Given that the second law of thermodynamics states that entropy always increases and the universe is currently not in a state of maximal entropy does that not imply that it must have started at a point in the past? And that it could not have existed forever?

Thermodynamics was developed in the mid 1800's while the big bang was first proposed in the 1920's if the above argument is correct did anyone else argue it before general relativity?

  • 1
    $\begingroup$ cool idea, no pun intended. $\endgroup$
    – Alex
    Jul 23, 2015 at 1:05
  • 1
    $\begingroup$ This argument only applies if you know the universe is finite, as if it is infinite, it can have infinite entropy, and therefore the second law of thermodynamics doesn't imply it started from zero. To my knowledge, before the big bang theory we had no evidence about whether the universe is finite or not. $\endgroup$
    – Omry
    Jul 24, 2015 at 11:02

2 Answers 2


Not really

Although it's reasonable, there are many ways to avoid such a conclusion, e.g.:

  1. There was a major statistical fluctuation that caused a decrease in entropy in the past (unusually among physics laws, the 2nd law is usually but not always obeyed). Such a fluctuation would be extremely unlikely, but in an infinite universe that has existed forever, it would inevitably occur.
  2. Some exotic mechanism can decrease entropy (c.f. black hole thermodynamics - a priori it's not impossible that a highly-disordered system tossed into a black hole just disappears, and total entropy goes down)
  3. Energy is not conserved, but is rather constantly produced (see steady-state universe), and that causes entropy to decrease.
  4. The laws of physics are changing, and the second law didn't become a law until some relatively-recent time. (This would imply energy is not conserved by Noether's theorem)

I'm sure even more ways to avoid such a conclusion are possible.

  • $\begingroup$ About your second point, surely black holes weren’t proposed until after the formulation of general relativity. $\endgroup$
    – Jono94
    Nov 7, 2023 at 2:56
  • $\begingroup$ @Jono94 I do believe that black holes were conceptualized before GR. Wikipedia gives John Michell's dark stars as an example. The precise formulations required GR, but I'm not sure how that is relevant to the question/answer. $\endgroup$
    – Allure
    Nov 7, 2023 at 8:26

The 2nd law, Kelvin form says " no process is possible whose only final result is to convert into work, a quantity of heat, extracted from a reservoir at uniform temperature!" Doesn't say anything about entropy.... 😀 The statement derived from Clausius theorem states the "the entropy of an isolated system can never decrease." Since the universe is closed, there is no heat in or out, so the entropy change of the U = 0. There is an ancient logic that says " the universe always was and always will be. Otherwise,at sometime there must have been nothing, but nothing [nothing includes no vacuum fields etc. (Only a human can get something for nothing ... lol 😀)] can ever produce something, otherwise it was not Nothing, but something, which we call the Unverse. Doesn't have anything to do with entropy.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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