Box-office failure as the price of authorship

In 1977, William Friedkin spent $22 million remaking Henri-Georges Clouzot's The Wages of Fear (1953). He changed the title to Sorcerer and set it in the present day, with a cast of relative unknowns from four different countries. He shot in the Dominican Republic, Ecuador, and Mexico. He built a bridge in a jungle for a sequence in which two trucks cross a rope bridge in a rainstorm — the sequence took weeks to film and cost hundreds of thousands of dollars.

Sorcerer opened the same week as Star Wars. It made $9 million on its $22 million budget and effectively ended Friedkin's status as a bankable director. He did not make another film with comparable studio support until To Live and Die in L.A. in 1985.

The injustice of the timing is almost comic. Friedkin lost a coin flip with George Lucas and paid for it with the rest of his career. But the timing was not the only problem. Sorcerer was a film with no obvious audience — slow, brutal, largely without redemption, organized around men doing something horrible to survive rather than something heroic. In 1977 that was hard to sell. In 2026 it is a masterclass in physical filmmaking.

The pattern

The years between 1972 and 1985 produced a cluster of expensive personal films that destroyed their directors' commercial standing. The scale and specificity of the failures varied, but the pattern was consistent: a director with major commercial credits was given unusual freedom on the basis of those credits, spent more than the studio expected, made something formally ambitious and commercially inert, and absorbed the consequences.

Peter Bogdanovich's At Long Last Love (1975) was a musical comedy with Burt Reynolds and Cybill Shepherd, shot live on set — the actors singing to live piano accompaniment rather than lip-syncing to pre-recorded tracks, as was standard. The experiment was a significant technical ambition and an equally significant commercial disaster. Variety called it "the worst musical ever made." It took Bogdanovich most of a decade to recover professionally, and he never fully did.

Francis Ford Coppola's One from the Heart (1982) was shot entirely on soundstages at Zoetrope Studios, his own production facility, using an early video assist system that allowed him to direct from a control room and composite the image in near-real-time. The film cost $26 million. It grossed $636,000 in its opening weekend. Zoetrope went bankrupt. Coppola spent the following decade directing films he did not necessarily want to make in order to pay off the debt.

Brian De Palma's The Bonfire of the Vanities (1990) was a different kind of failure — an adaptation of Tom Wolfe's novel that was misconceived from script stage and then publicly pilloried in Julie Salamon's book-length dissection The Devil's Candy, published the year after the film's release. The failure was administrative as much as creative, but the effect on De Palma's ability to get projects greenlit was the same.

What the failures gave us

The argument is not that the films were good and the studios were wrong — the argument is more complicated. Sorcerer is genuinely great. Heaven's Gate, which we have discussed at length elsewhere on this site, is genuinely great. At Long Last Love is not great but it is interesting: its failure taught Bogdanovich something about form, and his subsequent smaller films (Nickelodeon, Mask) show a director who had absorbed the lesson of his own excess. One from the Heart is not great but its failure accelerated the development of pre-visualization and compositing technologies that subsequent directors, including George Lucas and James Cameron, used to make technically ambitious films work on budget.

The industrial lesson is that the New Hollywood auteur system — in which directors were granted unusual creative control on the basis of box-office track records — was never sustainable at the budget levels it reached in the late 1970s. Heaven's Gate's failure was the proximate cause of its end. But the system had been under strain since Coppola's Apocalypse Now came in at $31 million in 1979 (budgeted at $12 million) and Spielberg's 1941 came in at $35 million the same year and underperformed. The studio system's withdrawal of auteur-level trust happened in slow motion, and then all at once.

What you lose when you manage failure away

The studio logic that emerged from these failures — the logic that has governed Hollywood since roughly 1982 — is rational from a risk-management perspective. Sequels, franchises, pre-sold IP, market-tested premises, opening-weekend tracking: each tool reduces the probability of a catastrophic miss. The system works. It produces films that make money reliably.

What it does not produce is Sorcerer. The film exists because Friedkin was allowed to be wrong on an expensive scale. The bridge sequence — the practical piece of filmmaking that most working cinematographers and directors cite when asked what they wish they could have made — was built, in a jungle, for a film that probably should not have existed by any rational financial calculus. It exists because a studio in 1977 trusted a director who had earned that trust and then discovered the trust had limits.

The films we surface in our deeper collections tend to come from the gaps in the managed system — the European co-productions with ambiguous financing, the American independents made before the residuals system made independent film economically complicated, the films from territories where the studio logic never fully took hold. This is not a nostalgic argument for the return of the New Hollywood. The New Hollywood had its own failures of judgment (misogyny, substance abuse, the collapse of story discipline). It is a structural argument: the conditions that made certain films possible are gone, and the films themselves are the record of what those conditions permitted.

Sources & further reading

• Bach, Steven. Final Cut: Art, Money, and Ego in the Making of Heaven's Gate, the Film that Sank United Artists (1985).
• Salamon, Julie. The Devil's Candy: The Anatomy of a Hollywood Fiasco (Houghton Mifflin, 1991) — on The Bonfire of the Vanities.
• Biskind, Peter. Easy Riders, Raging Bulls: How the Sex-Drugs-and-Rock-&-Roll Generation Saved Hollywood (Simon & Schuster, 1998).
• Criterion Collection — Sorcerer release page
• The Guardian — Sorcerer restoration review (2014)