Welcome to How’d They Do That? — a monthly column that unpacks moments of movie magic and celebrates the technical wizards who pulled them off. This entry explains how the Silent Era visual effect known as the “jeopardy shot” works.
As film historian Craig Barron puts it in the Criterion special feature “Visual Effects in The Gold Rush,” modern viewers have a tendency to look back at the Silent Era as somewhat archaic. And, to a degree, the contemporary movie-enjoyer is correct. Films made a century ago had a lighter and less “sophisticated” toolkit. But from a different perspective, as far as visual effects are concerned, that’s part of the era’s appeal. It was a time of innovation and creative solutions to impossible problems.
Impossible problems like how do we get Charlie Chaplin to stumble along a treacherous, Alaskan ravine without actually making him stumble along a treacherous Alaskan ravine?
The jeopardy shot
First released in 1925, The Gold Rush is, appropriately enough, a goldmine of early special effects work, from detailed miniature work to double exposures to fake snow. But today, we’re going to focus on one particular effect that occurs in one of the first shots in the film. It’s our introduction both to Chaplin’s Tramp (here dubbed the “Lone Prospector”) and the film’s setting: Chilkoot Pass, Alaska. The flat-footed Tramp stumbles and weaves along a thin cliff face, unbothered by the immense drop that awaits him should he slip. This is a treacherous place whose dangers seem beyond the notice of our bumbling hero. So, how did they pull off that shot? The production did shoot on location in the Sierra Nevada mountains. But would they really risk Chaplin’s life like that?
How’d they do that?
Long story short:
Jeopardy shots are a variation on the “glass shot,” an early form of physical matting that involves placing a large painted pane of glass between the subject and the camera to add, exaggerate, or replace scenery.
Long story long:
Glass shots are achieved by positioning a painted glass pane bound within a sturdy wooden frame between the camera and the set. Sometimes the framed glass would be secured onto a rostrum or a purpose-built “shed” to give it more stability and protection, especially if the shoot was outdoors. The shed (or a black canvas tent) also had the desired effect of shielding the glass against unwanted reflections.
The glass served as a painter’s canvas, precisely aligned with some real topographical element to sell the illusion. The vast majority of the glass was left blank, featuring only choice painted additions. When filmed, the painted elements and the live-action subjects combine into one piece of photography. The result is a composite image, wherein the painted portion of the glass appears to be consistent with the live-action scene.
As Mark Sawicki lays out in his 2007 book Filming the Fantastic: A Guide to Visual Effect Cinematography, you can easily simulate the optical physics of the trick for yourself. Hold your thumb and forefinger out as if to “crush” an object in the distance. As long as you keep one eye closed, the illusion of a giant hand will be sustained. If you open both eyes, it will be obvious that the object and your hand are on two separate planes. If you alternate which eye you keep closed, you will notice that the relationship between your hand and the object shifts (this is called parallax displacement). Because moving image cameras tend to only have one “eye” (and one position), as far as the camera is concerned, it has no depth perception and the painted object appears to be full-sized.
Two principles of optics sell the visual effect of the glass shot: the single lens and a large depth of field, the latter of which keeps the subject and the glass in focus. A third phenomenon known as the nodal point (the optical center of a lens where the incoming light is bundled in the optical axis) explains how cinematographers can perform simple camera movements with glass shots (like the one below, from Chaplin’s Modern Times) without ruining the effect.
To make a very long (and litigious) story short, Norman O. Dawn is generally acknowledged as the inventor of the glass shot process for moving images. (Though, as this American Cinematographer article notes, Dawn himself rejected “any assertion that [he] invented this technique —[he] merely built on to it and took advantage of conditions to advance an art in the making”). Despite this self-effacing comment, Dawn’s contributions are worth underlining. After all, he was likely the first individual in Hollywood credited for special effects work (a “cina-luminist” in 1917’s unfortunately-named Oriental Love).
A special effects pioneer whose name has been obscured through patent drama, Dawn’s first documented use of the glass process in movies is thought to have been in the 1907 travelogue Missions of California. As film scholar Leslie DeLassus describes in her interview with the University of Texas, Dawn was especially interested in using the process to restore architecture and vistas to be period-accurate, as with the 1927 film For the Term of His Natural Life, where a glass shot revives the decrepit Port Arthur Penitentiary. DeLassus argues that Dawn’s innovative use of the glass shot made spectacular scenes and set-pieces economically feasible in the wake of mega-expensive Studio-Era commercial features. His work also paved the way for cheap B-picture genres like sci-fi, horror, and adventure serials to add production value without blowing their budget.
One boon of the glass shot’s simplicity was that the director could see the finished product immediately on set. Rather than waiting for weeks to see the disparate elements come together, they could simply look through the viewfinder to see one unified image. The glass shot is, in this way, “the purest and least adulterated in-camera trick effect,” as DeLassus puts it. The beauty of the technique was that after the director called cut, no further work was necessary; the effect was achieved.
Yet another benefit of the glass shot was that it allowed performers to simulate stunts that would otherwise be life-threatening. And this brings us to the jeopardy shot, so-called because the effect could make actors appear to be in potential danger without actually putting them at risk. So, instead of risking the life of beloved actor Charlie Chaplin by making him scale a treacherous cliff face, a glass shot could place the Tramp in peril without endangering its star. The same also holds for the aforementioned rollerskating sequence in Modern Times, where Chaplin glides and swoops close to an unbarred ledge while blindfolded.
Other notable figures in the early days of the glass shot include Ferdinand Piney Earle (Dawn’s contemporary and patent rival); the prolific Emilio Ruiz del Rio; Walter G. Hall (who developed the glass shot off-shoot known as the Hall Process), and Ralph Hammeras, one of the first technicians to receive an Academy Award nomination in the field of special effects for his glass shots on 1927’s The Private Life of Helen of Troy. Hammeras’ stunning work with glass shots can be witnessed in the original The Lost World (1924) all the way through to 1963’s Cleopatra.
The glass shot was versatile, creatively liberating, and cost-effective, but it did have drawbacks. The most notable pitfall was the time taken up as the artist painted the glass. This could be especially troublesome if the shot took place outdoors, where changing light conditions could invalidate the shadow work and ruin the illusion of a seamless composite. As M. David Mullen (The Love Witch, Jennifer’s Body) proposes on this cinematography.com forum, “the reason glass shots disappeared from use so early on was that production time was too valuable to spend setting it up.”
One other wrinkle was that actors had to rehearse diligently if they were at risk of vanishing behind the glass-painted elements. Indeed, glass shots are likely one of the first instances of film actors having to imagine what they’re seeing. There is a straight line between painted panes of foregrounded glass and Sir Ian McKellen getting depressed about being trapped in green screen hell during the making of The Hobbit.
Roland Totheroh, The Gold Rush‘s cinematographer, was a frequent collaborator of Chaplin. And as Barron notes, during the Silent Era, a cinematographer was also expected to execute any and all visual effects. Most effects shots had to be accomplished in-camera to boot because this was a time before optical printers. The jeopardy shot in The Gold Rush, where Chaplin’s Tramp bobs along the precipice of a snowy cliff, was one such in-camera effect, opening up the scene and creating production value sacrificed by shooting on a sound stage.
Despite fading from the spotlight with the emergence of faster processes and new technologies, glass shots can still be found in a number of modern films (including 1982’s Conan the Barbarian, 1984’s Dune, 1994’s The Shadow, and 1997’s Seven Years in Tibet).
What’s the precedent for jeopardy shots?
The technique of placing painted glass between the subject and the camera was occasionally used in commercial still and portrait photography as far back as the Mid-19th Century. In fact, it was during his employment as a still photographer for the L.A.-based Thorpe Engraving Company in the early 20th Century that Norman O. Dawn first encountered the technique. As Birk Weiberg relays in his 2016 dissertation “Image as Collective: History of Optical Effects in Hollywood’s Studio System,” Dawn’s colleague Max Handschiegel (who would go on to invent the “Handshiegel Color Process“) suggested placing a pane of glass between Dawn’s camera and his subject (a building), to obscure an unsightly pole. Painting a tree on the pane of glass to obscure the pole, Dawn could produce a much more aesthetically pleasing image.
Indeed, the glass shot was another photographic trick that was grandfathered in and freely experimented with during the early days of moving images. It can be lumped in with a handful of still image effects, including multiple exposures and specialist lab processing, which significantly laid the groundwork for early cinematic special effects.
While this covers the more painterly origins of glass shots, it’s worth looking at what filmmakers did before jeopardy shots allowed cinematographers to mimic dangerous sets without putting performers at risk.
As I’m sure you’ve guessed, given that the attitude towards workplace safety during the Silent Era was an extremely sarcastic shrug … the risk of plummeting to certain death wasn’t exactly a problem for certain actors/stunt performers. Buster Keaton, Charlie Chaplin’s stone-faced American contemporary, is the most notable of these. Whereas Chaplin relied more on physical comedy, sight-gags, and facial expressions, Keaton’s style relied more on genuine peril, violence, and stunt work. To sum up: when you’re watching a Chaplin film, you rarely fear for his life; when you watch a Keaton film, genuine peril is part of the appeal.
To give one evocative example, Keaton really did jump from one building to another in his first feature film, 1923’s Three Ages. The initial plan was to complete the jump. But when he missed, he plummeted to the net waiting below, sustaining injuries that took him out of commission for three days. Instead of abandoning the stunt, he leaned into the mistake and did it again, incorporating awnings into the fall.
As Every Frame a Painting‘s tribute notes, Keaton’s approach to filmmaking was to never fake a gag. As far as Keaton was concerned, the best way to enthrall audiences was to show the real deal without cutting away. This is an attitude I’ve butted up against several times while writing this column. Every now and then, you encounter a special effect where the filmmakers just straight up did the thing you see on screen (Jackie Chan’s climactic pole slide in Police Story springs to mind). There is absolutely a bias involved when it comes to effects that eschew “trickery” for the real deal. As Norman O. Dawn himself notes (as quoted in Stephen Prince’s Digital Visual Effects in Cinema: The Seduction of Reality): the head honchos of the Classical Studio era “didn’t believe in telling anybody about effects … they didn’t want to let the exhibitors know that this was a cheap picture full of fakes.”
Our modern notion of what constitutes “fake” filmmaking is different from that of the Classical Studio era. But I think it’s fair to say that we still have a bias in our thinking about cinema that emphasizes live action. There is a reverence for “practical” effects over “digital” surrogates that I am certainly guilty of. But the awe reserved for unsimulated stuntwork, especially unsimulated stuntwork where bodily harm is part of the appeal, is something I’d be fine leaving in the past.