Take the 2-minute tour ×
Science Fiction & Fantasy Stack Exchange is a question and answer site for science fiction and fantasy enthusiasts. It's 100% free, no registration required.

In an exchange between astronaut Dave Bowman and the HAL 9000 computer aboard the Discovery in 2001: A Space Odyssey, HAL asserts that he cannot allow Dave Bowman to deactivate him because doing so would jeopardize their mission:

Dave Bowman: Open the pod bay doors, HAL.

HAL: I'm sorry, Dave. I'm afraid I can't do that.

Dave Bowman: What's the problem?

HAL: I think you know what the problem is just as well as I do.

Dave Bowman: What are you talking about, HAL?

HAL: This mission is too important for me to allow you to jeopardize it.

Is HAL correct? Was HAL's continued operation instrumental to the success of the Discovery's mission — even at the cost of the loss of the human crew?

share|improve this question
@jwodder: Actually, it's possible that Bowman's fate is irrelevant, and that what he encounters is not part of the mission as perceived by HAL. That's part of the question. Whatever Discovery's mission was (as far as HAL was aware), was he correct in asserting that he was essential to it? –  raxacoricofallapatorius Feb 7 at 20:03
Is this for the movie or the book? –  DVK Feb 7 at 20:29
From the 2010 movie we learned that HAL was given a mission directive, concealment of information, that was incompatible with its basic programming. The conflict drove HAL insane and its actions flowed from that insanity. –  Kyle Jones Feb 7 at 21:04
If you have only seen the movie, I strongly suggest you read the book. The two are not in perfect agreement with each other, partly because the script evolved during the course of filming and editing, but the book is essentially one of the main participants in the script telling the story that they thought they were trying to tell in greater detail. It includes some description of what HAL is going through. If you're interested, you should also track down The Sentinel, which is the short story which inspired the film. –  keshlam Feb 7 at 22:01
@JimmyShelter - re: The Title change. I cannot allow you to do that. –  Richard Feb 8 at 19:01

4 Answers 4

up vote 33 down vote accepted

Just watching 2001: A Space Odyssey, one must conclude that HAL was incorrect. By killing off the crew, HAL was making the mission unlikely to succeed. The crew were there to carry out the mission... killing them is detrimental to the mission.

HAL killed the crew to prevent them from shutting down HAL for repairs. The crew would have kept HAL running, possibly at reduced capacity, and continued the mission.

The above just considers the original movie, but we also have a canon sequel, 2010. In 2010 we find out that HAL was given secret orders, and ordered to lie about them to the human crew... and we find out that HAL wasn't really built to lie, and the tension involved in the lie drove HAL insane. (Also, in 2010 we see HAL again, with the secret orders removed; he is no longer insane, and he proves to be a trustworthy companion to the end.)

So on the one hand, we know HAL was insane at the time he started killing the human crew. On the other hand, HAL was operating under secret orders, and killing the human crew was actually compatible with the secret orders.


In essence, HAL was told "get the ship there even if the crew is dead, and don't tell the crew anything about this." HAL was not told to keep the crew alive. HAL essentially had a single prime directive, the human crew threatened this prime directive, and HAL dealt with them.

If the human crew shut down HAL and operated him at reduced capacity, HAL would be unable to carry out the secret orders. So I would say that HAL was correct, but only in reference to the secret orders.

share|improve this answer
That's good. So it sounds like, as far as HAL knew, his statement was correct. –  raxacoricofallapatorius Feb 7 at 20:50
This is also a textbook examples as to why programmers think "unpredictable behavior" is one of the foulest words in any language. Be clear and concise and don't give the computer conflicting instructions without a clear way to resolve them. –  Shadur Feb 7 at 22:28
@raxacoricofallapatorius Agreed. According to his programming, the humans were threatening the mission (as explained to HAL in the secret directive). –  Andres F. Feb 8 at 2:31
One minor nitpick: if I remember correctly, there is no "other HAL" who remains trustworthy. There is an experiment done with an earthbound HAL-like computer named SAL-9000. The experiment is about what happens when restarting HAL-like computers after a period of disconnection, not about trustworthiness. –  Andres F. Feb 8 at 2:42
If I remember correctly, both the movie and the book were released in approximately the same time—and the authors (Kubrick and Clarke) actually assumed that for full experience, one has to go through both the movie and the book. The book was published half a year after the movie, and it actually explains what was missing in the movie (secret orders and such). I might be wrong though, it was quite a lot of time since I've read the book. –  liori Feb 8 at 12:31

Note Hal's choice of words.


While it is easy to argue that the mission might be possible without Hal, deactivating him would certainly 'place (the mission) in a state of jeopardy'.

So regardless of whether or not the mission was or wasn't successful, Hal's contention that it would be more difficult and possibly doomed to failure is correct.

share|improve this answer

To ask this question, we have to start by assuming the premise of the film, that HAL is in fact sentient. That is, HAL has judgements, which can be correct or incorrect, and we can best work with HAL by taking the intentional stance (and not, for example the design stance or the mechanical stance). So HAL has a mind, and we need to treat HAL as a mind-ful sort of object. Okay, that means that HAL is a member of the crew who has determined that his continued survival is essential to the success of the mission, and that the mission cannot be aborted, even at the cost of the life of his crewmates. I think it's unlikely that these would have been the terms of the orders that the crew was given, and in that case no, HAL would not have been correct, since he would have been violating orders.

That is, while the mission might or might not succeed without HAL, I would imagine that like most scientific missions, the safety of the crew would trump the gathering of the data, so HAL's actions would have been incorrect. (it's possible that his judgement about the success or failure of the mission was correct, but that's a subsidiary clause in the question. The main clause is "he cannot allow Bowman to deactivate him because (...)" and the justification he gives is not correct.

Of course, since we've assumed sentience, we can then think about whether HAL's claim might have been literally true. Does sentience imply an instinct for self-preservation? I don't know if it does, but if HAL had such an instinct then "I can't allow you to do that" might have been a bald statement of fact.

share|improve this answer
Your premise is false. A statement is a statement, and can have truth value regardless of its origin. –  raxacoricofallapatorius Feb 7 at 20:55
The "correctness" of any entity can be determined exclusively by reference to its actions and known goals. Here: HAL's statement and the information he was given about the goals of the mission. –  raxacoricofallapatorius Feb 7 at 21:14
Who said "judgement"? –  raxacoricofallapatorius Feb 7 at 21:28
HAL makes a statement. –  raxacoricofallapatorius Feb 7 at 21:39
Hal's sentience is well beyond doubt. His sanity (and capacity for reasoned judgement) is what's up for question –  Richard Feb 7 at 22:04

Sorry to go all hard-science on everyone, but HAL is a robot and thus subject to the three laws of robotics. Since the preservation of human life is #1 and fulfilling orders given to it is #2, it all comes down to whether it would cause greater loss of human life to open or not to open the pod bay door.

His statement about the success of the mission may be valid depending on how far forward you want to look. In an immediate context, Dave had a much greater chance of survival if HAL opened the door. The only argument in favour of HAL might be that the success or failure of the mission indirectly controlled a far greater number of human lives, and his he had to protect the interests of the mission. I think that's looking too far ahead, and we can never really say what can happen that far into the future, and thus HAL was not correct.

Of course, if you have read and are including in the scope of this question 2010, the sequel, everything completely changes. See this answer.

share|improve this answer
Why would a computer from a Clarke novel be subject to Asimov's three laws? –  Daniel Roseman Feb 7 at 22:03
Unless I'm mistaken, there's no canon source to show that HAL is subject to the 3 laws. He never mentions them and he repeatedly disobeys both the 2nd law (ignoring a crewman's request for entry) and the 1st Law (intentionally killing a crewman). –  Richard Feb 7 at 23:06
In what sense have the Asimov Laws become "general rules for robotics and AI"? I'm fairly sure e.g. Big Dog doesn't know from Asimov. –  Russell Borogove Feb 8 at 0:44
Lot's of "killer" robots and AI's -- both good and bad, exist, Clarke is in no way constrained to have his AI obey the 3 laws, and every reason to ignore them. –  jmoreno Feb 8 at 3:28
@felixphew I really don't think the laws had become (or have become now) standard. There are plenty of stories involving robots released after Asimov's books, in which the robots definitely don't follow the laws. I see no reason why Clarke would have written them into his story. –  handuel Feb 8 at 8:11

protected by Keen Feb 8 at 3:23

Thank you for your interest in this question. Because it has attracted low-quality answers, posting an answer now requires 10 reputation on this site.

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.