Physics Stack Exchange is a question and answer site for active researchers, academics and students of physics. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

First of all, I am not a physicist, so I cannot guarantee things I say will make sense. I will try my best, though.

In classical mechanics we have the notion of inertial frame of reference. If my understanding is correct, such frames are exactly those for which Newton's laws of motion have the usual form. I believe I have also heard, that they are the frames that are "moving at a constant speed" with respect to some distinguished point in space.

So we have to be careful whether our frame is accelerated or not. Which makes me wonder:

Does there exist a set of physical laws (of same descriptive power Newton's laws have, but perhaps expressing relations in some other set of physical quantities) such that the laws from this set look the same, no matter how the frame of reference moves?

Does such a set of physical laws exist if we restrict to polynomially moving frames of reference? (By this I mean that the motion of the frame is described by a polynomial in $t$, where $t$ is time.)

I imagine this could simplify some calculations.

share|cite|improve this question
Are you asking for a synthetic version of Galilean geometry? It wouldn't be hard to cook it up. It would axiomatize the space-time points of Galilean space-time like Euclid/Hilbert axiomatized geometry. Then you can state the laws of motion without reference to an explicit frame, in terms of the curvature of the Galilean space-time trajectories. I don't think it would be so useful, because the notion of coordinates is not that bad, since the symmetry is expressed as explicit linear transformations on these, and checking covariance of the laws of motion isn't difficult. – Ron Maimon Dec 15 '11 at 10:11
Yes, I believe this is exactly what I'm looking for. The main reason for my interest in this is that choosing a "distinguished point" (which is supposed to be at rest) seems rather artificial to me. – Dejan Govc Dec 15 '11 at 14:50
It is artificial, but it is precisely as artificial as choosing coordinate axes for describing geometry. The mathematical structure is more symmetric than the description. It is probably philosophically best to consider the symmetry as external, and not force your language to be symmetric from the start. But for the special case of Galilean space-time, I am sure it can be done with no choice of special point (or coordinate axes) just by using Euclid's axioms, some betweenness axioms, etc. But axiomatizing 4d geometry Euclid's way is tough. – Ron Maimon Dec 15 '11 at 18:56
@Ron: I think this pretty much answers what I wanted to know. Thanks. – Dejan Govc Dec 16 '11 at 17:35
up vote 1 down vote accepted

You can indeed!!!! Relativity is not a good answer since you are asking for classical (Newtonian) Mechanics. There are two approaches. The first one is that of differential geometry. In it you realize that the problem of evaluating Newton's equations in non-inertial frames is that acceleration introduces curvature to the coordinates systems used. This is expressed with a mathematical symbol called the cristoffel tensor. I suggest you follow Schaum's Tensor Calculus or Grinfeld's Tensor Analysis and the Calculus of Moving Surfaces for more on this. In that case you arrive to the equation $$F^i=m\frac{\delta v^i}{\delta t}$$Another approach is that of analytical mechanics. In it you restrict yourself to a certain kind of systems (monogenic, holonomic, etc....) which lets you formulate mechanics in terms of what is called generalized coordinates. The formula you get is $$\frac{\textrm{d}}{\textrm{d}t}\frac{\partial \mathcal{L}}{\partial \dot{q}^i}-\frac{\partial \mathcal{L}}{q^i}=0$$

share|cite|improve this answer
Thanks! I've been studying some differential geometry lately, so this point of view is more than welcome. I'll have to think this through when I have some more time, but for now, +1. – Dejan Govc Apr 14 '15 at 21:54
@DejanGovc If you are studying differential geometry, the funny derivative with deltas is a funny composition of the normal derivative and the cristoffel symbol – Iván Mauricio Burbano Apr 15 '15 at 17:19
I have accepted your answer, since I think this point of view will be most useful to me. Could you perhaps recommend some literature explaining classical mechanics from this point of view? – Dejan Govc Apr 15 '15 at 17:32

Relativity achieves just what you describe, and does it for all physical laws, in all frames of reference, accelerated or not, and not even restricted as per your 'polynomial' stipulation.

To even make this work in uniform-motion frames, Special Relativity had to add a little complexity to Newton's equations, which otherwise would break down if the speeds involved in that uniform motion are very high.

Even more complexity was needed to encompass all reference frames, regardless of acceleration, or influence of gravity. But the resulting laws and equations are universally valid. (perhaps not at the quantum/particle level, however).

I'm not sure this bit: 'with respect to some distinguished point in space' is accurate WRT what Newton said, but regardless, it is not necessary after Relativity.

share|cite|improve this answer
Thanks, although this doesn't quite answer my question, it certainly helps clear a misconception or two of mine. I will upvote as soon as I have the capacity to do so. – Dejan Govc Dec 15 '11 at 14:53

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.