“in a classical deterministic system, with perfect information and arbitrary computing power you can predict the future in principle, but not in practice”

Does that mean you will be able to predict if any given Turing Machine will halt or not?

Does this lead to a contradiction?

In a purely deterministic world, fraudulent scientists would have the same respect than Nobel Prize winners. Because none of them would have the most minimum possibility to chose their own actions (good or bad).

In a world where a nondeterministic evolution is possible, we would be giving Prizes to people, because the creation of a scientific theory is not a deterministic process, but the outcome of a wise mixture of human intelligence, perseverance, and personal choices; whereas no prize would be given to a falling rock, because the rock is merely following a law in a passive way. We would recriminate fraudulent scientists, because they had the option to chose their actions and decided to make the fraud.

Natural selection – or even just the simpler truth that survival is not guaranteed – dictates that the intellectual processes of an organism must have some capacity to represent the world correctly, or else it will swiftly die. Meanwhile, the modern theory of computation (due to people like Turing) tells us that a relatively simple set of symbol manipulations is “computationally universal”, capable of doing anything that a modern computer can do. So above a very elementary threshold of computational ability, cognitive dispositions selected merely for compatibility with survival will also give rise to open-ended powers of rationality, bounded only by restrictions on memory, sensory bandwidth, etc.

In other words, the argument is:

1) The need to survive dictates that cognitive processes have some fidelity to reality.

2) Turing universality tells us that it’s a short step from “cognition with some fidelity to reality” to “cognition with an open-ended capacity to analyse data and draw correct conclusions”.

It is absolutely true that we may be caused to make mistakes. I mean, I believe in cause and effect, and I believe that people make many mistakes, therefore I believe those mistakes have causes! But I don’t believe that determinism implies the uselessness of science or of thought. Cause and effect can be our epistemic ally too.

Cheers,

James Goetz

Just as we cannot know scientifically (at this time) whether our lives are meaningful, in my opinion, we cannot know whether free will exists or not based on current science. Arguably, we’re no closer to resolving the free will debate than we were 100 years ago or in the time of Newton.

Most of us act, then, on the assumption that we have free will.

Whether we can answer that definitively, at some future time, remains to be seen.

First, many-worlds is not another interpretation of QM, as one reads sometimes, but a well-known misunderstanding of QM that cannot reproduce what we observe at our labs

Against Many-Worlds Interpretations 1990: Int. J. Mod. Phys. A 5, 1745–1762 by Kent, Adrian.

see also

http://www.mat.univie.ac.at/~neum/physfaq/topics/manyworlds

Note that Kent article is titled Interpretations, in plural, because there is not one MWI but a collection of mutually contradictory MWIs. The MWI by Deutsch (who you cite) is not the same than MWI by Everett, which is not the same than MWI by Hartle…

Second, science is an enterprise with no room for the kind of supernatural observers G introduced in many-worlds for justifying the kind of metaphysical process associated to deterministic knowledge

http://juanrga.com/en/knowledge/a1110222009v1.html

Therefore it would be a good idea to keep in mind the limits of the scope of science when discussing about science.

and third, I am not surprised that when you write about chaos you only cite deterministic chaos (where uncertainty of final states is due to our a small uncertainty in our knowledge of the initial state of a deterministic system), whereas you omit to cite the case of nondeterministic chaos, where the uncertainty about the system remains although you know the initial state with infinite precision.

This omission of fundamental results is still more glaring when a famous Nobel laureate wrote several popular books (including bestsellers) about the current state of the science of chaos.

Does free will exist if there is no brain to ponder it? Free will would seem to require life, minds, intelligence, and probably more than a single instance (to encourage interaction, socialization, the creation of civilization, culture, philosophical thought, science, and science blogs with comments). Epistemology and existentialism, anyone?

Let’s assume free will does exist. Did it always exist? If not, then when, why and how did it come into existence?

What manner of things possess free will? Does a chimp have free will? A snake? A fruit fly? A worm? A nematode? An amoeba? Does free will require a complex nervous system? Or self-awareness?

Consider entropy and Time’s Arrow. Our own limited existence, including our free will, is merely an eddy of local, temporary order in the rush toward the heat death of the universe. It is highly localized, extremely constrained.

What is free will within the context of the evolution of the universe? Is it nothing more than an temporary emergent property, possessed by only an insignificant number of small clumps of matter?

Is free will nothing other than a rounding error in the statistics of the universe?

My brain hurts. Can we stop now?

But if we postulate the will is not an entity but a process, we are left either with deterministic processess or interministic, i.e., probabilistic ones. By definition a deterministic process forming the “will” is unfree. But a probabilistic process means that the will is random. It seems inescapable to me that there’s a real problem in associating freedom of the will with the randomness of the will.

You could get around this by treating probabilistic processes as determinate, while acknowledging the plain truth that individual trials are not. Fair coins come up heads 50% of the time, a very specific value, which as the opening question highlighted, is in fact a key aspect to determinism. (I like to think of determinism coming in three varieties: mechanism, stochasm and history.) But it sppears this is not an option. This seems to be a shame, because if any individual act of will is an outcome of a probabilistic process, the peculiar determinateness of probabilistic processes can provide the bias predictability we associate with personal character, while the inherently probabilistic nature of individual outcomes, specific acts, account for the equally real unpredictability.

If the many clauses of the last sentence left it too obscure, think of it this way. The will plainly cannot be unconstrained. If Sean has an embarrassing need for latex for sexual fulfillment, he cannot will that he will be aroused in more socially acceptable. This is not reflection on Sean. I myself cannot reliably will myself to remember facts that I know! For both of us, I suppose exercising the will to decide to cultivate good habits would constitute freedom of the will, while the subsequent habits (should we be so fortunate as to succeed in our endeavors,) would not constitute will, but, well, habit. And for both of us, delaying gratification is in no sense a defiance of needs or desires imposed upon us by deterministic processes, even though it is entirely volitional.

As for alarm at the notion that a scientific understanding of the mind will leave old ideals of morality shamed, I’d say that’s because it’s true. On the one hand, miscreants who would simply be condemned as bad stand relieved of full responsibility. The insistence upon treating them as sinners would not just seem, but be, barbarous. And those of us fortunate to have met social expectations (publicly, anyhow) could not honestly congratulate ourselves upon our probity. All this would change society and undermine religion of course. The old joke is that hell was created so that heaven would have some entertainment. How could we be religious when delight in God’s justice is philistine backwardness?

The ensuing discussion about predeterminism, predestination and whether or not these contradict free will, were fascinating anticipations of Laplace’s dilemma, and ahead of their time philosophically. A key revelation from these discussions is (as indeed you observe) that it is possible for an omniscient being (or supercomputer) to be able to predetermine the outcome of a person’s choices without violating their free will (i.e. there is no predestination). This was the view held by the Catholic Church and opposed by Calvin (who concluded that there must be predestination). It appears that this makes one philosophical point on which you, Sean, and the Catholic church’s agree — even if you disagree that the existence of the Church’s particular “Laplace Demon”

It just seems to me that, within the context of the argument, what would resolve the issue in favour of the existence of free will would be a theory that explains it. Wouldn’t genuinely self-determined behaviour be exactly not this, ie not anything that you could explain, that wouldn’t conform to laws, that wouldn’t look like any form of behaviour in the inert universe.

I think the argument over Free Will is actually an argument over the limits of the scientific method. The reason it continues to go around and around is that it is impossible for physics and biology to explain the human capacity for voluntary action, and yet that thought means science cannot answer every question, and that is unthinkable (from within the system) so it returns to the begining again. These endless iterations leave plenty of time for coming up with quasi-technical terms that make it all look like a very important discussion.

As far as I’m concerned, we learn how to be free. You can’t learn a law of physics. Therefore, human freedom doesn’t have anything to do with physics.

http://bit.ly/thoughtknot

“If that’s the case, it’d make sense for there to be a slight delay while the decision result was translated into language, or into a visualized image, or into an action.”

As an experimental psychologist, I can tell you that 300ms is an eternity in behavioral response time studies. If it really took 300ms to turn a decision into a verbal response, then it would be really hard to get verbal responses in any paradigm less than, say, 350ms. But that’s not the case. You can’t wave it away that simply.

I don’t see any reason to assume awareness would be instantaneous. When a computer CPU makes a “decision”, that isn’t instantly reflected outside of the CPU (on the screen, in RAM, etc). That’d take few more nanoseconds.

After all, in the brain the decision would be made probably by some process of weighing alternative activation patterns, in a manner that isn’t necessarily tied into the verbal or visual centers. If that’s the case, it’d make sense for there to be a slight delay while the decision result was translated into language, or into a visualized image, or into an action.

The mind may not be dualistic in the mind/body sense, but there’s also no reason to assume that it all works together as an atomic unit, with a decision being instantly reflected throughout the brain and available to be expressed.