There's an interesting chart in the Wikipedia article on the argument against free will: http://en.wikipedia.org/wiki/Standar...inst_free_will
The chart shows these four mutually exclusive options to pick from regarding free will and determinism:
A. Hard Determinism: no free will, physical determinism is true
B. Compatibilism: free will, physical determinism is true
C. Hard Incompabitilism: no free will, physical determinism is false
D. Libertarianism: free will, physical determimism is false
If one holds the metaphysics that we are machines rather than organisms, none of the four options are satisfactory and the problem of free will is difficult. The reason it is difficult is because there are few, if any, agents recognized by this metaphysics who are able to make a choice.
However, if we look at our own experiences as organisms with the ability to choose, unburdened by the machine metaphor, it is easy to see which of the four options is correct:
A. Hard Determinism: Unscientific, our personal experience falsifies it
B. Compatibilism: Irrational, because it claims contradictory things are true
C. Hard Incompatibilism: Unscientific for the same reason as A
D. Libertarianism: True
It makes one wonder how anyone could choose A, B or C, and yet they do. A lot of this has to do with their philosophical approaches to quantum physics.
Those who choose A or B, don't believe that quantum indeterminism is real or maybe it will be overcome in the future. They can be dismissed as unscientific. They refuse to look at evidence.
Those who pick C, accept this indeterminism, but claim that what indeterminism implies is that "chance" or "randomness" is involved and a machine is not free just because there is randomness. Well, first of all, we are not machines. Second, quantum indeterminism is not "random" with a uniform distribution. The probabilities for different events to occur are not all the same. This non-random indeterminism doesn't help a machine get free will, but it does make this quantum indeterminism look more interesting, almost as if we could use an organism model even at the quantum level.