I'm happy to accept and use conservation of energy when I'm solving problems at Uni, but I'm curious about it to. For all of my adult life, and most of my childhood I've been told this law must hold true, but not what it is based on.

On what basis do we trust Conservation of Energy?

  • $\begingroup$ Classically, it comes from Newton's second law. And that's an axiom of physics (you may also say that it is the definition of force). Nowadays, we say it comes ftom time-invariance. $\endgroup$ Mar 1 '12 at 7:59
  • $\begingroup$ Related: physics.stackexchange.com/q/19216/2451 and links therein. $\endgroup$
    – Qmechanic
    Mar 1 '12 at 8:10
  • $\begingroup$ @Manishearth: Is there/Can there be symmetry-conservation identification without/before equations of motion? $\endgroup$
    – Nikolaj-K
    Mar 1 '12 at 8:25
  • $\begingroup$ @NickKidman I don't know. I'm not to clear about the exact derivation of conservation of energy from Noether's theorem, I just know that the corresponding symmetry is time. That's why I didn't post it as an answer. $\endgroup$ Mar 1 '12 at 8:36
  • $\begingroup$ @NikolajK No. A more complete statement would be that the total energy in a system does not change when it evolves according to the Euler-Lagrange equations. If you go "off shell", energy is not necessarily conserved. $\endgroup$ Jul 15 '14 at 19:54

Let me expand a bit on Manishearth's answer. There's an idea going back a long time called the principle of stationary action. See http://en.wikipedia.org/wiki/Principle_of_stationary_action for a description that isn't too mathematical. In the 18th and 19th centuries century the mathematicians Lagrange and Hamilton found ways of using this to describe mechanics. Then in the early 20th century the mathematician Emmy Noether discovered that in Lagrangian mechanics if a symmetry of the equations existed this meant there was a corresponding conservation law. As Manishearth says, one example of this is that time symmetry means that energy must be conserved.

Strictly speaking, the symmetry involved is "shift symmetry of time". This means that if I do an experiment, the time I do it doesn't matter so I'd get the same result tomorrow as I do today. If this is true Noether's theorem means that energy must be conserved.

Experimentally we find that repeating experiments does indeed give the same results, and we also find that everything observed so far obeys Lagrangian mechanics. This suggests that energy is indeed conserved. Strictly speaking this is an experimental observation not a proof, but few doubt that the principle applies as the universe would be a strange place if it didn't.

Wikipedia has lots of articles on Langrangian Mechanics and Noether's theorem, but they're a bit intimidating for the non-mathematician. If you're interested in knowing more Googling should find you plenty of more accessible articles.

  • 1
    $\begingroup$ I believe I'm studying Lagrangian mechanics this semester, as part of my Physics degree! Awesomesauce. Thanks very much! $\endgroup$
    – Pureferret
    Mar 1 '12 at 9:13
  • $\begingroup$ You're welcome - it's amazing (or maybe it isn't!) how many of us are on both the Physics and SciFi Stack exchanges :-) $\endgroup$ Mar 1 '12 at 10:05
  • 1
    $\begingroup$ @JohnRennie It's worth pointing out that conservation of energy actually doesn't hold in the GR, because the metric is not invariant with respect to time, hence Noether's theorem doesn't guarantee conservation of energy anymore. You can see my answer here for a pretty simple to understand example of loss of energy in the context of cosmology. $\endgroup$
    – ticster
    Jul 15 '14 at 20:14
  • $\begingroup$ But... Don't a lot of things that we know/believe say that time is not symmetrical? Entropy alone implies that some experiments will give different results in the future. And wouldn't there be different results pre and post Inflation? And things that are part of our observable universe today, will not be tomorrow. Or am I misunderstanding something important here? $\endgroup$ Aug 18 '15 at 14:19
  • 1
    $\begingroup$ @RBarryYoung: the technical meaning is that the action exhibits time shift symmetry. As a simplification, you can think of this as saying that the laws of physics don't change with time. $\endgroup$ Aug 19 '15 at 17:01

Historically, energy conservation was enforced by postulating new physics each time an apparent violation was discovered. This makes energy conservation not so much an empirical observation - rather it is an organizing principle that we impose successfully to explain how Nature behaves. This it is made true by definition!

Indeed, most real processes are dissipative, i.e., they actually lose energy. It is one of the great accomplishment of 19th century physics that in spite of this, energy conservation was postulated and used successfully to build a coherent theory of thermodynamics, which ultimately lead to a great unification of physics. (The last bit of this, the unification of gravity and quantum mechanics, is still a hard research problem.)

Observed dissipation doesn't contradict energy conservation, as the lost energy is, on a fundamental level, still there - it just moved from the part of a system described by our methods to unmodelled parts (the ''environment'') that picks up this energy. This is why real processes usually move to a state of least free energy (where the free part of the energy depends on how a system is embedded into the environment).

  • $\begingroup$ Does the term enviroment here include a possible microscopic level, which is not part of the model (as in fluids description and particle description), or is this term restricted to, say, the outside of some spatial bounds? $\endgroup$
    – Nikolaj-K
    Mar 3 '12 at 20:02
  • 1
    $\begingroup$ The first. The environment always consists of the unmodelled high frequency details of the interaction (low frequency details of significance would make the model inaccurate), and except for very tiny systems, these high frequency details stem (a) from having ignored microscopic details of the system itself, (b) from having ignored high frequency details of the forces that apply at the boundary of the system. If you have a well-isolated system, (b) is negligible and all dissipation is caused by (a). [turbulence: energy moves to higher and higher frequency, until it can be no longer resolved.] $\endgroup$ Mar 3 '12 at 21:00
  • $\begingroup$ @Arnold Neumaier: This is not true in the sense you say: there are only a finite number of things you need to add before energy conservation is just true. If we kept on finding more and more things to add, it wouldn't be a law anymore. The major insight was that heat and motion are both energy, and this fixed up the conservation law once and for all historically. The little violations due to neutrino emission were fixed in the 1930s. $\endgroup$
    – Ron Maimon
    Mar 4 '12 at 2:12
  • 1
    $\begingroup$ @Ron: 'Finite' depends on how one counts. - To prove my point: We only recently adjusted the mass of neutrinos in order to save energy conservation. - Dark energy is still postualted without proof to save the lase of energy conservation on astronomical scales. And it will always be like that, as we cannot give up energy conservation as an organizing principle without ruining the whole setting of modern physics. $\endgroup$ Mar 4 '12 at 11:27

Conservation of energy is a property that a particular physical system may have. Most often, one determines if a system conserves energy by studying the symmetries of the Lagrangian. As others have said, conservation of energy is associated with the Lagrangian being symmetric in time.

But there is no a priori reason that all possible Lagrangians conserve energy. For example, consider the Lagrangian of the universe. The universe, as we now know, is expanding, meaning it is certainly changing as a function of time. Thus, on a very global scale, the energy of the universe isn't conserved. But this applies to only the very largest of scales. Locally, we don't notice the expansion of the universe, and energy is conserved to excellent precision.

But, taking a step back, saying that conservation of energy can be derived from a symmetry of the Lagrangian is a bit of a circular argument. If you write down a Lagrangian that is invariant under time symmetry, then you can define an energy that doesn't change in time. That's true.

But I guess none of this yet answers your question, which was "On what basis do we trust conservation of energy." The answer to that is the vast experimental evidence, from day-to-day experiences to precision physical measurements. Based on experiment, our local laws of physics don't change as a function of time.


It seems to me the obvious answer is that it agrees with all the experiments.

I am an experimentalist after all, (the neutrino is perhaps the best "missing energy" story.)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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