeance, and once he realizes her commitment, she becomes an attractive partner. They pledge their partnership in a rolling-pin waltz, and the sequence ends with a shot that echoes the earlier capstone, but including Mrs. Lovett.

     

There’s a lot more to be said about The Last of Sheila and Sweeney Todd, but I invoke them just to show that Sondheim’s talents can create robust, innovative, sometimes disturbing cinema.

 

I could imagine someone criticizing Sondheim as the ultimate middlebrow artist. Adapting foreign films (Smiles of a Summer Night, Scola’s Passion) and Grand Guignol to the American musical; turning Seurat’s Grande jatte into a living tableau (and naming the painter’s model Dot!); imposing games with time and viewpoint on a showgirls’ reunion; investing fairy-tale optimism with sinister implication–all can seem too clever by half. An objector might say that a Sondheim show doses schmaltz and hokum with just enough formal ingenuity to let audiences feel clever. But I think that sort of having-it-both-ways defines a great deal of popular entertainment, and it has its own value. Bergman and Seurat and Little Red Riding Hood remain unharmed by plays that take them as pretexts for ravishing music and cunning theatrical games. Formal ingenuity is not a small thing.

Sondheim opened new vistas for the musical to explore. I suppose you can call Lin-Manuel Miranda’s Hamilton middlebrow too, making rap and hip-hop safe for people who can afford a Broadway ticket. But if you think Miranda opened new paths, note that he thinks that Sondheim pointed the way. He took advice from Sondheim while writing the show, and he paid tribute in a preface to an interview:

He is musical theater’s greatest lyricist, full stop. The days of competition with other musical theater songwriters are done: We now talk about his work the way we talk about Shakespeare or Dickens or Picasso.

You don’t have to go that far to see that artists at all levels of taste innovate, and their efforts sometimes oblige us to see fresh expressive possibilities in their artforms. If we are all nerds now, we can become connoisseurs of the experimentation–sometimes subtle, sometimes bodacious–that  pervades popular entertainment. Sondheim, meticulous and generous and tirelessly exploring, coaxes us to do that.


I’ve drawn most of my Sondheim quotations from Finishing the Hat: Collected Lyrics (1954-1981) with Attendant Comments, Principles, Heresies, Grudges, Whines and Anecdotes (Knopf, 2010) and Look, I Made a Hat: Collected Lyrics (1981-2011) with Attendant Comments, Amplifications, Dogmas, Harangues, Digressions, Anecdotes and Miscellany (Knopf, 2011).  Other remarks come from Craig Zadan, Sondheim & Company (Harper & Row, 1994) and  Meryl Secrest, Stephen Sondheim (Knopf, 1998). Google Book Search should help you locate my citations by phrase. Steve Swayne provides a thorough analysis of the role of cinema in his career in How Sondheim Found His Sound (University of Michigan Press, 2005), 159-213.

More general books about the craft traditions Sondheim adopts and revises are Philip Furia’s The Poets of Tin Pan Alley: A History of America’s Great Lyricists (Oxford, 1992) and Jack Viertel, The Secret Life of the American Musical: How Broadway Shows Are Built (Farrar Straus Giroux, 2016. I’ve praised and applied Viertel’s account in this earlier entry.

Sondheim discusses the composition of “Someone in a Tree” in this interview with Frank Rich. The second part is especially revealing about Sondheim’s combining music and lyrics through a vamping figure. That accompaniment, detached from the sung melodies, depends on a “gradual change” principle that reminds me of the minimalist music of Glass (Einstein on the Beach, 1975) and Reich (Music for 18 Musicians, 1976) at the same period. A musicologist would probably correct me, but if the affinity holds good it would further show Sondheim’s pluralistic appropriation of many traditions. Thanks to Jeff Smith for discussions of this and for help with other musical matters.

The John Dickson Carr quotation comes from his 1946 essay, “The Grandest Game in the World.” The most complete version of it is in The Door to Doom and Other Detections, ed. Douglas G. Greene (Harper, 1980), 334-335.

Many fine appreciations of Sondheim appeared around his birthday, but I especially like this one by Jennie Singer, packed with clips.

For a long time Kristin and I have studied innovations in popular storytelling, as in her analysis of narrative in the New Hollywood, my book on the same area, her monograph on P. G. Wodehouse, our book on Christopher Nolan, my studies of Hong Kong cinema and 1940s Hollywood, and comments over the years on this blog (e.g., Paranormal Activity and Happy Death Day). For more on Into the Woods, go here. 

The Last of Sheila (Herbert Ross, 1973). This shot harbors a clue–actually, more than one.

P. S. 18 April 2021: Sondheim has been extraordinarily generous in sitting for interviews describing his creative process, and many are stimulating. Alert reader and master interviewer Brian Rose kindly sent me a link to the remarkable 2020 interview with Adam Guettel that you might enjoy.

Let’s play God, imperfectly: BLOOD SIMPLE on the Criterion Channel

Blood Simple (1984).

DB here:

Over the last couple of years I’ve been writing a book on common strategies of popular storytelling in film and other media. I go on to trace how those strategies get worked out in detective stories and thrillers. If you follow this blog, you know that these genres are ones I enjoy and like studying.

So I was happy to offer as one installment of our Criterion Channel series, Observations on Film Art, a short analysis of storytelling strategies in Blood Simple. In it I suggest that although the film has the trappings of a neo-noir–the somewhat downmarket characters, the seedy milieu, the chiaroscuro lighting–in its narrative techniques it’s closer to a Hitchcock thriller. That’s because its manipulation of point of view, one of the resources of popular storytelling, is close to the “partial and misleading omniscience” of the thriller genre and Hitchcock’s narration in particular.

Put it another way. We aren’t restricted to what only one character knows, as in a detective story like The Big Sleep (1946). Instead, Blood Simple steers us selectively from one character to another. So we always know more than any one of them does. That creative choice increases suspense–knowing the dangers that lurk ahead–but it also  summons up an ironic detachment from them, as we watch them make their foolish mistakes. In a word, if you know the film: fish. Here’s another: lighter.

This shifting viewpoint doesn’t give us absolute knowledge, though. There is still some information that slips through the cracks. So we can enjoy superior knowledge in long stretches, while still getting some sudden surprises, or even shocks. (Consider the climax with the perforated wall and the knife at the window.) In their first feature, the Coens prove themselves already fully in control of finely-tuned cinematic storytelling. We’re a step ahead of the characters, but the film is a step ahead of us.

The Coens are also aware of our pleasure in their control. The characteristic Coen awareness, a sly recognition of letting the audience share their power over our access to the story world, is everywhere in evidence. I didn’t have time to mention the shot that everybody remembers, the tracking shot down the bar that simply lifts over the drunk sprawled on the bar top. We become aware of how the movie’s unfolding narration is absolutely ruling how we see this world, and the Coens make a gag out of it. Even the camera-god has a sense of humor.

If you have a chance to watch it, Jeff Smith, Kristin, and I hope you enjoy it–and, of course, the film, which is endlessly rewatchable.


As usual, thanks to the team at Criterion: Peter Becker, Kim Hendrickson, Penelope Bartlett, and their superb postproduction boffins. We recorded the commentary under Covid conditions, with the expert guidance of Erik Gunneson, Meg Hamel, and James Runde.

For more on the Coens’ mastery of storytelling technique, check out the analysis of their fine The Ballad of Buster Scruggs.

You can sample other blog entries on mysteries and thrillers in several entries, in particular our discussions of Hitchcock (of course), and the genre as a whole (here and here and here). An early version of one chapter of my book in progress is devoted to the great Rex Stout. Another chapter will revise what I said about Gone Girl. I also discuss Hollywood’s approach to crime and mysteries in the book Reinventing Hollywood: How 1940s Filmmakers Changed Movie Storytelling.

Blood Simple (1984).

 

A fast-paced cinematic impeachment trial

DB here:

Donald Trump did not lose the 2020 U. S. presidential election. He was inaugurated on 4 March. Joseph Biden was arrested and executed.

Or, if you like: Trump will be sworn in on 20 March, the anniversary of the founding of the Republican Party (in Ripon, Wisconsin). Or Trump is President already and working with the military to prepare the mass executions of Democrats. Meanwhile, the Pope was arrested, or will be.

These are among the many lies and delusions  promulgated by fantasists, particularly QOnan (not a typo). But these are also narratives.

Narrative in many manifestations, from jokes to comic books, is one of my keen interests. While mostly studying storytelling in film, I noticed some years back how the term and concept of “narrative” became a part of political discourse. I used that buzzword as a path into analyzing the yarn-spinning around presidential campaigns (here and here) and, inevitably, stages in the fascist coup attempted by the fumbling but indefatigable Trump regime (here and here and here and here). The coup effort continues, now at the level of Republican state legislatures.

The 6 January assault on the US Capitol, an atrocity many of us followed in real time, has been recast as different stories. The versions range from treating it as an escalation of Trump’s seething rallies to brazenly caricatural accounts, notably the version promulgated by my senator Ron Johnson (R-Wisconsin), a man of invincible ignorance. The versions I want to look at are those laid out in last month’s impeachment trial (Trump’s second) by the Prosecutors from the House of Representatives and the ex-president’s legal Defense team.

The trial began on 9 February with consideration of whether the Senate could impeach a President no longer holding office. That issue was settled by a vote that impeachment could proceed. The trial proper began on 1o February. The charge was that Trump incited the riot that led to the invasion of the Capitol on 6 January 2021.

On 10 and 11 February, the House Impeachment managers presented their case. On 12 February, the Trump attorneys presented their defense, after which both sides entertained questions from senators. The final day, 13 February, was taken up with debates about whether witnesses and further testimony would be invited. After that was resolved, the proceedings ended with a vote on conviction. Sixty-seven votes would have been enough to convict Trump, but only 57 votes for that outcome were cast. Trump was acquitted.

Trump’s original speech to his followers is transcribed here. Official transcripts of the Senate sessions are available on the Congressional Record site. A video record of the proceedings, with good quality on the video clips, is on the C-Span site. This also includes a rolling transcript.

Both sides used film to make their arguments. The House team promised what the press called “a fast-paced cinematic case,” and the Defense responded with pieces of cinema of their own. What resources of cinema did they use? And how do the uses differ?

 

Narratio as you like it

In Film Art: An Introduction, Kristin and I contrast narrative form with other options. We suggest that a film (or a literary text or a TV show, etc.) could be organized in various ways. Take whales. Using categorical form, you might provide a taxonomy of whales. Using rhetorical form, you could make an argument about whale conservation. With associational form, you might provide a poetic meditation on whales as metaphors for strength and grace. With abstract form, you might film whales in ways that turn them into pure patterns of imagery and sound. And using narrative form, you might tell a story about how a stubborn young girl helped get a stranded whale back into the sea.

Any given film could combine these principles. An argument for conserving whales could include sequences that give us information about genus and species, while other sequences could poeticize the great behemoths in the Whitman manner, or let us appreciate their abstract beauty. Since we all like stories, there will be a strong temptation to “narrativize” when we can. Even a film that tries to make an argument is likely to include some passages involving agents and actions, goals and obstacles.

A rhetorical film doesn’t have to include a narrative component, but it may well do so. Our example in Film Art, Pare Lorenz’s The River, relies on the classic rhetorical  structure of problem/ solution. But inserted in that structure we find stories about how the Dust Bowl came to be a disaster area and how the government has taken steps to improve things.

A court case is a prototype of rhetoric. Aristotle accorded little importance to storytelling in forensic rhetoric, but later thinkers recognized the powerful role that it plays. Theorists came to call narratio the portion of the argument that tells the story, or rather competing stories, of the matter in dispute.

All of which is a long-winded way of saying that we shouldn’t be surprised that the House Prosecutors and the White House Defense team would float some stories. But what stories were told? How did the advocates tell them? Do the presentations push beyond narrative to other types of form? And what implications do these strategies have for how political actors use moving-image media?

To analyze the cinematic strategies at work, it’s useful to know how they further the broader cases being made. So for each side, I’ll try to show how its use of cinema grows out of its general rhetorical/narrative strategies. Beyond that, we are in a strange re-creation of the situation dramatized in Fritz Lang’s staggering Fury (1936), a film for our moment if ever there was one. There people accused of lynching a man must watch a courtroom screening of their actions during the crime.

     

Needless to say, there was much less self-examination shown by Republican senators when confronted with powerful sounds and images showing the fury they unleashed and did nothing to forestall.

 

Witnesses, plenty, for the prosecution

Analyzing a piece of rhetoric demands that the audience be considered. Since rhetoric is centrally about persuasion, how is the argument shaped to suit the audience’s inclinations?

Almost no one expected that Trump would be found guilty, so were the House’s efforts to mount a case a hollow exercise? I think not. The House advocates, it seems to me, were addressing two audiences: the general public and, to put it grandiosely, history. Representative Jaime Raskin and his colleagues treated the matter as a criminal case, offering a detailed reconstruction of motive, method, and opportunity. Assuming that future analyses would be skeptical and rigorous, looking for holes in the Prosecution account, they tried to make it tight, evidence-based, and the most plausible explanation of the events of 6 January.

Their account can be revised in the light of new information (such as recent findings of who funded the rally and how it was coordinated with the White House). Still, it aspired to be a reliable first draft of the most accurate and comprehensive story of the event. The managers injected emotion into their account by emphasizing the casualties of the melée, the narrow escape that legislators made, and the damage to American ideals when the seat of government was ravaged.

The Prosecution accordingly constructed a complex, time-looping, “novelistic” narrative. Taking advantage of the primacy effect, they laid out their story in a dense fifteen-minute video at the start of the first day, during the argument about constitutionality. I’ll consider that montage shortly, but the crucial point is that it concentrated on the attack, with immersive and brutal footage.

Over the next two days, the House managers spiraled out from the attack footage by means of a plot structure that starts with a crisis. This strategy introduces us to the action just before the story’s climax and then fills in what led up to it.

The Prosecutors’ first step was to establish the story’s antagonist, the “inciter in chief.” Trump was put front and center in a flurry of quotes pulled from moments before, during, and after the raid, ending with him declaring to his followers: “We love you, you’re very special.” Who opposed this central figure? Interestingly, the Protagonists initially cast the Capitol police, particularly the black officers who were stunned by the racism they confronted.

Settling on Trump as the gang leader who fled the scene, the House team set out to prove, through detailed timelines, that the attack was “well-orchestrated.” The presentations were divided neatly into parts and chapters. The sections are mostly chronological, showing events leading up to the crisis, though further flashbacks are embedded. The first full day of the case provided a “moving spotlight” narration on the events, roaming back through time and skipping around in space. Thanks to cellphones, we can get a godlike view of the whole event–a crucial choice of viewpoint for the prosecution. I reconstruct the plot as follows.

The immediate provocation: Three messages (The Big Lie, Stop the Steal, Fight Like Hell) were hammered on after the election and then restated as talking points during the rally.
    Flashbacks [establishing motive and method]:
     Pre-election: Trump announced his resistance to any reports of losing.
     Election: On the big night, he tried to stop the counting and declared that he won.
     Post-election: He refused to concede, fought in the courts, and tried to round up phantom ballots. After failing to win over reluctant senators and he Department of Justice, he saw Pence as his last chance to overturn the electoral results.
When Pence demurred, Trump sought another avenue for action. He invited his followers to come to DC on 6 January to “stop the steal.” “Will be wild!” “He knew what he was doing.”
     The means: Violence. Trump has a history of support of militias and white supremacists (e.g., Proud Boys). His alt-right supporters reciprocated through social media, championing Trump’s Big Lie and declaring 6 January as Armageddon.
[The opportunity:] Back to the 6 January rally, with a split viewpoint: Trump’s speech is intercut with crowd response, using footage from participants’ phones.
Just after the rally: A timeline from 12:20, when the march to the Capitol began, to 2:10, when marchers overwhelmed the police and broke in.
The attack: The timeline continues to the end of the day. Multiple viewpoints provide phone imagery, surveillance footage, and tracking devices to follow the mob’s movements through the building. The assault is crosscut with the responses of police, legislators, and staff. Pence and others were rushed to safety. The mob filled the Crypt, burst into the House, and ransacked the Senate chamber. Outside, rioters boasted of the destruction. “I bled on it.”
Parallel action: “What was happening at the other end of Pennsylvania Avenue?” A new timeline traces Trump’s action from 12:20 at the rally (“I’ll be there with you”)  onward. At 2:38 Trump acknowledged the attack in a tweet. Meanwhile, his staff and members of Congress pleaded with him to call the marauders back. At 6:01 he urged the mob to go home and “Remember this day forever!”
Dispersal of the mob: At 7:00 PM the Senate certified the Electoral College vote.

Thus ended the first day of the House managers’ incitement case. The first stretch of the next day was devoted to replaying events from different points of view, as a novel or film might take us into individual characters’ experience.

The rioters’ testimony: Trump had a bond with the rioters. They asserted that he summoned them. “Our President wants us here.” “He’ll be happy.” His control over them was demonstrated when he asked them to go and they dispersed.
    Flashbacks establishing the power Trump has over his followers:
     Over the years: Trump roused his crowds with insults, making news on TV and nodding to white supremacists. He praised the militias that invaded the Michigan capitol, a “state-level dress rehearsal” for the 6 January siege. Trump did not upbraid the men who plotted to kidnap the Michigan governor; he joked about it at a rally. “This is his essential M.O.”
Trump’s response after the riot: Even after the rampage Trump declared his speech “totally appropriate.” He showed no remorse and failed to condemn the attacks. Even his order to go home was couched sympathetically (“We love you”). His video of 13 January regretting the violence refused to admit that Biden won the election, thus maintaining the Big Lie. Even some Republicans, including former members of his administration, have castigated his failure of duty .
Consequences for the radical right: Most immediately, there was rejoicing on social media (this was “the first stab”) and the riot was proof-of-concept for further missions. There were reports that his followers planned to attack Biden’s inauguration and are continuing to target minority communities. Long-term, this is Charlottesville’s “Unite the Right” rally writ large. Trump has forged a powerful coalition of the fringe groups on the right
Implications for the stability of government: Trump was prepared to sacrifice members of the legislative branch. Through television reports, he knew that everyone, especially Pence and Pelosi, were in mortal danger. While lawmakers were saying goodbye to their families, Trump abandoned his duty. In addition, Trump had no concern for the police or the other staff caught up in the turmoil. Some people were killed and many seriously injured. Ex-service staff called the siege worse than what they had seen in the Middle East.
Implications for national security: Other countries rejoiced at the failure to protect the center of government. (Russia: “Democracy is over!”) Allies were shocked and horrified.
Anticipations of defense arguments (confutatio in the rhetorical trade): This isn’t a Free Speech issue; indeed, Trump launched an attack on the First Amendment. Nor has Trump been denied Due Process.
Conclusion (peroratio): Summing up, with emotional weight added. If this isn’t an impeachable offense, what is?

My outline can’t do justice to the fine-grained detail of the evidence provided in the speeches and the fusillade of images and sounds. I want just to indicate that the argument that a “high crime” was committed and needs to be punished is framed almost completely on the basis of a chronological chain of cause and effect–i.e., a narrative. That narrative indicates motive, method, and opportunity, but these categories of culpability lie beneath the plot we’re given. A man incited a riot by priming the Big Lie, building a violent following, scheduling a provocative gathering during a crucial moment in the transfer of state power, and ranting at a rally.

Beyond the commission of the crime, there is a plea not usually heard in a criminal case. Letting this prominent man off not only sends a harmful message but threatens the future of American democracy.

On the first day, the Prosecutors screened a video that condensed the narrative they propose. While the clip includes some footage not previously shown in the media, it doesn’t include all the new material, some of it quite powerful. Those shots are sprinkled through presentations shown in the following days. The later presentations also recycle many of the shots and sounds presented in this montage, usually to recontextualize them. This montage introduces some of the stylistic methods that would reappear in the days to come.

From Trump’s speech at the rally to his final benediction (“Great patriots. . . .Remember this day forever!”), the video traces a completed arc of action There is no voice-over narrator to guide us, although somber intertitles, white on black, frame the sections. Similarly, there is no music. We hear only the voices and noises in the scenes, although subtitles are brought in to clarify muddy lines and to stress major turns in the action (“They’re leaving!”).

The overall style is that of direct cinema, where the technical roughness conveys the authenticity of events caught on the fly. While some news reportage is included, there’s no editorial comment or punditry. The fact that most of the footage was recorded by the rioters themselves adds conviction. Later presentations will draw on surveillance video to provide a more detached perspective, but the immediate purpose is to immerse us in the participants’ viewpoints. Their blasé attitude toward chronicling their crimes will support the Prosecution’s claim that they believed they had the president’s permission to invade the building and fight for him.

From the start, the video stresses the almost mystical bond between Trump and his supporters. Like Lang’s courtroom scene in Fury, the rally is cut to show both the event and the crowd’s reaction. In eyeline-match editing Trump’s acolytes respond eagerly to his words. After he says go to the Capitol, they reply, “Take the Capitol!” Later his speech is played over shots of their march–the sort of “sonic crosscutting” pioneered by Lang, again, in M (1931). When POTUS says he hopes Pence has the courage to reverse the electoral count, we cut to a shot of Pence. Soon enough the rioters will take up the call: “Traitor Pence!” It’s all part of a single chain of causality.

The Prosecutors anticipated that the Defense would play up Trump’s pro forma aside that the demonstration should be peaceful. But what exactly were the demonstrators supposed to do at the Capitol? To show that violence was really their only plausible option, “Stop the Steal” becomes a unifying thread of the piece. Announced by Trump, picked up by the crowd, and flashed before us in signs and hollers during the insurrection, the phrase suggests not a strategic plan for later primary campaigns but a demand for an immediate tactic, something short and sharp to halt the electoral process.

Crosscutting to show simultaneous events is central to the video’s style. Crosscutting links Trump’s speech to his crowds on the march. It broadens out to trace, for the sake of suspense, many fronts in the assault on the Capitol. While Mitch McConnell rebuts the Big Lie, crowds burst over the barriers and pound on police. Images of the panicked legislators and staff alternate with events elsewhere in the facility, with the danger coming ever closer. Moreover, thanks to some slight sound bridges, the editing gives the impression that the mob is just out of frame in certain shots of the House and Senate chambers.We aren’t far from Griffith’s celebrated crosscutting in The Lonely Villa (1909) and other last-minute-rescue plots.

Just as important, the crosscutting anticipates the parallel timelines that the Prosecution will flesh out in the days to come. Another method, that of split-screen commentary and comparison, is established; it will become useful for yielding the God’s-eye view central to the Prosecution’s case.

     

The video saves the most harrowing footage for its climax, as we see the rampaging mob battling police in the corridors. Ashlii Bobbitt is shot dead. Handheld imagery from the packed crowd records brutality at close quarters. Shots of the raiders hurling themselves forward in unison are intercut with shots of a guard crying out when his head is pinned in a doorway.

     

The epilogue takes us back outside the building. Trump’s video calling for peace is undercut by savage assaults on police on the front steps. Throughout, occasional shots of seething bystanders on the sidewalk have functioned as an ominous chorus (“30,000 guns”/ “Next trip”). Now the video ends on the sidewalk, where an alt-rightist calls for assaults on state governments. A title summing up the seven deaths that resulted from the attack is followed by a shot of Trump’s tweet, bookending the opening, thanking the raiders.

In the days afterward, the Prosecution’s case became an audiovisual one; in the age of Powerpoint, verbal arguments are supported by the evidence of images and sounds. Slides of tweets, Facebook pages, news headlines, and other texts were added to a surge of video clips, along with audio recordings from the beleaguered officers. The direct-cinema quality of the material is amped up by the presentation of body-camera footage from an officer beaten down by the mob.

The fallen-camera convention of pseudo-documentary films (like, say, Cloverfield and Georg) wouldn’t be as powerful if we didn’t know that in violent situations like this, cameras–and camera wielders–do drop. In this instance, the approximate optical POV of the stomped officer makes his attackers seem even more brutal.

The Trump Defense team will claim that the senators know all about what happened and so don’t need to have the entire attack played over and over. But the Prosecutors maintain that in fact we still don’t know a lot, and at the time nobody had a view of the whole drama as it unfolded. That’s the rationale for the timelines, for all the crosscutting and side=by-side juxtaposition of images of things happening at the same time.

This omniscient viewpoint is crucial to the case, and is seen with particular density and vividness during Delegate Stacey Plaskett’s use of visual evidence to complement legal documents. She mobilizes Facebook posts, surveillance footage, audio recordings from barricaded staff, and scale models to trace the rioters’ progress through the building. The section is capped by a confession from perhaps the most famous perp.

These God’s-eye views give a new meaning to the crosscutting initiated in the opening montage, filling in both specific locations and moment-by-moment changes in the rioters’ positions. The bonding of the mob with Trump is recalled when one uses Trump’s nickname, “Crazy Nancy.” And the scrutiny of the footage yields details like the stun gun carried by Barnett, aka Bigo. (Bigo, who “will not back down,” is now apprehensive about his prison stay.) The sweeping range of knowledge is counterbalanced by moments of intimacy, such as the whispers of staffers into their phones.

One advantage of the House managers’ omniscient framing is to mark out two layers of complicity. The parallel timelines make Trump’s actions during and after the assault integral to the act of incitement in the rally. As with an arsonist who sets a blaze and then sticks around to watch and impede efforts to put it out, Trump’s dereliction of duty is presented as of a piece with what Mitch McConnell would later denounce as his “provocation.”

Apart from the feelingful appeals of the succession of speakers laying out the case, the aggressiveness of the action captured in imagery and sound (the thudding doors, the babble of obscenities) and the panicked voices of police and staffers calling for help intensify the emotional dimension of the case.

 

Playing the Trump card

The Defense team were also playing to the general public, particularly the right-wing media. But their chief audience was the Republican senators, who needed a peg to hang their inclinations on. Declining to offer point-by-point rebuttals of the Prosecutors’ case presented a wide-ranging batch of arguments, a buffet of options rather than a tightly focused throughline. There are some sketchy counter-narratives in there, but they’re accompanied by other points, some legalistic, some evidentiary, and some simply suggestive.

The governing emotion, I think, was barely controlled outrage. The House charge, the Defense maintained, reflected a deep hatred of the President and the ordinary Americans he represented. The impeachment was an act of “political vengeance.” While the Defense often said that it would be better if everyone simply calmed down, there was little doubt that the best response to Democratic intemperance was equally fierce counterattack.

Trump’s Defense team relied heavily on arguments of principle. They didn’t try to refute the Prosecutors’ narrative. Indeed, they simply chopped off large stretches of it. They agreed that the sacking of the Capitol was “horrific,” but all the violence there was irrelevant to Trump’s case. The real perpetrators must be rounded up and brought to justice. If he didn’t incite the attack, then he’s not responsible, and all the shocking footage, all the timelines and tracking of the mob, is so much sensationalism.

How to show he did not incite the attack? Mostly by appeal to categorical form. For one thing, the Defense wanted to explore the relevance of two legal concepts. These the Prosecution correctly anticipated in its pre-rebuttal.

First, the Defense team asserted that Trump was denied Due Process on many grounds. Evidence that would be challenged in court was put forth as reliable. Video clips were edited to omit statements in Trump’s favor, and some of the slides of tweets had errors in dating or attribution.

The House evidence of Trump’s encouragement of violence was likewise claimed to be selective and taken out of context. Trump is a man of peace and he has never stoked anger–unlike the Democrats who in many speeches have indulged in hateful rhetoric. Nor is questioning an election result inherently inciting; many Democrats claimed that the 2016 election was stolen.

All this mitigating evidence would have come to light had Trump been given Due Process. Witnesses would have been called; depositions would have been taken; the Defense could have interrogated those accounts. As presented, the case for incitement would not stand up in any court. It is a prime example of the category “Politically motivated injustice.”

Second, Trump was said to be simply exercising his First Amendment rights. Contrary to the opinions of the 140 law professors who declared that Trump’s remarks fail as Free Speech, his statements at the rally are, perhaps lamentably, within the range of current public discourse. His urging his followers to fight simply echoes what Democrats have told their crowds on many occasions.

The Defense videos, fewer in number than the Prosecutors but recycling many shots several times, were designed to drive home these points. Significantly, the first video was designed to establish the “political vengeance” motive.

Apart from ascribing a vendetta to the impeachment initiative, this montage sets up the impromptu category of “Currently acceptable political discourse.” Specifically, the clip plants the “fight” motif that will dominate the later proceedings. It also prepares us for the throbbing music that will underscore the big sequence to come.

The Defense began by bracketing off the entire insurrection, explaining that they wouldn’t be showing footage of it.

They [the Prosecutors] don’t need to show you movies to show you that the riot happened here. We will stipulate that it happened, and you know all about it. 

The Prosecutors’ embrace of the omniscient perspective sought to show that none of the legislators could have “known all about it” then or even now. But the Defense’s dismissal of the entirety of the sacking of the Capitol cleared the way for the case predicated on categorical form. In backing up the argument against conviction, the Defense’s major video sequence assembled samples of prevailing political rhetoric drawn from Democrats’ speeches and statements.


Unlike the Prosecutors, the Defense team relied largely on material from the news media. (Presumably these excerpts don’t count as Fake News.) And, again, the clip is accompanied by throbbing, ominous music evocative of an urban action thriller. If the House case resembled a direct-cinema documentary, the Defense case is stylistically closer to a political ad. All it lacks is a title telling the viewer the sponsor who approved it, and where to send money.

Some of these Democrats’ remarks, taken singly, are pretty abrasive. But although the extremism on display is regrettable, the Defense argued that the Democrats have a perfect right to make them. So does Trump. The Democrats were not telling their crowds to resort to violence, and neither was he.

There is a certain irony in the Trump team’s use of editing to reiterate a concept. It’s a technique we can trace back to other Griffith films, like A Corner in Wheat (1909), but it’s most associated with Soviet filmmakers like Eisenstein and Vertov. The pounding rhythm of “fight, fight, fight” has some of the same force we get in those percussive intertitles in Russian silent classics.

And maybe, the Defense asked, the Democrats aren’t so innocent? A sketchy narrative is invoked in the first sequence above, which arranges statements chronologically to suggest a crescendo of cries for impeachment from the beginning of Trump’s term. The editing of the last montage moves even more strongly toward narrative patterning. Didn’t Democrats’ incendiary rhetoric play some causal role in the summer 2020 civic unrest?

Unlike the Prosecutors, the Defense sequence gives no specifics, no timelines or maps, no link between these sound bites and particular actions in particular places. Many of the Democratic speakers are declaring that they will fight; they aren’t all urging their followers to do so. Unlike the MAGA-blazoned rioters in the Capitol, the street fighters in the videos go unidentified. The episodes of street fighting are a cinematic version of Republicans’ equivalency between Black Lives Matter protests and the assault on the Capitol. This loosely plotted story gives many senators a peg to hang their preferences on.

What happened when the Defense got down to linking up events more exactly? In contrast to the panoramic sweep of the Democrats’ narrative, the Defense scope is very narrow: essentially, just the rally. Apart from some parsing of what Trump meant by “very fine people on both sides” in Charlottesville, there is no mention of his long record of inflammatory remarks. Instead, the Defense constructed a story of what Trump was really asking his crowd to do in that particular speech. They did not try to explicate the many other statements on 6 January, such as the tweet at the top of this section, that congratulate the rioters. The Defense team rewrote the rally as an isolated event, a stand-alone scene whose principal lines demand careful interpretation.

Split screen is used to show the edit points of House videos and to suggest that what Trump said in toto was mitigating.

According to this story, on the day in question Trump was not demanding violence. His call for courage and fighting was simply a request for his supporters to challenge errant Republicans in future elections. If any Trump supporters did engage in the riot, it was because they misunderstood him.

Their error was no greater than that of the Prosecutors, who initially believed that Trump was calling for blood. Once the House managers realized their mistake, Defense attorneys reasoned, they found other pretexts for attacking him. They whipped up the myth of the Big Lie about electoral finagling. Why is it a myth? Because in previous certification ceremonies Democrats accused electors of irregularity as well. They didn’t get away with it because they lacked Senate support and were shut down by the presiding official–in one case, none other than then-VP Joe Biden. Nonetheless, questioning electors’ bona fides, the Defense argued, is not uncommon after a presidential election.

Another, rather sketchy Defense story invoked the possibility that the violence visited on the Capitol was conducted by groups independent of Trump’s control. Some rioters came planning to attack the building and used his rally as a pretext for their mission. They “hijacked the event.” Trump could not incite something that was going to take place anyway. Moreover, some of those people were anti-Trump activists wanting to make mischief. Little evidence of these claims is adduced, but again a proper trial would presumably have brought such information to light.

In all, the people to blame are the Trump followers who misunderstood his message, the alt-right renegades who wanted to make trouble, and the undercover provocateurs from the left. Trump’s remarks, protected by Free Speech, remain distinct from the actions of these troublemakers.

A larger narrative, ultimately more powerful for the public audience, involves motives. First there was Trump’s motive. Trump, on the basis of his speeches, was said to be the most pro-police president in history. He would never encourage his followers to put peace officers in danger. But what about the other side?

Just as the House Prosecutors sought to show that Trump’s desire to retain power made him ever more desperate, the Defense claimed that what fueled this entire proceeding was irrational hatred of a duly elected President. Democrats have “lusted” for impeachment since 2017. The attack on the Capitol gave them the chance to unseat and disgrace him–and keep him from running again. Another montage of quotations, echoing the initial sequence, was invoked. In sum, this is a political trial and the defendant is being railroaded.

The Prosecutors created an intricate narrative centering on Trump and his MAGA army, with those on the other side–from police and staff to legislators both Democratic and Republican–cast as victims. A deadly attack was summoned into existence to keep a man in power. The Defense avoided a long-form narrative but pivoted to categorical organization: Trump’s situation became an instance of denied Due Process, and his tirade exemplified Free Speech in the currently overheated partisan dialogue.

But the Defense didn’t wholly avoid telling a story. It focused on the rally as a single incident that has been misconstrued.  And their broader story, probably the overarching one, pits Trump and his patriotic followers against the vengeful Democratic party and its army of summertime hooligans determined to unelect him, and make him forever unelectable.

 

There’s a lot more to be said about the structure and style of these fraught sessions of the Senate. They will be scrutinized for years by students of law, politics, social change–and, I hope, media. We can bring to the table some tools that allow us to analyze how traditions and conventions of cinematic forms and styles interact with the exigencies of occasion and ideology. They alert us to the ways in which our buttons are pushed and our thoughts and feelings are steered this way and that.

I should say, in case you’re wondering, that I find the Defense case preposterous–sophistic, evasive, streaked with irrevelancies and contradictions. That’s because I side with Aristotle’s view that rhetoric should be a packaging system for sound thinking. Logic and science are truth-tracking, whereas rhetoric makes no discoveries. It is a skill set, a tool kit for presenting ideas and opinions in ways that people will accept. Rhetoric is needed because a lay audience needs help in grasping the premises and conclusions of specialized inquiry. That means, for Aristotle, rooting arguments in not only plausible inference and ethos (trust in the speaker’s moral character, as displayed in the speech) but also appropriate emotional appeals that will engage the audience in the pursuit of understanding.

Ideally, the ideas and opinions set forth are the ones with the best rational and empirical backing. But they may not be. Prejudice, stereotypes, and quick-and-dirty inferences are the rhetorician’s stock-in-trade. Still, we can be on guard against weak rhetoric. Educators could go back to teaching high schoolers the fallacies of informal reasoning; a huge swath of political babble, including the Trump Defense case,  exemplifies tu quoque and ad hominem. And when films and videos take rhetorical form, our skepticism can be reinforced by awareness of how movies work. Studying film can alert us to the ways we can be gulled, blinded, or reinforced in our own prejudices.

It can also show how a rhetorical case can rest on a foundation of more or less plausible explanation. That, I think, is where the Prosecutors’ case leads us, thanks to its strict structure of likely cause and effect. But no less than the Defense, the House managers used the resources of the medium to quicken the impact of the documents they assembled. Fortunately, thanks to modern media, those documents are themselves vivid records in the moment. They will remain, after many viewings for decades to come, powerfully convincing–at least, to people of open mind. The question is how many of those people will still be around.


Geoffrey O’Brien has an engaging discussion of the ways current conspiracy theories echo those of earlier eras in “Hitler in Antarctica” at the New York Review of Books site.

I discuss the role of the crisis structure in classical American film in Reinventing Hollywood: How 1940s Filmmakers Changed Movie Storytelling.

It may be relevant that I thought and still think that Bill Clinton, a man utterly lacking in personal honor, should have been convicted in his 1999 impeachment exercise.

P.S. 21 March 2020: The New York Times has assembled an astonishing montage of mostly new footage of the combat outside the Capitol, showing the police trying to hold the insurrectionists back. Twelve Republican representatives voted against awarding Congressional medals to these officers. One of those opposing the legislation, Rep. Bob Good, was discovered to have a connection to the riot when it was revealed that his district’s party director and her husband were part of the crowd converging on the facility.

     

Left: Sen. Josh Hawley (R-MO) in the Senate Gallery during the impeachment trial; drawing by Art Lien.  Right: Extract from Prosecutors’ video montage. The speaker says: “There’s gotta be something in here we can fucking use against these scumbags.”

Madalena, Rosalind, and Suzanna: More Rotterdam revelations

Madalena (2021).

DB here:

A mixure of moods and tones for our final communiqué from the International Film Festival Rotterdam. Its fiftieth year has been a lively one.

 

The Madalena mystery

Madalena (2021).

In earlier entries (especially here) I’ve noted that the thriller genre is well-adapted to festival circulation. It doesn’t require the budget of a blockbuster. It can attract major actors who want tricky parts to play. It can be shot on contemporary locations. And the appeal to suspense and surprise fits comfortably with edgy narrative strategies favored by art cinema. At the limit, a filmmaker can arouse our thriller appetites and then try a bait-and-switch that not only warps the genre’s conventions but sets us thinking.

The Brazilian film Madalena, by Madiano Marcheti, starts as a classic mystery. In a vast field of soy, reas stalk gracefully as a monstrous pesticide-sprayer grinds toward them. But among the rows lies a corpse.

What follows is more fractured and prismatic. A first section attaches us to Luci, a friend of Madalena’s who works as manager of a club. She also picks up work dancing for TV commercials, one set in that very acreage. Then we follow Cristiano, whose father owns the land and demands he hustle to harvest. A third section takes us with trans woman Bianca and her girlfriends, who sort through Madalena’s belongings before setting out for a day of driving, swimming, gossiping, and teasing one another, the memory of Madalena never far from their thoughts.

Marcheti skips some of the standard scenes. We never see the police investigation, or even the discovery of the body.  The crime plot has been a pretext to reveal a cross-section of life in the community, from the wealthy farmers to the cottages where the staff live. The resolution shifts the question of who did it to the broader impact of the death, and how it stands for a horrifying statistic: Brazil has the world’s biggest murder rate of transgendered people.

Throughout, sexualization of bodies is a central motif. Luci and her posse hang out at curbside, Bianca and her posse turn tricks and find boyfriends, and Cristiano, after sizing up the crowd at Luci’s bar, winds up dancing with himself in mirror reflection.

To say much more would spoil things, so I’ll just note that this story is filmed with a pictorial intelligence that one seldom sees these days.  The imagery of the soy fields is at once magnificent and ominous. Drones hover over it like birds of prey, and its horizon haunts the people’s lives.

     

     

Overwhelming as the landscape is, it doesn’t blot out the characters’ routines and the crises that disrupt them. Moving from Luci’s aimless days and nights to Cristiano’s panic to Nadia’s quiet  tribute to Madalena, a locket set adrift in the stream that runs along the field, the film pauses for intimate moments. It reminded me a bit of Varda’s great Vagabond (Sans toit ni loi), in which an enigmatic figure’s fate charts the range of human indifference, but also affords glimpses of sympathy.

An informative discussion of the film with Marcheti is provided by IFFR here.

 

As we too like it

As We Like It (2021).

This movie saw me coming a mile away. It does for As You Like It what Lurhmann did for Romeo and Juliet, but to an Asia-pop beat. Four romantic couples lose and rediscover one another in a magical milieu–not the Forest of Arden (currently under corporate development) but a district of Taipei with no Web connections.  In Heaven, a sign informs us, there is no Internet.

Accordingly, people must deliver messages in person, seek out each other by dint of shoe leather and motorbikes, and actually meet face to face. So Rosalind’s quest to find her father the Duke (a genial tycoon) intertwines with Orlando’s search for her. But of course she’s disguised as a boy and aided by Celia, a fortune-teller who’s the dream girl of Orlando’s sidekick Dope.

The film’s world is maximum kawai, pushing beyond camp to a fangirl fantasy of irresponsible sweetness. This candy-colored city, with its pink blimps and anime posters, spills over with tweens, teens, and twentysomethings shopping in malls, flirting at stalls, and sipping bubble tea.

     

In the process, old stuff becomes cool.  Tradition, in the form of calligraphy and handmade paper, is a retro decorator choice, while letting your date clean your ears old-style makes him a friend with benefits.

     

It might all seem sappy, but like Tati’s Play Time and Wong Kar-wai’s Chungking Express, the film seeks to distill authentic poignancy out of kitsch, schlock, consumer clichés, lethal cuteness, and the detritus of urban lives. Frivolity must be good for something; why else would God give us giggles?  Comic form, as Shakespeare acknowledged, redeems a lot of silliness, especially if the gags are hurled at us with the ruthless conviction that anything goes.

Did I mention that all the roles are played by female performers?

In a switcheroo on Elizabethan theatre, globalization inverts the Globe. The film, a final title tells us, is dedicated to Shakespeare “but also to the patriarchy who would not allow female actors upon the stage.” The frisson is akin to that of Tsui Hark’s Peking Opera Blues and The East Is Red, where gender-blurring yields both humor and genuine feeling. From instant to instant, you see a character go male or female or something in between; a painted-on mustache and a swaggering gait become cosplay, not deep definitions of you. Unless you want it to be.

     

Identities are fluid. Okay, says Orlando, so he falls in love with Rosalind, then Roosevelt, and refinds him/her as Rose. What’s in a name? You get to call yourself, and be, what you wish, and love whoever.

     

Not a fresh-minted message these days, but the sparkle comes with how it’s all carried off. Every scene finds a clever way to amuse or bemuse. When Rosalind as Roosevelt slips into a trim suit, she pads out the crotch with a towel, and teeny gull-like waves waft out. That’s soft power, the equivalent of a mystic ring. Eventually she has to go along when Orlando visits the men’s room. While he stands at the pissoir, she ducks into a toilet pretending to take a dump, her groans covering the sound of peeling open a maxi-pad.

     

The project was co-directed. Wei Ying-chuan, a graduate of NYU’s Educational Theatre division, is a founder of Shakespeare’s Wild Sisters Group in Taiwan. Chen Hung-yi’s feature The Last Painting was chosen for IFFR in 2017 and won a best feature award at Cines del Sur. The pair bring an unflagging energy to the task of creating a paradise of easy living and loving–bereft of villains, open to any piece of harmless fun and heartbreak. As We Like It is a must for every LBGTQ film event, but its hella dirty fun for any festival whatsoever. Couples welcome.

Again, the IFFR provides a fine discussion of the film with the directors, moderated by our old friend Shelly Kraicer.

 

St. Tropez, mon amour

Suzanna Andler (2020).

Eric Bentley once described great serious literature as “soap opera plus.” Anna Karenina, Othello, and the rest offer us tormented love affairs, sexual jealousy, hidden schemes, and forced confessions of betrayal, but it’s all endowed with wider significance through characterization, implication, style. But can we have soap opera minus?

In Daisy Kenyon, Anatomy of a Murder, and other films, Preminger moves in this direction, banking the fires of conflicts drawn from lurid bestsellers, but other filmmakers have gone further in de-dramatizing melodrama. Dreyer’s Gertrud and some of Oliveira’s adaptations offer examples. Here we have the classic fraught situations, but muffled and fragmented and punctured through long pauses and wayward, looping, maddeningly banal conversations.

Marguerite Duras made this artistic strategy peculiarly her own, notably in her masterpiece India Song (1974) and its counterpart Son nom de Venise en Calcutta désert (1976). In a curious reversal, she often prepared the film first and then published the text as a quasi-play, as if scraping away the luscious imagery and ripe sound would create something even purer, soap opera distilled to Racinian starkness.

Suzanna Andler, a Duras theatre piece from 1968, has now been adapted to film featuring Charlotte Gainsbourg and three other players. The result isn’t as severe as the play reads, since director Benoît Jacquot has filmed it in a gorgeous villa overlooking the Mediterranean. It remains, however, in the tradition of kammerspiel. The bulk of the action takes place in a salon and the terrace outside, with one sequence, also in the play, set on a rocky beach.

The situation is sheer bourgeois melodrama. Suzanna is in a loveless marriage with the philandering Jean. She has apparently stayed with him for the sake of their children and the wealthy life they lead. Now Michel, a young journalist, has tempted her into a love affair, and she has for the first time cheated on Jean–who seems okay with it. Today’s crisis, if this counts as one, is her need to decide: Will she lease this villa for the summer with the kids? Or will she accept Michel’s invitation to go to Cannes? In the course of about four hours, she may make up her mind.

If some of my synopsis seems hazy, it’s partly because the exposition comes out in bits as Suzanna and others chat about her past, and partly because what she says may not be wholly truthful. She sometimes admits to lying. And what was her relation to the never-seen Bernard Fontaine, who has just been killed in a car crash? The blurry backstory is one strategy Duras uses to tamp down the melodrama, which usually gives its plots clear-cut contours and definite revelations.

In filming the play, Jacquot has taken an approach that approximates the rigor of Duras’s aesthetic. He has shot the blocks of action using slightly different techniques. Not for him obvious alternatives like color/ black-and-white or a range of tonalities. The differences are made harder to spot because Jacquot has not given us separate chapters corresponding to the act divisions; the scenes blend, punctuated only by long shots. So there are stylistic spoilers coming up.

At the start, Suzanna is shown the house by the real estate agent de Rivière. This segment is filmed in distant shots that emphasize the landscape and straight-on views of the sitting room opening out onto the terrace and the sea. The agent is seen from behind or at a distance, while the few close views we get concentrate on Susanna.

     

Staying behind alone, she falls asleep and awakes when Michel enters. This is a second phase of the play’s first act, and now Jacquot’s camera setups take a more oblique view of the room. The full-length windows dominate again, but now at an angle that recalls the magic mirror of India Song (on which Jacquot was an assistant).

     

The couple is often seen at a distance, but now closer views of Suzanna emphasize the mirror motif.

     

At a high point, the camera celebrates a momentary reconciliation with a track in to their embrace (the first such florid move in the film, I think).

In the sequence corresponding to the second act, Suzanna meets her friend (and Jean’s ex-lover) Monique. On the beach they talk of their pasts. Now the conversation is rendered in many rapid, tight shots of the two women. The orthodox shot/ reverse shot setups are sometimes given a strange emphasis when instead of A/B alternation we get two (but only two) variants of a view of each one as she speaks (A1/A2, then B1/B2). So a cut like this::

     

. . . is followed by ones like these:

     

Back at the villa, Suzanna answers a phone call from Jean, and they discuss their plans, with the uncertainty typical of all the film’s conversations. This scene is handled in circular tracking shots around Suzanna, from a moderately close distance.

     

As the conversation ends, Michel returns. After he reveals some key information about his relation to Jean, he stretches out on the sofa. In a long take running several minutes, the camera swings around them in a half-circle, clockwise and counterclockwise, often adjusting to her shifts in position.

     

The changing angle also captures Suzanna perched against a painting of very 60s boomerang shapes that echo the camera’s trajectory.

As the action approaches what might be a climax, Michel drifts out to the terrace and sits on the balustrade above the sea. Suzanna approaches.

Telling you what happens next would truly be a spoiler. On seeing it, I thought it was something that Jacquot added to the play, but nope . . . it’s there in the text, and he’s perfectly faithful to it.

As if all this patterning doesn’t look finicky enough, the scene on the beach is punctuated by a single shot of the Quai de Passy with a Métro train rumbling by.

This bump comes exactly halfway through the film, at the moment Suzanna mentions the surge of attraction she felt when Michel looked at her on their first encounter. Believe it or not, the line in the play also comes midway through the text. This alien shot functions expressively, I think. It underscores the epiphany Susanna felt upon learning she might be loved. Another filmmaker might have stressed the moment with music under her monologue, but Jacquot goes for a formal bonus: breaking the visual texture just here further articulates the design of his film.

The rigorous geometry Jacquot has clamped down on the play is interesting in itself, and the abstract array of options adds, I think, to the hieratic quality Duras is after. Yet each style matches the tenor of the action it carries and doesn’t conceal the subdued feelings rippling through the scenes. This dimension depends on Gainsbourg–her slim silhouette, her microdress, and especially her face, with her alert chin and hard mouth. Her vacillations have nothing of the diva about them, but still she stands forth as a new avatar of The Confused Woman so beloved of art cinema (Voyage to Italy, L’Avventura, The Headless Woman). Without those closer shots, the film might fall flat.

Once asked what would be his ideal final shot, Jacquot replied: “A distinctive glance [un certain regard] in close-up.” His ending delivers that.

Duras is doing something similar to what Wei and Chen do in As We Like It. She is seeking genuine emotion in clichés (unfaithful husband, wrung-out wife, surly rescuer). But she  hasn’t exempted her characters from social critique. Hiroshima, mon amour renders the meeting of two lovers as an intertwining of two national histories. The colonialists of India Song, drifting through their sparsely attended embassy parties, trying to replicate salon society in the tropics, cannot hear the voices offscreen of the people they subjugate. Likewise, Suzanna’s anxiety may or may not register some distant tremors. In summer of 1968 her world is sliding into something she isn’t prepared for. Far away from St. Tropez, in Paris students are hoping to find their own beach, but they’re doing it by tearing up the pavement.


Again, thanks to Gerwin Tamsma, Monika Hyatt, Frédérique Nijman, and their colleagues at the International Film Festival Rotterdam for allowing us to visit their event virtually. Here’s to another fifty years of ambitious programming!

A very helpful edition of Suzanna Andler has been published in conjunction with the film’s release. It contains a lot of stimulating background information and critical commentary. Bentley’s comment about “soap opera plus” comes from The Life of the Drama (Applause, 1991), 14. Thanks to Kelley Conway for sharing with me the Jacquot interview  in “Réponses à tout,” Libération (14 May 2004), 1.

Jacqout’s rendition of Duras’s play exemplifies what I called in Narration in the Fiction Film “parametric narration.” This rare approach consists of playing out a range of expressive possibilities, scene by scene, in ways that both shape the ongoing plot and “anthologize” sharply contrasting cinematic techniques. Noël Burch first proposed this idea in his enduring Theory of Film Practice (1973), although Eisenstein and Bazin envisioned it. But then they envisioned everything.

P. S. later: The Rotterdam prize winners have been announced (per Variety).

As We Like It (2021).

David Bordwell
top of page

have comments about the state of this website? go here