The Thief and the Cobbler (1993, Richard Williams)
In not just the animation field but the scope of cinema itself, The Thief and the Cobbler stands as a truly unique monument to both the brilliance of unbridled, unbroken creativity and to the dangers of allowing all other concerns to lie subservient to it. It’s one of those cautionary tragedies of filmmaking like Erich von Stroheim’s Greed or Orson Welles’ The Magnificent Ambersons that can likely never be seen as its primary author intended. In production as an adaptation of the Nasruddin stories of Idries Shah in 1964 before gradually mutating into the simpler story of a boy, a princess, a few bad guys and three golden balls, the film would find release in no form until nearly thirty years later, at which point it had long since left the hands of director Richard Williams.
Talk to some people about The Thief and the Cobbler and you will hear about its mastery; to others, its contribution to a frantic distrust of auteurist animation that continues to this day thanks to its facile story and technical triumphs at the expense of depth or broad appeal. Williams got funding to finish the movie by taking a day job as the lead animator on Robert Zemeckis’ Who Framed Roger Rabbit… but by his very own admission, he was never much of a character animator, and Cobbler displays mostly a knack for bravura visuals that’s applied to little else beyond the “oh, wow” potential. It’s not immersive enough to become more interesting than the tale of its own haggard production.
The story, such as it is, bears little evidence of its origins. After Shah’s sister wanted cash, Nasruddin was canned and the surrounding film was kept. This occurred some time in the early ’70s, at which point Williams was funding the project through odd jobs ranging from Pink Panther title sequences to a Raggedy Ann feature but was already recording voice artists for the roles of characters like the dazzling villain Zigzag, performed by Vincent Price. In so many stories like this, we all side with the director by default, but his reasoning for hiring Price, for instance, indicates that he was more a Gilliam than a Welles: he felt comforted by Price’s presence while animating a character he personally had come to despise. More to the point, in the classic Heaven’s Gate mode, he ignored all financial and practical concerns while allowing his obsession with the visual perfection of his film to wear its saga on and on for decades, so long that eventually you imagine he had to notice that golden balls and one-eyed monsters weren’t the makings of much of a movie. There is little purpose in even laying out the plot itself; to the degree that it’s even coherent, it has no emotional arc or body to speak of, and if you’ve read any fantasy story you can likely fill in the blanks.
Nevertheless, Williams went on and fixated on details; the film may be the most detail-oriented in all of cinema, but only in a technical sense. Why did the film take thirty years to make and still remain, by most every conceivable standard, unfinished? To begin with, the characters in the film, especially the “cobbler” Tack, exhibit smooth, almost mechanically “perfect” movements that seem today like computer-assisted graphics but are in fact hand-drawn. That’s a consequence of drawing a full 24 frames per second instead of just 12, a cost-intensive luxury that would already have doubled the production time. Moreover, several scenes don’t use the traditional foreground-background model of animation, whereby characters and props are crafted separately from backgrounds that are either mobile or sterile in their own plane. Instead, Williams frequently has backgrounds moving at the same rate — and in three full dimensions — as the action of a shot; you can watch these scenes and know beyond a doubt that they alone must have taken years to complete (and in some cases, they are not complete).
Which isn’t to say that Williams was completely incorrect. The film does indeed look extraordinarily beautiful. Tack is an inexcusably inexpressive character, many leagues below the Buster Keaton and Dopey caricatures to which Williams is clearly striving, but his eerily fluid acting is a marvel to watch. And the flat, colorful appearance of the film, based on 15th century Persian artwork and later copped shamelessly by Jeffrey Katzenberg’s execrable Aladdin, is stirringly gorgeous; as with something like Lawrence of Arabia, ignore the narrative and pretend you’re watching a series of glorious moving picture postcards and the film is an experience you’re unlikely to forget. And in those moments when the entire picture moves, when it seems to burst beyond the flat frame of all other hand-drawn animation, there is something audacious and absorbingly cinematic about it — to see those backgrounds moving with all else, to see stairs and walls and people all in unison, and to sense the bustling ingenuity at work, is honestly exhilarating.
Unfortunately, as Williams hunkered down at his drafting board and fell into a rabbit hole of perfectionism, he missed the bigger picture. Not only are the most narratively important sequences of the film unfinished, which makes those that are done feel like empty showpieces, they falter at the hands of a blank-faced inadequacy in crafting believable characters that extends to Williams’ heavy-handed work in Roger Rabbit, which is all technical alchemy and has no weight or liveliness. Tack, the Thief and Princess YumYum all move as though they are dancing atop preset patterns; Zigzag slightly escapes this problem with his frenetic design and movement, but only by a hair, with all opportunity to move outside of a predetermined plane nixed by the meticulous flatness of the director. When the film is over, you come away feeling hollow, as though you have seen something pretty impressive but have no takeaway.
After Williams secured financing in the late ’80s, he wreaked havoc on the Warner Bros. arrangement and continued to take little notice of passing deadlines. In the end, the worst possible outcome ensued: the picture was repossessed and forced into completion by an insurance company, who assigned a dunderheaded TV producer to the film. One thing about Fred Calvert, though: he got things done. Heavily critical of Williams’ working methods, if they could even be called that (no script, no storyboards, only a vague idea of the goals being worked toward), Calvert shot through the rest of production in a matter of months, imposing some Disneyesque touches in the absence of Williams, who was completely removed from the last months of production and post: a few songs, dialogue for the formerly mute title characters, and more ’90s strength for the heroine YumYum. Though Calvert did his best to bring something that Williams’ film had been missing, the result was equally soulless and far more pandering: just a third-rate Disney clone that happened to boast unusually strong effects animation and to have a very distinct “look.”
Adding insult to injury, though, by the time the film was theatrically released in the United States in 1995 under the ludicrous title Arabian Knight, it was widely mistaken as both a film with computer-assisted animation — all that work to no real end, especially since it has no significant amount of distinctive personality to distinguish it from CG — and a third-rate imitation of Aladdin. Production drawings, trailers and fragments of the film had been in circulation for decades by the time Aladdin went into production with the involvement of several people who had been privy to and even employed by Williams’ pet project. Its success was thus a spit in the eye to a man who, while probably not the creative genius may claim, certainly is a hard and dedicated worker who believed in his film more than anyone involved with Aladdin believed in anything aside from toy sales and box office figures.
The workprint that Williams screened in 1991 for Warner Bros. executives was bootlegged and is the source of most of the celebrated versions of the film exhibited online today, including the one I screened for this project. It was widely reputed to be just fifteen minutes away from completion, but this is a misleading statistic. To start with, those fifteen minutes are spread uncomfortably throughout the picture so that watching it is a miserably stilted experience (though still preferable to sitting through more than a sampling of Arabian Knight with its ludicrous voiceover and forced cuteness; I had to turn it off after less than ten minutes). We lived with similarly haphazard versions of Fritz Lang’s Metropolis for years, but that film had the advantage of being historically important and — this is key — humane, inventive and exuberant. Williams’ work — which is more a portfolio of his efforts than a movie — can claim just one of those, and its innovation seems solely to be making things difficult for the sake of making them difficult.
Not surprisingly, Williams doesn’t talk about The Thief and the Cobbler much. Its survival has been predicated on the years of anticipation in the animation community for it, fizzled into a weak stateside release by (irony) then-Disney subsidiary Miramax, and by the online expansions and tweaks of the workprint that are now visible in decent quality on Youtube. You sense so much of the painstaking labor that went into the film that it’s hard not feel like a dick calling it a terribly hollow experience. But we must be careful here: a disappointment does not a bad film make, and it’s impossible to dismiss and certainly to dislike any film that contains so much ingenious stuff, even if it’s not very competently strung together. There are numerous impressive sequences, especially in the first half.
For someone without a more than passing interest in the history of feature cartoons and production schedules gone awry, the film is likely to be a slog. (That’s especially true of the children who are ostensibly the target audience of a movie like this.) None of the characters has any heart or presence; they’re all like machines, and in turn I’m not sure I want a hand-drawn cartoon to remind me of CGI as much as this does. And perhaps it’s the long-buried literary geek in me, but: stories should not have three antagonists. In the final analysis, Cobbler has a strange mixture of a sense of life and a stilted surrealism that never really lands. And the last half hour, full of breathless action, climaxes atop climaxes and an overwhelming amount of “business,” is dull on par with the Lord of the Rings movies. But if this is as “finished” as the film got, I’m not sure how well we can really judge it anyway, positively or otherwise. The early chase scene through a checkerboard-Escher maze is one of the coolest things I’ve ever seen, no exaggeration, but it’s by far the best moment in the film — and that’s all it really is, “cool.”