Ian,
If hypothetically someone had in their possession all possible knowledge about you, and how you would react under any given circumstance (or alternatively consider an omniscient God), and therefore could, with 100% accuracy (or very close to it if we consider the intrinsic randomness), would this by definition mean that all your behaviour is "determined" or follows rules?
Well, since there intrinsic randomness, no. But if, for purposes of the thought experiment, we ignore that, then yes.
Also, suppose someone were to choose to move their limbs in an arbitrary manner (not randomly, we cannot act randomly). Would you still maintain their behaviour is following some rule?
I don't understand the question. If that person is not acting randomly, then what is arbitrary about his movements? You even cited arbitrariness of choices as your definition of random.
Now here's a thought experiment for you:
Imagine you are in a situation where you have two choices (A and B). Let's say you use your free-will to select choice A.
Now, you claim that this selection process is not deterministic, since that would imply that it was inevitable that you would choose A, and not B. Your conception of free-will is that even though you chose A, you could have chosen B instead.
You also claim that the selection process is not random, because that would imply that the choice was made arbitrarily, within the constraints of some relative probability for the two choices.
But what does it mean to say that the choice was not arbitrary? Would you agree that this simply means that there was a
reason you chose A, rather than B?
And if that is the case, then how was the selection of A not inevitable? If the reason that you chose A did not logically imply that you
would choose A, then in what sense can you claim that it is, in fact, the reason that you chose A?
Either the choice was arbitrary, or there was a reason for it, in which case it was inevitable. This has nothing to do with naturalism, or with any assumptions of being able to model the process based on our observations. It is simply a logical statement that is true for any hypothetical selection process.
Loki,
...because the bit about it being neither deterministic nor random does not seem to actually be part of his definition of free-will,...
--------------------------------------------------------------------------------
You're being very generous here.
I think I am just being consistent. I criticized Ian before on the basis that saying "Free will is neither deterministic nor random" is not a definition of free-will, but rather simply a statement about what it. Now I am simply sticking to that position. He has asserted that his conception of free-will is that an immaterial "self", which is not part of this Universe, interacts with the brain, and makes our decisions. I think that we need to acknowledge this definition before we can make any progress in determining whether any additional statements he makes about it are coherent or not.
Peskanov,
Would you agree that a probabilistic model for human behavior should be possible? Keep in mind that I am probably every bit as convinced as you are that human behavior is not deterministic
--------------------------------------------------------------------------------
How is so?
Do you think that a deterministic model of the brain is impossible?
That depends on how accurate you want the model to be. Even an individual neuron cannot be accurately modelled without taking noise into account. And there are many examples of statistical mechanics playing a major role in brain processes. For example, the visual system exhibits a behavior known as stochastic resonance, which is a well-known property of noisy nonlinear systems.
In my experience with artificial neuronal nets, complexity arises from the model, which is fully deterministic.
Do you think quantum indeterminism plays any meaningful role in brain's behaviour?
Not as such. What I am talking about is just ordinary thermal fluctuations. A neuron operates by transferring ions across membranes. Those individual ions are subject to thermal noise, and operate according to statistical mechanics. This has the net result of making the firing timing of the neuron a stochastic process, rather than a deterministic one. In some cases, it even results in the neuron firing when it otherwise would not.
Now, we could always model this noise by coupling our neural network to a pseudo-random number generator, but I submit that for all practical purposes, that is still a stochastic model. In fact, that is exactly how we model stochastic differential equations in practice, right now. Also, it is entirely possible that we may be able to construct a deterministic AI, but that would not be a model of the human brain. It would be something else.
Dr. Stupid