The mere fact that David Lynch's Dune was made at all, and in the Hollywood of the early 1980s to boot, is something of a miracle. Would that it was a better adaptation of the source material, or just a better movie, period. It seems best thought of as an SF-tinged descendant of conventional Hollywood historical costume epics — The Ten Commandments, Ben-Hur, Cleopatra, Spartacus — as it has both the best qualities of those projects (epic scope, ambitious plotting, fun casting) and their worst (ponderousness, pretentiousness, miscalculations of pace and tone).
Frank Herbert's now-classic novel used the struggle for resources in the Middle East as the jumping-off point for a blend of soft SF and James A. Michener-esque historical fiction that still remains unsurpassed. Two feuding noble houses in a far-future universe enter into a tricky agreement to transfer control of a strategically important planet, a wasteland that no one would bother with were it not for the invaluable, druglike substance harvested from its sands. When the Atreides, receivers of the planet, are betrayed by the Harkonnen, its former custodians, the dauphin of the Atreides goes into hiding with his mother, and finds he's being received as a messiah by the indigenous population.
As with any great work, there's so much more than a synopsis alone relates. Herbert's book meshed a wealth of details — big and little, plot- and setting-related — such that one couldn't ignore any one of them without hollowing things out. There was no way to wedge all that into even a three- or four-hour movie, let alone the two-and-fifteen of Lynch's theatrical version. Lynch also worked against the terrible odds of the sheer fact that this was a major studio production. There was just no way to tell a story this complex using the contemporaneous vocabulary of mainstream commercial filmmaking, and everything Lynch wanted to bring to it was either dialed down or pushed out altogether.
Most any SF or fantasy movie that's a) made for a broad audience and b) features a substantial amount of worldbuilding ends up being a compromise. It either has to explain itself at length to become comprehensible, or it has to just throw the audience into its experience and trust in their curiosity and intelligence. The former is safer, but runs the risk of making everything talky and static. The latter ... well, nobody spends $40 million on a movie, then just crosses their fingers and hopes everyone gets it. Universal Pictures's Sid Sheinberg (who also infamously presided over the mutilation of Ridley Scott's Legend and Terry Gilliam's Brazil) and producer Dino De Laurentiis were not in the habit of just crossing their fingers. And so much of the movie is riddled with clunky narrative devices like the opening sequence where Princess Irulan (Virginia Madsen), daughter of the universe's Padishah Emperor Shaddam IV (José Ferrer), explains how the spice that drives the universe can only be found on the desert planet Arrakis, and the power struggle about to unfold over it.
A plot is already in motion as the film begins. The Emperor feels threatened by the growing popularity of Duke Leto Atreides (Jürgen Prochnow), whose uprightness and humane decency draw loyalty. His outward plan is to assign House Atreides custody of Arrakis, swapping out the despised and despicable Baron Vladimir Harkonnen (a leprous Kenneth McMillan), but to secretly allow the Harkonnens to activate a sleeper in House Atreides, engineer their downfall, and retake spice production for themselves. At the center of this plot is the Duke's young son, Paul Atreides (Kyle Maclaughlan), the loving and loved product of Leo and his concubine Jessica (Francesca Annis). He is to be assassinated as part of the plot, as he unknowingly embodies a centuries-long program by Jessica's sisterhood, the Bene Gesserit, to produce a being of superhuman ability.
The best parts of the film are powered by the charisma of its cast: Maclaughlan, of course (he was also the best part of Lynch's fulsome Blue Velvet); Annis and Prochnow, palpably affectionate; and a feisty Patrick Stewart as Leto's personal guard Gurney Halleck. I also had a soft spot for Dean Stockwell as Dr. Yueh, Leto's physician, squeezed by the Harkonnens and who finds a way to squeeze back; and Sting makes an enjoyably hissable foil for Paul, even if his role is essentially a cameo. But too much of the rest of the cast is either underwritten (Sean Young as Chani, Paul's lover) or there mostly to embody repulsion (McMillan). Of all that I dislike about the film, its compulsive association of grotesquerie with evil is the worst, doubly so given that Lynch's previous movie purported to be a refutation of that very concept (The Elephant Man).
If the narrative mode for the movie was the costume epics of old, so were the visuals: lustrous palaces and audience chambers, cavernous grottoes, and Lynch's favorite atmosphere, grimy industrial spaces. Individual moments of Gothic gruesomeness and stark beauty stand out: the sandworms bursting through the dunes; the Guild Navigators folding space in their cathedral-like ship; Paul's little sister Alia, wielding a knife and drunk with the blood of her enemies. Nearly every shot exhibits the $40 million (easily $150 million today) spent on the movie. Problem is, the rest of the time, it doesn't look like they spent a fifth of that. Many of the optical effects — e.g., the ornithopter traveling across the desert — look shaky and thrown together even by 1984 standards. One drawback of the 2021 4K remaster of the film is how baldly it exposes those problems, even as it clarifies the gorgeous matte paintings and wardrobe design. It's depressing how a movie can look both lavish and cheap, so ambitious and visionary at one turn and so banal and mingy the next.
Dune has fruitlessly tempted filmmakers for decades. Before Lynch came along, Robert P. Jacobs tried to launch a version with David Lean at the helm, a project cut short by Jacobs's death. Then Alejandro Jodorowsky spearheaded a notoriously mythic version, one unabashedly deviant from Herbert's original story. What that project could have been was explored in a documentary that constitutes a fascinating one-eyed peek into an alternate cinematic universe. I think I would have admired the Jodo-Dune without actually liking it, because of how it would have been a Jodorowsky project at the expense of just about everything else. Jodorowsky works best when he is simply doing his own thing, without fetters, and not trying to reinvent or expand on someone else's vision.
After Jodorowsky's project fell apart, the book attracted the attention of producer Dino di Laurentiis, who originally tried to bring in director Ridley Scott, fresh off the success of Alien, along with production designer H.R. Giger. Tantalizing glimpses of Giger's concept work can be found in the anthologies of his art, but after Scott dropped out due to personal concerns, Lynch came aboard. His devotion to the project was clear: he worked for a year on six successive drafts of the screenplay, first with his Elephant Man co-writers Eric Bergren and Christopher de Vore, an then later on his own. But Lynch profoundly disliked Giger's designs — something that always struck me as odd given how Giger and Lynch often seemed to be tapping directly into the same underground veins — and instead brought in a new team of Tony Masters (of 2001: a space odyssey), Ron Miller, and costume designer Bob Underwood (Excalibur).
I originally aligned Lynch's Dune more closely to the psychedelic approach Jodorowsky would have taken, but that was only for lack of knowing more about what Jodorowsky would have done. That and the two artists have profoundly different sensibilities: Jodorowsky's psychedelia is more flamboyant and symbolic, even if more about the mere fact that it wields symbols than because its use of symbolism adds up to anything. Lynch's psychedelic visions are inner-directed and esoteric, personalized to the point of being impenetrable. But that was the lesser problem compared to him being at the mercy of people with the privilege of final cut. The disowned TV version, and the various fan edits that have circulated, offer more length, but not necessarily more clarity.
It's hard to tell what Lynch's real vision for the film would have looked like, but the rest of Lynch's work inclines me to think he was better suited to the movie as a designer than as a director or screenwriter. Earlier versions of the script still used irritating tricks like inner-monologue voice-overs to convey key information, and are about as murky as the finished product. I also suspect much of the final cut's ponderousness stemmed from Lynch's nominally conservative camera and editing, meaning his style left behind opportunities to compress storytelling or advance action. The latter half of the movie gets torpedoed worst by this, as whole segments of the book evaporate down to a few badly mouthed lines.
Most every troubled adaptation of some work of clout gets followed by a new attempt, typically a generation or so later. Fifteen years later, more or less right on time by that measure, came another cinematic version of Dune, John Harrison's four-hour TV version, for SyFy in 2000. It makes a valiant effort to preserve more of the story and is well worth seeking out, but suffers from a TV-sized budget and equally rinky-dink special effects. Now, another fifteen-plus years later, Denis Villeneuve's 2021 adaptation looms, promising to fuse the storytelling chops of Harrison's take with the vision of Lynch's version. It might well be so, especially in the wake of how fine Villeneuve's Blade Runner 2049 turned out to be.
Herbert spoke kindly of Lynch's Dune — he saw enough of his work in it to find it praiseworthy — but is also reputed to have said that many people have tried to film the book, and all of them have failed. "Unfilmable" was the adjective applied to the book both before and after Lynch came along for it. But the further we go into cinematic history, the more the unfilmable becomes filmable. Not just because we have examples of how to do it right, but cautionary examples for how to do it wrong — and mixed examples, where some things work and others fail. And also because the visual vocabulary available to the average filmmaker expands year by year, making what was once impossible to encapsulate in images now just a tougher exercise than most. It ought not to be impossible to film Dune well, once Lynch demonstrated it was possible to do it at all.