The Ghostly Illusion of Freewill
(A brief introduction to the problem of freewill for Cafe Philosophy magazine)
By BRENT SILBY
During my childhood I was fascinated by videogames. One game that stands out in my memory is Pacman. It wasn’t the gameplay that interested me so much as the behavior of the ghosts. As you watch them roam around the maze, you get the feeling that they are intelligent. They seem to be making decisions about how best to catch Pacman. But how free are their decisions? One of the interesting things I noticed was that I could play exactly the same game over and over if I moved Pacman in precisely the same way each time. The ghosts always followed the same behavioral pattern and didn’t deviate from that pattern until I changed my pattern. Experimenting with Pacman in this way revealed to me something about the ghosts’ behavior. True, they make decisions, but their decisions are firmly and predictably determined by the way I move around the maze.
Now, this may seem obvious. Of course the ghosts have to follow rules. If they didn’t, they might never catch Pacman. The ghosts do not have freewill. They can’t decide not to chase Pacman. They do exactly what their program tells them to do in any given situation. However, they do provide an illusion that they are making free choices.
We can accept that the ghosts in Pacman do not have freewill because they inhabit a simple deterministic world, but what about more sophisticated entities such as humans? Surely we have freewill. After all, our behavior is much more complex, and we certainly feel free?
Intuitively, many declare that we most certainly do have freewill. But we have to remind ourselves that the world we inhabit is also a deterministic system—more complicated than the Pacman world, but deterministic nonetheless. There are laws of nature that are consistent throughout the Universe. These laws determine the behavior of all natural systems from planetary motion to the firing of neurons in the human brain.
Now, in analyzing the neural activity of the brain, philosophers such as Paul Churchland (1996) conclude that the brain is a type of computer. It is not a digital computer like a Pacman machine; rather, it is device known as a connectionist network. Interestingly, since computer scientists can simulate connectionist networks using traditional digital computers (Copeland & Proudfoot, 2000; Fischetti, 2011), it is possible that future scientists will be able to represent the neural activity of the entire brain as a digital computer program. Such a program would be unimaginably more complex than the Pacman program. However, despite its complexity, scientists could, in principle (although unlikely in practice) use it to make accurate predictions about a person’s behavior. They would do this in the same way that Toru Iwatani predicts the behavior of Pacman’s ghosts—by collecting complete information about the current state of the world, analyzing the current state of a person’s memory, then working through the brain’s program to discover the person’s next action.
Imagine, for example, that scientists have a printout of Fred’s brain. This printout contains the complete brain program and a full memory download. Armed with this printout, along with complete knowledge of the laws of physics and current state of the world, scientists predict that at midday on July 23rd Fred will murder his business partner, Robert. Imagine now that Fred does indeed commit murder just as predicted. The scientists knew that Fred was going to commit the crime before Fred made the decision. In this way, it seems that Fred’s decision was not free. In fact, with all the conditions in place, it would have been impossible for Fred to choose anything other than to murder Robert. Why? Because if he was to choose anything else, the scientists would have predicted that outcome. You can imagine resetting the world to it’s state at 6am on July 23rd and watching the whole scenario play out identically—just like resetting the Pacman machine and playing an identical game.
Given that we live in a Universe governed by a consistent set of rules; and given that our brains are computational devices that operate according to the laws of physics, it seems we cannot escape the conclusion that we do not have freewill. We experience an illusion of freewill, but essentially, we are as free as the ghosts roaming around Pacman’s maze.
Churchland, P.M. (1996), The Engine of Reason, the Seat of the Soul, The MIT Press, 1996
Copeland, B.J. and Proudfoot, D. (2000) “Turing’s Neural Networks of 1948” alanturing.net, available at: http://www.alanturing.net/turing_archive/pages/Reference Articles/connectionism/Turing’s neural networks.html, [Accessed 27 December 2011]
Fischetti, M. (2011) “IBM Simulates 4.5 percent of the Human Brain, and All of the Cat Brain”, Scientific American (online), November 2011, available at: http://www.scientificamerican.com/article.cfm?id=graphic-science-ibm-simulates-4-percent-human-brain-all-of-cat-brain, [Accessed 2 January 2012]