Simple argument for a deterministic Universe-Please refute

[Faith In Reas]

A friend of mine came up with an annoyingly simple argument for a Newtonian, deterministic, causally necessary universe. I can't see how to refute it

Based on 2 things-
The "Principle Of Least Action"
A particular definition of "Event".

That's it!

"Principle Of Least Action"-
The "easiest" "event" always happens first.

"Event"-
Smallest possible unit of change. (The smallest possible difference between 2 states of affairs)

So-
You have;
S.O.A (state of affairs) "A" @ T1
then-
"Event"
then
S.O.A (state of affairs) "B" @ T2

That's it!
If you commit to the "2 things"
S.O.A "B" must neccessarily follow S.O.A "A".

ie "Causation" could be seen as the "one to one association of (dynamic) states of affairs with the "easiest" events to which they give rise. (nothing metaphysical being postulated here)

In principle, therefore (even if you object to the use of the term "Causation"), the ever changing world can be fully described in terms of the cascade of infinitesimal "events", each of which (according to the "Principle Of Least Action") is the first to arise from the S.O.A preceding it.

PLEASE refute this!

 

[Raze2dust]

what if there are two least action paths?

 

[f95toli]

You are basically using the same argument that was used by many of the philosophers/scientists that came after Newton (e.g. Laplace); and a "Newtonian" universe would presumably we deterministic.

However, the problem here is your first assumption: The principle of least action in its original form (which is what you are stating) does not really apply to quantum mechanical systems since they are inherently stochastic; all we can ever calculate is the probability that something will happen and even when there is a "most likely" event that won't happen every time; i.e. even if there is e.g. a 99% chance that event A will take place this still leaves a 1% chance that event be will happen instead.
I.e. we can never be sure which event will happen.

Hence, if you by determinism mean "the ability to predict the future", no the world is not deterministic.
(but of course that does not rule out "hidden" determinism, meaning the world is deterministic but we can't predict what will happen).

 

[Faith In Reas]

what if there are two least action paths?

Then you look closer. One must happen first. That is the smallest/simplest event.

 

[f95toli]

Then you look closer. One must happen first. That is the smallest/simplest event.

No, there is no "must". The classical least action path is often the most likely outcome; but other outcomes are usually possible (but you need QM to calculate the probabilites)

 

[Faith In Reas]

No, there is no "must". The classical least action path is often the most likely outcome; but other outcomes are usually possible (but you need QM to calculate the probabilites)

I meant in the sense that whatever event occurs, one will occur first. If it seems two indentical events are occurring, then if u look closer, one of them will be "easier" to occur than the other. That one will be the one that occurs first.

BTW-I'm playing Devils advocate here.

 

[Cvan]

I meant in the sense that whatever event occurs, one will occur first. If it seems two indentical events are occurring, then if u look closer, one of them will be "easier" to occur than the other. That one will be the one that occurs first.

BTW-I'm playing Devils advocate here.

Collapsing Wavefunctions upon OBSERVATION (interpretation), Quantum Mechanical interpretations and representations, Quantum weirdness, symmetry breaking, 'matter' prevalent over 'antimatter', second law of thermodynamics as a statistical-- but not without deviations from the standard on small scales-- law.. Virtual particles being created and destroyed within Planck times..

I had a similar line of thinking to you at one point, and it always sticks in the back of my head. Quantum Mechanics and other active and unresolved problems in physics have swayed me away from those ideas though. While the implications of a deterministic universe are startling, and may very well in the large scheme of things be plausible--there's no reason to dwell on the impossibly. In the event that you are correct, and the universe is deterministic in it's nature, well--what would that change?

 

[f95toli]

one of them will be "easier" to occur than the other. That one will be the one that occurs first.

No, according to classical physics that would be correct. But according to QM the "easiest" event is merely the most likely event; in many cases (including most situations in the macroscopic world) the probability for the "classical outcome" will dominate but it will never be 100%.
There are plenty of examples of quantum mechanical systems where the statistical distribution of possible outcomes is quite wide, and while the centre of the distribution will be the "principle of least action" most outcomes will NOT be near that path.

(I actually used work on one such system, I was studying macroscopic quantum tunnelling in a Josephson junction with a 2D potential).

 

[robertm]

It is not easiest for all the air in a room to suddenly concentrate on one side, leaving a vacuum on the other. However, there is no reason other than probability that the air doesn't normally do this. My point being that the easiest event doesn't always happen, only most of the time.

The macro universe usually works in a classical sense, however the quantum universe is a whole different world.

 

[kenewbie]

It is not easiest for all the air in a room to suddenly concentrate on one side, leaving a vacuum on the other. However, there is no reason other than probability that the air doesn't normally do this.

Uh? Once one area of a system is denser than another, the colliding molecules that ricochet towards the dense area are more likely to be bounced back than those going directly into the empty space. Thus brownian motion will always work to stabilize a system.

k

 

[Lojzek]

So-
You have;
S.O.A (state of affairs) "A" @ T1
then-
"Event"
then
S.O.A (state of affairs) "B" @ T2

That's it!
If you commit to the "2 things"
S.O.A "B" must neccessarily follow S.O.A "A".

The principle of least action determines the movement if the states at times t1 and t2 are already known! You assumed that B follows from A to prove that B follows from A! Note that the state at t1 is not enough to predict the future: two boundary conditions are necessary.

Principle of least action must be adapted to predict the future from a state at a single time:

I will illustrate this on classical mechanics:

Let's say we know all the positions xi(t1) and velocities vi(t1). To predict the future, we must go over all possible positions xi(t2) and select the one for which least action principle (with boundary conditions xi(t1),xi(t2)) implies movement with vi(t1).