Saturday, July 21, 2007

Free will and Quantum Mechanics

Before quantum mechanics showed up on the physics scene, life was literally more predictable. Classical physics gave us a deterministic view of the universe and it seemed that, if we somehow knew all the variables in the universe, we could predict what would happen in the future. It was as if God created the universe with a predetermined set of starting conditions and variables, and then left it to run its deterministic course - just like a mechanical clock after having been wound up. But if all is predetermined, what can we say about free will? Does it mean that no matter what we do, our future has already been determined? We may have the illusion of free will but is it just that - an illusion? Then along came a saviour, in the form of quantum mechanics. With quantum mechanics' statistical view of the world we suddenly had indeterminism. For example, if we have a box full of atomic nuclei, we can predict a time when half of them will decay, but we cannot calculate precisely when an individual nucleus will decay. Also, we cannot precisely know both the velocity and the position of an electron, so we can never know the exact starting conditions of a quantum system. Albert Einstein refused to believe in a statistical universe - he thought that quantum mechanics was incomplete and that there were "hidden variables" that could account for the quantum weirdness. John Bell's theory in 1964, followed its experimental confirmation in 1982, has shown us that there are no such hidden variables, and we must live with the weird world of quantum mechanics. But there are still some die-hard physicists that hope for a "classical system" to explain the statistical nature of quantum mechanics. One such physicist is Nobel Prize winner Gerard t'Hooft. In 2006, he wrote a paper The Mathematical Basis For Deterministic Quantum Mechanics where he suggests that, even though particles appear to behave unpredictably, we could indeed predict their behaviour by tracking their underlying states.
When this paper was released in 2006, the magazine New Scientist reported that, if t'Hooft's theory was correct, it would spell an end to the concept of free will. I will assume that the current prevailing view of quantum mechanics is correct and that it is, indeed, probabilistic ... but it doesn't matter.

How do we define free will? One way is: A decision that is not strictly determined by what has gone on before.
British mathematician Roger Penrose linked quantum mechanics to consciousness by proposing that quantum behaviour in the protein structures found in cells (including brain cells) can give rise to consciousness. And consciousness is needed in order for us to have free will. I would like to write a little about Penrose's theory in a subsequent post. But suppose that a decision in my brain is triggered by a quantum event. Since this event is purely random, how could it be due to my free will? By trading determinism with quantum uncertainty, we have transformed the decision from a causal one to a random one. Furthermore, other questions arise when we try to link quantum mechanics with free will:

Even though quantum mechanics may not be deterministic, it does have causality - i.e. there is always a reason or a cause for any phenomenon. For example, if we fire an electron through a double-slit system towards a screen, we will have no idea where precisely on the screen the electron will end up - no matter how carefully we aim (in fact, we don't even know through which slit the electron passed through); but we do know that the electron has ended up on the screen because we caused it to go there. So life is not completely random.

Although, at the quantum level, particle interactions are probabilistic in nature, on larger scales the effects of the uncertainty principle become negligible so that interactions in the macro (real) world become deterministic. Our brain cells as a whole are not of a quantum nature so we don't expect them to behave randomly.

If t'Hooft's theory is correct, and there are hidden variables, then quantum randomness is ruled out but not free will. Something can be deterministic but not computable. Even classical mechanics, with its much-vaunted determinism, is unpredictable. An example is the n-body problem of determining the behaviour of n interacting masses where n is equal to or greater than 3. These masses can be billiard balls or heavenly bodies or anything. We cannot compute with exactness how they will behave. So even a supposedly deterministic system has elements of unpredictability (randomness?).

So, in my opinion, we cannot draw any conclusions about free will from quantum mechanics, and I can't understand why people continue to link the two concepts together.