One issue that both commentators mentioned was the vast number of nerves and synapses in the brain would seem to argue against the proposition that an action was determined by a single molecular event (opening or closing an ion channel.)
In order to address this issue, please consider what is the function of the brain. First and foremost it is a reaction organ. It is a more sophisticated pathway, but exactly similar in function, to other reflexes such as the knee jerk. Incoming information is processed (fires through neural networks) and a response is triggered. Again, the Either/Or of General Loan comes to mind. Either he pulls the trigger or he doesn't. There is no in between state.
It is true that this is most likely not a sequential process. That is, there are probably many neural pathways firing in parallel. However, we would argue that the outcome of the billions of synapses still must funnel through the final common pathway and this, in turn, depends on one ion. (I cannot argue that the threshold for the action potential might not be different at different times. In fact, it is doubtful that exactly the same physical pathway is followed with each action.) What is important, though, is that it can only be one final nerve firing. If the nerve fires, the action takes place. If it doesn't, it doesn't. It makes no difference if there are billion or a trillion nerves involved. It is much like nuclear fission. There has to be the threshold of a single neutron between no reaction and boom.
On the other hand, if one agrees that the primary purpose of the brain is to generate reactions to external stimuli, then it would make evolutionary sense that the same stimulus would generate the same response. (The difference between a squid axon response and a human's response to Beethoven's Apassionata is merely a matter of degree.) Thus, it does not bother me that the threshold of a single nerve might change. As long as the outcome is consistent (not that humans are ever consistent.)
Now I appreciate the creativity of a scenario that invokes broken determinism. However, the real point of the argument for Free Will is that it is something transcendental that determines something physical, i.e. the biochemical reactions of the neural network. Now clearly not all of these neural reactions could be under the control of something transcendental. For instance, you would exclude synapses in the peripheral nervous system including the eye (which, as we showed before, does some primary processing of visual input in the retina.) But how deep of a level do you go in the brain to invoke the interaction? Where are the synapses that are under transcendental control? In the pineal gland (thank you Rene)? (I have the same problem with miracles).
For instance, while it might be true that the recognition of a molecule in the nose might involve quantum mechanical tunnelling, the transduction of that signal almost certainly involves G proteins just as in the eye (see posts on this in the Information series). Or if you are really interested:
The process of how the binding of the ligand (odor molecule or odorant) to the receptor leads to an action potential in the receptor neuron is via a second messenger pathway depending on the organism. In mammals the odorants stimulate adenylate cyclase to synthesize cAMP via a G protein called Golf. cAMP, which is the second messenger here, opens a cyclic nucleotide-gated ion channel (CNG) producing an influx of cations (largely Ca2+ with some Na+) into the cell, slightly depolarising it. The Ca2+ in turn opens a Ca2+-activated chloride channel, leading to efflux of Cl-,Let them eat chloride ions.
Yes, I suppose that Chemistry is a subscience of Physics. Being trained as a chemist I just don't want to think about it (those uppity physicists). However, pertinent to this discussion, when physicists start talking about Free Will and Quantum Theory, they argue in the abstract. They do not talk about real things, at least in my opinion. I recognize to a degree the restraints put on determinism by the Heisenberg Uncertainty Principal. And, I have much to read on this as suggested by Unreal Nature. So, I consider this an open discussion. Interestingly enough, Steve Wolfram has something to say about Free Will. We can all get in the ring.
Finally, let us propose a gedanken experiment. Take all the nerves in the brain involved in a given outcome (alright, General Loan pulling the trigger). Array those nerves along a time axis. There may be billions at first, but as the "decision" point is approached, the number will decrease until there is only one neuron left. We can say this because we employ the Laplace demon to slice time in such a way that every single (re)action in the universe occurs at a different time. Along the time axis going forward after the single neuron there is either no impulse (i.e. didn't make the gate), or an increasing number of neurons firing to produce the result.
I appreciate anyone sticking with me this far. I agree with The Growlery, it is an exhausting exercise. I know I have not truly addressed the objections. But then:
You say daemon, I say demon,
You say tomato, I say tomatoe