There's a natural, simple thought that the movie The Matrix encourages. This is that there's something bad about being inside the Matrix. That is, there's an important respect in which people inside the Matrix are worse off than people outside it. Of course, most people inside the Matrix are ignorant of the fact that they're in this bad situation. They falsely believe they're in the good situation. Despite that, they are still worse off than people who really are in the good situation.
I said this is a natural, simple thought. When we look more closely, though, this natural, simple thought starts to get very complicated and unclear. Many questions arise.
First question: Who is the Matrix supposed to be bad for? Is life inside the Matrix only bad for people like Trinity and Neo who have experienced life outside? Or is it also bad for all the ordinary Joes who've never been outside, and have no clue that their present lives are rife with illusion? The movie does seem to suggest that there's something bad about life in the Matrix even for these ordinary Joes. It may be difficult to face up to the grim realities outside the Matrix, but the movie does present this as a choice worth making. It encourages the viewer to sympathize with Neo's choice to take the red pill. The character Cypher who chooses to reinsert himself into the Matrix is not portrayed very sympathetically. And at the end of the movie, Neo seems to be embarking on a crusade to free more people from the Matrix.
What do you think? If you had the power to free people from the Matrix, would you use that power? We can assume that these people's minds are "ready," that is, they can survive being extracted from the Matrix without going insane. But let's suppose that once you freed them, they did not have the option of going back. Do you think they'd be better off outside? Would you free them? Do you think they'd thank you?
Or do you side with Cypher? Do you think that life inside the Matrix isn't all that bad—especially if your enjoyment of it isn't spoiled by the knowledge that it's all a machine-managed construct?
Second question: Does it matter who's running the Matrix, and why? In the movie, the machines are using the Matrix to keep us docile so that they can use us as a source of energy. In effect, we're their cattle. But what if we weren't at war with the machines? What if the machines' purposes were purely benevolent and philanthropic? What if they created the Matrix because they thought that our lives would be more pleasant in that virtual world than in the harsher real world? (Iakovos Vasiliou discusses a scenario like this in his essay.) Or what if we defeated the machines, took over the Matrix machinery ourselves, and then chose to plug ourselves back in because life inside was more fun? Would these differences make a difference to whether you regard life inside the Matrix as bad? Or to how bad you regard it?
In his third essay, Christopher Grau discusses Robert Nozick's "experience machine." Nozick thinks that there are things we value in life that we'd be losing out on if we plugged into an experience machine. He thinks there are things we lose out on even if the operators' intentions are benevolent and we plug in of our own free choice. Do you think that's right? Would you say the same thing about the Matrix?
Our answers to these questions will be useful guides as we try to determine what it is about the movie's version of the Matrix that makes us squeamish.
II
In order to figure out what's so bad about being in the Matrix, it will help to do some conceptual ground-clearing.
When they think about scenarios like the Matrix, some people have the thought:
If in every respect it seems to you that you're in the good situation, doesn't that make it true—at least, true for you—that you are in the good situation?
This line of thought is never fully endorsed in the movie, but the characters do sometimes flirt with it. Consider the conversation Neo and Morpheus have in the Construct:
Neo: This isn't real...
Morpheus: What is "real"? How do you define "real"? If you're talking about what you can feel, what you can smell, what you can taste and see, then "real" is simply electrical signals interpreted by your brain...
Consider Cypher's final conversation with Trinity:
Cypher: ...If I had to choose between that and the Matrix...I choose the Matrix.
Trinity: The Matrix isn't real.
Cypher: I disagree, Trinity. I think the Matrix can be more real than this world...
Are the claims that Morpheus and Cypher are making here right? Is the world that Trinity and Cypher experience and seem to interact with when they're inside the Matrix just as real (or more real?) than the world outside?
The standard view is "no," the Matrix world is in some important sense less real. As Morpheus goes on to say, the Matrix is "a dream world." The characters are just experiencing a "neural interactive simulation" of eating steak, jumping between buildings, dodging bullets, and so on. As Neo says when he's on the way to visit the Oracle, "I have these memories from my life. None of them happened." In fact, he never has eaten steak, and never will. It just seems to him that he has.
And presumably that's how things would be even if no one ever discovered that it was so; even if no one ever figured out that the Matrix was just a "dream world."
Philosophers would express this standard view by saying that facts like:
Some people get uneasy with this talk about "objective facts." They say:
Well, what's true for me might be different than what's true for you. When I'm in the Matrix it really is true for me that I'm eating steak and so on. That might not be true for you, but it is true for me.
Let's try to figure out what this means.
Some of the time, people use expressions like "true for me" in a way that doesn't conflict with the view that the facts in question are objective.
For instance, all that some people mean by saying that something is "true for them" is that they believe it to be true. When you're in the Matrix you do believe that you're eating steak; so in this sense it will be "true for you" that you're eating steak. And what you believe to be true will often be different from what I believe to be true; so in this sense something could be "true for you" but "false for me." When a philosopher says that it's an objective fact whether or not you've ever eaten steak, she's not disputing any of this. She accepts that you and I may disagree about whether you've ever eaten steak. She's not even claiming to know who's right. She may be ignorant or mistaken about your past dietary habits, and she knows this. You may have better evidence than she, and she knows this too. All she's claiming is that there is a fact of the matter about whether you've eaten steak—regardless of whether you or she or anybody else knows what that fact is, or has any beliefs about it. And this fact is an objective one. If it happens to be true that you've eaten steak, then it's true, period. It's not "true for you" but "false for me." What you and I believe, and who's got better evidence for their belief, are further separate questions.
Usually when two people disagree about some matter, they agree that the fact they're disputing is an objective one. They agree that one of them is right and the other wrong. They just disagree about who. For some matters, like ethical and artistic matters, this is less clear. It is philosophically controversial whether ethical and artistic truths are objective, and whether the same truths hold for everyone. But for our present discussion, we can set those controversies aside, and just concentrate on more prosaic and mundane matters, like whether you've ever eaten steak, whether your eyes have ever been open, and so on. For matters of this sort, we'd expect there to be only one single common truth, not one truth for you and a different truth for me.
Now, sometimes we speak incompletely. For example, we'll say that a kitchen gadget is useful, when we really mean that it's useful for certain purposes. It may be useful for cutting hard-boiled eggs but useless for cutting tomatoes or cheese. We'll say that the cut of certain suits makes them fit better, when we really mean that it makes them fit certain people better. It doesn't make them fit people with unusual body shapes better. And so on. In cases like this, if one way of completing the claim is natural when we're talking about you, and another way when we're talking about me, then we might be tempted to talk of the claim's being "true for you" but "false for me." For instance, suppose you're cutting eggs for a salad and I'm cutting the tomatoes. We're each using the same kitchen gadget, you with good results and me with frustrating results. If you say "This kitchen gadget is useful," I might respond "That may be true for you, but it's not true for me." There's no conflict here with the view that facts about usefulness are objective. Really there are several facts here:
So the ways of talking about things being "true for me" etc. that we've considered so far don't conflict with the view that the facts we're dealing with are objective.
People who dislike objective facts want to say something stronger. They want to say it really is true for the characters inside the Matrix that they've eaten steak. They're not just making a claim about what those characters think is true. When those characters think to themselves, "I've eaten steak hundreds of times, and so has my friend Neo," what they're thinking really is supposed to be true. At least for them. For Neo and Trinity and others it may not be true.
One way to flesh this idea out is with a philosophical theory called verificationism. (Sometimes this theory is called anti-realism.) If you're a verificationist about certain kinds of fact, then you reject the idea that those facts are objective. For example, a verificationist about height would say that how tall you are depends on what evidence there is about how tall you are. It's impossible for all the evidence to point one way, but the facts about your height to be otherwise. The facts have to be constrained by the evidence. Sure, the verificationist will say, people sometimes make mistakes about their height. They sometimes have false beliefs. But those mistakes have to be in principle discoverable and correctable. It doesn't make sense to talk about a situation where everybody is permanently and irremediably mistaken about your height, where the "real facts" are so well-concealed that no one will be able to ferret them out. If the "real facts" are so well-concealed, says the verificationist, then they cease being facts at all. The only height you can have is a height that it's in principle discoverable or verifiable that you have. (Hence the name "verificationism.")
When we're discussing the Matrix and examples like it in my undergraduate classes, and students start talking about things being "true for" them, but "false for" other people, they're usually trying to sign onto some kind of verificationism. They'll say things like this:
If all my evidence says that there is a tall mountain there, then in my personal picture of the world there is a tall mountain there. That's all it can mean, for me, to say that there's a tall mountain there. The mountain really is there, for me, so long as it appears real, and fits my conception of a tall mountain.
I'm always surprised to hear students voicing approval for this view. It's a pretty strange conception of reality. Some philosophers do defend the view. But I'd be really surprised if 30% of my university students really did think this is the way the world is. As a group, they don't usually tend to hold strange conceptions of reality; I don't find 30% of them believing in astrology or body-snatching aliens, for instance.
Mount Everest is 8,850 meters tall. Most of us think that Mt. Everest had this height well before there were any human beings, and that it would still have this height even if no human beings or other thinking subjects had ever existed. But it's not clear that a verificationist is entitled to say things like that. If there had never been any thinking subjects, then there wouldn't have been anybody who could have had evidence that Mt. Everest existed. So according to the verificationist, then, there wouldn't have been anybody for whom it was true that Mt. Everest is 8,850 m tall. It looks like the verificationist has to deny that Mt. Everest would still have been 8,850 m tall, even in situations where no thinking subjects had ever existed. This is what makes verificationism such a strange view.
Perhaps the verificationist will respond: Granted, in the situation we're envisaging, nobody actually has evidence that Mt. Everest is 8850 m tall. But the evidence is still available. (Mt. Everest will cast shadows of certain lengths at certain times of the day, and so on.) And if people had existed, they could have gathered and used that evidence. Maybe that's enough to make it true that Mt. Everest is still 8,850 m tall in the situation we're envisaging.
Things get tricky here. For instance, it's not clear that the verificationist is entitled to say that Mt. Everest would still cast those shadows, even if no observers had existed. But rather than pursuing these tricky details, let's instead think about examples where the relevant evidence isn't even available.
The usual varieties of verificationism say that for there to be a 8,850 m tall mountain, it has to be publicly verifiable that the mountain exists and is 8,850 m tall. That is, there has to be evidence that somebody somewhere could acquire that demonstrates that it is 8,850 m tall. A different version of the view would focus instead on what I myself am able to verify. This view might say that it's "true for me" that the mountain is 8,850 m tall only if I could verify that it's 8,850 m tall. It'd be "true for you" that it's 8,850 m tall only if you could verify that it's 8,850 m tall. And so on. We can call this second version of the view "personal verificationism," since it says that what's true—well, true for me—always depends on what I myself would be able to verify. If there's some fact that will forever be concealed from me, then it's not really a fact; at least, not a fact "for me." It may be a fact for other people, but that's a separate issue.
When professional philosophers discuss verificationism, they usually have the public version in mind. And the two versions do share many of the same features—and problems. However, I'm just going to talk about the personal version of the view. I think that people who aren't professional philosophers, like the students in my undergraduate classes, usually find the personal version more natural and attractive.
What does it mean to say that certain evidence is "available" or "unavailable"? One way of drawing this line would make it turn on whether you can obtain the evidence through your own active efforts: e.g., are there tests you can run that would give you the evidence you need? Or you might have a more liberal conception of what it is for evidence to be "available." On this more liberal conception, evidence will count as "available" even if it could just happen to fall into your lap, by chance. It doesn't have to be in your power to make the evidence appear.
Let's think about someone for whom evidence is unavailable even on this more liberal conception of "available." Suppose there's a character in The Matrix that it's impossible for Morpheus to "waken." Maybe this character believes in the "dream world" too strongly, and would just go insane and die if the "dream" ever started to unravel. Let's call this character Jeremy. According to the standard view, Jeremy has many false beliefs about his surroundings. He believes that he goes to work everyday on the 40th floor of an office building, that the sun streams into his office most mornings, that he often eats steak for dinner, and so on. All of these beliefs are false. In fact, there are no office buildings anymore; Jeremy has never seen the sun; he's never eaten steak; and he's spent his entire life in a small pod. But these are facts that Jeremy will never know. What's more, he's incapable of knowing them. If Morpheus told Jeremy the truth, Jeremy wouldn't believe him; and if Morpheus tried to show Jeremy the truth, Jeremy would go insane and die. So there are many truths about Jeremy's life that Jeremy will never be able to know.
That's what the standard view says. According to the verificationist, though, if it's impossible for Jeremy to know something, then that thing can't really be a "truth" about Jeremy's life. At least, it won't be a truth for Jeremy. What's true for Jeremy is that he really does work on the 40th floor of an office building, and so on. And this doesn't just mean that Jeremy thinks he works on the 40th floor etc. It means it really is a fact—a fact for Jeremy—that he works on the 40th floor of an office building. It may not be true for Morpheus that Jeremy works on the 40th floor of an office building, but it is true for Jeremy.
What do you think? Does that sound plausible to you?
Let's think about the comings and goings of people in the past. According to the standard view, on a given evening in the past, these people will either have been at a party in New York, or they won't have been there. Suppose they were there. But today only a little bit of evidence remains that they were there. Suppose you have it in your power to destroy that evidence, and manufacture evidence that they were elsewhere. Would you then have it in your power to change the past? That is what the character O'Brien in George Orwell's novel 1984 thinks:
An oblong slip of newspaper had appeared between O'Brien's fingers. For perhaps five seconds it was within the angle of Winston's vision... It was another copy of the photograph of Jones, Aaronson, and Rutherford at the party function in New York, which he had chanced upon eleven years ago and promptly destroyed. For only an instant it was before his eyes, then it was out of sight again...
"It exists!" he cried.
"No," said O'Brien.
He stepped across the room. There was a memory hole in the opposite wall. O'Brien lifted the grating. Unseen, the frail slip of paper was whirling away on the current of warm air; it was vanishing in a flash of flame. O'Brien turned away from the wall.
"Ashes," he said. "Not even identifiable ashes. Dust. It does not exist. It never existed."
"But it did exist! It does exist! It exists in memory. I remember it. You remember it."...
O'Brien was looking down at him speculatively. More than ever he had the air of a teacher taking pains with a wayward but promising child.
"There is a Party slogan dealing with the control of the past," he said. "Repeat it, if you please."
"'Who controls the past controls the future: who controls the present controls the past,'" repeated Winston obediently.
"'Who controls the present controls the past,'" said O'Brien, nodding his head with slow approval. "Is it your opinion, Winston, that the past has real existence?... Is there somewhere or other a place, a world of solid objects, where the past is still happening?"
"No."
"Then where does the past exist, if at all?"
"In records. It is written down."
"In records. And—?"
"In the mind. In human memories."
"In memory. Very well, then. We, the Party, control all records, and we control all memories. Then we control the past, do we not?"
Now, presumably O'Brien knows he's tampered with the evidence. So perhaps he can't change what's true for him about the past. But on the verificationist view, it does seem like he'd be able to change the past for other people.
What do you think? Does that sound plausible? Winston eventually comes to accept this view of reality. But to the reader it's supposed to sound like a lie.
What if the machines in The Matrix said to Neo and Morpheus, "Hey, why do you keep harping about this war between humans and machines? It never happened. At least, for all these people in their pods we're making it true that it never happened. Once we've removed every shred of evidence, and made it impossible for them to verify that there was a war between humans and machines, then we really will have changed the past for those people. They won't be deceived. Their past really will have happened the way it seems to them." Does that sound convincing? Or does it too sound like a lie?
What about facts for which there's simply no evidence either way? Morpheus says they don't know who struck first in the war between humans and machines. Maybe it's not important. And maybe the machines don't know either. Maybe all the evidence is lost. But presumably one of us did strike first. Presumably there is a fact about this, even if there's no evidence remaining. The verificationist has to deny this.
I hope all of this will make verificationism sound somewhat implausible to you. They aren't meant to be conclusive considerations. Philosophical discussions of verificationism get very complicated. The verificationist has to overcome many technical difficulties: e.g., how to draw the line between evidence that's available and evidence that's not. How to explain when evidence enables us to verify a hypothesis and when it doesn't. Whether verificationism itself is something we can verify. We can't go into these issues. If you're still inclined towards verificationism, I hope you'll at least grant that the view does go against our common-sense conception of reality, and that as a result it requires careful supporting argument. If you're going to hold the view in good intellectual conscience, there are a lot of difficulties and objections that need to be overcome.
III
I propose we set verificationism aside at this point; and see whether doing so helps us get any closer to determining what it is about the Matrix that makes it seem bad.
So now we'll say it is an objective fact whether you work on the 40th floor of an office building. We'll grant that it can seem to you in every respect that you're in "the good" situation (outside the Matrix), without it's thereby being true that you're in that situation.
OK. But this doesn't yet tell us why being inside the Matrix should be bad. Why is it important to really be in the situation we're calling "good"? Why isn't it good enough for us that we seem to be in the "good" situation? Isn't the experience or illusion of being in the good situation already pretty good? Why should it make our lives any better to really be there? (Especially if, as in the movie, the way the real "good" situation is is much less pleasant than the way things seem to be in the so-called "bad" situation.)
As Cypher says:
You know, I know that this steak doesn't exist. I know that when I put it in my mouth, the Matrix is telling my brain that it is juicy and delicious. After nine years, you know what I realize?... Ignorance is bliss.
Would it really make Cypher's life any better if he were really eating steak? Is it really eating steak that we value, or just the experience of eating steak? Wouldn't most people be satisfied with the experience—especially if it's indistinguishable from the real activity? Recall our friend Jeremy who spends his whole life inside the Matrix. How much is he missing out on, just because he never really gets to eat a steak? We're granting that there are truths about Jeremy's life that he'll never be able to know. But it's not obvious yet that any of them are truths he cares about. Perhaps the only things that Jeremy, and most of us, really care about are what kinds of experiences we're going to have, now and in the future. As Cypher recognizes, people who are stuck in the Matrix can still do pretty well by that score.
As we saw before, Nozick thinks that most of us wouldn't choose to spend the rest of our lives plugged into an "experience machine." He thinks there are things we value in life over and above what experiences we have. For instance, we value doing certain things, and not merely having the illusion or experience of doing them.
I agree with Nozick. For some matters, I think we genuinely do care about more than just what experiences we end up having. It would be implausible to claim this is always so. With regard to eating steak, the experience probably is all that we really value. But I think we feel differently about other matters. I'm going to try to persuade you that this is so, too.
Notice that what we're talking about here is the question: What do we actually value? Not the question: What should we value? Some readers may be willing to concede that we should care about more than our own experiences. (It's so selfish!) But it may appear that, as a matter of fact, our own experiences are all we really do care about—at least most of us. I'm going to argue that this isn't so. Most of us do in fact care about more than just what experiences we end up having.
There's a widely-held picture of human motivation that makes it difficult to see this. That picture goes like this. Ultimately, it says, everyone always acts for selfish motives. Whenever we do something on purpose, it's our own purpose that we're trying to achieve. We're always pursuing our own ends, and trying to satisfy our own desires. All that any of us are really after in life is getting more pleasant experiences for himself, and avoiding painful ones. Sometimes it may seem that we're doing things for other people's sake. For instance, we give money to charity, we buy presents for our children, we make sacrifices to please our spouses. But if you look closer, you'll see that even in cases like these, we're still always acting for selfish motives. We only do such things because it makes us feel good and noble to do them, and we like feeling noble. Or we do them because when people we care about are happy, that makes us happy too, and ultimately what we're after is that happiness for ourselves. Hence, since the only aim we have in life is just to have pleasant experiences, Nozick's experience machine gives us everything we want, and it would be foolish not to plug into it.
Now, I grant that some people may be as selfish as this picture says. But I doubt that many people are. The picture rests on two confusions, and once we clear those confusions up, I think there's no longer reason to believe that the only thing that any of us ever aims for in life is to have pleasant experiences.
The first confusion is to equate "pursuing our own ends, and trying to satisfy our own desires" with "acting for a selfish motive." To call a motive or aim "selfish" isn't just to say that it's a motive or aim that I have. It says more than that. It says something about the kind of motive it is. If my motive is to make me better off, then my motive is a selfish one. If my motive is to make you better off, then my motive is not selfish. From the mere fact that I'm pursuing one of my motives, it doesn't follow that my motive is of the first sort, rather than the second.
Ah, you'll say, but if my aim is to make you better off, then when I achieve that aim, I'll feel good. And this good feeling is really what I'll have been trying to obtain all along.
This is the second confusion. It's true that often when we get what we want (though sadly not always), we feel good. It's easy to make the mistake of thinking that what we really wanted was that good feeling. But let's think about this a bit harder. Why should making someone else better off give me a good feeling? And how do I know that it will have that effect?
Consider two stories. In story A, you go to visit the Oracle, and in her waiting room you see a boy bending spoons and a girl levitating blocks. You feel this inexplicable and unpleasant itch. Someone suggests as a hypothesis that the itch would go away if you gave the girl a spoon too. So you do so, and your itch goes away.
In story B, you walk into the same room, and you don't like the fact that the girl has no spoon. You would like her to have a spoon too. So you take a spoon and give it to the girl, and you feel pleased with the result.
In story A, your aim was to make yourself feel better, and giving the spoon to the girl was just a means to that end. It took experience and guesswork to figure out what would make you feel better in that way. In story B, on the other hand, no guesswork or experience seemed to be necessary. Here you were in a position to straightforwardly predict what would bring you pleasure. You could predict that because you had an aim other than making yourself feel better, you knew what that aim was, and usually you feel pleased when you get what you want. Your aim was to give a spoon to the girl. Your feeling of pleasure was a consequence or side-effect of achieving that aim. The pleasure is not what you were primarily aiming at; rather, it came about because you achieved what you were primarily aiming at. Don't mistake what you're aiming at with what happens as a result of your getting what you're aiming at.
Most often, when we do things to make other people better off, we're in a situation like the one in story B. Our pleasure isn't some unexplained effect of our actions, and what we're primarily trying to achieve. Our pleasure comes about because we got what we were primarily trying to achieve; and this makes it understandable why it should come about when it does.
Once we're straight about this, I think there's no argument left that the only thing anyone ever aims for in life is to have pleasant experiences. Some people do aim for that, some of the time. But many cases of giving to charity, making sacrifices for one's spouse, and so on, are not done for the pleasure they bring to oneself. There's something else that one is after, and pleasure is just a pleasant side-effect that sometimes comes along with getting the other things one is after.
Nozick said that most of us do value more than our own experiences, that there are things that we value that we'd miss out on if we plugged into the Matrix. I think Nozick is right. He's right about me, and he's probably right about you, too. We can easily find out. I've devised a little thought-experiment as a test.
Suppose I demonstrate to you that your friends and I are very good at keeping secrets. For instance, one day when Trinity isn't around, we all make lots of fun of her. We read her journal out loud and laugh really hard. We do ridiculous impersonations of her. And so on. It's hilarious. But of course we only do this behind Trinity's back. When she shows up, nobody giggles or snickers or anything like that. You're completely confident that we'll be able to keep our ridicule a secret from Trinity. She'll never know about it.
Suppose I also demonstrate to you that I am a powerful hypnotist. I can make people forget things, and once forgotten they never remember them. You're convinced that I have this power.
Now that you know all of that, I offer you a choice. Option 1 is I deposit $10 in your bank account, but then your friends and I will make fun of you behind your back, the way we made fun of Trinity. If you choose this option, then I will immediately use my hypnotic powers to make you forget about making the choice, being teased, and all that. From your point of view, it will seem that the bank made an error and now you have $10 more in your account than you had before. So in terms of what experiences you will have, this option has no downside. You won't even have to suffer from the expectation of being secretly teased, because I'll make you forget the whole arrangement as soon as you make your choice.
Option 2 is we keep things as they are. I pay you nothing, and your friends are no more or less likely to make fun of you behind your back than they were before.
So which would you choose?
When I offer my students this choice, I find that at least 95% of them choose Option 2. They think that the teasing would be a bad thing, even though they'd never know it was going on.
If the teasing doesn't seem so bad, then change the example. Say that in Option 1, your lover is cheating on you, but you never know about it. Or say that we're torturing your mother, but you never know about it. In every version, your experiences are smooth and untroubled, plus you get a little extra money. Which option would you choose?
If you find Option 2 more attractive, then that's support for Nozick's claim. The experience machine wouldn't give you everything you value. Option 1 gives you no experiences of being teased. It gives you no evidence that your lover is cheating on you, or that your mother is being tortured. But you don't just want to have experiences of things going well for yourself and your mother. You value really not being teased, really having a faithful lover, and really having an untortured mother.
Now, we do have to compare what we'd get by plugging into the experience machine to what we'd get if we don't plug in. I've only been arguing that we'd miss out on some things we'd value if we plugged in. I haven't said that it would never be reasonable to plug in. In some cases, the good of being plugged in could outweigh the bad. If the real world is miserable and nasty enough, it may make sense to plug in. Perhaps for Cypher, the real world is too nasty. All I'm saying is that plugging in won't give us everything we want. Our experiences aren't all that we value. So there is some bad to plugging in. There may also be some good to plugging in. Dreams and immersive role-playing do give us some of the things we value in life. I'm just saying they don't give us everything. Some aspects of how the world really is are important to us.
I haven't been able to say yet how important, though. It's hard to know what the right balance point is. How bad does the real world have to be, before it makes sense to make Cypher's choice, and plug back into the blissful experience machine? This is a hard question. In part, it will depend on whether the Matrix or the experience machine involve any hidden costs. And this is something we haven't yet settled.
IV
Before we can determine what are the major costs of living inside the Matrix, we have to confront one last complication.
We said that for most people inside the Matrix, the experience of eating steak may be enough. We said they probably don't care about whether they've ever really eaten steak. Let's pause over this for a moment. What do these characters mean by "eating steak"?
Suppose you grew up with a friend you called "Jiro." You didn't realize it, but that isn't really your friend's name, at least not the name his parents gave him. His name is really "Takeshi." "Jiro" is his uncle's name. But you got the names mixed up when you were little, and no one bothered to correct you. So all your life you've been saying "Jiro" to talk about Takeshi. Isn't it plausible then that in your mouth, "Jiro" now means Takeshi?
Similarly, Jeremy has grown up inside the Matrix program, and on various occasions he's interacted in certain ways with other parts of the Matrix program, ways he described as "eating steak." Now perhaps all he means by "eating steak" is just interacting in those certain way with the Matrix. He's done that many times. So perhaps he really has managed to eat steak on many occasions. At least, he's managed to do what he calls "eating steak." It's not clear that there's anything more that Jeremy would like to be doing, but isn't. Is there?
The philosophical issues here are fascinating, but they get complicated really fast. I myself think that for some of Jeremy's concepts, the story we just sketched may be right.
Interestingly, this doesn't seem to be the movie's own attitude. Recall what Cypher says:
You know, I know that this steak doesn't exist.
And when Morpheus and Neo are fighting in the sparring program, Morpheus asks:
Do you think that's air you're breathing?
Cypher and Morpheus are both rejecting the view that the Matrix simulations really provide what they mean by "steak" and "air." That is, they're rejecting the view that all they mean by "steak" and "air" is interacting in certain ways with the Matrix program.
As I said, the philosophical issues here can get really complicated. One way to avoid these difficulties is to concentrate on what would be bad about living in the Matrix for the first generation of Matrix inductees: people who grew up outside the Matrix, and have just been freshly plugged in. Presumably what they mean by "eating steak" has to do with cow flesh, not with patterns in the Matrix simulation. Presumably what they mean by "air" is made up of nitrogen and oxygen, not 1s and 0s.
I want to try a different strategy. We can suppose we're talking about people who have spent all their lives so far inside the Matrix. I want to try to find something we value that goes beyond what experiences we're having, and where we can agree that the people inside the Matrix really would value that same thing. They wouldn't just value having some Matrix substitute. And yet this will be something that people inside the Matrix don't have. They only seem to have it.
If we can find something like that, then we'll have found something that really does deserve the name of "what's bad about living in the Matrix."
I can think of three possibilities.
The first has to do with certain kinds of scientific knowledge. I'd guess that physicists in the Matrix have some fundamentally false beliefs about the underlying make-up of their world, what the "laws of nature" are, and so on. For some people, figuring such matters out is important. They value learning the truth about those matters. But not everybody feels that way. For your average non-physicist, the possibility that we're mistaken about questions like these isn't going to provoke existential anxiety, or set them off on a crusade like the one Neo undertakes at the end of the movie.
The second candidate for being what's bad about living in the Matrix has to do with interpersonal relationships. One thing we place a lot of value on in life are our interactions with other people. Most of us want our friends' feelings to be genuine. For instance, it would be bad if the person who acts like your best friend really despises you. Even if you never found out about it. Most of us also want the important people in our lives to be real. We don't want them to be programming constructs, like Mouse's "Woman in Red." Perhaps for some people, programming constructs are enough. They may not care whether their friends and lovers really have an inner life of their own, and have their own thoughts and emotions, and genuine feelings towards them. It would be enough if their friends and lovers acted the their parts well. I think that for most of us, though, this would not be enough. Most of us really would like to have the real thing. It would suck if the children you devote so much love and attention to are really just parts of a computer program, and don't have any capacity to benefit from, or to appreciate, your efforts.
Here's another thought-experiment. Suppose that tomorrow we're going to wipe your memory clean and ship you off to a new colony. You'll be able to live a decent life there; you just won't have any memory of your past. Nor do you get to take any of your money or personal belongings along with you. But today, before we wipe your memory clean, we allow you to spend the money you have left to arrange a nice life for yourself in the new colony. For instance, if you spend $1,000 we'll set it up so that the apartment you get there doesn't have cockroaches. And so on. How would you spend your money?
What if there were two options on the menu. If you choose Option 1, you'll get an extremely realistic set of friends and lovers in the new colony. You won't be able to distinguish them from the real thing. But really they'll just be empty shells animated by a (non-intelligent) computer program. They won't have any inner life of their own. (In the terminology of role-playing games, they'll be NPCs.) You know this now, but when you get to the colony you will have forgotten it. If you choose Option 2, you get friends and lovers who are real people.
Most people I know would choose Option 2, even if it were somewhat more expensive, and so kept them from buying other nice things for their new life. E.g., they'd choose Option 2 even if it meant they'd have to put up with more cockroaches.
So one thing that many of us value in life is that the other people we form emotional attachments to are real people, and that they care about us in the ways they seem to. In Nozick's experience machine, this seems to be lacking. His experience machine sounds like a one-person Matrix. You just get to enjoy your own experience script. You don't get to interact with other people. (See the discussion of "solitary Matrices" in Richard Hanley's essay.)
In the real Matrix, on the other hand, it seems like people do get to interact with many other real human beings. So a lack of interpersonal relationships may be a bad thing about Nozick's machine, but it doesn't seem to be a bad thing about the Matrix we see in the movie.
I think our third candidate for what's bad about living in the Matrix is more apt. In the movie, humans in the Matrix are all slaves. They're not in charge of their own lives. They may be contented slaves, unaware of their chains, but they're slaves nonetheless. They have only a very limited ability to shape their own futures. As Morpheus puts it:
What is the Matrix? Control. The Matrix is a computer-generated dream world, built to keep us under control...
Now—to me anyway—the most disturbing thing about this isn't that the machines are farming us for energy. We're not told enough about how the energy-farming works to make it seem very bad. Perhaps the machines are only taking energy we were making no use of, anyway. Perhaps the machines ensure that—except for the rare occasions when an Agent takes over your body and gets it killed—we live longer and healthier lives in the Matrix energy-farm than we would in the wild.
No, what seems awful about our enslavement in the Matrix is rather that we have so little control over our futures. According to Agent Smith, the Matrix was designed to simulate the end of the 20th century, because the machines have found that keeps their energy-farm running smoothly. Generations of us have now lived out their lives in the Matrix. So generations of us have all experienced life in this simulated end-of-the-20th-century. What happens when the simulation gets up to 2003? Do the machines erase our memories and reset everything back to 1980? The movie doesn't say. But presumably they do something like that. This means there are real limits to how much we can accomplish. If your ambitions in the Matrix are relatively small-scale, like opening a restaurant or becoming a famous actor, then you may very well be able to achieve them. But if your ambitions are larger—e.g., introducing some long-term social change—then whatever progress you make towards that goal will be wiped out when the simulation gets reset. Any long-term efforts of this sort would be an exercise an futility.
And what if our ambitions don't please the farmers? For instance, what if we are computer scientists working to create artificial intelligence? The machines would probably find it easiest to just keep sabotaging our attempts. After all, they wouldn't want us to re-enact the war between humans and machines, inside the Matrix. That would be bad for their crops. And they certainly wouldn't want us to create benevolent AIs, AIs who would figure out about the Matrix and fight on our side. So the machines will tinker with our history, and see to it that grand, noble ambitions of this sort never get realized.
Of course, they'll also see to it that none of our grander baser ambitions get realized, either. They probably just disconnect or reprogram anyone who's hatching plans for mass genocide.
But if given the choice, I think most of us would like humans to be in charge of our own destiny. We don't want our long-term efforts to be futile. We don't want to be living out someone else's plan for our lives. Sure, there will always be some limits to what we can do. Very likely we'll never be able to vacation in the center of the sun. But we'd like to have as much control over our destiny as we can. We don't want other intelligent agents deciding such things for us. Especially when those agents' first priority is how well their energy-farms are doing; that may not correlate well with how well-off our lives and society are.
So it seems rotten if the machines control our fates and our civilization. One thing we place a lot of value on is being in charge of our own lives, not being someone else's slave or plaything. We want to be politically free.
And plausibly, what people mean by "political freedom" and "being in charge of our own lives" is the same inside the Matrix as outside it. We're not indifferent between the real thing and some Matrix simulation of it. We want to have the real thing. When we're inside the Matrix, we haven't got it. We just don't realize that we haven't got it.
So I think this is the best answer about what's so bad about living in the Matrix.
For me, at least, it's a surprising answer. The Matrix raises so many interesting metaphysical and epistemological issues. If you're of a philosophical bent, like me, then those issues will be intellectually compelling. But there's a difference between what we find intellectually compelling and what we place the most value on in life. Intellectual matters will be only one value among many. For most of us, the worst thing about living in the Matrix would not be something metaphysical or epistemological. Rather, the worst thing would be something political. It would be the fact that Life in the Matrix is a kind of Slavery, of the sort of we've discussed.
I think that is what's really bad about living in the Matrix we see in the movie. That is what motivates Neo and Morpheus and Trinity to fight the machines, and try to free everyone they can.
If the Matrix weren't a kind of enslavement—and it still involved interacting with other real people—then maybe it wouldn't be so bad after all.
James Pryor
Return to topI. Brains in Vats
The Matrix presents a version of an old philosophical fable: the brain in a vat. A disembodied brain is floating in a vat, inside a scientist's laboratory. The scientist has arranged that the brain will be stimulated with the same sort of inputs that a normal embodied brain receives. To do this, the brain is connected to a giant computer simulation of a world. The simulation determines which inputs the brain receives. When the brain produces outputs, these are fed back into the simulation. The internal state of the brain is just like that of a normal brain, despite the fact that it lacks a body. From the brain's point of view, things seem very much as they seem to you and me.
The brain is massively deluded, it seems. It has all sorts of false beliefs about the world. It believes that it has a body, but it has no body. It believes that it is walking outside in the sunlight, but in fact it is inside a dark lab. It believes it is one place, when in fact it may be somewhere quite different. Perhaps it thinks it is in Tucson, when it is actually in Australia, or even in outer space.
Neo's situation at the beginning of The Matrix is something like this. He thinks that he lives in a city, he thinks that he has hair, he thinks it is 1999, and he thinks that it is sunny outside. In reality, he is floating in space, he has no hair, the year is around 2199, and the world has been darkened by war. There are a few small differences from the vat scenario above: Neo's brain is located in a body, and the computer simulation is controlled by machines rather than by a scientist. But the essential details are much the same. In effect, Neo is a brain in a vat.
Let's say that a matrix (lower-case "m") is an artificially-designed computer simulation of a world. So the Matrix in the movie is one example of a matrix. And let's say that someone is envatted, or that they are in a matrix, if they have a cognitive system which receives its inputs from and sends its outputs to a matrix. Then the brain at the beginning is envatted, and so is Neo.
We can imagine that a matrix simulates the entire physics of a world, keeping track of every last particle throughout space and time. (Later, we will look at ways in which this set-up might be varied.) An envatted being will be associated with a particular simulated body. A connection is arranged so that whenever this body receives sensory inputs inside the simulation, the envatted cognitive system will receive sensory inputs of the same sort. When the envatted cognitive system produces motor outputs, corresponding outputs will be fed to the motor organs of the simulated body.
When the possibility of a matrix is raised, a question immediately follows. How do I know that I am not in a matrix? After all, there could be a brain in a vat structured exactly like my brain, hooked up to a matrix, with experiences indistinguishable from those I am having now. From the inside, there is no way to tell for sure that I am not in the situation of the brain in a vat. So it seems that there is no way to know for sure that I am not in a matrix.
Let us call the hypothesis that I am in a matrix and have always been in a matrix the Matrix Hypothesis. Equivalently, the Matrix Hypothesis says that I am envatted and have always been envatted. This is not quite equivalent to the hypothesis that I am in the Matrix, as the Matrix is just one specific version of a matrix. And for now, I will ignore the complication that people sometimes travel back and forth between the Matrix and the external world. These issues aside, we can think of the Matrix Hypothesis informally as saying that I am in the same sort of situation as people who have always been in the Matrix.
The Matrix Hypothesis is one that we should take seriously. As Nick Bostrom has suggested, it is not out of the question that in the history of the universe, technology will evolve that will allow beings to create computer simulations of entire worlds. There may well be vast numbers of such computer simulations, compared to just one real world. If so, there may well be many more beings who are in a matrix than beings who are not. Given all this, one might even infer that it is more likely that we are in a matrix than that we are not. Whether this is right or not, it certainly seems that we cannot be certain that we are not in a matrix.
Serious consequences seem to follow. My envatted counterpart seems to be massively deluded. It thinks it is in Tucson; it thinks it is sitting at a desk writing an article; it thinks it has a body. But on the face of it, all of these beliefs are false. Likewise, it seems that if I am envatted, my own corresponding beliefs are false. If I am envatted, I am not really in Tucson, I am not really sitting at a desk, and I may not even have a body. So if I don't know that I am not envatted, then I don't know that I am in Tucson, I don't know that I am sitting at a desk, and I don't know that I have a body.
The Matrix Hypothesis threatens to undercut almost everything I know. It seems to be a skeptical hypothesis: a hypothesis that I cannot rule out, and one that would falsify most of my beliefs if it were true. Where there is a skeptical hypothesis, it looks like none of these beliefs count as genuine knowledge. Of course the beliefs might be true — I might be lucky, and not be envatted — but I can't rule out the possibility that they are false. So a skeptical hypothesis leads to skepticism about these beliefs: I believe these things, but I do not know them.
To sum up the reasoning: I don't know that I'm not in a matrix. If I'm in a matrix, I'm probably not in Tucson. So if I don't know that I'm not in a matrix, then I don't know that I'm in Tucson. The same goes for almost everything else I think I know about the external world.
II. Envatment Reconsidered
This is a standard way of thinking about the vat scenario. It seems that this view is also endorsed by the people who created The Matrix . On the DVD case for the movie, one sees the following:
Perception: Our day-in, day-out world is real.
Reality: That world is a hoax, an elaborate deception spun by all-powerful machines that control us. Whoa.
I think this view is not quite right. I think that even if I am in a matrix, my world is perfectly real. A brain in a vat is not massively deluded (at least if it has always been in the vat). Neo does not have massively false beliefs about the external world. Instead, envatted beings have largely correct beliefs about their world. If so, the Matrix Hypothesis is not a skeptical hypothesis, and its possibility does not undercut everything that I think I know.
Philosophers have held this sort of view before. The 18th-century Irish philosopher George Berkeley held, in effect, that appearance is reality. (Recall Morpheus: "What is real? How do you define real? If you're talking about what you can feel, what you can smell, what you can taste and see, then real is simply electrical signals interpreted by your brain.") If this is right, then the world perceived by envatted beings is perfectly real: they have all the right appearances, and appearance is reality. So on this view, even envatted beings have true beliefs about the world.
I have recently found myself embracing a similar conclusion, though for quite different reasons. I don't find the view that appearance is reality plausible, so I don't endorse Berkeley's reasoning. And until recently, it has seemed quite obvious to me that brains in vats would have massively false beliefs. But I now think there is a line of reasoning that shows that this is wrong.
I still think I cannot rule out the hypothesis that I am in a matrix. But I think that even I am in a matrix, I am still in Tucson, I am still sitting at my desk, and so on. So the hypothesis that I am in a matrix is not a skeptical hypothesis. The same goes for Neo. At the beginning of the film, if he thinks "I have hair", he is correct. If he thinks "It is sunny outside", he is correct. And the same goes, of course, for the original brain in a vat. When it thinks "I have a body", it is correct. When it thinks "I am walking", it is correct.
This view may seem very counterintuitive at first. Initially, it seemed quite counterintuitive to me. So I'll now present the line of reasoning that has convinced me that it is correct.
III. The Metaphysical Hypothesis
I will argue that the hypothesis that I am envatted is not a skeptical hypothesis, but a metaphysical hypothesis. That is, it is a hypothesis about the underlying nature of reality.
Where physics is concerned with the microscopic processes that underlie macroscopic reality, metaphysics is concerned with the fundamental nature of reality. A metaphysical hypothesis might make a claim about the reality that underlies physics itself. Alternatively, it might say something about the nature of our minds, or the creation of our world.
I think the Matrix Hypothesis should be regarded as a metaphysical hypothesis with all three of these elements. It makes a claim about the reality underlying physics, about the nature of our minds, and about the creation of the world.
In particular, I think the Matrix Hypothesis is equivalent to a version of the following three-part Metaphysical Hypothesis. First, physical processes are fundamentally computational. Second, our cognitive systems are separate from physical processes, but interact with these processes. Third, physical reality was created by beings outside physical space-time.
Importantly, nothing about this Metaphysical Hypothesis is skeptical. The Metaphysical Hypothesis here tells us about the processes underlying our ordinary reality, but it does not entail that this reality does not exist. We still have bodies, and there are still chairs and tables: it's just that their fundamental nature is a bit different from what we may have thought. In this manner, the Metaphysical Hypothesis is analogous to a physical hypotheses, such as one involving quantum mechanics. Both the physical hypothesis and the Metaphysical Hypothesis tells us about the processes underlying chairs. They do not entail that there are no chairs. Rather, they tell us what chairs are really like.
I will make the case by introducing each of the three parts of the Metaphysical Hypothesis separately. I will suggest that each of them is coherent, and cannot be conclusively ruled out. And I will suggest that none of them is a skeptical hypothesis: even if they are true, most of our ordinary beliefs are still correct. The same goes for a combination of all three hypothesis. I will then argue that the Matrix Hypothesis hypothesis is equivalent to this combination.
(1) The Creation Hypothesis
The Creation Hypothesis says: Physical space-time and its contents were created by beings outside physical space-time.
This is a familiar hypothesis. A version of it is believed by many people in our society, and perhaps by the majority of the people in the world. If one believes that God created the world, and if one believes that God is outside physical space-time, then one believes the Creation Hypothesis. One needn't believe in God to believe the Creation Hypothesis, though. Perhaps our world was created by a relatively ordinary being in the "next universe up", using the latest world-making technology in that universe. If so, the Creation Hypothesis is true.
I don't know whether the Creation Hypothesis is true. But I don't know for certain that it is false. The hypothesis is clearly coherent, and I cannot conclusively rule it out.
The Creation Hypothesis is not a skeptical hypothesis. Even if it is true, most of my ordinary beliefs are still true. I still have hands, I am still in Tucson, and so on. Perhaps a few of my beliefs will turn out false: if I am an atheist, for example, or if I believe all reality started with the Big Bang. But most of my everyday beliefs about the external world will remain intact.
(2) The Computational Hypothesis
The Computational Hypothesis says: Microphysical processes throughout space-time are constituted by underlying computational processes.
The Computational Hypothesis says that physics as we know it not the fundamental level of reality. Just as chemical processes underlie biological processes, and microphysical processes underlie chemical processes, something underlies microphysical processes. Underneath the level of quarks and electrons and photons is a further level: the level of bits. These bits are governed by a computational algorithm, which at a higher-level produces the processes that we think of as fundamental particles, forces, and so on.
The Computational Hypothesis is not as widely believed as the Creation Hypothesis, but some people take it seriously. Most famously, Ed Fredkin has postulated that the universe is at bottom some sort of computer. More recently, Stephen Wolfram has taken up the idea in his book A New Kind of Science, suggesting that at the fundamental level, physical reality may be a sort of cellular automata, with interacting bits governed by simple rules. And some physicists have looked into the possibility that the laws of physics might be formulated computationally, or could be seen as the consequence of certain computational principles.
One might worry that pure bits could not be the fundamental level of reality: a bit is just a 0 or a 1, and reality can't really be zeroes and ones. Or perhaps a bit is just a "pure difference" between two basic states, and there can't be a reality made up of pure differences. Rather, bits always have to be implemented by more basic states, such as voltages in a normal computer.
I don't know whether this objection is right. I don't think it's completely out of the question that there could be a universe of "pure bits". But this doesn't matter for present purposes. We can suppose that the computational level is itself constituted by an even more fundamental level, at which the computational processes are implemented. It doesn't matter for present purposes what that more fundamental level is. All that matters is that microphysical processes are constituted by computational processes, which are themselves constituted by more basic processes. From now on I will regard the Computational Hypothesis as saying this.
I don't know whether the Computational Hypothesis is correct. But again, I don't know that it is false. The hypothesis is coherent, if speculative, and I cannot conclusively rule it out.
The Computational Hypothesis is not a skeptical hypothesis. If it is true, there are still electrons and protons. On this picture, electrons and protons will be analogous to molecules: they are made up of something more basic, but they still exist. Similarly, if the Computational Hypothesis is true, there are still tables and chairs, and macroscopic reality still exists. It just turns out that their fundamental reality is a little different from what we thought.
The situation here is analogous to that with quantum mechanics or relativity. These may lead us to revise a few "metaphysical" beliefs about the external world: that the world is made of classical particles, or that there is absolute time. But most of our ordinary beliefs are left intact. Likewise, accepting the Computational Hypothesis may lead us to revise a few metaphysical beliefs: that electrons and protons are fundamental, for example. But most of our ordinary beliefs are unaffected.
(3) The Mind-Body Hypothesis
The Mind-Body Hypothesis says: My mind is (and has always been) constituted by processes outside physical space-time, and receives its perceptual inputs from and sends its outputs to processes in physical space-time.
The Mind-Body Hypothesis is also quite familiar, and quite widely believed. Descartes believed something like this: on his view, we have nonphysical minds that interact with our physical bodies. The hypothesis is less widely believed today than in Descartes' time, but there are still many people who accept the Mind-Body Hypothesis.
Whether or not the Mind-Body Hypothesis is true, it is certainly coherent. Even if contemporary science tends to suggest that the hypothesis is false, we cannot rule it out conclusively.
The Mind-Body Hypothesis is not a skeptical hypothesis. Even if my mind is outside physical space-time, I still have a body, I am still in Tucson, and so on. At most, accepting this hypothesis would make us revise a few metaphysical beliefs about our minds. Our ordinary beliefs about external reality will remain largely intact.
(4) The Metaphysical Hypothesis
We can now put these hypotheses together. First we can consider the Combination Hypothesis, which combines all three. It says that physical space-time and its contents were created by beings outside physical space-time, that microphysical processes are constituted by computational processes, and that our minds are outside physical space-time but interact with it.
As with the hypotheses taken individually, the Combination Hypothesis is coherent, and we cannot conclusively rule it out. And like the hypotheses taken individually, it is not a skeptical hypothesis. Accepting it might lead us to revise a few of our beliefs, but it would leave most of them intact.
Finally, we can consider the Metaphysical Hypothesis (with a capital M). Like the Combination Hypothesis, this combines the Creation Hypothesis, the Computational Hypothesis, and the Mind-Body Hypothesis. It also adds the following more specific claim: the computational processes underlying physical space-time were designed by the creators as a computer simulation of a world.
(It may also be useful to think of the Metaphysical Hypothesis as saying that the computational processes constituting physical space-time are part of a broader domain, and that the creators and my cognitive system are also located within this domain. This addition is not strictly necessary for what follows, but it matches up with the most common way of thinking about the Matrix Hypothesis.)
The Metaphysical Hypothesis is a slightly more specific version of the Combination Hypothesis, in that in specifies some relations between the various parts of the hypothesis. Again, the Metaphysical Hypothesis is a coherent hypothesis, and we cannot conclusively rule it out. And again, it is not a skeptical hypothesis. Even if we accept it, most of our ordinary beliefs about the external world will be left intact.
IV. The Matrix Hypothesis as a Metaphysical Hypothesis
Recall that the Matrix Hypothesis says: I have (and have always had) a cognitive system that receives its inputs from and sends its outputs to an artificially-designed computer simulation of a world.
I will argue that the Matrix Hypothesis is equivalent to the Metaphysical Hypothesis, in the following sense: if I accept the Metaphysical Hypothesis, I should accept the Matrix Hypothesis, and if I accept the Matrix Hypothesis, I should accept the Metaphysical Hypothesis. That is, the two hypotheses imply each other, where this means that if one accepts the one, one should accept the other.
Take the first direction first, from the Metaphysical Hypothesis to the Matrix Hypothesis. The Mind-Body Hypothesis implies that I have (and have always had) an isolated cognitive system which receives its inputs from and sends its outputs to processes in physical space-time. In conjunction with the Computational Hypothesis, this implies that my cognitive system receives inputs from and sends outputs to the computational processes that constitute physical space-time. The Creation Hypothesis (along with the rest of the Metaphysical Hypothesis) implies that these processes were artificially designed to simulate a world. It follows that I have (and have always had) an isolated cognitive system that receives its inputs from and sends its outputs to an artificially-designed computer simulation of a world. This is just the Matrix Hypothesis. So the Metaphysical Hypothesis implies the Matrix Hypothesis.
The other direction is closely related. To put it informally: If I accept the Matrix Hypothesis, I accept that what underlies apparent reality is just as the Metaphysical Hypothesis specifies. There is a domain containing my cognitive system, causally interacting with a computer simulation of physical-space time, which was created by other beings in that domain. This is just what has to obtain in order for the Metaphysical Hypothesis to obtain. If one accepts this, one should accept the Creation Hypothesis, the Computational Hypothesis, the Mind-Body Hypothesis, and the relevant relations among these.
This may be a little clearer through a picture. Here is the shape of the world according to the Matrix Hypothesis.
At the fundamental level, this picture of the shape of the world is exactly the same as the picture of the Metaphysical Hypothesis given above. So if one accepts that the world is as it is according to the Matrix Hypothesis, one should accept that it is as it is according to the Metaphysical Hypothesis.
One might make various objections. For example, one might object that the Matrix Hypothesis implies that a computer simulation of physical processes exists, but (unlike the Metaphysical Hypothesis) it does not imply that the physical processes themselves exist. I will discuss this and other objections in later sections. For now, though, I take it that there is a strong case that the Matrix Hypothesis implies the Metaphysical Hypothesis, and vice versa.
V. Life in the Matrix
If this is right, it follows that the Matrix Hypothesis is not a skeptical hypothesis. If I accept it, I should not infer that the external world does not exist, or that I have no body, or that there are no tables and chairs, or that I am not in Tucson. Rather, I should infer that the physical world is constituted by computations beneath the microphysical level. There are still tables, chairs, and bodies: these are made up fundamentally of bits, and of whatever constitutes these bits. This world was created by other beings, but is still perfectly real. My mind is separate from physical processes, and interacts with them. My mind may not have been created by these beings, and it may not be made up of bits, but it still interacts with these bits.
The result is a complex picture of the fundamental nature of reality. The picture is strange and surprising, perhaps, but it is a picture of a full-blooded external world. If we are in a matrix, this is simply the way that the world is.
We can think of the Matrix Hypothesis as a creation myth for the information age. If it is correct, then the physical world was created, just not necessarily by gods. Underlying the physical world is a giant computation, and creators created this world by implementing this computation. And our minds lie outside this physical structure, with an independent nature that interacts with this structure.
Many of the same issues that arise with standard creation myths arise here. When was the world created? Strictly speaking, it was not created within our time at all. When did history begin? The creators might have started the simulation in 4004 BC (or in 1999) with the fossil record intact, but it would have been much easier for them to start the simulation at the Big Bang and let things run their course from there. When do our nonphysical minds start to exist? It depends on just when new envatted cognitive systems are attached to the simulation (perhaps at the time of conception within the matrix, or perhaps at time of birth?). Is there life after death? It depends on just what happens to the envatted systems once their simulated bodies die. How do mind and body interact? By causal links that are outside physical space and time.
Even if we not in a matrix, we can extend a version of this reasoning to other beings who are in a matrix. If they discover their situation, and come to accept that they are in a matrix, they should not reject their ordinary beliefs about the external world. At most, they should come to revise their beliefs about the underlying nature of their world: they should come to accept that external objects are made of bits, and so on. These beings are not massively deluded: most of their ordinary beliefs about their world are correct.
There are a few qualifications here. One may worry about beliefs about other people's minds. I believe that my friends are conscious. If I am in a matrix, is this correct? In the Matrix depicted in the movie, these beliefs are mostly fine. This is a multi-vat matrix: for each of my perceived friends, there is an envatted being in the external reality, who is presumably conscious like me. The exception might be beings such as Agent Smith, who are not envatted, but are entirely computational. Whether these beings are conscious depends on whether computation is enough for consciousness. I will remain neutral on that issue here. We could circumvent this issue by building into the Matrix Hypothesis the requirement that all the beings we perceive are envatted. But even if we do not build in this requirement, we are not much worse off than in the actual world, where there is a legitimate issue about whether other beings are conscious, quite independently of whether we are in a matrix.
One might also worry about beliefs about the distant past, and about the far future. These will be unthreatened as long as the computer simulation covers all of space-time, from the Big Bang until the end of the universe. This is built into the Metaphysical Hypothesis, and we can stipulate that it is built into the Matrix Hypothesis too, by requiring that the computer simulation be a simulation of an entire world. There may be other simulations that start in the recent past (perhaps the Matrix in the movie is like this), and there may be others that only last for a short while. In these cases, the envatted beings will have false beliefs about the past and/or the future in their worlds. But as long as the simulation covers the lifespan of these beings, it is plausible that they will have mostly correct beliefs about the current state of their environment.
There may be some respects in which the beings in a matrix are deceived. It may be that the creators of the matrix control and interfere with much of what happens in the simulated world. (The Matrix in the movie may be like this, though the extent of the creators' control is not quite clear.) If so, then these beings may have much less control over what happens than they think. But the same goes if there is an interfering god in a non-matrix world. And the Matrix Hypothesis does not imply that the creators interfere with the world, though it leaves the possibility open. At worst, the Matrix Hypothesis is no more skeptical in this respect than the Creation Hypothesis in a non-matrix world.
The inhabitants of a matrix may also be deceived in that reality is much bigger than they think. They might think their physical universe is all there is, when in fact there is much more in the world, including beings and objects that they can never possibly see. But again, this sort of worry can arise equally in a non-matrix world. For example, cosmologists seriously entertain the hypothesis that our universe may stem from a black hole in the "next universe up", and that in reality there may be a whole tree of universes. If so, the world is also much bigger than we think, and there may be beings and objects that we can never possibly see. But either way, the world that we see is perfectly real.
Importantly, none of these sources of skepticism — about other minds, the past and the future, about our control over the world, and about the extent of the world — casts doubt on our belief in the reality of the world that we perceive. None of them leads us to doubt the existence of external objects such as tables and chairs, in the way that the vat hypothesis is supposed to do. And none of these worries is especially tied to the matrix scenario. One can raise doubts about whether other minds exist, whether the past and the future exist, and whether we have control over our worlds quite independently of whether we are in a matrix. If this is right, then the Matrix Hypothesis does not raise the distinctive skeptical issues that it is often taken to raise.
I suggested before that it is not out of the question that we really are in a matrix. One might have thought that this is a worrying conclusion. But if I am right, it is not nearly as worrying as one might have thought. Even if we are in such a matrix, our world is no less real than we thought it was. It just has a surprising fundamental nature.
VI. Objection: Simulation is not Reality
(This slightly technical section can be skipped without too much loss.)
A common line of objection is that a simulation is not the same as reality. The Matrix Hypothesis implies only that a simulation of physical processes exists. By contrast, the Metaphysical Hypothesis implies that physical processes really exist (they are explicitly mentioned in the Computational Hypothesis and elsewhere). If so, then the Matrix Hypothesis cannot imply the Metaphysical Hypothesis. On this view, if I am in a matrix, then physical processes do not really exist.
In response: My argument does not require the general assumption that simulation is the same as reality. The argument works quite differently. But the objection helps us to flesh out the informal argument that the Matrix Hypothesis implies the Metaphysical Hypothesis.
Because the Computational Hypothesis is coherent, it is clearly possible that a computational level underlies real physical processes, and it is possible that the computations here are implemented by further processes in turn. So there is some sort of computational system that could yield reality here. But here, the objector will hold that not all computational systems are created equal. To say that some computational systems will yield real physical processes in this role is not to say that they all do. Perhaps some of them are merely simulations. If so, then the Matrix Hypothesis may not yield reality.
To rebut this objection, we can appeal to two principles. First, any abstract computation that could be used to simulate physical space-time is such that it could turn out to underlie real physical processes. Second, given an abstract computation that could underlie physical processes, the precise way in which it is implemented is irrelevant to whether it does underlie physical processes. In particular, the fact that the implementation was designed as a simulation is irrelevant. The conclusion then follows directly.
On the first point: let us think of abstract computations in purely formal terms, abstracting away from their manner of implementation. For an abstract computation to qualify as a simulation of physical reality, it must have computational elements that correspond to every particle in reality (likewise for fields, waves, or whatever is fundamental), dynamically evolving in a way that corresponds to the particle's evolution. But then, it is guaranteed that the computation will have a rich enough causal structure that it could in principle underlie physics in our world. Any computation will do, as long as it has enough detail to correspond to the fine details of physical processes.
On the second point: given an abstract computation that could underlie physical reality, it does not matter how the computation is implemented. We can imagine discovering that some computational level underlies the level of atoms and electrons. Once we have discovered this, it is possible that this computational level is implemented by more basic processes. There are many hypotheses about what the underlying processes could be, but none of them is especially privileged, and none of them would lead us to reject the hypothesis that the computational level constitutes physical processes. That is, the Computational Hypothesis is implementation-independent: as long as we have the right sort of abstract computation, the manner of implementation does not matter.
In particular, it is irrelevant whether or not these implementing processes were artificially created, and it is irrelevant whether they were intended as a simulation. What matters is the intrinsic nature of the processes, not their origin. And what matters about this intrinsic nature is simply that they are arranged in such a way to implement the right sort of computation. If so, the fact that the implementation originated as a simulation is irrelevant to whether it can constitute physical reality.
There is one further constraint on the implementing processes: they must be connected to our experiences in the right sort of way. That is when we have an experience of an object, the processes underlying the simulation of that object must be causally connected in the right sort of way to our experiences. If this is not the case, then there will be no reason to think that these computational processes underlie the physical processes that we perceive. If there is an isolated computer simulation to which nobody is connected in this way, we should say that it is simply a simulation. But an appropriate hook-up to our perceptual experiences is built into the Matrix Hypothesis, on the most natural understanding of that hypothesis. So the Matrix Hypothesis has no problems here.
Overall, then, we have seen that a computational process could underlie physical reality, that any abstract computation that qualifies as a simulation of physical reality could play this role, and that any implementation of this computation could constitute physical reality, as long as it is hooked up to our experiences in the relevant way. The Matrix Hypothesis guarantees that we have an abstract computation of the right sort, and it guarantees that it is hooked up to our experiences in the relevant way. So the Matrix Hypothesis implies that the Computational Hypothesis is correct, and that the computer simulation constitutes genuine physical processes.
VII. Other Objections
When we look at a brain in a vat from the outside, it is hard to avoid the sense that it is deluded. This sense manifests itself in a number of related objections. These are not direct objections to the argument above, but they are objections to its conclusion.
Objection 1: A brain in a vat may think it is outside walking in the sun, when in fact it is alone in a dark room. Surely it is deluded!
Response: The brain is alone in a dark room. But this does not imply that the person is alone in a dark room. By analogy, just say Descartes is right that we have disembodied minds outside space-time, made of ectoplasm. When I think "I am outside in the sun", an angel might look at my ectoplasmic mind and note that in fact it is not exposed to any sun at all. Does it follow that my thought is incorrect? Presumably not: I can be outside in the sun, even if my ectoplasmic mind is not. The angel would be wrong to infer that I have an incorrect belief. Likewise, we should not infer that envatted being has an incorrect belief. At least, it is no more deluded than a Cartesian mind.
The moral is that the immediate surroundings of our minds may well be irrelevant to the truth of most of our beliefs. What matters is the processes that our minds are connected to, by perceptual inputs and motor outputs. Once we recognize this, the objection falls away.
Objection 2: An envatted being may believe that it is in Tucson, when in fact it is in New York, and has never been anywhere near Tucson. Surely this belief is deluded.
Response: The envatted being's concept of "Tucson" does not refer to what we call Tucson. Rather, it refers to something else entirely: call this Tucson*, or "virtual Tucson". We might think of this as a "virtual location" (more on this in a moment). When the being says to itself "I am in Tucson", it really is thinking that it is in Tucson*, and it may well in fact be in Tucson*. Because Tucson is not Tucson*, the fact that the being has never been in Tucson is irrelevant to whether its belief is true.
A rough analogy: I look at my colleague Terry, and think "that's Terry". Elsewhere in the world, a duplicate of me looks at a duplicate of Terry. It thinks "that's Terry", but it is not looking at the real Terry. Is its belief false? It seems not: my duplicate's "Terry" concept refers not to Terry, but to his duplicate Terry*. My duplicate really is looking at Terry*, so its belief is true. The same sort of thing is happening in the case above.
Objection 3: Before he leaves the Matrix, Neo believes that he has hair. But in reality he has no hair (the body in the vat is bald). Surely this belief is deluded.
Response: This case is like the last one. Neo's concept of "hair" does not refer to real hair, but to something else that we might call hair* ("virtual hair"). So the fact that Neo does not have real hair is irrelevant to whether his belief is true. Neo really does has virtual hair, so he is correct.
Objection 4: What sort of objects does an envatted being refer to. What is virtual hair, virtual Tucson, and so on?
Response: These are all entities constituted by computational processes. If I am envatted, then the objects that I refer to (hair, Tucson, and so on) are all made of bits. And if another being is envatted, the objects that it refers to (hair*, Tucson*, and so on) are likewise made of bits. If the envatted being is hooked up to a simulation in my computer, then the objects it refers to are constituted by patterns of bits inside my computer. We might call these things virtual objects. Virtual hands are not hands (assuming I am not envatted), but they exist inside the computer all the same. Virtual Tucson is not Tucson, but it exists inside the computer all the same.
Objection 5: You just said that virtual hands are not real hands. Does this mean that if we are in the matrix, we don't have real hands?
Response: No. If we are not in the matrix, but someone else is, we should say that their term "hand" refers to virtual hands, but our term does not. So in this case, our hands aren't virtual hands. But if we are in the matrix, then our term "hand" refers to something that's made of bits: virtual hands, or at least something that would be regarded as virtual hands by people in the next world up. That is, if we are in the matrix, real hands are made of bits. Things look quite different, and our words refer to different things, depending on whether our perspective is inside or outside the matrix.
This sort of perspective shift is common in thinking about the matrix scenario. From the first-person perspective, we suppose that we are in a matrix. Here, real things in our world are made of bits, though the "next world up" might not be made of bits. From the third-person perspective, we suppose that someone else is in a matrix but we are not. Here, real things in our world are not made of bits, but the "next world down" is made of bits. On the first way of doing things, our words refer to computational entities. On the second way of doing things, the envatted beings' words refer to computational entities, but our words do not.
Objection 6: Just which pattern of bits is a given virtual object? Surely it will be impossible to pick out a precise set.
Response: This question is like asking: just which part of the quantum wavefunction is this chair, or is the University of Arizona? These objects are all ultimately constituted by an underlying quantum wavefunction, but there may be no precise part of the micro-level wavefunction that we can say "is" the chair or the university. The chair and the university exist at a higher level. Likewise, if we are envatted, there may be no precise set of bits in the micro-level computational process that is the chair or the university. These exist at a higher level. And if someone else is envatted, there may be no precise sets of bits in the computer simulation that "are" the objects they refer to. But just as a chair exists without being any precise part of the wavefunction, a virtual chair may exist without being any precise set of bits.
Objection 7: An envatted being thinks it performs actions, and it thinks it has friends. Are these beliefs correct?
Response: One might try to say that the being performs actions* and that it has friends*. But for various reason I think it is not plausible that words like "action" and "friend" can shift their meanings as easily as words like like "Tucson" and "hair". Instead, I think one can say truthfully (in our own language) that the envatted being performs actions, and that it has friends. To be sure, it performs actions in its environment, and its environment is not our environment but the virtual environment. And its friends likewise inhabit the virtual environment (assuming that we have a multi-vat matrix, or that computation suffices for consciousness). But the envatted being is not incorrect in this respect.
Objection 8: Set these technical points aside. Surely, if we are in a matrix, the world is nothing like we think it is!
Response: I deny this. Even if we are in a matrix, there are still people, football games, and particles, arranged in space-time just as we think they are. It is just that the world has a further nature that goes beyond our initial conception. In particular, things in the world are realized computationally in a way that we might not have originally imagined. But this does not contradict any of our ordinary beliefs. At most, it will contradict a few of our more abstract metaphysical beliefs. But exactly the same goes for quantum mechanics, relativity theory, and so on.
If we are in a matrix, we may not have many false beliefs, but there is much knowledge that we lack. For example, we do not know that the universe is realized computationally. But this is exactly what one might expect. Even if we are not in a matrix, there may well be much about the fundamental nature of reality that we do not know. We are not omniscient creatures, and our knowledge of the world is at best partial. This is simply the condition of a creature living in a world.
VIII. Other Skeptical Hypotheses
The Matrix Hypothesis is one example of a traditional "skeptical" hypothesis, but it is not the only example. Other skeptical hypotheses are not quite as straightforward as the Matrix Hypothesis. Still, I think that for many of them, a similar line of reasoning applies. In particular, one can argue that most of these are not global skeptical hypotheses: that is, their truth would not undercut all of our empirical beliefs about the physical world. At worst, most of them are partial skeptical hypotheses, undercutting some of our empirical beliefs, but leaving many of these beliefs intact.
New Matrix Hypothesis: I was recently created, along with all my memories, and was put in a newly-created matrix.
What if both the matrix and I have existed for only a short time? This hypothesis is a computational version of Bertrand Russell's Recent Creation Hypothesis: the physical world was created only recently (with fossil record intact), and so was I (with memories intact). On that hypothesis, the external world that I perceive really exists, and most of my beliefs about its current states are plausibly true, but I have many false beliefs about the past. I think the same should be said of the New Matrix Hypothesis. One can argue, along the lines presented earlier, that the New Matrix Hypothesis is equivalent to a combination of the Metaphysical Hypothesis with the Recent Creation Hypothesis. This combination is not a global skeptical hypothesis (though it is a partial skeptical hypothesis, where beliefs about the past are concerned). So the same goes for the New Matrix Hypothesis.
Recent Matrix Hypothesis: For most of my life I have not been envatted, but I was recently hooked up to a matrix.
If I was recently put in a matrix without realizing it, it seems that many of my beliefs about my current environment are false. Let's say that just yesterday someone put me into a simulation, in which I fly to Las Vegas and gamble at a casino. Then I may believe that I am in Las Vegas now, and that I am in a casino, but these beliefs at false: I am really in a laboratory in Tucson.
This result is quite different from the long-term matrix. The difference lies in the fact that my conception of external reality is anchored to the reality in which I have lived most of my life. If I have been envatted all my life, my conception is anchored to the computationally constituted reality. But if I was just envatted yesterday, my conception is anchored to the external reality. So when I think that I am in Las Vegas, I am thinking that I am in the external Las Vegas, and this thought is false.
Still, this does not undercut all of my beliefs about the external world. I believe that I was born in Sydney, that there is water in the oceans, and so on, and all of these beliefs are correct. It is only my recently acquired beliefs, stemming from perception of the simulated environment, that will be false. So this is only a partial skeptical hypothesis: its possibility casts doubt on a subset of our empirical beliefs, but it does not cast doubt on all of them.
Interestingly, the Recent Matrix and the New Matrix hypothesis give opposite results, despite their similar nature: the Recent Matrix Hypothesis yields true beliefs about the past but false beliefs about the present, while the New Matrix Hypothesis yields false beliefs about the past and true beliefs about the present. The differences are tied to the fact that in Recent Matrix Hypothesis, I really have a past existence for my beliefs to be about, and that past reality has played a role in anchoring the contents of my thoughts that has no parallel under the New Matrix Hypothesis.
Local Matrix Hypothesis: I am hooked up to a computer simulation of a fixed local environment in a world.
On one way of doing this, a computer simulates a small fixed environment in a world, and the subjects in the simulation encounter some sort of barrier when they try to leave that area. For example, in the movie The Thirteenth Floor, just California is simulated, and when the subject tries to drive to Nevada, the road says "Closed for Repair" (with faint green electronic mountains in the distance!). Of course this is not the best way to create a matrix, as subjects are likely to discover the limits to their world.
This hypothesis is analogous to a Local Creation Hypothesis, on which creators just created a local part of the physical world. Under this hypothesis, we will have true beliefs about nearby matters, but false beliefs about matters further from home. By the usual sort of reasoning, the Local Matrix Hypothesis can be seen as a combination of the Metaphysical Hypothesis with the Local Creation Hypothesis. So we should say the same thing about this.
Extendible Local Matrix Hypothesis: I am hooked up to a computer simulation of a local environment in a world, extended when necessary depending on subject's movements.
This hypothesis avoids the obvious difficulties with a fixed local matrix. Here the creators simulate a local environment and extend it when necessary. For example, they might right now be concentrating on simulating a room in my house in Tucson. If I walk into another room, or fly to another city, they will simulate those. Of course they need to make sure that when I go to these places, they match my memories and beliefs reasonably well, with allowance for evolution in the meantime. The same goes for when I encounter familiar people, or people I have only heard about. Presumably the simulators keep up a database of the information about the world that has been settled so far, updating this information whenever necessary as time goes along, and making up new details when they need them.
This sort of simulation is quite unlike simulation in an ordinary matrix. In a matrix, the whole world is simulated at once. There are high start-up costs, but once the simulation is up and running, it will take care of itself. By contrast, the extendible local matrix involves "just-in-time" simulation. This has much lower start-up costs, but it requires much more work and creativity as the simulation evolves.
This hypothesis is analogous to an Extendible Local Creation Hypothesis about ordinary reality, under which creators create just a local physical environment, and extend it when necessary. Here, external reality exists and many local beliefs are true, but again beliefs about matters further from home are false. If we combine that hypothesis with the Metaphysical Hypothesis, the result is the Extendible Local Matrix Hypothesis. So if we are in an extendible local matrix, external reality still exists, but there is not as much of it as we thought. Of course if I travel in the right direction, more of it may come into existence!
The situation is reminiscent of The Truman Show. Truman lives in an artificial environment made up of actors and props, which behave appropriately when he is around, but which may be completely different when he is absent. Truman has many true beliefs about his current environment: there really are tables and chairs in front of him, and so on. But he is deeply mistaken about things outside his current environment, and further from home.
It is common to think that while The Truman Show poses a disturbing skeptical scenario, The Matrix is much worse. But if I am right, things are reversed. If I am in a matrix, then most of my beliefs about the external world are true. If I am in something like The Truman Show, then a great number of my beliefs are false. On reflection, it seems to me that this is the right conclusion. If we were to discover that we were (and always had been) in a matrix, this would be surprising, but we would quickly get used to it. If we were to discover that we were (and always had been) in the Truman Show, we might well go insane.
Macroscopic Matrix Hypothesis: I am hooked up to a computer simulation of macroscopic physical processes without microphysical detail.
One can imagine that for ease of simulation, the makers of a matrix might not both to simulate low-level physics. Instead, they might just represent macroscopic objects in the world and their properties: e.g. that there is a table with such-and-such shape, position, and color, with a book on top of it with certain properties, and so on. They will need to make some effort to make sure that these objects behave in a physically reasonable way, and they will have to make special provisions for handling microphysical measurements, but one can imagine that at least a reasonable simulation could be created this way.
I think this hypothesis is analogous to a Macroscopic World Hypothesis: there are no microphysical processes, and instead macroscopic physical objects exist as fundamental objects in the world, with properties of shape, color, position, and so on. This is a coherent way our world could be, and it is not a global skeptical hypothesis, though it may lead to false scientific beliefs about lower levels of reality. The Macroscopic Matrix Hypothesis can be seen as a combination of this hypothesis with a version of the Metaphysical Hypothesis. As such, it is not a global skeptical hypothesis either.
One can also combine the various hypothesis above in various ways, yielding hypotheses such as a New Local Macroscopic Matrix Hypothesis. For the usual reasons, all of these can be seen as analogs of corresponding hypotheses about the physical world. So all of them are compatible with the existence of physical reality, and none is a global skeptical hypothesis.
The God Hypothesis: Physical reality is represented in the mind of God, and our own thoughts and perceptions depend on God's mind.
A hypothesis like this was put forward by George Berkeley as a view about how our world might really be. Berkeley intended this as a sort of metaphysical hypothesis about the nature of reality. Most other philosophers have differed from Berkeley in regarding this as a sort of skeptical hypothesis. If I am right, Berkeley is closer to the truth. The God Hypothesis can be seen as a version of the Matrix Hypothesis, on which the simulation of the world is implemented in the mind of God. If this is right, we should say that physical processes really exist: it's just that at the most fundamental level, they are constituted by processes in the mind of God.
Evil Genius Hypothesis: I have a disembodied mind, and an evil genius is feeding me sensory inputs to give the appearance of an external world.
This is Rene Descartes's classical skeptical hypothesis. What should we say about it? This depends on just how the evil genius works. If the evil genius simulates an entire world in his head in order to determine what inputs I should receive, then we have a version of the God Hypothesis. Here we should say that physical reality exists and is constituted by processes within the genius. If the evil genius is simulating only a small part of the physical world, just enough to give me reasonably consistent inputs, then we have an analog of the Local Matrix Hypothesis (in either its fixed or flexible versions). Here we should say that just a local part of external reality exists. If the evil genius is not bothering to simulate the microphysical level, but just the macroscopic level, then we have an analog of the Macroscopic Matrix Hypothesis. Here we should say that local external macroscopic objects exist, but our beliefs about their microphysical nature are incorrect.
The evil genius hypothesis is often taken to be a global skeptical hypothesis. But if the reasoning above is right, this is incorrect. Even if the Evil Genius Hypothesis is correct, some of the external reality that we apparently perceive really exists, though we may have some false beliefs about it, depending on details. It is just that this external reality has an underlying nature that is quite different from what we may have thought.
Dream Hypothesis: I am now and have always been dreaming.
Descartes raised the question: how do you know that you are not currently dreaming? Morpheus raises a similar question:
Have you ever had a dream, Neo, that you were so sure was real. What if you were unable to wake from that dream? How would you know the difference between the dream world and the real world?
The hypothesis that I am currently dreaming is analogous to a version of the Recent Matrix Hypothesis. I cannot rule it out conclusively, and if it is correct, then many of my beliefs about my current environment are incorrect. But presumably I still have many true beliefs about the external world, anchored in the past.
What if I have always been dreaming? That is, what if all of my apparent perceptual inputs have been generated by my own cognitive system, without my realizing this? I think this case is analogous to the Evil Genius Hypothesis: it's just that the role of the "evil genius" is played by a part of my own cognitive system! If my dream-generating system simulates all of space-time, we have something like the original Matrix Hypothesis. If it models just my local environment, or just some macroscopic processes, we have analogs of the more local versions of the Evil Genius Hypothesis above. In any of these cases, we should say that the objects that I am currently perceiving really exist (although objects farther from home may not). It is just that some of them are constituted by my own cognitive processes.
Chaos Hypothesis: I do not receive inputs from anywhere in the world. Instead, I have random uncaused experiences. Through a huge coincidence, they are exactly the sort of regular, structured experiences with which I am familiar.
The Chaos Hypothesis is an extraordinarily unlikely hypothesis, much more unlikely than anything considered above. But it is still one that could in principle obtain, even if it has miniscule probability. If I am chaotically envatted, do physical processes obtain in the external world? I think we should say that they do not. My experiences of external objects are caused by nothing, and the set of experiences associated with my conception of a given object will have no common source. Indeed, my experiences are not caused by any reality external to them at all. So this is a genuine skeptical hypothesis: if accepted, it would cause us to reject most of our beliefs about the external world.
So far, the only clear case of a global skeptical hypothesis is the Chaos Hypothesis. Unlike the previous hypothesis, accepting this hypothesis would undercut all of our substantive beliefs about the external world. Where does the difference come from?
Arguably, what is crucial is that on the Chaos Hypothesis, there is no causal explanation of our experiences at all, and there is no explanation for the regularities in our experience. In all the previous cases, there is some explanation for these regularities, though perhaps not the explanation that we expect. One might suggest that as long as a hypothesis involves some reasonable explanation for the regularities in our experience, then it will not be a global skeptical hypothesis.
If so, then if we are granted the assumption that there is some explanation for the regularities in our experience, then it is safe to say that some of our beliefs about the external world are correct. This is not much, but it is something!
David Chalmers
David Chalmers' website: http://www.consc.net
(Some philosophical notes on this article can be found here.)
Return to topThe significance of The Matrix as a movie with deep philosophical overtones is well recognized. Whenever the movie is discussed in philosophy classes, comparisons are made with Descartes’ Meditations, particularly the dream argument and the evil genius scenario, both of which are intended to generate skeptical doubt. How do we know, for example, that we are awake now, rather than merely dreaming? How do we know that our thoughts are not being manipulated, and that our perceptions of ‘reality’ are accurate? The Matrix makes these doubts stand out vividly.
However, The Matrix raises many other interesting philosophical issues, and ones that are worthy of further discussion. This essay explores some of the moral issues raised in The Matrix. The first is the issue of the moral status of the created beings, the ‘artificial’ intelligences, which figure into the universe of The Matrix. The second is the issue of whether or not one can do anything wrong in circumstances where one’s experiences are non-veridical; that is, where one’s experiences fail to reflect reality.
I. The Moral Status of Programs
There is a reality to the Matrix. The substance of that reality may differ dramatically from the substance we label ‘real’ — the ‘real’ world is the desert reality that Morpheus reveals to Neo. But it is clear that, out of the grip of the Matrix, though still having certain dream-like experiences, Neo and his enlightened friends are dealing with actual sentient programs, and making decisions that have actual effects for themselves as well as the machines and the programs. What is the moral status of the sentient programs that populate the Matrix, or, for that matter, the moral status of the machines themselves? The universe of The Matrix is populated with beings that have been created — created by programmers or created by the machine universe itself. The agents, such as Smith, Neo’s pursuer, are prime examples. These beings come into and go out of existence without comment on the part of whoever controls the switches — and without any moral debate on the part of the humans who also would like to see the agents destroyed. There seems to be an implicit view that their existence is less significant, their lives of less moral import, than the lives of ‘naturally’ existing creatures such as ourselves. An obvious explanation for this attitude is that humans are long accustomed to thinking of themselves as being at the center of the universe. The geographic point changed with Copernicus. However, our view of our dominant place in the moral universe has stayed fixed. But, once again, science — and particularly, now, cognitive science holds the potential for challenging this certainty. And science fiction such as The Matrix, which explores differing directions for these potentialities, also brings challenges to this worldview. What The Matrix offers is a vivid thought experiment. It is a thought experiment which makes us ask the sort of ‘what if?’ question that leads to a change in self conception. It forces us to see where our well accepted moral principles would take us within one possible world.
We know that killing human beings is wrong. It is wrong because human beings have moral standing. Human beings are widely believed to have this standing in virtue of consciousness and sentience. For example, a rock has no moral standing whatsoever. Kicking a rock does not harm it, and no moral rights are violated. It is an inanimate, non-conscious object incapable of either thought or sensation. Animals, however, are generally taken to have some moral standing in virtue of their sentience. Kicking an animal for no compelling reason is generally taken to be immoral. Human beings have greater standing in virtue of their higher rational capacities. They can experience more varied and complex harms, and a wider range of emotional responses – such as resentment – in virtue of their rationality. How one came into existence is not taken to be morally significant. Some people are the products of natural conception, and some are the result of conception in the laboratory. This makes no difference to the possession of those qualities we take to be morally significant – consciousness and rationality. And, surely, the substance from which someone is created is completely irrelevant to the issue of moral status. If a person’s consciousness could somehow be transferred to a metallic or plastic robotic body, the end result would still be a person.
It would seem, then, that the fact that one is created, or artificial, is in no way relevant to one’s moral standing. And, if this is the case, then the world of The Matrix presents underappreciated moral complexities. Agents such as Smith, while not very pleasant, would arguably have moral standing, moral rights. Of course, Neo has the right to defend himself — Smith is not, after all, an innocent. Indeed, if the religious theme is pursued, he is an agent of darkness. But any innocent creations of the machines — beings brought into existence to populate the Matrix — also would have moral rights. Just as it would be wrong to flip a switch and kill an innocent human being, no matter how that human being came into existence, it would be wrong to flip a switch and kill a sentient program. As long, of course, as that program possessed the qualities we regard as morally relevant. And this is where one of the primary issues raised by the possibility of artificial intelligence becomes important to the question at hand. Do these programs possess consciousness? Since we are considering the world of The Matrix, let’s look at what evidence seems to exist in the movie. While we don’t have much information about the machines themselves, their agents are on ample display.1
Smith, of course, and his colleagues seem remarkably without affect. Yet, at critical points they do display emotions: anger, fear, and surprise. They seem able to plan and to carry through on a plan. Smith also displays a capacity for sadistic pleasure — at one point he displays this, when he forces Neo’s mouth shut. Smith also displays extreme fear near the end of the movie, when Neo leaps through him. The agents display many, if not all, of the responses we associate with consciousness and sentience. But this brings us to another skeptical challenge posed in The Matrix. How can we be sure they do posses minds, and are not mere automata, albeit highly complex ones? Though the movie invites this reflection, it is important to see where this challenge can take us. The "how can I be sure?" question can extend beyond the agents to our fellow human beings. Since a person’s conscious experiences are essentially private, one cannot be directly aware of another’s experiences. We might try, as St. Augustine suggested, to solve this problem by appeal to analogy: I do directly experience my own mental states — I know that I am a conscious, aware, being. I also know on the basis of observation that I am structurally similar to other human beings. Thus, I reason by analogy, that they must experience mental states as well.2 And, indeed, The Matrix invites such a comparison when the agents display behavior consistent with the experience of certain psychological states.3
Given, then, that we believe what we are invited to believe it would follow that the sentient programs, the cyber persons, do possess those qualities we associate with moral standing. They have moral rights on the basis of consciousness and sentience and rationality. Thus, their moral standing is the same as that of human beings.
It is possible that human beings have some additional value — a kind of antiquarian value. We are, so to speak, "the originals." The original Mona Lisa, for example, has value in excess of its copies. But this kind of value is not moral value and does not reflect on the moral standing of the object, or the moral significance of the lives themselves. The Mona Lisa does have value, but no moral standing since it is a mere painting; it lacks consciousness. It may be damaged, but not harmed in the way that humans and sentient creatures can be harmed.
Perhaps the machines view humans this way. To the machines, the value of humans is mainly instrumental. They are valued as a source of energy, but they may also have some antiquarian value. Humans are merely relics of a past they themselves helped to destroy. If that’s the case, the machines have turned the tables. They are making the same moral mistake humans apparently made in the context of The Matrix, in viewing other rational life forms as simple instruments, to use and destroy as one wishes. Indeed, both sides of the conflict seem to have displayed some moral blindness. The humans, in using and destroying, and the machines, certainly, in their subjection of the humans. But both sides view themselves as fighting for survival, and I imagine that Smith and Smith’s creators, as well as Neo and his friends, would argue that moral qualms like these are a luxury.
II. Manipulation and Immorality
The world that the pre-enlightened Neo inhabits is one made up by machines. The machines have created a humdrum existence for humans, to keep them happy and pacified and free of the knowledge that they are being used as a source of energy for the machines. Most humans believe that this world is real, but they are mistaken. Within this world they build lives for themselves, have relationships, eat lovely dinners, and at least seem to both create and destroy. To some extent this existence is dream like. It isn’t real. When the unenlightened person thinks he’s eating a steak, he isn’t. Instead, the machines generate mental experiences which correspond to the experience of eating a steak, but which are non-veridical – that is, the person is not actually eating a steak. There is no real or actual steak. The human being’s actions, in that respect, have no real or actual consequences in a world that exists independently of his or her mind. However, even in this unenlightened state, the humans do have some control, since what they ‘do’ in the Matrix has consequences which are realized in the real world. Getting smashed by a truck in the Matrix kills the person in reality. The Matrix offers a ‘brain-in-a-vat’ experience, but one where the experiencer does have some control.4 The enlightened can, in principle, understand the rules of the Matrix and learn to exert that control with full understanding.5
But, as the steak example illustrates, there are many other ‘actions’ they perform that seem to have no effects in the real world. The pre-enlightened Neo and most of the humans living in the Matrix seem to be deluded. One issue raised by this is the extent to which they can be held responsible for their actions in the Matrix. Suppose, just for the sake of argument, that something like wearing fur is immoral. Is simply making a choice to wear fur, along with the belief that one is wearing fur, enough to make one guilty of wrongdoing? Is it really only the thought that counts, morally? A competing view is that the choices people make must result in actual bad consequences in order for them to be guilty of wrongdoing; or, actual good consequences in order for them to be considered to have acted rightly. So, the issue is that of whether or not the moral quality of a person’s actions — its rightness or wrongness — is determined solely by his or her subjective states, or whether, instead, actual consequences figure into this determination.
In the Matrix if fur is worn it is virtual fur, and not real — though the wearer does not realize this. Again, this is because he or she is being mentally manipulated. But is this a genuine delusion? Certainly, an insane person who fails to have a grip on reality, and is deluded in this sense, is thought to have diminished moral responsibility for what he or she does while deluded. Such a person is generally held to not be morally responsible in those circumstances. He is not punished, though he may be confined to a mental hospital and treated for his insanity. The explanation is that the actions performed while insane are not truly voluntary. If the persons who live in the Matrix are similarly deluded, then it would seem that they are not responsible for what they ‘do’ in the Matrix.
Some writers have argued that one cannot be held responsible for what happens in a dream, since dreams themselves are not voluntary, nor are the ‘actions’ one seems to perform in a dream.6 Other writers, such as Henry David Thoreau, had the view that what we seemed to do in a dream reflected on our character; and the contents of dreams could reveal true virtue or vice.7 Even if the actions one performs in a dream have no actual good or bad consequences, they reveal truths about one’s emotional make-up, and one’s inner desires, and these, in turn are revealing of character. But, as we’ve discussed, the Matrix isn’t a dream. The unenlightened exist, rather, in a state of psychological manipulation. The actions they seem to perform don’t always have the effects (in reality) that they have reason to expect, based on their manipulated experiences. But even in the Matrix we can argue that they make voluntary choices. They are not irrational. They are not like the insane. Neo believes what any rational, reasonable person would believe under the circumstances. The pre-enlightened are analogous to persons who make decisions based on lies that others have told them. They act, but without relevant information. It’s that condition that Neo would like to rectify at the end of The Matrix.
The view I favor is that without actual bad effects the actions of those in the Matrix are not immoral. But, again, this claim is controversial. Some would argue that it’s simply "the thought that counts"; that it is the person’s intentions which determine the moral quality of what he or she does. Immanuel Kant, for example, is famous for having claimed that all that matters, intrinsically, is a good will – actual consequences are irrelevant to moral worth.8 However, it would then be the case that forming bad intentions in one’s dreams is also sufficient for immorality, and this seems highly counterintuitive. If that’s true, then the intention to do something immoral along with the belief that one has so acted, is enough to make one guilty of moral wrongdoing. Instead, it seems more plausible that it must also be the case that there is some actual bad brought about, or at least the realistic prospect of some actual bad consequences, and thus non-veridical ‘wrongdoing’ in the Matrix is not actual wrongdoing.
This seems to be clearly the case in a dream. In a dream, when the dreamer decides to do something bad that decision doesn’t impact on the real world. But the Matrix is not really a dream. If we assume that the virtual world of the Matrix is complete — that is, completely like the real world before the machines took over — then the virtual ‘harms’ are still real in that they are realizized in terms of actual unpleasant mental states. The virtual fur coat is the result then of a virtual animal getting killed, but a virtual animal with all the right sorts of mental states — in this case, pain and suffering. If this is the case, then the killer, though mistaken in thinking the dead animal ‘real’ has still produced bad effects in the form of genuine pain and suffering. And thus, the action is immoral even though non-veridical. However, if the world of the Matrix is incomplete, the issue becomes more complicated. If Cypher’s virtual steak comes from a virtual meat locker, and the meat locker is the end of the line — and the acquisition of the steak does not involve the killing of a virtual animal with all the same psychology of pain and suffering a ‘real’ animal feels, then no moral harm has been done.
But note that Thoreau’s point still holds even though the Matrix is not exactly like a dream. That is — even if a person hasn’t actually done anything bad, or caused any real harm to another sentient life form, we may still make a negative evaluation of the person’s character.
But my guess is that the Matrix is a complete alternate reality created in the image of the pre-machine reality. And the Matrix, if it does offer such a complete replication of the pre-machine reality, is truly a self-contained world. It has its own objects, its own people, animals and … ethics. The systematic deception of the humans doesn’t change this.
Julia Driver
Footnotes
1. The issue of the moral status of the machines themselves should be kept distinct from the issue of the moral status of the sentient programs. I will focus on the latter issue here in discussion, simply because the movie provides more information about the behavior of these constructs. But the same points would hold for the machines themselves – if they have those qualities that are morally significant, consciousness and rationality, then they also possess moral standing.
2. St. Augustine, The Trinity (8.6.9). Again, this line of reasoning is controversial since it relies on a single case analogy.
3. A lot hinges on what we take to be ‘structurally similar’. Some would argue that while the sentient programs are not themselves structures, the machines are, and thus the machines may possess consciousness, though the programs cannot. However, I believe the sentient programs can be structurally similar if that’s understood functionally – their code has structure which provides functional equivalence to the physical states that underlie our mental states. But, this issue would be extremely controversial, and there isn’t enough time to delve into it more fully here.
4. See Christopher Grau’s introductory essays on this site for more on dream skepticism and brain-in-a-vat skepticism.
5.The unenlightened, on the other hand, are constantly being "Gettiered". A woman may have justified true belief that her husband is dead, because she has just ‘seen’ him smashed by a truck. But being in the Matrix she lacks true knowledge because she is deceived in the true manner of his death.
6. See, for example, William Mann’s "Dreams of Immorality," Philosophy (1983), pp. 378-85.
7. Thoreau writes about this in A Week on the Concord and the Merrimack (1849).
8. This also is controversial, but see Kant’s Foundations of the Metaphysics of Morals, trans. Lewis White Beck, and critical essays ed. by Robert Paul Wolf (NY: MacMillan, 1969):
Nothing in the world — indeed, nothing even beyond the…world — can possibly be conceived which could be called good without qualification except a good will…The good will is not good because of what it effects or accomplishes or because of its adequacy to achieve some proposed end; it is good only because of its willing, i.e., it is good of itself. (pp. 11-12)
Return to topThe Matrix provides a fine resource for illustrating philosophical ideas. Many films have themes that one can philosophize about, or that serve as useful illustrations of philosophical ideas, such as the wonderful films Sophie’s Choice or The Sheltering Sky. But The Matrix offers more than this. It belongs in a special class of films including Blade Runner, Total Recall, Crimes and Misdemeanors, A Clockwork Orange, The Unbearable Lightness of Being, and The Truman Show. All of these films are intentionally philosophical. Each shows how richly philosophical themes can be developed through cinema. Perhaps the best of these films is The Matrix.
I.
No doubt, the most striking philosophical theme found in The Matrix concerns skepticism about knowledge of an external world. The dream world Neo inhabited was a perfectly comfortable “reality”—except for the fact that it was not reality. Life from inside it completely shielded one from what Morpheus aptly called “the desert of the real,” that desolated shell of a planet on which countless humans were unknowingly ensconced in slimy wombs. But there are many other philosophical themes explored within The Matrix. One is the concept of freedom. Freedom is mentioned at various points in the film.1 It mattered a great deal who did what freely. For instance, it was important that Neo freely chose to take the red pill and not the blue pill. Had he taken the blue pill, he’d have been returned to that humdrum dream world of vapid city dwellers. He’d never have taken the path that eventually led him to his heroic defeat of the agents, and that left him at the end of the film entertaining the prospect of saving the human race. At various other points Neo made choices freely, and, as with taking the red pill, it was the quality of having made them freely that gave them the importance they had. For instance, Neo freely decided to risk his life for Morpheus; instead of fleeing when his own life was in danger, he returned to save Morpheus from cranial meltdown at the hands of those treacherous agents in their zoot suits. Also, Neo freely followed the white rabbit that led him tumbling down that rabbit hole. And he remained in the car when Trinity and Switch gave him the opportunity to bail. By remaining in the car, Neo freely chose to resist the agents. He chose on his own not to get out and walk away down that street, down that well worn path that, Trinity reminded him, led to nowhere special. In choosing to remain in the car, he freely embarked upon a path that would lead to an exciting future, to an exciting life.
But it was not just Neo’s freedom that mattered. Freedom was an issue for the others as well. During Cypher’s attempted mutiny, Trinity reminded him that all of Morpheus's rebels had freely chosen the red pill, and so none could claim that they were in their dire straights undeservedly. All the same, Cypher regretted his choice. He felt duped; he did not regard his choice to take the red pill as free. As he saw it, he was scammed. In fact, he was of the opinion that he’d have had more freedom as a steak-eating, satiated participant in The Matrix, oblivious to the “truth” about the ugly shell that would have held him in perpetual slumber.
Freedom also mattered a great deal when it was not possessed. It seems that this was the case with those countless human drones, all contained in their artificial wombs. As Morpheus and company saw it (save for Cypher), their poor, ignorant kin were victims, blind to their lack of freedom—maybe even happy in their plodding little lives within the Matrix, working in cubicles all day—but victims all the same, enslaved in the service of generating battery juice for those battery-powered A.I. meanies. Even the leader of the agents’ posse,