- Consciousness and computation make it unlikely that we are living in a simulation.
- Virtual objects differ from real ones because of how we interact with them.
- Life in virtual reality lacks the satisfaction of needs that occurs in real life.
The metaverse is supposed to be a brave new world that mingles reality with computer-generated virtual reality. But getting lost in it is easy, as we see from the Facebook company’s stumbling attempts to reinvent itself as Meta. David Chalmers’ new book Reality+ asks important questions about the philosophical significance of virtual technology, concluding: (pp. xvii, 17):
- We cannot know whether we are in a virtual world, a computer simulation.
- Virtual reality is genuine reality.
- You can lead a fully meaningful life in a virtual world.
All of these claims are implausible.
We are not living in a virtual world
Chalmers’ main argument that we might be living in a computer simulation is (p. 101):
1. It’s more likely than not that conscious human-like simulations are possible.
2. It’s more likely than not that if conscious human-like simulations are possible, many human-like populations will create them.
3. There’s a good chance (25 percent or so) that we are computer simulations.
The general idea is that future generations of programmers will produce such a large number of computer simulations that our own experienced lives probably occur in one of them.
This statistical argument fails because both of its main premises are implausible. It may well turn out that someday computers will have consciousness (Thagard 2021). But it is unlikely that their consciousness will be just like ours because the physical mechanisms of computers are so different from the neural and flesh-and-blood mechanisms that produce human consciousness. The assumption that computer consciousness will be the same as human consciousness assumes substrate independence, the claim that mental states can operate in a broad range of physical systems. But I argue that energy considerations negate this claim (Thagard 2022):
- Real-world information processing depends on energy.
- Energy depends on material substrates.
- Therefore, information processing depends on material substrates.
Hence, it is unlikely that there will ever be exact simulations of human consciousness.
Even if such simulations could be produced, I doubt that there will ever be the large number of them that Chalmers’s second premise assumes. Simulating even one human consciousness, let alone that of all the billions of people currently living, would take vast amounts of programming effort, computer time, and energy supplies. Future humans will have too much trouble surviving pandemics, climate change, and autocratic leaders to generate countless simulations of previous generations.
Instead of the statistical argument for our being in a virtual world, we should ask: What is the best causal explanation of our current existence and experiences? In another blog post, I argue that the hypothesis that we operate in the real universe is far more reasonable than the simulation hypothesis.
Virtual reality is not reality
Suppose that you put on your virtual reality headset and use it to explore Jurassic World. Afterward, you feel relieved that the dinosaur chasing you was not real, or was it? Chalmers insists (p. 105) that the entities in virtual reality really exist, as structures of binary information, bits. He considers five criteria for reality: existence, causal powers, mind-independence, non-illusoriness, and genuineness. He concludes (p. 116) that if we’re in a perfect, permanent, simulation, then the objects we perceive are real according to all five of these criteria.
But we are not living in a simulation, so it’s clear that virtual entities are enormously different from real ones. Real entities are ones that can do things to us and that we can do things to. Dinosaurs in a computer game cannot bite us, and we cannot shoot them. Digital bits exist in computer chips, but do not exist in the real world that science tells us consists of many kinds of objects, including sub-atomic particles, atoms, molecules, rocks, planets, stars, and organisms. With all of these, we can interact in ways very different from how we interact with bits. The best explanation of the vast amount of evidence for established theories in physics, chemistry, and biology is that the entities they talk about are real, whereas the best explanation of the images and sounds in a computer game is that the entities depicted are merely simulated. Virtual reality is different from reality.
A purely virtual life is not meaningful
Chalmers tries to reassure readers that they should not be distressed if life is just a simulation because life can still be good. He considers various possible sources of value, including pleasant experiences, the satisfaction of desires, connections to other people, and other values such as knowledge and freedom. I think that the objective sources of value are needs, which make it easy to see the difference between virtual and real lives.
Humans have biological needs for oxygen, water, food, shelter, and health care, but also psychological needs for relatedness to other people, competence to accomplish tasks, and autonomy to do things without control from others (Ryan and Deci 2017). Future virtual reality could give you the illusion of satisfying needs by providing you with experiences such as eating exquisite food in a beautiful palace, but it would not thereby satisfy your needs if you were still hungry and trapped in a dangerous cave. Needs satisfaction is a matter of biological reality, which virtual reality barely approximates.
Similarly, social connections provided by computer games, romance novels, or romantic movies may provide people with an approximation to the experience of satisfying the need for relatedness to other people, but the fundamental psychological, biological needs would remain unsatisfied. Virtual love is not love, just as virtual food is not food. Analogously, simulated competence and autonomy fail to satisfy the relevant needs. People can legitimately find meaning in their lives through valuable pursuits of love, work, and play, but virtual reality just provides paltry play.
Chalmers is not a reality denier, unlike the skeptics that I challenged in a previous post. Rather, he is a reality diluter, making it weaker by trying to break down boundaries between what is real and what is only virtual, simulated, or imagined. Reality deserves better.
Chalmers, D. J. (2022). Reality+: Virtual worlds and the problems of philosophy. New York: Norton.
Ryan, R. M., & Deci, E. L. (2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. New York: Guilford.
Thagard, P. (2021). Bots and beasts: What makes machines, animals, and people smart? Cambridge, MA: MIT Press.
Thagard, P. (2022). Energy requirements undermine substrate independence and mind-body functionalism. Philosophy of Science, 89, 70-88.