3

In classical mechanics, specifying the initial coordinates and velocities of all particles uniquely determines the system's future; we do not need to specify accelerations or higher derivatives.

  1. Can this be proven classically using Lorentz invariance, isotropy and homogeneity of spacetime, etc?
  2. If not, can it be proven by taking an appropriate classical limit of a quantum mechanical theory?
Qmechanic
  • 201,751
user111187
  • 331
  • 1
    No, it can not be proven because as theories both special relativity and quantum mechanics are designed to have classical mechanics as a limit, so you can at most, recover what somebody put willfully into these theories when they were discovered. The equations of motion in classical mechanics are validated by one thing and one thing only: experiments. That's just as true for any other physical theory. In this case nature doesn't seem to care about "jerk", I can't tell you, though, what the experimental limits on that are. – CuriousOne Sep 20 '14 at 13:06
  • Possible duplicates: http://physics.stackexchange.com/q/4102/2451 , http://physics.stackexchange.com/q/18588/2451 and links therein. – Qmechanic Sep 20 '14 at 13:36
  • @Qmechanic - Related, but not necessarily duplicates. Those related links address why "we do not need to specify accelerations or higher derivatives". It does not address why "specifying the initial coordinates and velocities of all particles uniquely determines the system's future" (which is not necessarily the case). – David Hammen Sep 20 '14 at 13:41

1 Answers1

2

In classical mechanics, specifying the initial coordinates and velocities of all particles uniquely determines the system's future ...

Uniqueness and existence are not guaranteed in classical mechanics. Physicists gloss over the cases where uniqueness is not guaranteed. Initial positions and velocities suffice if the resultant accelerations are Lipschitz continuous. The classical universe does not necessarily obey the Lipschitz condition.

One issue is that classical mechanics ultimately deals with point masses, and their use can result in infinities (which obviously violate the Lipschitz condition). The hand wave solution to this issue is that point masses don't really exist; they're just a nice mathematical construct. The mathematical singularities that can result from the use of point masses never actually occur in physical reality.

Another issue is that one can construct a Lagrangian landscape that violates Lipschitz continuity. Norton's dome (discussed here at physics SE) is a nice example of how violations of Lipschitz continuity lead to non-uniqueness. The hand wave solution to this issue is that the cuspy nature of Norton's dome represents something that doesn't exist in physical reality. Nature abhors singularities.

... we do not need to specify accelerations or higher derivatives.

This is a separate question, one that has already been answered nicely in the questions "Why are there only derivatives to the first order in the Lagrangian" and "Why are differential equations for fields in physics of order two."

Community
  • 1
David Hammen
  • 41,359
  • Why can't mathematical physics get over the fact that point masses don't exist, not even in theory? Every professional textbook on classical mechanics clearly says that point masses represent an approximation of the center of mass movement of real objects, assuming that no rotational and internal degrees of freedom are excited! No Lipschitz conditions necessary, we simply switch to a correct description when things get down to a point! Having said that, I find the mathematical exploits of classical mechanics fascinating, even though they have little relevance in the real world. – CuriousOne Sep 20 '14 at 13:45