Reality+: Virtual Worlds and the Problems of Philosophy

by David J. Chalmers

demon, how can you know that anything is real?

The evil-demon argument calls into question everything you know about the external worl...

How does the evil demon do its work? Descartes is not clear on the details.

Presumably the demon has to keep a complicated model of a fictional world in its head to make sure the subject has matching experiences over time. Every time I return to Australia or visit an old friend, my experience needs to be consistent with previous visits. The demon will also need models for places I’ve read about and places I’ll eventually visit, as well as everything I read about in newspapers or watch on TV. The model will have to be constantly updated. This will be a lot of work, although perhaps the work is nothing for an all-powerful demon.

An especially insidious version of Descartes’s evil demon gets inside people’s minds and directly tampers with their thoughts. In the modern version, this demon could be an evil neuroscientist. Perhaps the demon manipulates your brain so that you believe that you’re in Antarctica. Descartes says that a deceiver might even manipulate his thoughts so that “I too go wrong every time I add two and ...

The mind-tampering demon threatens to lead to a sort of internal-world skepticism in which you can’t even trust your own rationality or reasoning anymore. This sort of evil demon is fascin...

I’m concerned here with scenarios in which my external world is manipulated, not scenarios in which my internal world is manipulated directly. I’ll return to mind-tamperi...

If Descartes’s evil demon lives in the computer age, its task is a lot easier. It can simply offload the modeling work into a computer. It can run a computer simulation of the world and connect subjects to the simulation so that they experience the world as it evolves.

In the 20th century, the American philosopher Hilary Putnam and others updated Descartes’s idea with equipment from modern science. The evil demon was replaced by an evil scientist, and the person deceived by the evil demon was replaced by a brain in a vat. Like the brains that float in jars in the Steve Martin movie The Man with Two Brains, the brain is kept alive with a carefully balanced mix of nutrients. Putnam tells us that the brain’s nerve endings are “connected to a super-scientific computer.” The computer sends electronic impulses to the brain, bringing about the illusion that everything is normal. The brain experiences a richly detailed and well-populated world, but in fact it is alone in a laboratory.

The simulation idea captures an element at the core of all of the great Cartesian scenarios: The evil demon must do its work by simulating a world.

A lifelong dream can be seen as a sort of simulated world.

The brain in a vat is connected to ...

Making the simulation a computer simulation helps us to pin down the scenario in more concrete terms with...

The brain-in-a-vat idea is one version of the simulation hypothesis. It involves an impure simulation, in which a brain is connected to the simulation from the outside. The simulation hypothesis also includes other versions, such as pure simu...

You might think that the switch from evil demons to brains in vats to simulations is a mere change in packaging, but there is one respect in which the use of modern technology makes the argument more powerful. Because the evil-demon hypothesis is so fanciful, Descartes was reluctant to put too much weight on it. It was important to him that his skeptical concerns be grounded in reasonable doubts that he should take seriously, given his beliefs. He gave more weight to the deceiving-God hypothesis because he believed in an all-powerful God and thought it reasonable that God would have the power to deceive us. Because this was a realistic hypothesis, it gave him greater reason for doubt.

too. In philosophy, an argument is a chain of reasoning that supports a conclusion.

Here’s a formal argument for skepticism about the external world. You can’t know you’re not in a simulation. If you can’t know you’re not in a simulation, you can’t know anything about the external world. __________________________ So: You can’t know anything about the external world.

Bertrand Russell once said, “The point of philosophy is to start with something so simple as not to seem worth stating, and to end with something so paradoxical that no one will believe it.”

Why believe the first premise? I’ve made an initial case in chapter 2. In a good-enough simulation, the world would look and feel to you exactly as today’s world looks and feels to you now. And if a simulation would look and feel the same as reality, it’s hard to see how we could know we’re in a simulation rather than reality.

Why believe the second premise? Pick anything you think you knew about the external world. You thought you knew that Paris is in France, or that there’s a spoon in front of you. But if you’re in a simulation, then your beliefs about Paris and the spoon come from the simulation, not from reality. Paris and the spoon are simulated.

The world outside the simulation may be entirely different. There may well be no Paris and no spoon in reality outside the simulation. So to know that Paris is in France or that there’s truly a spoon in front of you, you h...

The reasoning here is a bit like this: If your phone is a knockoff, you don’t really have an iPhone. So if you can’t know that your phone isn’t a knockoff, you can’t know that you have an iPhone. In this case, we start from the plausible claim: If you’re in a simulation, there’s no spoon in front of you. By the same sort of reasoning as in the iPhone case, we get to: If you can’t know you’re not in a simula...

Our Reality Question about virtual reality was: Is virtual reality real or an illusion? If you answer by saying Virtual reality is an illusion, you’ll probably accept the second premise. Here’s why. Given this answer, you’ll also accept Simulations are illusions, since simulations are a kind of virtual reality in the broad sense. In fact, you’ll probably accept If you’re in a simulation, everything you experience in the external world is illusory. So if you can’t rule out the simulation hypothesis, you can’t rule o...

Strictly speaking, the argument doesn’t stop you from knowing a few things about the external world. Some things are true as a matter of logic or mathematics. You can know that all dogs are dogs, for example. You can know that if there is one table here and a different table there, there are two tables. But these are all trivialities. To be strictly correct, we could adjust the conclusion to “We can’t know anything substantial about the external world.”

If we accept the premises, the argument leads us to global skepticism about the external world—that is, the view that we don’t know anything substantial about the external world. Maybe we can still know that two plus two is four, but that’s not a huge consolation. What can we do to avoid the shocking conclusion?

Descartes himself didn’t want to be a skeptic. In fact, he wanted to establish a foundation for all knowledge. So after casting all our knowledge into doubt with his skeptical arguments, he tried to build it back up, piece by piece.

Descartes needed to start with a piece of knowledge he couldn’t doubt. He needed to uncover something about reality that would be true even if he was having sensory illusions, even if he was dreaming, even if he was being fooled by an evil demon. He found a candidate: his own existence.

Descartes’s famous argument for his own existence, presented most explicitly in his 1637 Discourse on Method, went like this: Cogit...

like an argument. The premise of the argument

I am thinking.

The conclusion is...

As with most arguments, the real work is done by the premise. Once you grant that, the conclusion I exist seems...

How does Descartes know he’s thinking?

Even if you’re in the grip of a sensory illusion, you’re still thinking. Even if you’re dreaming, you’re still thinking. Even if you’re being fooled by an evil demon, you’re still thinking. Even if you’re a brain in a vat, you’re still thinking. Even if you’re in a simulation, you’re still thinking.

Descartes reasoned that he could not doubt that he’s thinking. Even if he doubted that he was thinking, his doubt was itself a sort of thinking. To doubt that one is thinking is internally inconsistent: The doubting itself shows that the doubt is wrong. Once Descartes knew he was thinking, it was a small step to knowing his own existence...

Some question the cogito part. How can Descartes be so sure that he’s even managing to doubt? That is, how does he know that he’s not a mindless automaton?

Others question the step to sum. Is it so obvious that thinking requires a thinker? According to the 18th-century German philosopher Georg Lichtenberg, Descartes should have said, “There is thinking, therefore thought exists.” That way, he could have known that thoughts exist, but he should not have been so sure about himself.

Still, a lot of people accept Descartes’s Cogito, ergo sum. It’s hard to doubt that I’m thinking. The evil-demon scenario doesn’t really call my own mind into doubt, and it’s not easy to generate scenarios that do. As a result, even some skeptical philosophers are prepared to say ...

Speaking for myself, I don’t think there’s anything special about thinking per se. Descartes could have said, “I feel, therefore I am,” or “I see, therefore I am,” or “I worry, therefore I am.” All of these are claims about his mind that he...

In my view, the best statement of the cogito is “I am conscious, therefore I am.”

it’s arguable that this is what Descartes really meant. He explicitly defines thought as including everything we’re conscious of, and says that it includes the senses and imagination as well as the intellect and the will.

If we grant Cogito, ergo sum, that gives Descartes a foundation. The hard part is what comes next. How do we get from knowledge of ourselves and our own minds to knowledge of the external world?

Descartes argued that he had an idea of God as a perfect being. His idea of God is an idea of a being who is perfectly good, perfectly wise, and so on. In fact, he argued that the idea itself is a perfect idea, and therefore it could not have come from anywhere except from a perfect being. That is, the idea of God must have come from God. If this argument works, it gets us from knowledge of our own mind to knowledge of something outside us and much greater than us. It gets us to God.

Once we’ve gotten to God, the argument goes, getting to the world around us is easy. Since God is a perfect being, he would not allow us to be deceived. So, given that God exists, there can be no evil demon and no dreams or sensory illusions that persist for one’s whole life. God will ensure that our impressions of the external world are, by and large, accurate. For him to do anything else would be imperfect. So the external world exists and is much as we thought it was. Hallelujah!

One obvious problem: Why couldn’t the idea of a perfect being come from somewhere other than a perfect being?

In fact, why couldn’t an evil demon give someone the idea of a perfect being?

Another problem: Even if there is a perfect being, how can we be sure we’re not being deceived? Maybe deceiving us is part of the perfect being’s master plan! For example, maybe we all need to go through a period of deception before we’re finally enlightened about reality. We’re imp...

you don’t need to be perfect to create the idea of perfection. In fact, it was precisely because we’re imperfect that we had to create Sim Descartes in a simulation; creating him in a nonsimulated world was too hard for us.

Admittedly, for all we know, there really is a perfect God out there who created us and indirectly created the simulation.

Descartes could argue that our own idea of perfection came from God, so Sim Descartes got the idea indirectly from God, and his argument for a perfect being still stands. ...