Children of Divorce (1927)

Kristin here:

In 2008, David and I attended Il Cinema Ritrovato for the sixth time. The Hollywood director being featured that year was Josef von Sternberg. We saw some of the gorgeous prints on show, most notably (for me), a chance to re-watch the underrated Thunderbolt (1929), his first sound feature. To us the big auteur of the year, however, was Lev Kuleshov. Astonishingly, we only blogged from the festival once that year. It was a busy summer.

We and a great many other festival-goers lined up to see one of the rarer items on the program: Children of Divorce, credited to Frank Lloyd but with reportedly about half the footage re-shot anonymously by von Sternberg. We and a considerable number of those festival-goers did not get into the auditorium. For some reason this rare, legendary film was shown in the smallest venue, the Mastroianni, which was packed, with the two side aisles full of standees. (See the bottom image of our blog from the festival, as we caught a glimpse of those lucky enough to get in before  retreating to find something else to watch or perhaps to wander around the always enticing Film Book Fair.)

Eight years later, in 2016, Flicker Alley released a Blu-ray/DVD combo of the Library of Congress’ 4K restoration of Children of Divorce. That two-disc set is out of print, but last month a MOD (manufacture-on-demand) version was made available. It is a single disc, Blu-ray only, and its main supplement is a pretty good one-hour documentary on Clara Bow produced in 1999 for TCM. (The original booklet is not included.) Somehow we missed the original release, but now I have a chance to catch up with this elusive film.

Naturally I wanted to find out if any information on which scenes of the film von Sternberg re-shot were available, I looked at various internet sources, including books in our library the reviews of the original 2016 Flicker Alley release and books. I found nothing on the subject, not even from film buffs speculating on the basis of style which scenes were his. I suppose one reason why so little discussion of von Sternberg’s contribution is that for many viewers and purchasers of the Flicker Alley discs, this a Clara Bow and Gary Cooper film. Those online reviews from 2016 focus on them rather than on the two very different directors of Children of Divorce. The third star, Esther Ralston is excellent as Jean, and in 1929 she would go on to play the title character in The Case of Lena Smith, Sternberg’s last silent film, which survives only in a fragment. Still, she doesn’t have the lingering reputation and devoted following that her co-stars still enjoy.

Spoilers ahead, including a revelation of the ending.

 

Von Sternberg’s account

In trying to discover von Sternberg’s contribution, we seem to be entirely dependent on von Sternberg’s sketchy recollections of his work on the film in his memoir, Fun in a Chinese Laundry (1965). Needless to say, one cannot take his account as the unvarnished truth, but it may contain some elements of that commodity.

According to von Sternberg, B. P. Schulberg, head of Paramount, asked him to watch a film that the studio considered unreleasable. Could von Sternberg could improve it by adding the sort of clever intertitles that he was then known for?

I looked at the film as he requested. Its title was Children of Divorce, and it had been made by a prominent director, Frank Lloyd, normally an effective director of commercial films. This one was a sad affair, containing theatricals by Gary Cooper and Clara Bow, the “It” girl of her day. I reported back to the executive who had backed this venture with a million dollars, and told him that no skill of mine could restore life to the film by injecting text into the mouths of the players. I suggested that half the film be remade.

That Lloyd should fall down so badly seems odd, given that he had been cranking out films at a rapid pace since 1915. These included films that would be considered star vehicles and adaptations of popular literature, such as the first version of The Sea Hawk in 1924. He would soon go on to win two early best-director Oscars, one for the Greta Garbo vehicle The Divine Woman (1928) and one for Cavalcade (1933), as well as being nominated for what is probably his best-known film, Mutiny on the Bounty (1935)–which won Best Picture despite not winning any other of its seven other nominated categories. Cavalcade often shows up on lists of the worst films ever to won the highest prize, but nevertheless, it seems odd that a director so respected within the industry that he won an Oscar for a film made the year following Children of Divorce‘s release should fall down so badly in creating the latter. We shall probably never know what cause this strange failure.

I must admit to never having seen a Frank Lloyd film, but von Sternberg’s assessment of him as an “effective director” seems lukewarm. It suggests that although Lloyd was a good, solid Hollywood practitioner, he had created an unwonted lemon.

In assessing the verity of von Sternberg’s account, we should pause over von Sternberg’s claim that Children of Divorce had had a million-dollar budget. That was well above the average cost of a film in those days. (Von Stroheim’s Foolish Wives, with its giant Monte-Carlo set, had gone over budget to become the first million-dollar film in 1922, and Universal head Carl Laemmle made the best of the situation by blazoning the extravagant budget in publicity for the film.) The original costs estimate sheet on Flicker Alley’s site, along with other documents from the film’s production, puts the total budget at $334,000–pre-re-shoots–a far more plausible sum. Possibly, writing in the wake of the scandal over the cost overruns leading to an estimated $44 million budge for Cleopatra (1963), he had to boost the paltry-sounding budget of a late-1920s feature at least to seven figures to show that he was doing an important favor for Schulberg.

In response to von Sternberg’s suggestion that half the film be remade, Schulberg (according to von Sternberg) said that he could not allot another five weeks to making changes in a project that still might flop. Von Sternberg writes, “I carelessly replied that I could remake half the film in three days and turn over a successful version to him.”

Schulberg asked, “What can I do about the sets? They’ve been torn down and the stages are full.”

Again according to von Sternberg, “I told him to have a tent erected for use as a stage and to dig out the sets from storage, and not to bother about anything else.” Schulberg agreed.

Von Sternberg’s entire account of the process of re-shooting the scenes runs as follows:

The tent went up and the transfusion began. A rainstorm came and lasted three days and three nights. We waded through the new scenes, now and then dodging a heavy burst of water that penetrated the canvas overhead. The crew that helped me and the poor actors that I mercilessly put through their new paces had to take a prolonged rest cure when I had finished with them. To match the scenes I wished to retain I had to use the style of the replaced director. The assignment was completed on time and, after removing the old scenes and replacing them with the new material, I showed the film to the flabbergasted executives.

This tells us discouragingly little. The weather and the cast’s exhaustion make for a dramatic tale but tell us nothing about the film itself. The only statement of interest is “To match the scenes I wished to retain I had to use the style of the replaced director.” Von Sternberg certainly knew the well-established norms of Hollywood filmmaking and if he deliberately set to blend his work into an existing film, his style could be difficult to detect. After all, we don’t know exactly what was wrong with the original version. Bad performances? Von Sternberg’s reference to “theatricals by Gary Cooper and Clara Bow” suggests that this may have been the case. Bad storytelling? Von Sternberg doesn’t mention whether he re-wrote any of the scenes he revised.

The new version premiered on April 2, 1927. Von Sternberg was not paid for his labors (he says), but he did prove himself as a director. Schulberg allowed him to direct an entire feature himself. Later that year, on August 20, his first silent masterpiece, Underworld, came out. This item in Variety of March, 1927, seems to confirm the basic claim of von Sternberg’s account. Several trade papers mentioned that Sternberg had re-shot some scenes, but none specifies which ones.

Briefly, Children of Divorce concerns two girls whose recently divorced parents have dumped them in a convenient French convent to be cared for. The parents return to the free life of single people, visiting their daughters infrequently. Kitty, a frightened, lonely girl, befriends the kindly Jean. Kitty introduces Jean to Teddy, an old neighbor of Kitty’s.  Teddy and Jean hit it off and promise to marry as adults.

Kitty grows up and becomes Clara Bow. She is poor and must give up her love for the noble but impoverished Prince Ludovico (Einar Hansen) to pursue Teddy, now grown up to be Gary Cooper and very rich. His meeting with the grown-up Jean (Esther Ralston), herself now described as the richest woman in the USA, rekindles their love. Despite knowing this, Kitty uses the occasion of a drunken party to trick Teddy into marrying her, and they have a daughter.

Teddy wants to divorce Kitty and marry Jean, but insists that the daughter must not be as they were, a child of divorce. Kitty thinks this would free her to marry Ludo, but he informs her that his religion would not permit him to marry a divorced woman. She determines to hold onto Teddy. Jean decides to marry Ludo, thus quashing Teddy’s hopes of marrying her. Pure misery abounds–all the result, one way or another, of divorce.

 

What did von Sternberg do?

Naturally one would wish to know which scenes are Lloyd’s and which Sternberg’s. In her brief program notes for the film in the Bologna catalogue, Janet Bergstrom confidently comments, “You can’t miss the shadow language of his scenes.” Perhaps not, but there are actually very few Sternbergian shadowy shots in the film, and those are often brief touches. Here are the main examples. The second scene of the film shows the young Kitty (later to grow into Clara Bow) during her first night in the French convent where her newly divorced mother has dumped her (bottom). In another such atmospheric shot, Teddy reads Jean’s letter refusing to marry him if he divorces Kitty (top).

Other such shots occur occasionally. When Teddy and Kitty’s daughter totters winsomely down the stairs during a party, the elaborate railing casts a shadow over her. Jean’s first sight of her fixes her determination that for the sake of the child she will not agree to marry Teddy if he divorces Kitty.

   

These are the only shots that one could argue contained strongly Sternbergian shadows. The opening scene in the convent as Kitty’s mother leaves her has a shadow of an offscreen window on the wall at the center of the shot, but surely one could not be certain that an image directed by Lloyd could not have such a modest shadow simply to create the atmosphere of the setting.

Indeed, it seems like a typical shot from an A picture of the mid-1920s.

I do not think that one can determine which shots and scenes were redone by von Sternberg just by the lighting. “To match the scenes I wished to retain,” von Sternberg must have used the three-point lighting system that had been established as the norm during the second half of the 1910s. He did a very good job, and there are many scenes that look like they could have been shot by Lloyd or Sternberg, adhering to that same norm and of course using the same sets and costumes. Which of the two directors made this image, with its exemplary use of the three-point lighting system devised by Hollywood practitioners in the second half of the 1910s?

Von Sternberg’s reference to Lloyd as an “effective” director may sound like diplomatic, lukewarm praise, but he might have meant that any good Hollywood director with an experienced cinematographer could have produced such an image, with its perfect blend of key, fill, and back lighting.

Similarly, we might be tempted to identify this glamor shot of Ralston as directed by von Sternberg, but it seems a fairly conventional approach to filming a beautiful star in the late 1920s.

And here’s an impressive depth staging in the scene where Luco tells Kitty that he could never marry a divorced woman for religious reasons.

It is tempting to think that the stark juxtaposition of a sharply in-focus foreground cut off from the soft-focus background by the edge-lit fringes on the curtains is a von Sternbergian touch, but we cannot be certain that Lloyd and a good cinematographer could not create the same effect.

 

A possible, probable von Sternberg scene

The big climactic scene sure looks like a very skillful director made it. One is tempted to say that that director was von Sternberg.

The action brings to a head Kitty’s machinations. Having been rejected by Luco, she realizes that her refusal to let Teddy and Jean be together is making them miserable and is not the best thing for her daughter, either. She writes a suicide note addressed to Jean, hugs her daughter for the last time, and heads into her bedroom clutching a bottle of poison.

Here is a breakdown of the scene. It begins with Kitty reading a newspaper. We see her point-of-view, revealing that Luco is going to marry Jean. Luco has already rejected her, but this scene is the moment when she learns that he is to marry her friend. The newspaper shot is followed by a closer view of Kitty.

      

She looks up and to her right, her face grim. It is notable that Bow has been de-glamorized for this scene, with her eyebrows minimized and her hair swept back from her face in a matronly style quite different from her usual bouncy curls. The camera pulls in toward her until it is in extreme close-up.

     

The pull-in reveals a tear that has slid down from her right eye. It is worth pointing out that this is Clara Bow, noted for her flapper image, always lively and carefree. Clearly she could play drama and even melodrama. Perhaps this is an instance of von Sternberg’s famous ability to direct female performances.

The intense close view is interrupted by a cut to Kitty dropping the newspaper. A long shot follows as she stands and moves to sit down at the desk. A cut-in to a medium shot shows her thinking and beginning to write.

     

This leads to her point-of-view on what she writes. We learn only that it is a letter to Jean and that Kitty has had a realization. A return to the medium-shot framing shows her reacting to a knock at the offscreen door and saying, “Come in” as she hides the letter under the desk-blotter. An eyeline-match cut shows her daughter entering. A pan follows as she crosses to the desk and Kitty kneels to embrace her.

      

A cut-in shows Kitty emotionally embracing the girl. During the course of this, the girl’s hat falls off. In a return to the medium-long shot, Kitty carries the girl to her nurse. Another shot shows her returning to fetch the fallen hat and hugging it while displaying despair.

      

A more distant framing shows her putting the hat back on the child. The nurse’s cheery expression as she helps put the hat on contrasts greatly with Kitty’s unnoticed emotion. Once the others leave, Kitty returns to sit at the desk and pulls out the unfinished letter. Again we see her point-of-view as she completes the letter and we realize that it is an apology and suicide note, telling Jean to marry Teddy and raise the daughter that should have been hers. A return to the medium shot shows her putting the letter into an envelope.

      

A rather clumsy and unnecessary cut-in to a tighter shot of Kitty might suggest a blend of Lloyd and von Sternberg footage is occurring, but the lighting on the set, Kitty’s hair-do and costume are identical, so the footage is probably all from the same director. A return to the previous medium shot shows Kitty hesitating as she finishes addressing the envelope and presses a button to summon a servant. A cut back to a more distant shot shows her rising and starting rightward toward the door.

     

A cut to the doorway shows Kitty handing the envelope to a maid, who leaves. Kitty than moves away from the camera to reach for a small cabinet on a table. A cut-in with match-on-action shows her unlocking the cabinet and taking out an object barely recognizable as a small bottle.

      

A cut to the mirror above the cabinet shows Kitty looking at herself, with the image of her face going out of focus. A cut to a medium-long shot leads to the camera panning with Kitty as she moves to a door at the rear.

     

She opens the door and walks into the distance as the camera tracks slowly back.

   

That’s a pretty flashy scene, and it certainly seems more likely than not that von Sternberg directed it.

Once again Flicker Alley deserves credit for bringing us another important film from the silent era.

 

---

Thanks to our friends at Flicker Alley for facilitating this entry.

 

 

The End of the World (1916).

DB here:

The pull-quote might be “Gripping entertainment and a vivid introduction to storytelling strategies characteristic of Danish silent cinema!”  (Too long for a poster, though.) It appears in my essay on a remarkable silent film you may not know. I bet you’d like it.

Danish cinema has gripped my interest for about fifty years. Like most cinéphiles, I started with Dreyer, moved on to Christensen, and then just tried to keep up with trends leading to Scherfig, Vinterberg, Winding Refn, Anders Thomas Jensen, and Dogme. There always seemed to be a new comedy or noir or psychological drama or just weird-ass experiment to keep my loyalty (most recently, the well-crafted Another Round).

One of our first blog entries, on 20 October 2006, was devoted to an anthology on the great film company Nordisk. Soon I was chattering about von Trier’s editing in The Boss of It All and surveying a big batch of recent releases.

Now this national cinema’s silent-era history is coming steadily online. The Danes are too modest to brag about the enormous accomplishment of making so many beautifully restored classics available for anyone to watch. But here they are, accompanied by thematic essays from critics and historians.

Like other Little Cinemas That Could (Hong Kong, Taiwan. Iran), Denmark attracts me because it has shown what can be done a lot of imagination on smallish budgets. Or sometimes, biggish budgets. That’s an impulse that emerged in the 1910s when Nordisk was struggling to keep a foothold in the international market during the Great War. One result was a pair of remarkable spectacles.

A Trip to Mars (Himmelskibet, 1918) is a massive, nutty plea for peace and international—make that interplanetary—understanding. The Martians are more or less like us, except they don’t kill other creatures, which leaves them time to assemble in carefully picturesque crowds and invest in ambitious infrastructure projects.

The other big Nordisk production was The End of the World (Verdens Undergang, 1916). A comet is plunging toward earth. Can we avoid collision? Or at least survive?

All the conventions of the cosmic disaster movie (Armageddon, Independence Day, 2012) are already in place. We have the innocent family, the corrupt capitalist squeezing money out of catastrophe, the scientists trying to calm the public, and of course the separated lovers who must find one another in the midst of chaos.

The special effects range from passable to truly impressive, as in the model of the village under fiery bombardment, surmounting today’s entry. The comet’s approach is cleverly suggested as a blip in the sky, and the shots of the heroine’s drowned neighborhood are splendid.

     

Just as remarkable are other technical achievements. The lighting in the underground passages of the capitalist’s mansion, with its Caligariesque steps, could teach the Germans a few tricks, and the miners’ fierce assault on the plutocrats is cut with rowdy, immersive vigor.

     

     

August Blom had made his reputation with Asta Nielsen dramas and another would-be blockbuster (Atlantis, 1913). He’s often considered a stolid director, but The End of the World seems to me an underrated achievement. Dismissed by many critics as over-produced, its ambitious spectacle is probably more to our current taste for overwhelming scale. For us, it seems, too much is never enough.

So I recommend to your attention this remarkable movie. As usual, I throw in a case for the 1910s as one of the great and glorious eras of film history. You can handily sample further evidence in the film links alongside the essay.

---

Thanks to Thomas Christensen and his colleagues at the Danish Film Archive. It was fun!

There’s always more to say about the Danes. Outside our blog entries, I’ve written about Nordisk and the “tableau aesthetic” and on early Dreyer in another essay on the Danish Film Institute site.

The End of the World (1916).

Kristin here:

To the general film-going public, old films are in black-and-white. They may be vaguely aware that before The Wizard of Oz and Gone with the Wind, color film was invented.

The history of film color, however, is vastly more complicated than that. Prof. Barbara Flueckiger, of the University of Zurich, has devoted much of her career to studying that history. With Eva Hielscher and Nadine Wieylisbach, she co-edited the 2020 collection, Color Mania: The Material of Color in Photography and Film (Zurich: Eds. Lars Müller/ Fotomuseum Winterthur). Barbara has also spent the past decade leading a team who have created a recently inaugurated and invaluable website that acts as a boundless resource for information on color processes.

We are delighted that Barbara has accepted our invitation to write a guest blog entry for us. She describes the website and gives a succinct outline of the history of film color, loaded with beautiful illustrative frames. Most of these were taken from original archival prints that reveal how seldom–especially in this age of digital home video–we see color films as they looked when they were released.

 

Barbara Flueckiger

From their earliest days, films were colored. During the first three decades, most color imagery was obtained by applying dyes to black and white prints, either by hand, through stencils, or as tinting and toning of the filmstrips. From the beginning, however, many ideas emerged to capture colors directly on film as so-called mimetic colors. That could be done either by optical and mechanical means, such as colored rotary filters, or by chemical interventions, often in combination with optical configurations of cameras. Several hundred analog color processes and film stocks were invented in the first 100 years of film history. Many of them were never successful commercially.

This history is largely unknown to the general audience as well as to many film scholars and historians.

To close this gap in our knowledge, in 2011 I started to develop the Timeline of Historical Film Colors, a comprehensive web resource. I wanted to document the development of film colors from their prehistory in still photography in the 19^th century to the latest developments in the analog domain. As of 2021, the platform contains hundreds of primary and secondary sources, patents, links, selected analyses, physical measurements and downloads, as well as more than 23,000 photographs of historical film prints and negatives. These items provide film historians, researchers, archivists, curators, film restorers and students easy access to a vast array of information. A tagging system connects the entries, galleries, photos and quotes to an underlying thesaurus containing certain topics, persons, aesthetic concepts, technologies, archives, genres, persons or companies. A comparison function allows side-by-side inspection of different prints of the same film.

 

Why film color?

High-resolution photographs displayed in galleries are a central part of the Timeline of Historical Film Colors. Early on I developed a method to photographically capture and document historical film materials in a standardized way. It uses a modular and calibrated camera set-up based on a DSLR camera with a macro lens and remote control from the computer to adjust all the parameters. It is crucial to show the full range of color processes in an aesthetically pleasing way, one that aims at recreating the visual impression on the bench, including the edge information and color distribution in the perforation area. These elements are vital for the identification of film stocks and the genealogy of prints.

These photos allow researchers and students to examine individual historical prints, since they often have to work with less-than-ideal digitizations on DVDs and Blu-rays that are just a faint echo of the historical source material. In recent years this photographic method has been adopted by my teams in the current research projects. Some archives, such as the Academy Film Archive, have started to use the method, and the BFI National Archive and the George Eastman Museum plan to do so soon.

During the last years my teams and I visited many archives in Europe, the US and Japan to take these photographs, such as the Harvard Film Archive, EYE Filmmuseum Amsterdam, National Film Archive Prague, Deutsche Kinemathek Berlin, the Academy Film Archive, the Library of Congress, George Eastman Museum, the BFI National Archive, Cinémathèque française Paris, the UCLA Film & Television Archive, Bundesarchiv Filmarchiv Berlin, Museum of Modern Art, DFF Deutsches Filminstitut & Filmmuseum Frankfurt, the National Film Archive of Japan and others.

 

The lost colors of film history

Most films produced before the mid-1930s have been passed on in black-and-white prints. It was not until the famous FIAF conference in Brighton 1978 that the colors of the first decades of film history began to attract some attention from insider circles focusing on silent film.

To this day, the lack of awareness of film’s colorful past has persisted. Early applied colors such as tinting, toning, hand-coloring, and stencil-coloring are ephemeral by nature, since each exhibition print was dyed separately, in a variety of shades and hues. Moreover, these prints were produced with highly flammable nitrate cellulose as a base. Many deadly cinema fires in the early decades of the 20^th century demonstrated the dangers of nitrate stock. Therefore, many original colored film prints have been hidden in cans sitting on the shelves in archives’ nitrate vaults. These facilities are fitted with special safety measures such as break-off walls and earth dams.

Eventually in the 1950s safety celluloid film stocks replaced nitrate. From that point on, new prints of colored early films were made on safety stock from the black-and-white camera negatives, intermediate negatives, or positive distribution prints. When colored distribution prints were used, the new copies were usually made only in black-and-white.

In the early 1980s a second threat to the history of colors in film became apparent. Martin Scorsese was among the prominent filmmakers and scholars who rang the alarm bell over the fading of so-called chromogenic stocks produced from the late 1930s to the 1980s. Due to the physical decay of mainly the cyan dye in these film stocks, original prints become nearly monochromatic, retaining mainly colors in the magenta to red spectrum. To this day, dye fading has remained one of the most pervasive problems for the search of authentic film colors.

 

Applied colors

During the first three decades, so-called applied colors dominated. Historians estimate that about 80% of film prints were colored by tinting, toning, or hand- and stencil-coloring.

Tinting means submerging black-and-white film positives into dye baths, so that the prints’ gelatin emulsion acquired a more or less uniform, mostly monochrome color. Tinting can be identified by the inspection of the perforation area that is also uniformly colored. Toning, on the other hand, is a complementary process whereby the silver image is replaced by colored metallic pigments (metallic toning) or dyes (mordant or dye toning). In contrast to tinting, toning leaves the perforation area mostly colorless.

For these coloring processes the individual prints had to be cut into segments that were then dyed in batches and reassembled into the final distribution print. As a result, individual prints can vary considerably in their color schemes.

Whether tinting and toning schemes vary due to cultural norms and tastes has remained a topic of debate. To a high degree it is also uncertain who made the decisions about the coloring, except for cases where scripts, production notes, or film negatives indicate the attribution of colors. In addition to colored prints there were so-called copyright books that show the color scheme by single frames attached to the pages of the booklets, deposited at the Library of Congress by distributor George Kleine. Subtle shades emerged that often make it difficult to distinguish between the two, because the black silver image gives way to nuanced interactions with the tinting dyes in middle tones.

In some cases, combining tinting and toning allowed for two colors to appear within a single image.

The range and variety are even greater in the case of hand and stencil coloring. These techniques generally required the application of  up to six dyes on each individual frame, either by hand through tiny brushes or by cut-out stencils. These laborious processes were demanding, given the small image area and the huge number of frames, generally 16 to 18 per second of running time. Hand-colored films show an uneven application of dyes with soft transitions between individual colors. For stencil coloring, each dye necessitated a separate, colorless print from which the stencils were cut out by needles or metallic styluses. As a result, shapes appear more or less sharp-edged. It was a mechanized version of hand coloring that allowed the coloring of feature-length films and higher numbers of distribution prints. Over the years, improved techniques were introduced to transfer the shapes from projected magnifications onto the film prints with the help of pantographs.

 

Needless to say, stencil and hand coloring were reserved for more ambitious or luxurious films. However, they also allowed for the creation of a higher reality effect in documentaries, travelogues, or fashion films by anticipating the development of mimetic colors. Exotic places, ethnicities, or historical settings were among the prevailing topics of stencil-colored films.

 

 

In fact, the richness and scope of stencil-colored films can be fascinating to the modern viewer. That holds true for both the bolder color in the first decade of the 20^th century or the more nuanced pastel shades that became increasingly prevalent in the 1920s.

 

A special case of applied colors is the Handschiegl process, a printing process developed by Max Handschiegl and Alvin Wyckoff, often used in Cecil B. DeMille’s films, especially for title cards. It produces highly detailed and precise colors with stunning effects.

 

Mimetic colors

Already in France in the 1860s, Charles Cros and Louis Ducos du Hauron separately wrote descriptions of many of the principles for achieving mimetic colors in still photography. As it turned out, however, it was a much more demanding task to develop solutions for moving pictures. Some of the problems related to the high throughput during projection of 16 or more single frames per second. Other problems resulted from much higher requirements for image size on the big cinema screen, where resolution and registration were paramount. Due to the rapid succession of frames necessary for the illusion of movement, minute deviations occurring between frames created disturbing amounts of flicker or color fringing. Contemporary commentators often labeled the result as “color bombardment” that caused “eye strain”.

To this day, mimetic colors combine two to four color components either in additive or subtractive admixtures. In the 19^th century their development followed psychophysical insights into the human visual system by Thomas Young and Hermann von Helmholtz. They showed that color impressions are the result of physiological sensors in the human retina sensitive to three different spectral ranges of the visible light.

 

Additive colors

Additive admixtures operate with colored light where the sum of the three additive primaries red, green and blue results in white light. The earliest attempts to create colors on the screen by optical means employed additive principles by rotary filters in front of the camera and projector respectively. These included the three-color Turner Lee and the most successful additive two-color process Kinemacolor.

 

In Kinemacolor, rotary filters in red and green spinning in front of camera and projector recorded and transmitted the color information by temporal synthesis. The impression of color was created in the eyes of the spectators. Based on contemporary reports and digital reconstructions, the poor quality and limited color spectrum were readily apparent. Due to the temporal shift between the two successive frames with the red and green color separations, Kinemacolor and all processes operating by the same principle created color fringes and a headache-inducing amount of flicker.

Mroz Farbenfilm. Urlaubfarbenfilm F. Apfelthaler (AUT 1932, Friedrich Apfelthaler). Credit: Österreichisches Filmmuseum. Video and reconstruction by Giorgio Trumpy, David Pfluger and Martin Weiss.

Attempts with temporal synthesis were followed by additive processes that employed spatial synthesis by the application of beam splitters. In this configuration, up to three color records were taken through filters simultaneously to eliminate temporal parallax. But this approach introduced spatial parallax instead, and this arrangement could also create color fringes by poor registration. Gaumont’s Chronochrome process was certainly the most convincing attempt to combine three color separations with this principle.

A third type operating with additive admixtures of light are the so-called screen processes. Additive colors are mixed, either by random, small-scale mosaic image elements or by organization into lines. Color impressions in these systems result from the fusion of the individual dots or lines into red, green and blue in visual perception. The effect resembles that of pointillist paintings where colors are divided into small dots.

Lenticular screen processes, by contrast, combine tiny lenses imprinted onto a black-and-white film strip with colored filters in front of the camera and projector.

Mosaic screens created by colored potato starch were popular in still photography with the Lumière brothers’ Autochromes, introduced in 1907 and widely used by professionals and advanced amateurs. The principle was later adopted in the Cinécolor process for color film but failed due to the uneven distribution of the starch particles. Among the line screen processes, Dufaycolor was the most successful one, widely used in documentaries and famously in Len Lye’s experimental films with direct animation painted directly onto the film strip and then captured and distributed on Dufaycolor film stock. Apparently, Lye was not convinced by the somewhat muted color palettes of the process.

 

 

 

 

Lenticular films such as Kodacolor were also mostly used for home movies with the exception of Thomson color for Jacques Tati’s Jour de fête (France, 1949).

None of the additive principles proved to be successful for the long term. Many of them required special installations in the cinema, and most delivered poor results, most notably dim images.

 

Subtractive colors

Finally, subtractive admixtures became the norm. The three primaries cyan, yellow and magenta filtered the light, with black being the sum of these three subtractive colors. Two or three differently colored emulsion layers are attached to the film base, on one side or both sides of the film.

Most early two-color films were using double coated film stock. The earliest one was Kodachrome Two-color developed in 1915, presented in 1916 with the short film Concerning $1000, but mostly in use in the 1920s for fashion films, for the dance film The Flute of Krishna (USA 1926) by choreographer Martha Graham and for the experimental film [Kaleidoscope] by Loyd A. Jones. Kodachrome Two-color film was shot through a beam splitter and combined two emulsions in orange-red and bluish green on either side of the film carrier, with beige, brown and golden tones in the spectrum between the two color components.

 

A large range of subtractive two-color processes on double-coated stocks emerged in the 1920s and the first half of the 1930s. These included Multicolor, Polychromide, Sirius Farbenfilm, Sennett Color, and Ufacolor. Many of them were short-lived and tested only with short films or commercials. Even failures, however, can help us understand the technical development and the basic principles applied. Prizma II, one of the many processes invented by William van Doren Kelley, was more widely used for travelogues produced by the Prizma Company, for commercials, and for the first feature film in color, The Glorious Adventure (Great Britain,1922, J. Stuart Blackton). At splices the two color components orange and cyan become visible. Overall, Prizma II has a pale and often dirty-looking appearance with occasional patches in orange-red popping out on flowers, fruit or costumes.

 

In the 1930s subtractive processes turned to three colors, most famously with Technicolor No. IV and the subsequent Technicolor No. V, which was printed from chromogenic camera negatives.  Founded in 1915, the Technicolor Company went through many failures and set-backs, with the exception of a short color rush in the late 1920s with the two-color dye-transfer process Technicolor No. III. Following the series Great Events with 12 short films produced by the Technicolor company to establish the process, mostly musicals and a few other genres exploited the two-color process during this short boom. But some of them are highly remarkable, with sophisticated camerawork by Technicolor’s own cinematographer Ray Rennahan, including the musical Whoopee! (USA 1930, Thornton Freeland) choreographed by Busby Berkeley, King of Jazz (USA 1930, John Murray Anderson, Pál Fejös), Doctor X (USA 1932, Michael Curtiz), and Mystery of the Wax Museum (USA 1933, Michael Curtiz).

 

While the technologies applied in Technicolor’s various processes changed considerably over the years, the beam splitter was one of the few constants. Both in Technicolor No. II and III, a beam splitter separated the two color records and captured them mirrored upside down on one black-and-white negative. The bulky and heavy Technicolor No. IV camera recorded the color separations on three black-and-white 35 mm negatives. From these three negatives matrices were produced as wash-off reliefs, ready for the dye-transfer of the three primaries onto the positive print. The result was a series of color images, along with the frame lines and the soundtrack as silver images.

For almost two decades Technicolor dominated the market for high-quality color films. Part of its success was due to a comprehensive package that included the camera, specialized cinematographers, and all the lab works executed exclusively in Technicolor’s own plants. One of the building blocks of Technicolor’s long-term dominance, however, was the so-called Color Advisory Service, famously led by color consultant Natalie M. Kalmus. She defined aesthetic guidelines for film productions shot with the process, informed by color norms related to concepts of “elevated taste,” located in a broader cultural context with references to the concept of “color consciousness.”

 

 

Despite all the efforts to control the color schemes, people often associate Technicolor with highly saturated, deep colors. On close inspection in our detailed analyses of color films, however, we have observed that many films adhere to the rules with mostly restrained color schemes and unsaturated backgrounds to guarantee optimal figure/ground separation. But there are also deviations from these self-imposed norms, surprisingly clashing hues even in films produced with Natalie M. Kalmus as color consultant.

Moreover, there is a great variability of different looks and color applications during the almost two decades. Individual color aesthetics were related to personal styles of cinematographers, directors or production companies, genres or changing preferences in fashion and design, and changing color compounds and recipes employed in the process. Technicolor’s idiosyncrasies – what we perceive as typical “Technicolor look” – are mostly due to the dye-transfer process itself. Pasty, dense colors in patchy structures create an almost opaque appearance on the screen, an effect somewhat like oil paint. When we work with the film elements on the bench in archives, we not only have to increase exposure considerably due to the density of the film stock, but we also notice the color layer’s almost sticky viscosity, often visible as a relief on the surface.

Compared to Technicolor, Gasparcolor produced much more saturated, brilliant and luminous colors. In fact, the process, developed in the early 1930s by Hungarian emigré Béla Gaspar in Germany, was possibly the most advanced and complex process at the time. In its principle–the silver dye-bleach process described by Raphaël E. Liesegang in the late 1890s – the silver acted as a catalyst for the local destruction of the dyes embedded in the three emulsion layers on the two sides of a reversal positive. It is thus a chromolytic reversal process. Due to the political circumstances during the Third Reich in Germany, Gaspar eventually had to flee.

Like Technicolor Gasparcolor required the recording of three color separations on black-and-white negatives. Since most of the Gasparcolor films were animations, these separations were captured in succession on adjacent film frames but could of course also have been shot through a beam splitter similar to Technicolor No. IV. In fact, only one documentary is widely known, Colour on the Thames (Great Britain, 1936), shot by Adrian Cornwell-Clyne. Among the films produced with Gasparcolor are famous avant-garde experimental films by Oskar Fischinger, Hans Fischinger, and Len Lye. Gasparcolor prints can easily be identified by the black perforation area and the colored soundtrack.

Both Technicolor and Gasparcolor prints stored in archives are in remarkably good shape, due to their stable colors. Ironically, chromogenic film stocks, the technical principle that ultimately won the competition and became the new standard, had the least stable dyes. Chromogenic means that the dyes need to be developed after exposure. Embedded in the emulsion of a single strip of film stock are three or more layers. These layers are sensitive to different spectra. All contain silver halides and the color-forming substances, so-called dye couplers that are subsequently developed into dyes. In a second stage the silver is bleached out and leaves the color information in the form of finely dispersed dye clouds in the three or more emulsion layers in cyan, magenta and yellow. The result is a highly translucent, glowing image whose fine-grain structure depends on the speed of the film stock. The slower the speed, the finer the grain.

In contrast to Technicolor the shooting of the chromogenic film could be done with normal cameras on one negative or camera reversal. Chromogenic films increasingly became the norm, starting with Agfa’s first negative-positive process Agfacolor. Emerging in the late 1930s, Agfacolor was promoted by German propaganda in a bid to counteract Technicolor’s dominance. Agfacolor had particularly soft colors in the pastel range with a typical, slightly darkened orange-tomato red. Difficulties in the blue range produced turquoise shades that become quite apparent in skies. Greens had a tendency to look brownish or blackened; shadows had a greenish tinge. Chromogenic multilayer film stocks were incredibly difficult to balance and to produce, requiring a high level of knowledge in physics and chemistry.

 

After World War II ended, the Allies were able to exploit German color-film patents. The result was the appearance of Fujicolor, Eastman Color, and many derivatives, such as Ferraniacolor, Ansco Color, and Sovcolor. The worldwide adoption of color in film production soon followed.

 

 

 

A plurality of styles emerged, less defined by technical limitations than by cultural contexts and individual preferences of filmmakers, art directors, costume designers, and cinematographers. Color aesthetics in film are not only created by hues, color schemes, and color contrasts, but also by lighting styles, by material properties of surfaces and textures, by depth of field, image composition, and by movement. The combination of these factors influences the image’s figure-ground relationships.

In the course of our research, we investigated a large corpus of more than 400 films – mainly from 1895 to 1995 – with a computer assisted workflow. A video annotation software has been developed based on our approach since 2017, when we figured out that tools available then were not well suited to the detailed annotation and visual analysis of film (color) aesthetics. The visual analysis and annotation software VIAN has been created by Gaudenz Halter in collaboration with the Visualization and MultiMedia Lab of the University of Zurich. The tools enable researchers to create detailed analyses including figure/ground separation and a large range of visualizations that make diachronic developments immediately evident or support the testing of new hypotheses.

A deepened understanding of color film technologies and aesthetics is an essential prerequisite for the scientifically sound digitization and restoration of color films, which is one of the most pressing topics today and therefore remains at the center of our research activities.

---

Acknowledgements

I would like to express my immense gratitude to Kristin Thompson and David Bordwell for publishing this blog post and for all the inspiration that guided my research.
A huge thank you to my teams ERC Advanced Grant FilmColors, SNSF Film Colors. Technology, Cultures, Institutions, ERC Proof-of-Concept VeCoScan.

Special gratitude is dedicated to all the film archives with their wonderful collections.

This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme, grant agreement No 670446 FilmColors.

The Blue Angel (1930)

Kristin here:

The end of this horrendous year is fast approaching, though we are still facing a long struggle to end the pandemic. On a happier note, it is also time for a little distraction in the form of my annual year-end round-up of the ten best films of ninety years ago. 1930 was also a grim year, with the Depression well established and nowhere close to its end. Herbert Hoover decided not to use federal governmental powers to help end the crisis (sound familiar?) and eventually was cashiered into a 32-year retirement, though he continued to oppose Roosevelt’s policies. At least we don’t face that long a stretch of down time with the current “president.”

I discovered that 1930 did not produce many masterpieces. There are some obvious titles on my final list, as always, but some of the films are good but not great. In any other year they would not have made the cut. (The contrast with 1927, for example, is pretty stark.) Presumably one major reason is that filmmakers were still struggling with the introduction of sound. Fritz Lang didn’t release a film that year, though he came roaring back with M in 1931, as assured a treatment of sound cinema as one could find during those early days. (For those of you who subscribe to The Criterion Channel, I contributed a video essay on sound in M to David’s, Jeff Smith’s, and my series, also called “Observations on Film Art.”) Eisenstein was off in Hollywood unsuccessfully trying to make a film for Paramount and about to fall in love with Mexico. With one notable exception, Hollywood’s important auteurs made no films or minor films or films that don’t survive.

This dearth of really great films will be short-lived. From 1932 on, I shall no doubt have the opposite problem. For now, here are my picks. All the films are worth seeing, though good copies or even any copies of some are hard to track down. (The idea that all films are available to stream online is as absurd as when, back in 2007 when I wrote “The Celestial Multiplex.”) Some of my illustrations in this entry are subpar, but 1930 does not seem to be a year that studios and home-video companies are mining for titles to restore. I hope this entry will give them a few ideas.

I try not to include more than one film by the same director in these entries, aiming for variety and coverage. This year’s most prolific great directors were Josef von Sternberg and Yasujiro Ozu, who together could have filled in half the ten slots, but I have stuck to my guideline. (When we reach 1932 and 1933, I may be tempted to make an exception for Ozu.)

Given that sound developed at a different pace in different countries, some of this year’s films are silent.

For previous 90-year-ago lists, see 1917, 1918, 1919, 1920, 1921, 1922, 1923, 1924, 1925, 1926, 1927, 1928 and 1929.

 

Dovzhenko’s Earth

Some of the films on my top-ten list this year may raise eyebrows, but Earth is probably my least controversial choice. It remains one of the last classics of the Soviet Montage movement.

The policy of merging small farms into collective ones had begun in 1927 as part of the First Five-Year Plan. In 1929-30, the push increased, with the seizure of the assets of the more prosperous peasants who resisted collectivization, termed kulaks. Major films were made with the aim of villainizing the kulaks and emphasizing the backbreaking labor involved in traditional, non-mechanized farming. Eager young peasants were hailed as heroes, and tractors appeared as the means to greatly increase wheat yields–wheat which was often seized and sent to the cities to enable rapid industrialization.

Eisenstein had promoted collectivization with his Old and New in 1929, and Dovzhenko followed a year later with Earth. While Eisenstein’s approach had been sarcastic and comic, Dovzhenko pursued his poetic approach to cinema. The opening celebrates the fecundity of the earth and the cycles of nature. The famous opening juxtaposes a peasant girl’s face with a large sunflower. Semion, an aged peasant, is calmly, even cheerfully awaiting his imminent death, lying on a blanket and surrounded by heaps of apples.

  

A kulak family is soon introduced, angry at the idea that their possessions will be seized. The head of the family declares that he will kill his animals rather than turn them over, and he tries to attack his horse with an axe.

The arrival of the village’s new government-provided tractor marks a shift to an emphasis on the advantages of collective farming. As it approaches, the kulaks stand watching and mocking the idea of a machine taking over. In one shot, an elderly kulak stands framed against the sky, his wealth indicated by the presence of a hefty cow. The tractor temporarily halts, but it turns out that the radiator is dry. The simple solution is for the collective members to urinate into it.

Earth is a visually beautiful film, but like other Soviet classics I have complained about, there is no decent print of it available. The old Kino disc groups it with Pudovkin’s The End of St. Petersburg (the cropping of which I mentiooned in the 1927 entry) and Chess Fever. The copy of Earth is too dark and indistinct. To get some frames for illustrations, I watched the visually superior copy on YouTube, but there the images are cropped to a widescreen format. Just as bad, there are ads scattered through it, including one in the middle of Vasyl’s twilight dance in the road just before he is shot!

As before, I must end by pointing out that an archival restoration and a Blu-ray release are in order.

 

Ozu debuts on the list

Although after 1937 there are years when Ozu did not make a film, he will likely feature on these lists as long as I continue to post them. In 1930 he made six features and a short. Only three survive: Walk Cheerfully, I Flunked, But … , and That Night’s Wife. I’ve chosen the last, though Walk Cheerfully is a viable candidate as well. Sound was adopted slowly in Japan, and Ozu held out until 1936 before releasing his first talkie, The Only Son, which will doubtless feature on my list for that year.

The British Film Institute released That Night’s Wife in a boxed set called “The Gangster Films,” along with Walk Cheerfully and Dragnet Girl. Yes, Ozu made gangster films, including the wonderful Dragnet Girl, the film that might lead me to make my exception in the 1933 entry.

That Night’s Wife, though, is not really a gangster film. Shuji, the father of a daughter with a potentially fatal illness, is not a gang member but a decent working man who stages a hold-up to get money for her medicine. Ozu handles this hoary plot premise with his usual unconventional approach.

The opening, with its shots of dark, empty streets through which the police chase Shuji, is film noir ahead of its time. We soon are introduced to his worried wife, Mayumi, tending the little girl. She’s cute but has a distinctly bratty streak so common to Ozu children. This and fairly frequent moments of subtle humor help Ozu avoid a purely melodramatic presentation of the situation.

Once Shuji arrives at home, Mayumi gets him to confess what he has done. He intends to turn himself in after the daughter’s crisis passes, but a gruff detective who has posed as a cabby to find out where he lives, appears. Being told that the daughter will only survive if she lasts through the night, the Detective agrees to stay and settles in to guard his prisoner. Mayumi reveals her inner toughness and determination to defend her husband. At one point the Detective dozes off and she steals his gun and points it and her husband’s pistol at him, urging Shuji to escape (above). That shot, by the way, shows a motif typical of early Ozu films: having posters, often for American movies, visible on the walls behind the characters.

That Night’s Wife doesn’t display the fully mature Ozu style, but no one would mistake it for a film by another director. He has already mastered the perfect match on action, and he uses one in the early scene where Mayumi sits down as the doctor tends the daughter. Ozu cuts in with a move across the axis of action that places us 180 degrees on the other side of her.

   

Ozu is not using 360-degree space consistently yet, but he’s playing with the idea. Twice he has transitions that move through a series of adjacent spaces, one of his key devices and one that will last throughout his career. Here we move from the worried Mayumi to Shuji, trying to elude the police after the theft.

   

   

   

There’s a lyrical motif mutating from shot to shot: worried wife, interior lamp, interior plant against wood, outdoor light with leaves, stone building with the shadows of leaves, and stone building with the subject of Mayumi’s thoughts, cowering. But causally the four “extra” shots tell us little about where Shuji is in relation to the home he seeks to reach.

I took these frames from the BFI set, but in the US Criterion released the same three films in its Eclipse brand under the more accurate title “Silent Ozu – Three Crime Dramas.” (I Flunked, But … is in the BFI’s four-film set, “The Student Comedies.” Criterion hasn’t released a DVD of it, but it is streaming on the Channel.)

 

The Germans’ sound advantage

Unlike other countries, Germany developed its own sound system, owned by the Tobis-Klangfilm company. Tobis was initially able to use court injunctions to keep American films out of the German market. In 1930, this pressure led ERPI and RCA, owners of patents on the major US sound-on-film systems, to form a cartel in conjunction with Tobis, which had producing and distribution subsidiaries throughout Europe. Tobis-Klangfilm continued to dominate European sound recording and reproduction until the beginning of World War II.

Partly as a result of this technical advantage, several early 1930s German films became classics, though not all of them are known abroad.

One internationally famous classic is The Blue Angel. I’ve mentioned that von Sternberg was among the most prolific of major directors in 1930. I debated whether to put this or Morocco on the list. If cinematography were the main criterion, Morocco would have been the choice. Not surprisingly, Lee Garmes’s camerawork and lighting is spectacularly good.

I initially saw The Blue Angel decades ago, in grad school. It was the usual hazy 16mm print of the American release version. Watching it again in the German version, remastered from a 35mm negative, was a revelation. Not that it will ever be one of my favorites. It has its problems. The motif of the clown, implied to be Lola’s degraded ex-lover, is heavy-handed. Emil Jannings’ lumbering performance as Prof. Rath is at odds with the casual realism of the other performances. The time-passing montage accomplished via a hair-curler ripping down a series of calendar pages creates a time gap of years passing and the characters–especially Lola–have changed radically. Her change from a kind attitude toward Rath to cruelty and indifference is logical, given his passive decline from professor to clown, but it is presented too abruptly. Still, in my opinion the film’s pluses outweigh its minuses.

The Blue Angel, seen in a good print, is wonderfully photographed as well (see top). It may not have Lee Garmes, but it has Günther Rittau (co-cinematographer with Hans Schneeberger). He had lensed Die Nibelungen, Metropolis, Heimkehr, and Asphalt. He brings a dark look to the film, as in these shots of Rath in dignified professor mode and later onstage in a caricature of his former self.

   

The camera movements are rare and unobtrusive. A visual motif set up early on shows the bustling classroom that Rath presides over, where despite the students’ dislike of him, they obey his strict commands. Later, in after Rath has visited The Blue Angel and met Lola, his blackboard is covered with caricatures and insults drawn there by the students to mock his obsession with Lola. A slow track back emphasizes the empty classroom as he sits at his desk, his authority over the students gone.

     

No doubt it’s a matter of taste, but I find Marlene Dietrich’s performance here more attractive than in Morocco–and it is, of course, central to the film’s appeal.  Her amused playfulness with Rath, her quietly sarcastic attitude toward her sexuality being The Blue Angel’s main attraction, and her frequent bits of business with props unrelated to the story all differ considerably from her Hollywood performances. She has a natural glamor here that will disappear in her roles for von Sternberg in Hollywood.

After this film, von Sternberg made Dietrich a Hollywood symbol of glamor, but as a result she tends to be less spontaneous and active, with the camera and lighting lingering over her. Basically her glamor became more sultry. The result has its own appeal (and Shanghai Express is a good bet for a place on the 1932 list), but it is revealing to see her performing in a way that she only recaptured occasionally, if at all, in her subsequent films.

The frames were taken from the Kino on Video two-disc set, which has both the German and truncated American versions, as well as some supplements. It was subsequently released on Blu-ray, but I haven’t seen that version.

I always try to call attention to little-known films rather than sticking to the conventional classics for the whole list–and there aren’t a lot of such classics this year anyway. This time it’s a charming, imaginative musical romance, Die Drei von der Tankstelle, or “The Three from the Filling Station,” directed by Wilhelm Thiele and starring Lilian Harvey and Willy Fritsch. It was so popular that it outgrossed The Blue Angel in Germany.

Three carefree young men, who start off driving in a car and singing about friendship, arrive in their lavish shared home to discover that they are bankrupt. Low on gas, they are inspired to start their own filling station. They sing a nonsense song, “Kuckuck” (the German equivalent of “cuckoo”) as their belongings are hauled away, and they give this name to their art-deco filling station.

Lilian (the heroine’s name as well) shows up in search of gas for her roadster, and the three men fall in love with her. The rest of the film involves their competition over her. Naturally she chooses Willy (Fritsch’s character name). The two were often paired in later German films of the 1930s and became the ideal movie couple.

The film maintains its playful, often silly tone until the end. As Lilian and Willy are united  before a closed theatrical curtain, she turns front and says, “People! The public!” There is an objection that this is not the way to end an operetta and the curtain opens to reveal the entire cast as well as additional dancers and singers performing a lively and somewhat chaotic grand-finale number.

   

I originally saw this film at the Cinémathèque Royale de Belgique (now the Cinematek). As it does not circulate in the US on home-video, I have taken these low-quality frames from a homemade DVD, probably originally dubbed from an unsubtitled German VHS copy. It is available on a German DVD, since it is considered a perennial classic there. Unfortunately it has no subtitles. (The German manufacturers do not seem to have cottoned onto what their counterparts in other European nations–notably France–have: that if you include optional English subtitles, you can sell more copies.)

I hope, however, that eventually someone at the DVD/Blu-ray companies will discover this very entertaining film and make it more widely available.

 

A German director abroad

F. W. Murnau’s City Girl is another of those films I saw in a poor print decades ago and respect a great deal more after seeing it in a good version and knowing more about film than I did then. I wouldn’t put it in the same league as some of his earlier films, but for a 1930 list, it stands out.

Murnau was already on the way out at Fox when he was making the film. After his departure, the film was “finished,” including some dialogue scenes that were inserted. As with several other films of the era, such as Borzage’s Lucky Star and Duvivier’s Au Bonheur des Dames (see below), we are lucky in that only the silent version survives. The restored dialogue scenes in Fejos’s Lonesome show how ghastly these added interludes could be. (My apologies to the restorers. Obviously somebody had to do it.)

The heroine of the title is Kate, a waitress in a busy Chicago diner. Disillusioned by the obnoxious male customers she must cope with and lonely in her sparse apartment (below), she dreams of the apparently peaceful countryside that she sees in pictures. Lem wanders in for a meal, having been sent by his tyrannical father to sell the harvest from their wheat farm in Minnesota. He becomes a regular customer as he stays in Chicago waiting for the price of wheat to rise. His friendly, courteous behavior confirms her idealization of rural life, and they fall in love. Once she arrives at the farm, however, her illusions are shattered by the father’s violent rejection and the loutish behavior of a gang hired to help with the harvest, against neither of whom Lem seems able to defend her.

As in Sunrise, City Girl creates a contrast of city and country, though on the whole the city here comes across as more oppressive than the relatively friendly treatment the couple in Sunrise receive there. Sunrise spends half the film presenting the aftermath to the couple’s reconciliation, while City Girl has a grimmer tone, and the happy ending arrives very suddenly and briefly.

According to Janet Bergstrom’s informative essay accompanying the “Murnau, Borzage and Fox” DVD boxed-set, Murnau initially wanted to film in Grandeur, a widescreen format of the day, but the budget wouldn’t allow it. (I suppose the effect might have worked, as the 70mm version of Days of Heaven demonstrates with a similar locale.) Perhaps it was just as well. Cinematographer Ernest Palmer has handled the landscapes well (top of this section) and given a distinctly documentary look to the harvest scenes.

The restored version of the film in the Fox boxed-set is excellent, but that box is long out of print. British company Eureka! has released City Girl on Blu-ray, which should look even better, available as an import on Amazon.com in the US (where it is listed as Blu-ray but is the same edition as is listed on amazon.co.uk as dual-format Blu-ray/DVD).

 

The French between silence and sound

René Clair has long been considered one of the directors whose early sound films were among the most imaginative and technically sophisticated. Indeed, that era was perhaps the high point of his career, with Sous les toits de Paris (1930), Le Million (1931), and À nous la liberté (1932) coming out in quick succession. The first two were made for the French subsidiary of Tobis, Films Sonores Tobis, which produced French films in its Epinay Studios. Its prosperity and technical sophistication are evident in the large sets constructed for Sous les toits.

The film centers on Albert, who makes his living selling sheet music to passersby in the street and leading them in sing-alongs. In the opening scene, a small crowd joins in a song called “Sous les toits de Paris” (above). He falls in love with Pola, who is claimed by Fred, one of the many thugs played by Gaston Modot across his career. She moves in with Albert, but when he is framed for a crime committed by Fred’s gang and sent to jail, she takes up with Albert’s best friend.

This light tale gives Clair the opportunity to show off the flexibility and even flashiness of his style. The opening shot is a long crane movement descending from chimneys against the sky to the crowd singing along with Albert. Clair also uses a variety of camera angles, as in the low-angle still above.

   

There are several carefully lit street scenes, including this one shot in depth with both characters’ backs to the camera as Albert spots Pola outside a bar.

The climactic scene involves a tense conversation between Albert and his pal as Pola looks on. Clair shoots the two men in shot/reverse shot through glass doors, so that we do not hear them but only watch their actions and faces in a deliberate reversion to silent cinema.

     

Sous les toits de Paris is still in print as a DVD from The Criterion Collection. (An early issue, number 161.)

At least in the US, Julien Duvivier is not as familiar as Clair, being known mainly for Pépé le Moko, but his reputation is rightly growing. David did his part in promoting Duvivier by contributing a video analysis of his 1941 Hollywood romance Lydia on The Criterion Channel.

I don’t know whether Duvivier was inspired by L’Herbier’s L’Argent (which featured on my 1928 list) to make Au Bonheur des Dames, but there are certainly similarities between them. L’Herbier’s film was adapted from the eighteenth novel in Zola’s Rougon-Macquart series, published in 1891, but the film updated the story to the 1920s. Au Bonheur des Dames was adapted from the eleventh novel in the series (1883) and also updated. L’Argent showed the harm done by large banks, while Au Bonheur des Dames deals with the damage wrought on traditional small shops by the rise of huge, corporate-level department stores. Zola had modeled his giant store on Le Bon Marché, while Duvivier was able to shoot interiors for the store “Au Bonheur des Dames” in the Galerie Lafayette–much as L’Herbier had shot in the Paris Bourse over a weekend.

The story concerns Denise (Dita Parlo in her first French film), a naive young woman who comes to Paris expecting to work and live in her uncle’s custom-tailor shop. Instead she finds him on the verge of ruin, since the new department store is located directly across the street. He refuses to sell to Mouret, who runs the department store and wants to expand. But all the shop-owners on the block have given up, and the demolition work removing them gradually drives the tailor mad. In the meantime Denise, fascinated despite herself with the glamor of the department store, gets a job as a model there. She attracts the romantic attention of Mouret and the lecherous attention of a manager and would-be rapist (see respectively in the center and right foreground in the frame above).

Also like L’Argent, Duvivier’s film has strong elements of the Impressionist movement, which was largely over by the late 1920s. The opening portrays Denise’s reaction to the bustle of Paris after she arrives by train by a rapid montage images superimposed around her bewildered face. Later, her uncle’s growing obsession with the demolition of his neighborhood is suggested by many subjective devices, including a prismatic shot of the workers.

  

There are also many shots staged in depth, including a cut between Denise’s cousin’s fiancé, who has been caught sneaking out to desert her, and a reverse angle (maintaining screen direction) from behind her as he looks guiltily back.

  

The film was made in both sound and silent versions. Only the latter survives. I watched the Arte Video/Lobster Films version, still in print. (A Facets release in the US is out of print.) The French DVD has English subtitles, as well as a short, informative video introduction by Serge Bromberg in lieu of program notes.

 

The lingering legacy of World War I

During World War I, the relatively small number of films about the war naturally supported the combat, demonized the enemy, and in the US, urged people to buy War Bonds.

Given the pointlessness and widespread death and injury caused by the war, a reaction soon occurred in filmmaking within the major countries involved. Abel Gance’s J’Accuse (1919) is the earliest film displaying a bitterness about the needless bloodshed of the war, followed soon by Rex Ingram’s adaptation of Vicente Blasco Ibáñez’s 1918 bestseller, The Four Horsemen of the Apocalypse. The film made it onto my 1921 ten-best list. There the emphasis shifts to family members and friends of different nationalities having to fight each other (a theme that Griffith had used in The Birth of a Nation). Four Horsemen ended with progressively distant views revealing an enormous cemetery of white wooden crosses, a motif also used briefly in J’Accuse that would become common in pacifist films of the period.

The next big film was King Vidor’s The Big Parade (see my 1925 list), one of the great box-office successes of the 1920s. The filmmakers went so far as to end the film with the hero reunited with his French love, but having lost a leg in the combat. From that point, killing off some or all of the major characters became the norm of subsequent significant anti-war films. (The Big Parade does kill off the hero’s two friends-in-arms.)

The 1930s carried on the condemnation of World War I, with two of the major films on opposite sides being released in 1930. They carry on the idea introduced in The Big Parade that the ordinary soldiers fighting on the German and American sides were basically alike and could recognize each other’s humanity.

G. W. Pabst’s Westfront 1918 starts late in the war, when the German soldiers in the trenches have become thoroughly disillusioned and unspeakably weary. The first big attack on their position turns out to be friendly fire from their own artillery. (The trope of incompetent commanding officers as among the true villains in World War I–and in some case other wars–has lingered in films ever since.) One of the four German soldiers who form the group protagonist falls in love with a French barmaid in a village near the front. In the final scene, huge numbers of casualties from the latest French attack are treated at a German war hospital, with French and German patients mixed together–as the dead had been in the battle itself (see the image above).

Pabst, like other filmmakers, uses the visit home on leave as a way to emphasize the war’s impact on the home front. Karl, the most important character of the four main soldiers, returns for the first time in eighteen months only to find his wife in bed with another man. Pabst introduces the home-front section with a long line of people waiting to buy food.  Karl’s mother sees him arrive but cannot greet him because after hours of waiting she had reached the front of the line.

The impressive final battle scene ends with the last of the central characters dead, and the carnage drives the lieutenant who commands them into mad raving.

The film makes a powerful statement designed to end all wars, and yet one cannot help wonder if those Germans inclined to resent the results of the Treaty of Versailles might have been inspired to a greater desire for revenge by it.

The American counterpart of Pabst’s film is Lewis Milestone’s adaptation of All Quiet on the Western Front. The 1929 translation of the German novel had become a huge bestseller in the US. Producer Carl Laemmle, who was born in Germany and visited friends and family there regularly, no doubt had a personal as well as a financial incentive to acquire the rights for Universal.

It would be easy to dismiss this big Hollywood production and ultimate Best Picture winner at the Oscars as a slick commercial venture. Certainly it has the glossy, expensive studio look that Westfront 1918 lacks, as in this scene, with its perfect three-point lighting carefully picking out each figure.

Being from a German novel, however, makes this the first significant Hollywood film to treat German soldiers sympathetically and to make them the lead characters. As in Westfront 1918, there is a tight little group of friends, three in this case, with the point-of-view figure being Paul (Lew Ayres). He is the one whom we follow on his visit home while on leave. There he visits his mother, as expected, but he also drops in on his old classroom and is disgusted to find that his teacher is glorifying war and preparing a new generation of young men to go off to become soldiers.

The opening of the film had shown Paul as a student being told of the glories of war and rushing off to sign up. This is the only one of the anti-war films I know of that portrays this sort of social indoctrination as a crucial cause of the horrors of war.

As with the other anti-war films of the 1920s and 1930s, the scenes of  are extended and realistic in their depiction of trench warfare. And as in Westfront 1918, all the main characters die. At the end we see them marching off, with vast fields of white crosses superimposed.

The French produced their great anti-war films only slightly behind the Germans and Americans. Raymond Bernard’s masterful but little-known Wooden Crosses (1932), which focuses largely on French troops, is no less powerful in its depiction of trench warfare. Jean Renoir’s Grand Illusion (1938) broke with tradition by downplaying battlefield action and focusing on the senseless loss of life and liberty and the effects of class differences in warfare.

All Quiet on the Western Front has been restored and released on Blu-ray a deluxe edition with DVD and supplements or with the supplements but no DVD.

 

Experimental cinema in 1930

Surprisingly enough, in a weak year for commercial feature films, there was a considerable number of significant experimental films made. The standard historical accounts when I was a grad student (and probably thereafter) suggested that sound made film production too expensive for the shoestring-budgets of experimental filmmakers. Clearly, though, many of them carried on. Some of the 1930 films are silent, but by no means all. Therefore I’ve decided to devote the tenth slot in my list to a brief summary of several of them. These are in no particular order.

Mechanical Principles was Ralph Steiner’s next film after the better-known H2O (1929). Despite the scientific-sounding name, Mechanical Principles is an abstract film made by shooting close view of moving gears and other devices for moving parts of machines (above). The shapes and movements are engaging in themselves, but there is also, I think, a subtle and amusing progression toward more odd and even absurd-looking mechanisms. I doubt anyone would learn anything about “mechanical principles” from the film. (Included in Flicker Alley’s excellent duel Blu-ray/DVD collection, “Masterworks of American Avant-garde Experimental Film 1920-1970.” The visual quality is distinctly better than the versions on YouTube.)

The wonderful German animator Oskar Fischinger specialized in making abstract animation in time to pieces of music. In 1930 he made Studies No. 2 through 7. YouTube contains only snippets of his films; presumably the rights holder, the Center for Visual Music, has diligently patrolled it.) Studies No. 6 and 7 are the most widely seen.  I am particularly fond of Study No. 6, to “Los Verderomes fandango,” by Jacinto Guerrero (below). These shorts use a broad range of musical accompaniment. Study No. 7 is done to Brahms’s “Hungarian Dance No. 5.” I haven’t seen all of the Studies, but I assume they all use the same technique, charcoal drawings on paper. There were VHS videotapes released and a laserdisc in 1996, all long out of print. The Center for Visual Music has brought out two DVDs with selections from his films, digitally remastered and with intriguing-sounding supplements.They are sold through the CVM.

The Bridge is a silent film, adapted from the Ambrose Bierce story “An Occurrence at Owl Creek Bridge.” Its director, Charles Vidor (né Karoly Vidor in Hungary, no relation to King Vidor), had come to the US in 1924 and worked as a chorus singer. He managed to self-finance The Bridge, which centers on a man condemned to execution as a spy. As he is being hanged by a military escort on a bridge, we see a complex combination of flashbacks to his childhood and fantasies of escape (below). The film was impressive enough to gain Vidor to gain an MGM contract to direct The Mask of Fu Manchu (1932). The high points of his long career are Cover Girl (1944) and Gilda (1946).

The Bridge is included on the fourth disc in the “Unseen Cinema” collection.

Walther Ruttmann’s Weekend (the original title is in English) is not a exactly a film, even though its subtitle is “Ein Hörfilm von Walter [sic] Ruttmann.” That roughly translates as “a radio-film,” and it was produced by the Reichsrundfunk Gesellschaft (national radio). On the DVD it is reproduced as a black screen with sound over, which one could play in projection. The list of sections given at the beginning runs: “Jazz der Arbeit / Feierabend / Fahrt ins Freie / Pastorale / Wiederbeginn der Arbeit / Jazz der Arbeit” (Jazz of Work / Quitting time / Travel in the Open Air / Pastorale / Recommencing Work / Jazz of Work). The “film” is a lively and often amusing montage of sounds, including trains, tools, engines starting, drum beats, brief snatches of dialogue, a cuckoo clock, a siren, a rooster, and so on. Its twelve-minute length is well gauged to avoid having the device wear out its welcome.

I hardly need to write anything about Dziga Vertov’s Enthusiasm: Symphony of the Donbass. For many years Vertov held his high reputation largely on the basis of Man with a Movie Camera, one of last year’s top ten. Enthusiasm was Vertov’s last Ukrainian film and apparently the first Ukrainian sound feature. Although the film’s credits emphasize that it was shot on location, including the interior of a coal mine, clearly nearly all the footage was shot silent and had a “symphony” of sounds woven together and laid over it.

The film has some of the playfulness and experimental cinematography so familiar from Man with a Movie Camera. That film is so engaging in part because it was primarily a lively look at Soviet everyday life, including movie-going. Enthusiasm, however, is far more overtly propagandistic, being largely a celebration of the early achievement of the First Five Year Plan’s coal quota in the Ukraine in four years. This was thanks to the “enthusiasts,” as the film puts it: Stakhanovites, that is, workers who did considerably more than their comrades through hard labor and efficiency. (They were named after Alexei Stakhanov, a coal miner; the movement started in the coal industry.) Some of the Soviet workers and peasants sincerely approached their work in this way, but in reality much of the achievements of the Plan were achieved through forced labor.

The restored film is included on Flicker Alley’s Blu-ray disc, “Dziga Vertov: The Man with the Movie Camera and Other Newly Restored Works.”

One of the finest experimental films of 1930 is Lázló Moholy-Nagy’s Ein Lichtspiel: Schwartz Weiss Grau (“A light play [i.e., a film]: Black White Gray”). It is derived from his constructive sculpture, Lichtrequisit einer elektrischen Bühne (“Light device for an electric stage”), which consists of many metal elements, some of them pierced with rows of holes, through which lights can be shown to create moving shadows. It later came to be known as the Light-Space Modulator. (The original is in the Busch-Reisinger Museum, Harvard.)

Moholy-Nagy made a film of the Modulator in action, with shifting solid shapes and shadows. It is sometimes hard to tell whether the camera or the device itself itself is moving.

Unfortunately no high-quality print seems to be available. Facets offers several of Moholy-Nagy’s films on separate DVDs, including Ein Lichtspiel–which is a little odd, given that it is only six and a half minutes long. (The modern soundtrack makes the object seem like a very elaborate set of wind chimes.) This DVD is probably the source of some of the better versions of the film posted on YouTube. It would be a pity to encounter the film in this way, but given the current situation, this is the best of the copies available there.

Kenneth MacPherson was co-founder of the Swiss-based Pool-Group, along with his wife, Bryher (née Annie Winifred Ellerman) and the poet, HD (Hilda Doolittle). Together they started Close-up, an important journal on international art and experimental cinema (1927-33). MacPherson made a few films, most notably the short silent feature Borderline. MacPherson was influenced by  French Impressionism and Soviet Montage, and their stylistic traits are used in Borderline, including subjective camerawork and bursts of very rapid editing to convey violence or psychological stress.

The plot deals with racial and sexual tensions in a small Swiss village when a black couple comes to live there. Pete, in a genial performance by Paul Robeson, takes back his fiancée (or wife?) after she has had an affair with a white man, Thorne. Thorne remains obsessed with her, which gradually drives his neurotic, racist wife, Astrid (HD, credited as Helga Doorn) mad. Although Thorne has rented a room over a tavern owned by people sympathetic to the black couple, the scandal leads the villages to demand that Thorne be evicted, and he leaves town.

Some other experimental films made in 1930 that I didn’t think were as good as these but are still worth watching if one is particularly interested in this genre:

Francis Bruguière’s Light Rhythms (on the third disc of the Unseen Cinema set), a more low-budget attempt at something like Ein Lichtspiel, using paper cut-outs.

Luis Buñuel’s L’Âge d’or. (The BFI Blu-ray/DVD combo is out of print, although amazon.com is still selling it. Kino Lorber put out a DVD, which I haven’t seen.) I know this is considered a wonderful film by many, but I dislike it and much prefer Un chien andalou.

James Sibley Watson’s Tomatos Another Day (on Kino International’s “Avant-garde 3” set) is a pretty dreadful film, a sort of “theater of the absurd” before its time. I suspect it is a parody of a modernist play of the day. If so, and if anyone knows what play it was, please message me on Facebook and I’ll add that information. Suffice it to say that the husband, his wife, and her lover say obvious things out loud and very slowly “You are my lover,” “I am alone,” that sort of thing. Then all three start talking in dreadful puns (“Tomotos’ another day”). One might say it’s so bad, it’s good.

Sergei Eisenstein and Grigory Alexandrov co-directed Romance sentimentale, a quite uncharacteristic imitation of French Impression or city symphonies. Finally getting a chance to see it decades ago was one of the great disappointments of my life.  It is available as a supplement on the Kino DVD of the restored Qué viva México.

---

A succinct account of Tobis-Klangfilm’s business operations can be found in Douglas Gomery’s “Tri-Ergon, Tobis-Klangfilm, and the Coming of Sound,” Cinema Journal 16, 1 (August 1976): 51-61. (Available on JSTOR.)

I have summarized the “wooden crosses” motif that is almost universal in anti-war films of the 1920s and 1930s (and occasional later films as well) in my video essay on Raymond Bernard’s Wooden Crosses on The Criterion Channel, where the film is permanently streaming.

Our friend and colleague Vance Kepley published an excellent study, In the Service of the State: The Films of Alexander Dovzhenko (University of Wisconsin Press, 1986).

January 27, 2021: Thanks to Cindy Keefer of the Center for Visual Music for kindly sending some corrections to my discussion of Oskar Fischinger:

 

 

All Quiet on the Western Front (1930)