Ah, armchair philosophers and the undying false dichotomy between determinism and non-determinism.

And of course the utter red herring of "Free Will".


It's kind of depressing how often "logic" and "computer science" are mentioned as the impetus for these discussions, and yet nobody seems to be actually educated as to the state of the art in those fields.

Goedel incompleteness has no bearing whatsoever on determinism, by the way. It only relates to formal systems, and the universe is quite clearly not required to be encompassed by any formal system.

Also, there is a third state known as indeterminacy which is far more important to this topic than any of the rest.


The universe is not conclusively non-deterministic in any strictly meaningful sense because that mandates a deeper understanding of its operating principles than we actually have. It can, at best, be said to be indeterminate. Also, the idea of a deterministic universe presupposes the existence of some formal axioms from which we could describe all operating phenomena of said universe; it is far from clear whether or not this is the case.

According to the best experimental models we have, the universe is probabilistic in nature, which is far from saying it is non-deterministic. We just don't have a complete model for explaining the observed distributions of probability.


Frankly, this is ground that has been trodden to death over the past few hundred years, and if any genuine insight is to be had from the whole affair, it's going to come from people who are deeply familiar with the existing body of thought on the subject. Everyone else is basically playing a game of catch-up with Romantic era philosophers who had a lot more time on their hands to think about this stuff. (If you can get all the way from Descartes to contemporary postmodernism by yourself without any additional insight, you're pretty damn smart.)

I have often found that a ram constraint in one of my programs caused me to be more creative and make a much better program. Would this coincidence apply to real life events as well? The less memory one has the more effort one must put into using one's brain, thus one becomes smarter? Is that what smart is?

I would say that in Computing and in many other fields geniuses are found when he/she has achieved something from nothing. I mean, he/she found out something when they had miminum resources at their disposal. So yeah, I think this is no coincidence.

On to the Determinism vs Non-Determinism:
There are evidences, in my vision, of both things occurring in our universe.

For example, when the sperm meets the egg meiosis may start. If it does, the outcome is totally unpredictable. The genes will be combined through cross-over and mutation at rates and conditions unmeasurable.
You could go deeper in that thought (which is heavy) and try to reason why each of our cells live. Why do they seek resources and multiply. We have been studying this and we understand how it works, but the motivation of nature in its biological processes is quite unpredictable.

An example of determinism would be the laws of physics. Gravity, for instance, has many explanations for why it happens, but the outcome is known and single: things are attracted to a bigger body.

Now, just to heat up the discussion.
For those that believe that the universe has an inner formula (unknown to us) and that everything can be synthesised deterministically, do you believe that we will reach a deterministic-polynomial solution to the NP-Complete problems? Or is this an exception to the rule?

Frankly, this is ground that has been trodden to death over the past few hundred years, and if any genuine insight is to be had from the whole affair, it's going to come from people who are deeply familiar with the existing body of thought on the subject.

You are right that this has been discussed by far too long time and that we need an insight.
However I believe this insight will not come from people like us, in a forum, ensuring our already formed opinions, nor from people that study the matter. Because those questions, IMO, are not based on knowledge nor on linear progressions of thought.

We really need an insight from a genius (with little resource and humble thoughts) or from an exterior superior force (?alien? / ?Gods?).
I believe in many enlightened people that are born from time to time, since philosophers thousands years ago, to George Boole and Alan Turing.

Ah, armchair philosophers and the undying false dichotomy between determinism and non-determinism.

I'm going to stop you right there. Is that what you actually meant to say? Because I'm pretty sure most people wouldn't object to defining determinism of some outcome as something like "there exists some model that, given the appropriate inputs, will accurately predict the outcome" and non-determinism as "there does not exist such a model," in which case it's not a false dichotomy at all.

[quote name='ApochPiQ' timestamp='1353379945' post='5002548']
Ah, armchair philosophers and the undying false dichotomy between determinism and non-determinism.

I'm going to stop you right there. Is that what you actually meant to say? Because I'm pretty sure most people wouldn't object to defining determinism of some outcome as something like "there exists some model that, given the appropriate inputs, will accurately predict the outcome" and non-determinism as "there does not exist such a model," in which case it's not a false dichotomy at all.
[/quote]Well it's hard to see through the smug cloud but maybe it's a false dichotomy because a "dichotomy" when drawn as a Venn diagram is exactly two non-overlapping circles, but you can instead choose to step away from those two and say "you know what, there's no way to tell either way, so I'm going to draw a new circle that overlaps both of yours and go do something useful".
Kind of like answering with mu ("unask the question").

If we assume there's a multiverse, made up of all the universes from all the possible initial conditions, then it's possible to hypothesise that different laws of physics apply to different 'universes'. In that case, perhaps some of them could be deterministic, and others non-deterministic, which makes both positions are true.

Everyone else is basically playing a game of catch-up with Romantic era philosophers who had a lot more time on their hands to think about this stuff.

To be fair, many of those philosophers also believed that humans were half material and half magic, that mice spontaneously came into existence from filth, knew nothing of probabilistic physics or the geometry of the universe, and fields like neuroscience (which turns some of their philosophising into real, testable science) were centuries from being discovered.
Perhaps starting from scratch with a modern scientific view of reality would allow a new generation of philosophers to tread completely new ground, or at least skip past a few dead ends?
Half of Descartes "breakthroughs" were just describing fictional systems of integrating magic with biology.... which we should really laugh at today.

Well it's hard to see through the smug cloud but maybe it's a false dichotomy because a "dichotomy" when drawn as a Venn diagram is exactly two non-overlapping circles, but you can instead choose to step away from those two and say "you know what, there's no way to tell either way, so I'm going to draw a new circle that overlaps both of yours and go do something useful".
Kind of like answering with mu ("unask the question").

Yeah, that does make sense. I still think it's worth differentiating a false dichotomy from a useless one, though.

I think one can ask the following questions:

1 - Can an external observer determine the state of the universe at any given time t given the initial state of the universe(at t=0) and the laws of physics?
Let's assume that the external observer cannot 'affect' the universe, he can only observe and conclude.
It all depends on, I think, the laws of physics. If we assume all the laws of classical mechanics are true, then the answer will be yes because none of the laws in classical mechanics have a random element. With the advent of quantum mechanics I am not sure if we can come to the same conclusion, because AFAIK quantum mechanics has random elements in it.
One way of looking at it is:
A function, say f(x) takes the current state of the universe x and returns the state of the universe at the next 'moment of time'
If the function is one to one then the external observer can determine the future state of the universe.
If the function is one to many then the external observer will not be able to determine the future state of the universe.
An example of a one to one function is f(x) = x+1, here every value of x will have only one corresponding value of f(x)(0->1 , 1->2 etc)
A one to many function would be something like f(x) = sqrt(x), here if x is say 9, f(x) can either be -3 or +3 therefore there is no way of determining what is the exact value of f(x) and therefore the external observer won't be able to determine the future state of the universe given a function like f(x).

2 - Can an internal observer determine the state of the universe at any given time t given the initial state of the universe and the laws of physics?
By internal observer I mean people like you, me or any random scientist. Can we predict the future given enough knowledge and computational ability?
If the laws of physics have random elements then ofcourse we can't using the same explanation as above.
But if the laws of physics are non-random?
What happens if I predict my own future then try to falsify it on purpose.
For example, I, knowing myself and all the laws of physics, predict that tomorrow morning I will pick up my cup of tea with my right hand. But tomorrow morning I will pick it up with my left hand instead, just to prove myself wrong. Ofcourse when I first predicted this, I would take that into consideration and conclude that because my future self would like to prove my past self wrong, so he will pick it up with the left hand instead. But then my future self, knowing I had predicted that he would pick it up with the left hand, would now pick it up with his right. And the loop goes on and on...
Imagine the following scenario:
There is a Robot that is programmed to predict his the future.
The robot knows the initial state of the universe and all the laws of physics.
What is the robot is going to do is simulate an 'alternate reality' using the initial state and the laws of physics.
In his alternate reality, he will have the simulate himself too(because he is a part of the universe).
His simulation of himself will also try to simulate the alternate reality.
His simulation of himself then will have a simulation of itself(which is the simulation of the simulation of the robot).
And the simulation of the simulation of the robot will again have a simulation of himself.
So using this sort of brute force algorithm of simulating the future, the robot will run into an infinite loop and he will not be able to predict the future.

This was only an example of how an algorithm for future-prediction can be made and does not prove that an internal observer cannot predict the future with a non-random ruleset for their might be another algorithm that can accurately predict the future. But until that algorithm is found, we cannot say that an internal observer can determine the future state of the universe.

All my thoughts on the subject in a nutshell.

AFAIK quantum mechanics has random elements in it

Too far, too fast.

How do we know that quantum events are random? Because we can observe no discernible patterns.

Is that sufficient to prove randomness? No, because we can't see the entire sequence (and even if we could, we might not be able to figure out the derivation - imagine looking at a random slice of 100 digits of pi...)

[quote name='MathAddict' timestamp='1353443140' post='5002745']
AFAIK quantum mechanics has random elements in it

Too far, too fast.

How do we know that quantum events are random? Because we can observe no discernible patterns.

Is that sufficient to prove randomness? No, because we can't see the entire sequence (and even if we could, we might not be able to figure out the derivation - imagine looking at a random slice of 100 digits of pi...)
[/quote]
I refrained from using quantum mechanics because I don't know much about the subject.
What I meant to say was - if physics depends on randomness, determinism is not possible. The quantum mechanics thing was just a real life example.


EDIT:
[media][/media]
According to that, there is a way to check whether 'this sort of classical underlying explanation of quantum mechanics can exist even in principle' and it turns out there isn't.

Disclaimer: I haven't actually watched the video with the Einstein doll.

Quantum mechanics doesn't say that randomness exists; it still allows for hidden variables but puts constraints on how they can manifest physically. Bell's theorem, for instance, shows that quantum mechanics cannot be described in terms of local hidden variables, but this is a much stricter definition and says nothing about the kind of hidden variables we imagine in, say, a computer simulation (e.g. a global pseudo-random generator).

There are also constraints about the kind of predictions we can make, related for instance to the fact that we can't make a simulation of the universe that a) fits inside the universe and b) contains a perfect simulation of itself. This too is sad but says nothing about whether the universe is actually deterministic.