Open the Pod Bay Doors · Episode 4 of 10 · 22 min
How Kubrick Built HAL 9000 — and Why HAL Actually Refuses
HAL is often remembered as a machine that rebelled, but this episode argues he did exactly what he was told by people who gave him two incompatible orders. It follows the making of the character, from a 1964 lunch to the advisers, IBM's conditions and Douglas Rain's quick recording, and clears away myths such as the IBM letter-shift.
Who's who
- Stanley Kubrick+2 more episodesDirector and co-writer of 2001
- Arthur C. Clarke+2 more episodesCo-writer and author of the novel that explains HAL's failure
- Marvin Minsky+1 more episodeMIT AI pioneer and the film's principal machine-intelligence advisor
- Irving John Good+1 more episodeCryptanalyst and statistician who consulted on supercomputers
- Douglas Rain+1 more episodeCanadian actor who voiced HAL
- Martin Balsam+1 more episodeActor cast first as HAL's voice and replaced
- Frederick I. Ordway IIINASA scientific and technical consultant
- Douglas TrumbullSpecial photographic effects supervisor
Where it happens
- Trader Vic's, New YorkWhere Kubrick and Clarke first met in April 1964
- Shepperton Studios, EnglandWhere principal photography began in December 1965
- Borehamwood, England+1 more episodeMGM-British studio where Discovery's centrifuge was built
- Urbana, IllinoisHAL's stated birthplace in the story
Timeline
- April 22, 1964Kubrick and Clarke meet at Trader Vic's in New York
- December 29, 1965Principal photography begins at Shepperton Studios
- April 2, 19682001 premieres in Washington, D.C.
- July 1968Clarke's novel is published months after the film
- April 1969The film wins the Oscar for visual effects
- 19822010: Odyssey Two supplies the diagnosis of HAL's failure
Still debated
Whether HAL is conscious is left open by the film. So is whether Clarke's novel should count as authoritative for what the film says, and whether Kubrick's withholding of explanation was artistry or a result of late cuts.
← PreviousNorbert Wiener's 1960 Warning About What We Ask Machines to DoNext →MYCIN Beat Stanford's Doctors in 1979 and Was Never Used
