The Cinema Laser
An edited version of "Shooting Westworld "- By Michael Crichton
The screenplay for Westworld was written in August, 1972, and subsequently offered to the major studios. Every one turned it down, except for Metro-Goldwyn-Mayer. That immediately presented a problem. MGM had a bad reputation among filmmakers; in recent years directors ... had complained bitterly about their treatment there. There were too many stories of unreasonable pressure, arbitrary script changes, inadequate post-production, and cavalier recutting of the final film.
... but then we didn't have a choice. We began preproduction in November, 1972.
Preproduction is the time preparatory to shooting when the creative elements are assembled, the script is polished, the sets are built, the locations picked, and the cast hired. It is a busy period for any film, but we had several peculiar problems with Westworld.
The first problem, to be blunt, was the studio. ...
An orderly preproduction was impossible. We didn't have out cast until forty-eight hours before shooting began. MGM kept demanding script changes right up to the day of shooting. ...
Our second problem was the budget. MGM had agreed to make the film only if it could be done for less than a million dollars. After a month of preparation, we decided it was impossible. Metro reluctantly increased the budget by $250,000, and we continued work. ...
... Of the total budget, roughly $250,000 went for cast salaries. We had forty speaking roles, and several expensive stars. Of the remaining million dollars, about $400,000 went for salaries of the crew during the six week shooting schedule. We had about eighty people working on cameras, costumes, props, hair, lighting, sound, transportation, and so forth. ...
After paying cast and crew salaries, we had $600,000 for everything else in the film: sets, props, special effects, extras, and so on. This still seems like a lot of money, until one figures it out item by item. For example, a Hollywood extra gets about forty dollars a day plus fringe benefits. Even without crowd scenes, you need several extras a day just to keep your locations from looking deserted. Before you know it, you're spending $50,000 for extras during the course of the film. And extras need costumes, and horses. It adds up fast.
Each specific budget category was so thinly financed as to seem impossible. Out of that $600,000, we could give the art director, Herman Blumenthal, only $75,000 for set construction. Anyone who has ever considered building a larger garage or finishing a basement playroom will understand the dimensions of this problem. For that $75,000, Herman had to build twenty sets covering nearly 200,000 square feet. And although he had certain advantages over standard constructin requirements, he also had certain special problems. His floors had to be built with such fine tolerances that a camera rolling over them would not wobble or bounce. Many interiors had to be aged. The detail work had to be excellent because minor flaws become glaring when the image is projected on an eighty-foot-wide screen.
In the final film, almost everything was used more than once. We used one medieval stairway three times, in different places. We used a single underground corridor nine times with six light changes, then tore out a false wall and used the same corridor, now widened, as teh robot-repair area. We used one hotel room twice, changing furniture and camera angles. In the case of the hovercraft, we built only half the set (which is symmetrical), but "flopped" half the film left to right in the optical printer, so that the final effect was a full hovercraft compartment. Sometimes these maneuvers forced some peculiar cutting patterns, but the average filmgoer probably never notices this.
Our final problem area was technical. For an inexpensive film, we were blithely using a wide range of special-effects techniques: front projection, rear projection, burn-ins, video replay, and blue screen. Gene Polito, the cinematographer, fortunately has a broad technical background from his early days as an engineer; he saved us more than once.
Three problems were especially tricky. One was the robot eyes. I wanted eyes that looked only slightly unreal, not strikingly bizarre. After some experimentation, we settled on eighty percent reflectant mirrored coontact lenses, which gave us flexibility to control the "kick" by lighting. They also had the virtue of permitting the actors to see through them.
The second problem was what we called the "gunslinger POV," the machine view of the world. We reviewed standard special-effects techniques and rejected them all; they were too familiar, and shared a "filmic" quality - no matter how strange, they still looked like photographic images. I didn't want that.
I knew it was possible to scan images with a computer, and then reconstitute those images in some other form. I didn't know whether these techniques could be applied to motion picture film, but the idea seemed worth exploring. I also liked the idea that the machine world-view would be literally that - a series of pictures created by computer. Several experts told us that this was impossible. Finally we found a computer-graphics artist, John Whitney, Jr., who said he could do it. Neither he nor anyone else could be sure exactly how it would turn out, but we decided to go ahead and try it.
The final technical problem was burning the gunslinger's face with acid. Nobody had the faintest idea of how that could be done; the script called for the skin of the face to bubble and dissolve. Frank Griffith, the makeup man, began experimenting, but by the time we started shooting we still didn't know how to make the effect work.
In all our planning, my overriding concern was to avoid a bizarre, science fiction appearance to the film. The story was strange and certainly suggested a strange treatment; I could easily imagine using wide-angle lenses, eccentric compositions, and disorienting cutting patterns. I decided to shoot the film straight, playing against the strangeness instead of emphasizing it.
major consideration here was the script itself. Most of the situations
in the film are clichés; they are incidents out of hundreds
of old movies. I felt that they should be shot as clichés.
This dictated a conventional treatment in the choice of lenses
and the staging. There were a couple of peculiar corollaries to
that decision. One was the use of arbitrary crane shots which
appear throughout the picture - the camera moves up and down for
no damned reason except that's the way the old movies were done.
Another was the use of slow motion for shootouts because, in the
years since Kurosawa, Penn and Peckinpah explored the technique,
slow motion has become its own sort of cliché.
Our shooting schedule was so tight that we had no allowances for errors, mistakes, or problems of any sort.
Our luck held for two weeks, then Yul Brynner got shot in the eye with wadding from a blank cartridge. It was a freak accident; we had taken every possible precaution to avoid precisly that, but it happened. Yul's cornea was scratched. He shrugged off the injury, which was minor enough, but it made it impossible for him to wear his silvered contact lenses. Whenever he tried, tears ran down one cheek, and the injured eye turned bright red from irritation. We had to shift the schedule radically to allow several weeks for his eye to recover.
In many ways, it was lucky that no one else was injured. I am a stickler for actors doing their own stunts; I think audiences can always tell if doubles are used. But there are always unexpected elements in a stunt.
Jim Brolin had recently broken his leg doing stunts, and was leery of further injury. His part required a fearless, effortless quality which he never felt during all his jumps and falls. To his credit, the audience will never suspect his apprehension. On the last day of shooting, he was to be bitten by a rattlesnake in the desert. The actual strike was done with reverse photography - the snake was attached to his arm, then pulled off. In the final film, it appears to strike him.
For most sequences, we used unmilked rattlers, but we milked the rattler that was attached to Jim's arm. Even the most thorough milking does not remove all the venom, however - a consideration that Brolin appreciated while the snake was having its fangs hooked into his shirt. Under the shirt he wore a leather and cotton pad designed to protect his arm. Suddenly he yelled that he had been bitten. And he had been, by the small teeth on the snake's lower jaw. Everyone had forgotten about that. He had no ill effects, but he did have a good scare.
Dick Benjamin hadn't done physical roles before. His first stunt required him to be thrown into a breakaway post taht would snap on impact. He hit the post so hard that half the set came crashing down on him - drapes, lights, rigging, everything. He emerged laughing. (In the final picture, you can see some of the electrical cables lying around his feet.)
In the desert, I needed shots of Benjamin riding along a ridge. We picked what seemed to be a suitable ridge, and set up our telephoto cameras about a mile distant, communicating with the actors by radio. The wrangler, Dick Webb, rode along the ridge and reported that it was pretty windy up there. We were surprised; it was pleasant enough down with the cameras. I asked how windy. "Maybe sixty miles an hour," Webb reported laconically. I asked if the shot could be made; he said yes. Dick got on the radio and said he'd try it. I was watching through binoculars and gulped when Benjamin appeared on the ridge. His clothes flapped like sails in a hurricane.
Many people have noticed that Benjamin isn't riding very fast on that ridge, as the sequence appears in the final film. That's the reason: he was trying to gallop along a sheer drop of several hundred feet, with a gale wind blowing.
Yul had been a circus performer in his youth, and he has great physical prowess. He did a number of forward and backward falls without difficulty, including one at the end of the film when he fell flat on his face with a slamming thump so unnerving that the whole set echoed.
His greatest problem was the sequence where his face was burned with acid. Frank Griffith had experimented with various techniques and finally settled on a mixture of makeup and Alka-Seltzer, which fizzed and bubbled when water touched it. That left a final problem: smoke. We had done a test with an extra who had small tubes of smoke glued to his face in various places. Once the smoke began to be squirted out, the extra coughed and cried and sputtered in a comical fashion. The smoke was enormously irritating to nose and eyes.
Still, it was our best solution, if Yul could avoid coughing. He had an upset stomach that day, and had taken Alka-Seltzer for it; as the makeup was applied and the plastic smoke tubes attached around his face, we joked that he had Alka-Seltzer inside and outside.
We did the sequence once
- water in the face, and then smoke pumped through tubing - and
it worked all right, but it happened too fast. Yul said he could
do it slower, and he did the second time around. Somehow we got
the shot, but it was a remarkable effort on his part. As soon
as I yelled "Cut!" he exploded in an attack of coughing.
The rapid work pace built an extraordinary spirit of camaraderie. The pettiness and temper tantrums which are a predictable part of film work simply didn't exist, for the most part. Dailies were a sort of ritual. Everyone attended them: actors, crew, everyone. I worried that this would invite ego flare-ups and grumblings, but that never happened. The actors were generous with one another, and the crew helped one another with problems. Nobody was fired during the course of production, which is a rare thing in itself.
As director, I was camera-cutting. This means I would never shoot a whole scene from a single angle. I'd shoot part of a scene from one angle; part from another angle, with little or no overlap. It was the fastest way to work, but also the riskiest. If the scene didn't work as shot, I had no extra footage to play with.
I was also picking certain sequences to dwell on, and shooting the others quickly. I knew I could not shoot the whole film well in the thirty days I had been given, so I picked the key scenes and concentrated on them. The rest of the scenes were plainly shot in haste, with hope that the audience would forgive me later.
It wasn't an ideal situation.
But I had known that, going in, and so had everyone else. We made
the best film we could in the time we had, and we kept our fingers
Two weeks after shooting, I saw the assembled film for the first time. It was horrible. It was boring, contrived, self-indulgent and slack. I left the projectin room in silent depression. All of our energy and enthusiasm had been wasted on a piece of silly garbage.
Dave Bretherto* was the only person in good spirits. He cheered me up enough to start editing. We went sequence by sequence, changing timing, replacing shots, adding and dropping things within the narrow limits that were possible with the minimal footage I had shot. The picture slowly improved. After a month, I thought we might have something decent after all.
We ran the film for the MGM executives. A few executives liked it, but the general feeling ws that it was a disaster. However, there was no talk of taking the picture away from me - everyone knew that I had camera-cut, and there was no spare film to play with. And the MGM management did agree to some additional shooting. I had once considered beginning the film with a TV commercial, and this idea was taken up. There was a writers' strike at the time, so the commercial was written by Steve Frankfurt, a New York adman.
The executives saw the film
again, a month later. They felt that it was improved, but still
a lousy picture. In any case, we'd all know in two weeks, when
the film was previewed before an audience.
I had plenty to worry about until then. The computer-generated footage was coming in, far behind schedule. John Whitney had found the process unexpectedly time consuming - it took eight hours to produce ten seconds of film. There were intricate problems with color and contrast balance. We were still doing a lot of testing. On the other hand, we were pleased with the general effect.
Fred Karlin was modifying
his score, but nobody knew how it was being changed. I heard it
for the first time when we did a temporary mix of dialogue, sound
effects and music. I thought the new score was terrific. We got
the last of our computer footage, and were ready for the preview.
Metro made no bones about the preview. There would be only one, at the studio, with a selected audience of "regular people," no movie people. If the reaction was good, MGM would release the film carefully; if the reaction was unfavorable, they'd dump it.
The critical question was how many viewers would call the picture either "excellent" or "good" on the preview cards. The picture began. There was a lot of coughing and shuffling; the laughs didn't work and came in the wrong places. I sank lower in my seat. ...
Then at the end there were
screams and applause and when the cards came back, we had a ninety-five-percent
rating, the highest anybody at the studio could remember. People
were slapping me on the back and saying they'd known it all along.
I went out and got very drunk.
Several other changes occurred in the course of shooting. Principally, the ending was changed. We deleted the final fight between Martin and the gunslinger. We tried it, but it seemed stagey and foolish, so we elected to drop it entirely. I had liked the idea of a complex machine being destroyed by a simple machine - the rack - but otherwise I didn't miss the lost sequence. Deleted scenes also included the salesroom sequences and the bank robbery.
The opening is changed. We were unable to get convincing model footage of the hovercraft, and so could not play the sequence with full-screen model shots. We had to use TV images. That was a post-production situation and is, by any standard, a compromise.
Still, a film is an exercise in what is possible. I am satisfied that we arrived at the best solutions to the problems we faced, given the available options.
Westworld was not intended to be profound. Neither was it inteded to be stupid, but out clear goal was entertainment. I like to think that audiences have fun with this film. We had fun making it.
October 13, 1973