I got the impression from several public lectures that, when it comes to the fine tuning problem, physicists tend to give multiverse together with anthropic principle as a plausible solution, but seldom anthropic principle alone.

However it seems the core of this kind of solution is really anthropic principle, that is, once we decide to invoke anthropic principle, the concept of multiverse becomes irrelevant(if fine tuning problem is our only concern), adding multiverse to the explanation does not seem to strengthen the solution by much. Is there anything I missed or misunderstood?

Of course it is possible that I overestimate the popularity of multiverse explanation in academia, after all I got this impression from public lectures and I never talked to a cosmologist in person, in this case I would be happy if anyone better informed can confirm this.


It seems plausible that the probability of the universe being just right for human life by pure chance is pretty low, and possibly absurdly low. That means if the universe only had one chance at existing the probability we'd be around to see it is vanishingly small.

The various multiverse theories suggest that huge numbers of disconnected universes get created, so even though the probability of human life in any single universe is small there are enough universes that it becomes almost inevitable that at least one of them contains humans. This is why the anthropic principle only makes sense if multiple universes exist.

Response to comment:

Ideally physicists test their theories with experiment. Sadly we can't create universes so we can't easily test our theories of how the universe came to exist. It seems likely we'll never be able to prove beyond reasonable doubt whether our ideas are correct.

Given this situation we look at the probability that any particular explanation for the universe could be true. If we accept the idea that we are here by chance then it is much much more probable that the universe had multiple goes than that it had only one go. When I say much much more probable no-one can put a figure on this but if we believe ideas about the string landscape the probability that a single universe got it right is around $1$ in $10^{500}$. This number is so small as to be indistinuishable from zero. The number of atoms in the observable universe is only around $10^{80}$.

Similar arguments have been made before. In 1802 William Paley observed that the probability of a living being appearing by chance was so low that there had to have been a divine watchmaker to create it. Scientists agreed this was improbable, and in 1859 Darwin explained that living objects didn't appear by chance but were the result of evolution having lots of goes at it.

The same applies to the origin of the universe. Asking us to believe a single universe got it right first time is at the six impossible things before breakfast level. It seems vastly more likely a multiverse is responsible.

  • $\begingroup$ Yes, this is the standard argument. Let's assume fine tuning is anthropic, then even if there is only one universe, the very existence of creature capable of asking fine tuning implies the universe has to be fine tuned already, I don't see how my argument is weaker than the one with multiverse. $\endgroup$ – Jia Yiyang Jan 1 '14 at 7:24
  • $\begingroup$ An arising consensus among experts like Lubos Motl, Nima Arkani Hamed, etc is that a certain amount of fine tuning might have to be accepted at the end. But it is too early to definitively say how much, it depends on the future results of still ongoing and upcoming experiments that look for new physics, the specific version or formulation of the hierarchy problem considered, etc ... $\endgroup$ – Dilaton Jan 1 '14 at 7:32
  • $\begingroup$ @JiaYiyang: I've edited my answer to respond to youir comment $\endgroup$ – John Rennie Jan 1 '14 at 8:04

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.