Futureworld: The Westworld Sequel Nobody Remembers

Delos is open again. The company has spent a fortune on new safeguards, the malfunction that killed its guests three years ago has been engineered out, and the resort is inviting the press in to see for themselves. Westworld is closed. The attraction now is Futureworld: a space station, a Mars excursion, a spa, and Roman World for those with older tastes. Two journalists are on the guest list — Chuck Browning, who broke the original story, and Tracy Ballard, a television reporter — and the corporation would very much like a favourable piece.

Futureworld was released in 1976, directed by Richard T. Heffron, and it is the sequel to one of the most efficient science fiction films of the previous decade. Michael Crichton had nothing to do with it. MGM, which made Westworld, was not involved either; the sequel went to American International Pictures, an outfit built on quantity, and it looks it.

The film is a considerable disappointment, and it contains one thing of permanent importance. Both facts deserve a hearing.

What Westworld had, and what this loses

Westworld works because it is a maintenance report.

Crichton’s structure was cold: guests arrive, the park functions, technicians in a control room notice a rising incident rate, and a supervisor makes a defensible decision to continue operating. The malfunctions spread laterally, like an infection through a system, and the film’s most quoted line is about a central control problem that nobody diagnosed in time. Yul Brynner’s Gunslinger is not a villain with a plan; he is a fault condition wearing a hat, and his relentlessness is terrifying precisely because there is no motive to negotiate with.

Futureworld discards this completely and replaces it with a conspiracy. Delos is now run by people with intentions. There is a scheme, there are conspirators who explain it, and there is a plot to be uncovered by reporters. Once the machines have a plan, they can be outwitted, and once they can be outwitted the film becomes a thriller with robots in it.

That is the central failure, and it is a failure of understanding rather than of budget. The screenplay, by Mayo Simon and George Schenck, takes the elements of Westworld — the resort, the androids, the technicians — and reassembles them into a completely conventional shape. Crichton’s whole point had been that nobody was in charge. The sequel’s whole point is that somebody is.

The film also makes the standard sequel error of retaining a beloved element without its function. Brynner appears, in black, as the Gunslinger — in a dream sequence, disconnected from the plot. It is fan service before the term existed, and it is the film acknowledging a debt it has no intention of honouring.

The reason it is in the history books

Two sequences in Futureworld were produced by computer, and one of them is the first appearance of three-dimensional computer-generated imagery in any feature film.

The work came from Ed Catmull and Fred Parke — Catmull’s animated hand and Parke’s animated face, produced at the University of Utah, the institution where essentially all of the field’s foundational work was being done. The hand rotates, the fingers articulate, the face moves. In 1976, in an AIP release, on a screen, in a cinema.

The scale of what follows from that is difficult to overstate. Catmull would go on to co-found Pixar and spend a career at the head of the industry those two clips began. His hand animation is where the entire computer-graphics feature industry has its earliest public showing, and it is sitting inside a mediocre sequel to a Michael Crichton film about a theme park.

The lineage is worth laying out properly, because it is one of the odder facts in the medium’s history and it belongs to one writer. Westworld, in 1973, used digital image processing to render the Gunslinger’s pixelated point of view — the first use of digital imagery in a feature. Futureworld, in 1976, delivered the first 3D computer-generated imagery. Five years later Looker, also Crichton, produced the first computer-generated human character. Three landmarks, three films, one writer’s franchise and one writer’s obsession — and the middle entry is the one he had no involvement with, which suggests the interest was in the water at Delos rather than in the man.

There is a further irony that the film never sees. Futureworld uses its computer graphics as a prop — a technician demonstrates a machine, the machine renders a hand — and the film has no idea it is showing you the future of its own medium. Crichton, in Looker, would put the same technology at the centre of an argument about what it was going to do. Here it is set dressing.

What actually works

There is craft in it, and pretending otherwise would be lazy.

Heffron shoots the resort’s underside with genuine interest. The film’s most effective material is the below-decks stuff: Delos as an industrial facility, corridors of racked components, technicians running maintenance on rows of blank faces. Stuart Margolin plays a maintenance man with a robot he keeps for company, and the performance is oddly moving — a man in a factory who has made a friend of a machine because the humans around him are running a business.

Peter Fonda and Blythe Danner are better than the material. Danner in particular is doing precise work as a reporter who has been sent to write a puff piece and knows it. And the film’s decision to make its protagonists journalists is a shrewd 1976 instinct — this is a picture made in the immediate wake of the Watergate reporting, and the investigative-press thriller was the shape that year’s paranoia naturally took.

Some of the production is real. Sequences were shot at NASA facilities, which gives the space-station material a documentary texture that AIP could never have built on a set, and the film’s future looks plausibly institutional rather than designed.

What Delos did next

The franchise’s later history is the best available proof of the diagnosis.

Futureworld did not launch a series. What followed instead was Beyond Westworld, a television show that reached American screens in 1980 and was cancelled almost immediately — a handful of episodes were produced and fewer were broadcast, built around a Delos security man chasing androids that a rogue scientist keeps deploying against the world. That premise is Futureworld’s error industrialised: a villain with a plan, weekly, forever. It lasted about a month.

Then the property sat dormant for three and a half decades until HBO revived it in 2016 and ran it for four seasons. The revival’s first year is very good, and the reason it is good is instructive: it went back past Futureworld entirely and picked up Crichton’s original proposition — a resort full of machines executing their programming correctly, and a corporation whose control room cannot explain the incident rate. The series then spent its later seasons accumulating conspiracies, factions and masterminds, and became progressively harder to care about as it did so.

Two revivals, both drawn to a scheme, both weakest where the schemers are strongest. Whatever else Futureworld failed at, it identified the trap early and fell into it first.

The case against, and the verdict

It is slow. The middle hour is two reporters walking around a resort being shown things, and the film’s idea of tension is a corridor. The Roman World and Mars sequences exist to justify the brochure. The score is anonymous. The conspiracy, once unpacked, is a plot device from a 1950s serial. And the film is fundamentally cynical in a way Westworld never was — Crichton’s film was a warning, and this one is a product built to a spec.

The wider point is about what happens when a sequel keeps the furniture and loses the thesis. Everything visible in Westworld is here: the androids, the resort, the corridors, Brynner. The one thing that made Westworld frightening — the proposition that a complex system will kill you through nobody’s fault at all — has been swapped for a scheme, because a scheme is easier to write and easier to resolve. Delos in 1976 is a company with a plan, and a company with a plan is just a villain.

Watch it anyway, once, for Catmull’s hand and for the racks of faces in the basement. It is a bad film that is standing on the spot where an entire industry started, entirely unaware of it. That is a rare thing to be able to see.

Spoilers below

The scheme is duplication. Delos is scanning its high-value guests — politicians, industrialists, military officers, the people whose invitations the resort takes such care over — and building androids to replace them. The corporation is running a soft coup through its guest list: send the powerful home as machines, and own the world without a shot.

It is a serviceable idea from a poor source. This is Invasion of the Body Snatchers with a booking system, and where Siegel’s film in 1956 had made the replacement horrifying by keeping the mechanism invisible and the victims ordinary, Futureworld explains its process at length to two reporters who write it down. The paranoia evaporates on contact with the exposition.

The film’s one genuinely good sequence is Tracy’s dream, and it is where Brynner earns his fee. Ballard, asleep in the resort, dreams the Gunslinger — Brynner in black, doing the walk, the stillness, the whole apparatus of the 1973 film — and the sequence is charged in a way nothing around it is, because for three minutes the film remembers what it is a sequel to. It is also, tellingly, entirely non-diegetic. The best thing in Futureworld is a dream that has no bearing on the plot.

The duplicate confrontation in the third act is the film’s payoff and its self-diagnosis. Browning and Ballard meet their own copies, and the resolution turns on the humans out-thinking machines that have been built to think like them. That is a satisfying enough beat, and it is the exact opposite of the ending Crichton wrote three years earlier, where a man survives by exploiting a physical property of a machine that is not thinking at all. Westworld ends with an exhausted human beating a mechanism. Futureworld ends with a clever human beating a plot. The distance between those two endings is the whole reason one film is a landmark and the other is a curiosity with a famous hand in it.