John Henry

Assuming that the universe is fully interconnected, so any closed system is subject to external influences, and given that any system of modelling is a part of the universe and therefore smaller than it, is it impossible to make any predictions with complete certainty?

To give an example of what I mean, you could have a simple model of a toy railway, which would predict that it would start and stop when you pressed the appropriate buttons, and would keep going round the same loops until you interfered. Such a model would not predict that your brother might come and kick the train off the rails. You could extend the model to include your brother and make predictions involving the possibility of him kicking the train off the rails. However, the new model would not take into account the fact that he might get hit by a meteor, etc...

Eventually, for a completely reliable model of where your train would go, you would need to model the whole universe, which would surely be impossible to do with anything smaller than the unioverse itself.

TheKak

You can see meteors coming and predict when they will get here, so I dont think your model would have to be as big as the universe.

__________________
Roses are red, violets are blue, I'm a schizophrenic and so am I.

Mantus

I suppose that is the difference between predicting and knowing the future.

Prediction involves odds, while knowledge of the future involves certainty.

Giltwist

This does lead to an interesting thought. THe laws of physics demand that every object experiences gravity from every other object in the universe. Also, the human mind is the most powerful computer known to man. These two together easily lend to the credibility of. at least limited, precognizence.

CSflim

Well I guess the two most important objections to this brand of determinism are scientific ones.
1. Chaos Theory
2. Quantum Mechanics

Chaos theory shows how exceedingly difficult it is to make predictions for non-linear systems. You might intuitivly belive that if we have a model of a system, with inititial values 99.9% accurate, you would expect to get precistions as accurate or say reasonably accurate say 90% accurate. Chaos theory shows that this is not the case. In order to get accurate results, you would in fact need to have incredibly accurate initial measurements...and even then your predictions will only be valid for a short time into the future. If you want to make predictions for an arbitrary length of time, you will need initial measurements 100% accurate! (Bear in mind that a number to signify the position of a single particle would have to be infinitely long!).

Quantum Mechanics shows that nature appears to be random at heart...not nicely deterministic like Newtonian mechanics. Some people attribute this randomness to our lack of information about what is "going on down there", others believe that unpredictability is an objective property of nature. The jury is still out on this one, although the current scientific thinking tends towards randomness as something real.

Both of these things paint a pretty grim picture of our ability to predict relatively small things, let alone an entire universe.

__________________

stingc

Chaos isn't a problem at a fundamental level. The systems normally studied in that field are still deterministic at least in principle.

But quantum mechanics (probably) removes any hope of classical determinism.

thingstodo

I think the chaos theory is pretty cool. The first time I heard of it was reading Jurasick Park. The whole concept makes predicting anything at least some sort of a gamble, even a sure thing. And the more time involved, the greater th chance for an unexpected outcome.

A butterfly in China flaps its wings....

__________________
If you're wringing your hands you can't roll up your shirt sleeves.

Stangers have the best candy.

Giltwist

But the point is, over very short periods of time on very limited scales, chaos theory does NOT preclude fairly accurate foreknowledge.

Journeyman

I've long toyed with the redefinition of the universe as a computer. Set your initial formulas, input how much value any unit has, set a starting point with all units spatially and chronologically in synch: Press enter.

So it could be that we're living in the most powerful computer, it's just not "known to man" as such.

CSflim

Exactly.
We can all make "fairly accurate" predictions over short lengths of time. But that wasn't the question posed by the thread starter.

__________________

Giltwist

Sure it is. He asked if it is possible to make any possibilities with certainty, and I said you can, if you make your field of interest small enough. Of course, if you make it too small, quantum mechanics takes hold and you have Heisenburg's uncertainty. But at the right size, you can make predictions as "certain" as you believe things can be.