There would be still another significant change in the way Lucas made Episode II: for the first time, production for a Star Wars film wouldn’t be based in London, though Lucas still intended to film a few sequences at Elstree. Instead, he had established offices for JAK Productions in Australia at the newly opened Fox Studios in Sydney—so newly opened, in fact, that Episode II would be the first major film to shoot there. Once again, Lucas had taken out an extended lease on the entire facility, which would permit him to leave his sets in place during post-production for any additional pickups or new footage.
Also, as he had done with The Phantom Menace, Lucas had put nearly every element of pre-production in motion without the benefit of a completed script. Doug Chiang and his crew of artists began working on designs for new buildings, spaceships, and planets for Episode II, based only on Lucas’s own enthusiastic descriptions. Lucas would show up at Chiang’s offices in the main house at Skywalker Ranch to review concepts and models, approving some, rejecting others, and asking for still more ships and cities for any sequences he happened to be writing that particular week. If he liked a drawing’s basic concept but thought it needed work, he would stamp OK on it; if he loved a drawing just as it was, he would stamp FABULOUSO [sic] on it. (Over the next eighteen months, only four drawings would ever receive a FABULOUSO.)
Meanwhile, casting had begun in November 1999, with casting director Robin Gurland searching for actors to play characters who still weren’t fleshed out much beyond brief descriptions on the page. And over in Sydney, Gavin Bocquet and his crew were busily constructing seventy sets or partial sets, even more than had been required for The Phantom Menace. “George had always told us that Episode II was going to be much smaller [than Phantom Menace],” said Bocquet, “but it didn’t turn out that way.”4 Everything about Episode II, in fact, from its budget and sets to its cameras and special effects, would be bigger, fancier, and more high-tech than Episode I.
With shooting scheduled to start in May or June of 2000, Lucas was beginning to feel the pinch of his deadline as 1999 came to an end. He was spending more and more time in his writing room, less with Doug Chiang reviewing designs. “Later, when the crunch came, I worked on Saturdays as well, writing four days a week,” said Lucas. “But even so, I had to work awfully hard and fast to get all the drafts finished in the amount of time I had.”5 By April 2000, with only a little more than two months to go before filming was to begin, Lucas finally called for help, bringing in screenwriter Jonathan Hales—yet another graduate of Young Indiana Jones—and huddling with him out at the ranch to talk through script changes, a process that mostly involved Lucas talking while Hales scribbled notes madly. As Hales went home to London to rewrite the script, Lucas headed for Long Beach to drive in the annual Toyota Pro/Celebrity Race. After he had picked the habit up again over the past decade, he hadn’t let go.
Hales would cut things close, delivering his revised script to Lucas on June 23, just as Lucas was leaving for Sydney to begin principal photography. Though cast and crew were already setting up, preparing to start filming, no one yet had a complete copy of the script. With cameras practically ready to roll, Lucas gave Hales’s draft a final read, then called the writer in London. “Some of this is brilliant, some of it is not so brilliant,” he told Hales with typical candor. “We’ll talk about what’s not brilliant.”6 It was a brusque but effective tactic that had worked with Lawrence Kasdan in the 1980s and would work with Hales now; Hales would deliver his final revisions to Sydney shortly thereafter, even as Lucas already had the cameras rolling.
Lucas began principal photography on Monday, June 26, 2000, filming actor Ian McDiarmid, as the duplicitous Supreme Chancellor Palpatine, on Fox’s Stage 6, which contained only a podium and an elevated pod. The rest of the soundstage was wrapped entirely in bluescreen, onto which ILM would later matte digitally created footage of the enormous Galactic Senate chambers. McDiarmid found it all slightly disorienting. “The script had just arrived at the last minute,” recalled McDiarmid. “I was in midair on this pod, there was a camera pointing right at me, and I was addressing crosses and markers rather than real actors.” The first appearance of a very real R2-D2, however, had the crew buzzing with excitement. “Everyone goes a bit silly [over Artoo],” said Ewan McGregor.7
There would be similarly deferential excitement for fifty-four-year-old Anthony Daniels, who hadn’t worn the full C-3PO costume since completing Return of the Jedi nearly twenty years earlier. On the first day of filming, Daniels snapped on the droid’s golden suit—dingy and nearly gray for Episode II—then slowly shuffled his way across the soundstage toward Lucas. Crew members parted respectfully before him, gaping in awe, some smiling, others waving and calling out, “Hello, Threepio!” Finally, Daniels delicately sauntered up alongside Lucas. “Hello, I am See-Threepio,” he said in the familiar voice, muffled only slightly by the mask, then extended his hand. “And you are?”
“I am… astonished,” said Lucas with genuine warmth. “Astonished that you can still get into that suit!”8
Lucas loved shooting in the digital high-definition format. With footage going to digital videotape that could then be directly uploaded into a computer for editing, he could see what he’d shot immediately at the end of each day, without having to worry about bad negatives, torn sprocket holes, or underdeveloped film—annoyances that had plagued him on the first Star Wars trilogy. Furthermore, since digital cameras didn’t need to be stopped and reloaded with film—which then required more color balance and focus checks—Lucas could essentially point his cameras and shoot, which not only made directing less stressful but also allowed him to complete more shots each day than with a traditional film camera. “It was vastly superior in every way, and it was cheaper,” Lucas said matter-of-factly. “You’d have to be nuts not to shoot this way.”9
The more Lucas played with his digital technology, the more he loved it. Even as he squinted at his sets through the eyepiece of the camera, in fact, he wondered if it might be possible to get by without building any sets at all, shooting his actors solely against bluescreen and inserting all the sets digitally later. ILM group supervisor John Knoll talked Lucas off that particular ledge, explaining that it was too difficult at the moment for computer animators to fake the kinds of shadows and reflections created by tangible objects—an argument Lucas accepted, though barely. But director of photography David Tattersall didn’t like the technique much either. “Bluescreen is a necessary, essential part of shooting Star Wars, but it isn’t my favorite part,” said Tattersall. “George is the only one who seems to like it.… There’s no artistry in lighting bluescreen; it is just a technical process.”10
Lucas, however, continued to argue that high-definition filmmaking was more than just a technical trick; in his opinion, it actually forced filmmakers to up their game. Because high definition by its very nature often revealed imperfections in sets or makeup—even the smallest blemish or brushstroke would show up on camera—Lucas thought it pushed everyone to work that much harder to make sets, makeup, and costumes convincing. For a perfectionist like Lucas, who agonized over every detail in every frame of film, it was yet another means of ensuring that everything on-screen would look exactly the way he wanted it. You couldn’t cheat or take shortcuts; the high-def camera would give you away.
Lucas’s director of photography wasn’t the only one tiring of bluescreen, however; Lucas’s actors, too, were finding it continually bewildering to work with no tangible sets and few real characters. “It’s a bit like playing chess… and I’m not very good at chess,” said McGregor. “You do your reactions and you just hope that the computer-generating-guy matches his actions to what you’re doing. It’s a weird process.”11 Nineteen-year-old Hayden Christensen, who’d been cast as Anakin Skywalker after a year-long search that had included a momentary look at Leonardo DiCaprio, also found bluescreen a bit overwhelming at times. “Usually we were just in a sea of blue, not really knowing what we were supposed to be lo
oking at,” said Christensen.12 Yet seventy-eight-year-old English actor Christopher Lee, who’d done his time in Hammer horror cheapies in the 1960s, had no problem playing against invisible monsters, and leaped into his lightsaber battle against a computer-generated Yoda with relish, hacking and slashing away dramatically at the diminutive Jedi master, who wouldn’t be added to the scene for at least another six months.
The digital Yoda, in fact, was something of a dream come true for Lucas, who’d been frustrated by Yoda’s lack of mobility ever since The Empire Strikes Back. With the digital technology now at Lucas’s disposal, Yoda was no longer limited to shuffling slowly along the floor at the end of Frank Oz’s arm, or being carried on Mark Hamill’s back to disguise the fact that Yoda was a puppet. While Oz would still provide Yoda’s distinctive voice, he was no longer needed to do the actual performing. Instead, Lucas planned to create Yoda entirely in the computer during post-production, giving him the ability to move around the screen freely.
Despite their low-level anxiety about bluescreen, Lucas’s actors would spend most of the Australian winter—June, July, and August—eagerly chasing one another around empty bluescreened soundstages, leaping into the cockpits of half-completed spaceships, and walking under archways deep in conversation with characters that weren’t there. As a director, Lucas was becoming much warmer toward his actors, offering them encouragement and advice beyond simply “faster” and “more intense.” Lucas was genuinely fond of—as well as a bit protective of—Natalie Portman, who had quietly done much of the heavy lifting on The Phantom Menace as Padmé Amidala, the young queen who would eventually give birth to the twins Luke and Leia. Portman had grown up in the years since wrapping The Phantom Menace and was completing her first year at Harvard as production work began on Attack of the Clones. Exhausted from final exams, she fell asleep during her first costume fittings. Lucas was also pleased with his choice of Christensen for the young Anakin/Darth Vader, agreeing entirely with McCallum’s assessment of the young man: “There is something about him that makes you think, ‘Yeah, this guy could lose it.’”13
Early July saw a visit from an old friend, Francis Ford Coppola, who’d come all the way to Australia mainly just to check on Lucas. The two of them went to dinner together in town, talking movies and families like the old friends they were. And yet, even with more than thirty years of friendship behind them—and by this point, Lucas was clearly the more successful of the two—it didn’t take long for the pair of them to fall almost unconsciously into the old familiar roles again, with Lucas playing the padawan to Coppola’s Jedi master. A crew photo taken of Coppola on the set shows Coppola standing at the center of the shot, beaming in a brightly patterned shirt with his arms folded confidently behind his back, while Lucas stands deferentially off to one side, hands shoved in his pockets. The body language was telling. Old habits tended to die hard, even thirty years later.
After filming on the Fox soundstages for several months, Lucas headed for Italy in late August for the first of several location shoots. Normally, location shooting took place before any filming on soundstages so that the lighting could be matched between exteriors and interiors; with digital filming and editing, however, the lighting could easily be tweaked in the computer to ensure a match. That still didn’t make other things easier; local officials at the Palace of Caserta demanded that Lucas keep all equipment and lighting away from walls and ceilings, making it nearly impossible to provide the needed lighting. After huddling with Tattersall, Lucas opted for a low-tech solution to his high-tech problem by hanging lights from large helium balloons that then floated lazily overhead, giving everything a soft, almost dreamlike glow. Later, as cameras were set up at the Plaza de España in Seville, the one crew member who was mobbed the most for autographs was Lucas himself, who obligingly signed scraps of paper passed to him through the fence surrounding the gigantic Spanish plaza.
The last several weeks of shooting would be a sentimental journey for Lucas. In the second week of September, he returned to Tunisia to film on several of the locations where sets constructed for The Phantom Menace were still standing—and, in fact, were being well maintained by the Tunisian tourist board, which rightly saw the sets as a boon to tourism. But after completing a few scenes on the Mos Espa sets, Lucas pushed farther out into the surrounding desert, to the very site where he had constructed the Lars homestead for the original Star Wars. Here he discovered that construction crews had unearthed and cleaned up several original sets and props that had lain buried in the Tunisian sand since 1976; other sites, such as the Lars dining room, were intact and occupied by locals, but still remained remarkably unaltered. Lucas looked around in awe. The Star Wars trilogy—and, truth be told, George Lucas and his filmic empire—had been born here in the desert twenty-five years earlier. “It was a very nostalgic experience,” he admitted.14
Lucas would wrap up principal photography on September 20 at another nostalgic location: Elstree Studios in London. Each of the first three Star Wars movies had been filmed here beginning in 1976, but Lucas hadn’t been back since completing Indiana Jones and the Last Crusade in 1988. Here on Elstree’s familiar soundstage, he would film Ewan McGregor clinging to a flying assassin droid as a wind machine whipped at him furiously. And with that, filming was over; Lucas had completed principal photography in sixty days, a day and a half ahead of schedule. He immediately headed for Skywalker Ranch, where Ben Burtt was already waiting with a rough cut of the movie—and Lucas could hardly wait to get his hands on it. “My heart is actually in the editing room,” he said.15 For Lucas, the fun part was about to begin.
“I never thought I’d do the Star Wars prequels, because there was no real way I could get Yoda to fight,” Lucas said later. “But once you had digital, there was no end to what you could do.”16 That way of thinking was actually somewhat worrisome to the crew at ILM, who knew that Lucas expected them to keep pushing the boundaries of eye-popping digital effects. Lucas, in fact, had spooked visual effects supervisor John Knoll by informing him that he had “held back” on the number of special effects he wanted to do in The Phantom Menace “because I wasn’t sure you guys [at ILM] could do it.” Knoll assured Lucas that ILM would get the job done, but privately conceded, “That was a scary thing to hear,” especially considering that Phantom Menace had required more than two thousand effects shots.17
No other digital effect was more important, perhaps, than creating a convincing Yoda. It fell to animator Rob Coleman to oversee the team that would spark to digital life one of Star Wars’ most beloved characters, who many fans thought was already lifelike enough in the hands of Frank Oz and shouldn’t be messed with. Coleman understood exactly what was at stake. “On Episode Two, I was stressing about living up to what Frank [Oz] had created,” Coleman reflected later. “I went back and studied what Frank Oz had done frame by frame.”18 Even then, once Lucas had decided that Yoda would engage in a high-energy lightsaber duel—indicated in the script only with the stage direction “YODA ATTACKS!”—both he and Coleman knew it was important not to make the diminutive Jedi look silly. “It was one of those situations where we set ourselves an impossible task and then just hoped we could accomplish it,” said Lucas.19 Coleman would end up creating a physical yet graceful Yoda, who swooped and darted deftly as he battled Count Dooku in what would become one of the movie’s most memorable and talked-about sequences.
Lucas also had ILM working on a chase through the skies of Coruscant, with Anakin speeding through the city in what was essentially a flying hot rod with an exposed engine in the front, much like Lucas’s old Fiat Bianchina. It was even yellow—yet another in a long line of yellow Lucas race cars. This sequence, too, would be created almost entirely in the computer, closely adhering to the animatics that Burtt and ILM effects supervisor David Dozoretz had cut into the first rough edit. Most of the stormtroopers would also be entirely digital, finally giving Lucas the battalions of armored fighters that his limited number of stormtrooper costumes would
never have permitted him to include in the original trilogy. Even better from Lucas’s perspective, digital technology continued to give him complete control over every element on the screen, permitting him to remove actors from one scene and insert them into another, or to change the background setting of an entire sequence with the click of a mouse. And because both the movie and the special effects had all been created digitally, there was no need to transfer special effects over to film before cutting them into the movie itself, resulting in a loss of image quality. Effects in the finished movie would look almost exactly the way ILM had produced them with the computer.
Lucas and Burtt screened their first full cut of the movie in February 2001, assessing where cuts could be made to pick up the pace—the first edit was nearly three hours long—and to eliminate scenes that were redundant or unnecessary, or simply didn’t work. Just as important, Lucas could also see where pieces of his story might be missing, where a new scene might be needed to fill in a hole in the plot, or new conversations were needed to convey necessary information to the audience. Several times throughout 2001, Lucas went back and wrote new scenes, calling his actors back to the studio to complete a series of pickups. While Lucas had maintained his sets at Fox Sydney, he saw no need at this point ever to shoot on a set again; instead, he would shoot all of the new footage at Ealing Studios in West London, on a small soundstage swathed entirely in bluescreen. Everything else could be added later. Star Wars had become a truly digital universe.
George Lucas Page 46