The Incoherence Of The Free Will Hypothesis

Before we can ask ourselves whether we possess a free will, we have to ask ourselves what free will is and what criteria we might accept as proof of its existence.  If we pose the hypothesis that free will exists, then that hypothesis must be testable, which is to say falsifiable.  What experiment would you use to prove that free will exists or does not?  

At first it seems easy to construct such an experiment.  Suppose that we developed a machine that could perfectly predict human behavior via a sophisticated mathematical model of psychology. We'll call this machine the Sorter. The makers of this machine suppose that if the Sorter succeeds in predicting the behavior of its subjects, then free will does not exist. However, the converse doesn't obtain. The failure of the Sorter to predict human behavior does not mean that free will doesn't exist. The hypothesis may live another day, but one could always argue that a better Sorter might still prove it false. With this line of reasoning we would have to conclude that we can never know if one possesses free will because a better, more accurate model of human behavior might always be possible.

Having stated that, we must step back and ask the question again: what is free will?  What have we proven false if the Sorter succeeds?  The nature of the experiment implies that free will is the ability to act in a way that cannot be predicted by any model, no matter how sophisticated.  This is tantamount to stating that what governs human behavior is either pure chance or some principle outside the descriptive powers of science.  

Neither of these thoughts is very satisfying.  If pure chance governs our behavior, then there is nothing we can do to change the outcome of our lives.  More significantly, if society is the aggregation of human behavior, then it cannot achieve momentum toward a common goal.  Over small intervals this may appear to be the case, but even a coin toss can come up heads ten times in a row.  Over large sample sizes regression to the mean is inevitable.  

The other option is that human behavior is essentially miraculous.  What we are saying is that science cannot predict it, nor can it ever predict it.  Thus, it must exist outside of logic, cause and effect, or any principles for decoding the universe that the human mind could ever conjure.  In this case the concept is incoherent by definition.  That is, we are defining free will as something the human mind can never understand.  The first question I might ask is how this is distinguishable from pure chance.  While we can suppose the existence of this super-natural principle that we cannot even describe, we have to live with the knowledge that we cannot control it to achieve our goals and that we cannot ever know if it exists or if our lives are really are victims of probability.  Either way, it is not related to human consciousness or subject to human desire.