QM and Determinism

1. Oct 1, 2014

Johan0001

"Determinism often is taken to mean causal determinism, which in physics is known as cause-and-effect. It is the concept that events within a given paradigm are bound by causality in such a way that any state (of an object or event) is completely determined by prior states" -- from wiki

Does it follow that the next state of a system is already mapped out by the history built up of prior states.
Thus without observation or interference we can determine the next state to happen, if we had all the information of prior states?

Can an event occur without any observation/measurement thereof?
This question seems nonsensical , but when one talks of observation/measurement / events in QM they
are crucial in understanding their definitions WRT determinism.

How does QM relate to determinism , if everything can be predicted given that we have collected all the information available to us in the universe.

I'm quite confused on how it fits together? I.E. QM and determinism.

.

2. Oct 1, 2014

atyy

A deterministic theory is one in which it is possible to say, given full knowledge of the state at one point in time, it is possible in principle to know the state at all other points in time.

Quantum mechanics is not a deterministic theory. In quantum mechanics, only future "expectation values" or "average values over many repetitions of the same experiment" can be known given full knowledge of the state at an initial time.

3. Oct 2, 2014

naima

And the same is true for the past. Given the result of a measure, one cannot know exactly how the state of the incoming particle was prepared

4. Oct 2, 2014

Johan0001

Quantum mechanics is not a deterministic theory. In quantum mechanics, only future "expectation values" or "average values over many repetitions of the same experiment" can be known given full knowledge of the state at an initial time.

Then which theory would describe what we observe most accurately, Determinism or QM , because they have radically different fundamentals , which seem to contradict each other.

5. Oct 2, 2014

naima

QM is a theory and determinism is a philosophical notion.

6. Oct 2, 2014

atyy

At present, QM, which is a non-deterministic theory, is our most fundamental theory, in the sense that all observations to date are consistent with QM. However, determinism and QM do not contradict each other.

This is because a non-deterministic theory can be an approximation to a deterministic theory. It could be that we don't have enough experimental control, so that what we consider a "state" is actually a different state each time we do "the same" experiment. So it is conceivable that there is a non-deterministic theory that underlies QM. If this is the case, then by improving our experimental finesse, we may one day observe phenomena that cannot be well described by QM.

It is also possible for a deterministic theory to be an approximation to a non-deterministic theory. For example, deterministic classical mechanics is a good approximation to quantum theory over some regime.

So at this stage, we cannot say whether determinism or non-determinism is more fundamental, since each can arise from the other.

7. Oct 2, 2014

microsansfil

Indeed, there is a philosophical definition of determinism:

From Stanford Encyclopedia of Philosophy:

Determinism: The world is governed by (or is under the sway of) determinism if and only if, given a specified way things are at a time t, the way things go thereafter is fixed as a matter of natural law.

A theoretical model can not be qualified "deterministic" related to the notion of causality ?

Patrick

8. Oct 2, 2014

Neandethal00

QM is a theory and determinism is a philosophical notion.

May we label determinism as Observation?
Unfortunate thing is physicists are content with nondeterministic nature of QM.
Very few are bothered by why QM is nondeterministic.

9. Oct 2, 2014

the_pulp

QM is not deterministic in the sense that, given the full information of a state, you cannot predict the output of a an experiment over this state. Nevertheless, there are a lot of interpretations of "why" this happens, and these interpretations vary in wether they are:
Deterministic or Stochastic
Local or Nonlocal
If there are hidden variables or not in the instrument of the experiment or in the state itself
...
The problem is that there are nice characteristics and not nice characteristics (in terms of what is the average man intuition) and there is no interpretation that has all of the nice characteristics and none of the not nice ones.

The interpretation that I like the most is very similar to the interpretation called "Time Symmetric Interpretation" (for a quick review of the different interpretations, see http://en.wikipedia.org/wiki/Interpretations_of_quantum_mechanics ) and it is deterministic. In a word, in this terms, we cannot say that QM is deterministic or not.

this does not abolish the general notion that given the full knowledge of a state, we can not know for certain the result of a experiment
the practical interpretation applied in real life (and what I see in the forum) is "Shut up and calculate", meaning that all this discussion is irrelevant to the actual state of science and that perhaps will never be of any use

Im just an amateur reader of QM and such so take my comments as just a little of information that should be validated or corrected by professionals.

10. Oct 2, 2014

Staff: Mentor

If QM is deterministic or not is purely a matter of interpretation eg BM is completely deterministic, probabilities appear due to lack of knowledge of initial conditions.

That being the case it's meaningless to talk about determinism in QM without reference to a specific interpretation.

Looking just at the formalism Gleasons theroem shows you can't have a probability measure of just 0 or 1, ie determinism, if you have non-contextuality which is its rock bottom essence - in this sense Gleason is a stronger version of Kochen-Specker which anyone interested in this should look into.

With our modern understanding of decoherence the real issue is the so called problem of definite outcomes. We end up with an improper mixed state after decoherence, but exactly how is a particular outcome singled out.

Thanks
Bill

11. Oct 2, 2014

Staff: Mentor

In modern times observations and decoherence are usually taken to be the same thing. The rock bottom question is the definite outcomes issue I mentioned above.

Thanks
Bill

12. Oct 3, 2014

vanhees71

It's indeed a question of definition, what you mean by "deterministic". In my definition, which I think is the most common definition among physicists, a physical model is deterministic, if the complete knowledge of the state of the system implies the knowledge of the values of all possible observables of the system, and it is (at least in principle) possible to gain complete knowledge about the state of any system. All of classical physics is deterministic (i.e., Newtonian and relativistic discrete and continuum mechanics, classical electromagnetism).

Quantum theory is not deterministic, because the complete knowledge about the state of the system does not imply the knowledge about the values of all observables. This is the content of the general Heisenberg-Robertson uncertainty relation in the minimal statistical interpretation, which in my opinion is the only one justified by physics, i.e., without additional metaphysical elements reflecting the personal view of the followers of any representation going beyond that. According to quantum theory in the minimal representation the indeterminacy of observables according to the preparation of the system in a certain state is objective, i.e., these observables really do not take any determined value, but you can predict the probability of finding a certain value of any observable.

All of today's physics is, however, causal. This means that the complete knowledge about the state of a system in the past implies its evolution in the future, i.e., the state at later times is given by dynamical laws. On a fundamental level this causality is even local in time, i.e., it is sufficient to know the state of the system at one instant of time, $t_0$ to calculate the state of the system at any later time, $t>t_0$.

For a very detailed and careful analysis of this, see the introductory chapter of

Schwinger, Julian: Quantum Mechanics, Symbolism of Atomic Measurements, Springer, 2001

13. Oct 3, 2014

moriheru

In a deterministic theory, given some information, one can predict the future of a state. In QM one only has probabilityamplitudes or expectation-values of the nth state of say some particle.

14. Oct 3, 2014

Johan0001

bhobba wrote -
With HUP we would never really know the initial conditions precisely in the microscopic world, but is it really necessary to know ?

Is it not accurate enough to say that , the most probable future state is the one with the highest probability.
"The most probable macro state is the one with the highest number of micro states"

If we can only absorb/bounce photons off particles then , what do we bounce
off photons.

atyv wrote -
I tend to believe that there is no 2 states that can be exactly the same when repeating the experiment over and over, no matter how strict our repeated experiment is done.
If something has changed in the universe , between the 2 repeated experiments , then another state has evolved in the second attempt of the "same experiment" around which is a result of previous states of the universe.

My question is , how practical is this approach , does it matter if we cannot define each state exactly. Why cant we just say :
The next state is exactly what the highest probability predicts.
"The most probable macro state is the one with the highest number of micro states"
What could change this prediction?

15. Oct 3, 2014

Staff: Mentor

Why do you think we can only know about 'things' by bouncing photons off them?

Indeed QFT tells us photons 'bouncing' is way off the mark.

Do you know what a tautology is?

Thanks
Bill

16. Oct 3, 2014

aleazk

Do you know references in which I can read a little more about the physical implications of contextuality? I mean, if nature is non-contextual, then Gleason's theorem provides a quite strong case in favor of indeterminism in QM, and from the very core of its mathematical foundation.

17. Oct 3, 2014

atyy

No one is suggesting that present observations be modelled by a deterministic theory. It is impractical, and given that no violation of QM has been observed, we should stick to it.

However, QM is not a completely stochastic theory. It has deterministic time evolution as well as stochastic time evolution. The time evolution between measurements is deterministic. When a measurement is made, the time evolution is stochastic.

18. Oct 3, 2014

Staff: Mentor

http://en.wikipedia.org/wiki/Quantum_contextuality

Thanks
Bill

19. Oct 3, 2014

atyy

That would not be the right interpretation. The interpretation is that if we want a formalism that is non-contextual, and uses rays in Hilbert space to label measurement outcomes, then the Born rule of QM is unique.

Why would we want such a formalism? Well, it works! Also, there are indications that QM is "easier to handle" than other alternatives with the same "expressive" power.
http://arxiv.org/abs/quant-ph/0604155
http://arxiv.org/abs/0711.4770

Another intriguing approach is the Piron approach, with a recent result by Soler, that bhobba can tell you about :)

20. Oct 3, 2014

the_pulp

What I will say is just philosophy, but I dont see that it will be much more philosophy than some other posts.

We can say that nature is not only deterministic between measurements but also deterministic during measurements but also it is deterministic during measurement. We could say that what really happens is that the instrument used to measure has a lot of degrees of freedom (lets say, just to be graphical, the exact positions of each and every atom of the instrument) inaccesible to humans and if we knew all of them, we could really know the exact result of the experiment. Then we will have other problems (Nature will be no local -perhaps this can be saved with retrocausality but I dont want to miss the point-) but nature could really be deterministic and QM (and its born rule) is just the aproximation that should be used when dealing with instruments with a lot of degrees of freedom.

I am not saying that this is the truth. I just say that up to now (and up to what I understood -much of it with a lot of your help-) we cannot say that QM says that nature is deterministic o stochastic. Here, when I say nature, I mean "what 'really' is behind QM, if there 'really' is something behind QM".

Last edited: Oct 3, 2014
21. Oct 3, 2014

Staff: Mentor

We accept much more philosophy in the QM forum than anywhere else, so you're OK.
To repeat Bhobba: "the_pulp has it right"
- T'Hooft has given us a proof by example that a completely deterministic interpretation can be consistent with QM.
- There are already too many is no shortage of stochastic interpretations.
- No experiment can distinguish them (although Bell has pushed the local determinists back into their final unassailable redoubt - superdeterminism - and it's not a place where I'd want to live).

Last edited: Oct 3, 2014
22. Oct 3, 2014

Jano L.

Do you mean also quantum theory is causal? I do not understand how that can be. Consider the following example from quantum theory.

Let a silver atom (initially with the spin state $|x+\rangle$ - eigenket of the $\hat s_x$ operator associated with the eigenvalue $+\hbar/2$) move along the axis $y$ and let it pass through the Stern-Gerlach magnets oriented in such a way that the silver atoms are separated into two groups with different $z$ while their $x$ coordinates stay the same. Let us denote the two $z$ values $z_+,z_-$.

When the $z$ coordinate of the silver atom is measured after passing the Stern-Gerlach magnet to be $z_+$, the spin state of the atom before the measurement of atom's position is believed to have changed into $|z+\rangle$. This new state cannot be calculated from the Schroedinger-Pauli equation for the spin wave function and the initial spin state $|x+\rangle$; books on quantum theory only talk about calculating that there is 50% probability the state will change into $|z+\rangle$ and 50% probability it will change into $|z-\rangle$... Have I missed something?

23. Oct 3, 2014

Staff: Mentor

Taking into account decoherernce everything is causal up until you get the improper mixed state. What it doesn't say is which element of the mixed state is selected by observation to go into which beam in your example - that is the point causality breaks down - maybe.

The maybe is we have interpretations such as BM where a specific element of the mixed state is there prior to observation - it all depends on how you look at it. And BM is not the only one - there is Nelson stochastics, primary state diffusion and probably others.

Thanks
Bill

24. Oct 4, 2014

microsansfil

Hello,

What is the falsification scientific method for probabilistic ? Is there a relationship beetween causality and probabilistic in the frame of QM ?

The classical physical concept of causality have to be revisited in light of the QM ?

Patrick

25. Oct 4, 2014

vanhees71

Of course, everything in physics is (asssumed to be) causal. Your example is not too difficult. You have a classical initial-value problem for the Schrödinger equation. The initial state is a wave packet of particles with determinate spin-x component, running along the y-axis through a inhomogeneous magnetic field adjusted to measure the spin-z component. What happens is that the wave gets split into two wave packets, leading to a (nearly perfect) entanglement between position and spin-z component. You'll find 50% of the particles in the one wave packet having (nearly) 100% spin-z up and 50% of the particles in the other wave packet having (nearly) 100% spin-z down.

Of course, the spin-z component is not determined before filtering out one beam, but the state evolution is causal. The Schrödinger equation is a causal differential equation, as it must be, because it describes a physical dynamical process.

For details of a correct quantum mechanical treatment, taking into account the correct magnetostatic field, fulfilling the constraint $\vec{\nabla} \cdot \vec{B}=0$, including the "spin-flip probabilities", see

PHYSICAL REVIEW A 71, 052106 (2005)
Quantum mechanical description of Stern-Gerlach experiments
G. Potel, F. Barranco, S. Cruz-Barrios, and J. Gómez-Camacho
http://dx.doi.org/10.1103/PhysRevA.71.052106
http://arxiv.org/pdf/quant-ph/0409206

Most textbooks are pretty sloppy in this. You can, however treat the problem correctly analytically to a high accuracy in an advanced quantum-mechanics lecture.

Last edited: Oct 4, 2014