by John Calhoun
The "hows" of camera movement - the technology and
techniques whereby camera operators, cinematographers, and directors
can create moving shots in a film - are easy enough to identify
and define. It's when you start trying to talk about the "why" and
the "when" that the discussion gets vague and a bit
complicated - as hard to pin down as the famous tracking shot
that opens the noir classic Touch of Evil (1958).The impulse
toward movement is not hard to understand; we are, after all,
talking about motion pictures. But if you don't want to be indiscriminant
in moving the camera, questions arise: When is it called for?
When should it stop? And what does a camera on the move mean
to the spectator?
"My philosophy is, good camera movement happens when you're
not even aware of it," says Owen Roizman, ASC. "It
just feels right. My approach has been to always let my instincts,
the scene and what the actors are doing dictate whether or not
the camera should move." For William A. Fraker, ASC, it's
all about story: "Camera movement's great, but it's got
to be about exposition. You have to involve the audience visually,
but to do that, you don't need fancy movement and 360-degree
dolly shots." According to Haskell Wexler, ASC, "Camera
movements are used to lead the eye, to give people a feeling
- an emotional one, a logical one, a dramatic one - just the
way lighting and framing are." In fact, both Wexler and
Roizman maintain that camera movement can't be considered separately
from those other key elements of a cinematographer's art. John
Toll, ASC, puts it succinctly: "The shot has to fit into
the overall visual design of a film to have impact."
Still, it is movement that makes a movie a movie. Other visual
arts are lit and composed, but only in cinema (and its video
offspring) does the ability exist to reframe a continuous image.
And ever since 1896 or so, filmmakers have been looking for ways
to do so. In 1916, D.W. Griffith and cameraman Billy Bitzer created
what is one of cinema's earliest camera moves: the slow, simultaneous
track in and down on the massive Gates of Babylon set in Intolerance.
The obvious reason for the shot was to first establish the scale
of the set, and then to move closer to verify that actual human
activity was taking place in it. This is what Steadicam inventor
Garrett Brown refers to as "the 3-D effect." In an
excerpt from an upcoming series of articles on "The Moving
Camera" for Zerb: Journal of the Guild of Television
Cameramen, Brown explains, "When the camera begins to
move, we are suddenly given the missing information as to shape
and layout and size. The two-dimensional image acquires the illusion
of three-dimensionality and we are carried across the divide
of the screen, deeper and deeper into a world that is not contiguous
to our own."
As the silent era reached its apogee in the 1920s, artists around
the world were finding new ways to incorporate camera movement
into the grammar of film. In the Soviet Union, Sergei Eisenstein
was doing pioneering work with montage; in Germany, F.W. Murnau
was experimenting with other methods of telling a story through
purely visual means. Murnau's 1924 film The Last Laugh,
shot by Karl Freund at Ufa Studios, is so visually expressive
that it contains not a single intertitle. In one sequence, the
camera was operated from a bicycle, and in another, Freund strapped
the camera to his chest to simulate a drunken character's point
of view. When Murnau came to Hollywood to make Sunrise in 1927,
he imported his roving style, working with cinematographers Charles
Rosher, ASC and Karl Struss, ASC to unleash the camera from its
bonds.
During his audio commentary on the recently released DVD of Sunrise,
John Bailey, ASC lovingly describes what he calls "one of
the most famous shots in all of silent film." The lead character,
a farmer played by George O'Brien, walks into a swamp at night
to tryst with a licentious woman from the city. As the farmer
moves through the foggy set, Bailey details, "The camera
is following him, not on a dolly, but on an overhead track with
a platform suspended from it." O'Brien moves in a more or
less figure-eight pattern, followed by the camera; the stalking
movement mirrors the farmer's compulsive entry into dangerous
psychological terrain.
For another famous sequence in Sunrise, the filmmakers
placed the camera on a trolley that appears to move continuously
on tracks from a country location to the city, which was built
on the Fox lot. Murnau and his cinematographers were always looking
for reasons to free the camera, whether to express an emotion
(as in the swamp scene), or to get the characters from point
A to point B in an eloquent way (the trolley scene).
French director Abel Gance was another filmmaker for whom "the
tripod was a set of crutches, supporting a lame imagination," writes
Kevin Brownlow in his book The Parade's Gone By. Recalling
Gance's biggest production, Napoleon (1927), which was
shot by Jules Kruger, Garrett Brown marvels, "There are
things in Napoleon that are so ahead of their time. [Gance]
stuck the camera on every possible type of vehicle; he did things
in that film that weren't taken up again for 30 years."
However, everything seemed to change when movies began to talk.
The demands of technology and recording dialogue started this
transformation, and the "invisible" classical style
that soon developed finished it. The odd talent like Busby Berkeley
was still devising means to move the camera between a grouping
of dancers' legs, and a stray crane shot - such as Scarlett O'Hara
searching through the bodies at the depot in Gone With the
Wind, or the swoop down a staircase into a close-up on the
key in Ingrid Bergman's hand in Notorious - still got
through. But when Fraker says, "If you look at the great,
great films, there's very little camera movement," it's
the 1930s and '40s he's talking about.
Page
1
|