# Explanation in the Multiverse

by on May 3, 2011

When comparing explanations for any given event, most of us – often without realizing it – weigh probabilities. If you come home to find that your TV is missing, the best explanation might be that you’ve been burglarized. It may be true that aliens used a futuristic device to make your TV invisible, but knowing that (a) burglary is common and (b) aliens in general are completely unknown to us, then clearly the burglar hypothesis is the  more probable (and thus superior) explanation.

This sort of probability weighing is not only important for every day life, but it is also important for statistical testing of scientific hypotheses  (which use p-values or Bayes theorem), principles like Occam’s razor (i.e. more assumptions = less probable),  and the old adage, “extraordinary claims require extraordinary evidence” (i.e. a claim which doesn’t fit our background knowledge requires  evidence to increase its probability).  Unfortunately, an infinite multiverse could mess this all up.

Very often I hear people trying to answer the fine-tuning argument for God by using various multiverse proposals, and usually these debaters tend to ignore (or not be aware of) further problems caused by this proposal. As alluded to in the opening of this post, one major problem is that infinity – the sort actualized in a infinite multiverse – doesn’t get along with probability.

As I explained in my post Ethics in the Multiverse, if there are an infinite number of universes, then this suggests that all possibilities, no matter how unlikely, are actualized an infinite number of times. So in my example about the missing TV, both the alien explanation and the burglar explanation are possible, but we chose the burglar explanation because it seems to be much more probable. But if we consider the infinite multiverse, then there are an infinite number of universes where burglars stole your TV and there are also an infinite number of universes where aliens made your TV invisible! So, as you can see, we run into a huge problem when we attempt to predict which explanation is more likely.

As physicist Brian Greene writes in The Hidden Reality, “Every possible outcome allowed by quantum calculations, however unlikely – a .1 percent quantum probability, a .0001 percent quantum probability, a .0000000001 percent quantum probability – would be realized in an infinitely many universes simply because any of these numbers times infinity equals infinity. Without a fundamental prescription for comparing infinite collections, we can’t possibly say that one collection of universes is lager than the rest and thus the most likely kind of universe for us to witness, we lose the capacity to make definite predictions” (p. 185).

But Greene’s concern goes much further than missing TVs. If we lose our ability to calculate probabilities, then all our evidence – even that which leads us to various multiverse proposals – is unreliable. Quite a paradox. Perhaps then it’s fortunate that Greene concludes the section with optimism. According to him, most physicists do suspect that this problem will be solved eventually.

But until then, it’s a problem worth considering when arguing about fine-tuning.

• Piprod01

But why does the multiverse hypothesis require an infinite number of universes? Why not, just an extremely high number?

• Bob (Piemaster1123)

Good point. The question is, if there is just an extremely high number, how many do we need? That will depend on how the multiverse is formed. For example, if we consider the multiverse formed by taking any decision and having at least one universe in which each possibility is formed, we have to have at least one universe for every decision ever made in the history of time. This would be an absurdly large number, so we typically say it is infinite, but it is not actually infinite, just really huge.

That does seem to solve the probability issue… I’d like to hear what someone else has to say about it.

• Antybu86

Because most multiverse hypotheses usually either necessitate them or contain no factor to limit them. Eternal inflation, for example, poses that as a high-energy vacuum of space expands, mini “island universes” arise within in it. Since thee is no end to the expansion, there is no end to the amount of universes that will arise.

• Anonymous

But this seems to suggest that there are never actually infinite universes at one time, there are still always a finite number of universes that never stop increasing. It’s not possible to add a finite amount of something to a finite amount of something and get an infinite amount of something.

• Piprod01

Just a thought, if there is an infinite number of island universes, wouldn’t that guarantee that a Universe identical to ours exist? And the Andy Burke of that Universe would bring up the exact same objections? It’s all entirely speculative, but it’s still.

If we’re to say the multiverse had a beginning of some kind, we wouldn’t have any problems with infinities.

That’s not to say that I think the best objection to fine-tuning arguments is to propose any multiverse, I think it’s better to point out where we’re treading the line of speculation – the notions of fiddling with constants.

• Anonymous

So does this only apply to “true” probabilities (meaning probabilities that are actually determined by quantum chance)? Because obviously we know from living in this world that it actually is more likely that your house was burglarized than that aliens came, because most of these things are not actually determined by chance. But then again, if it’s possible that aliens stole your TV, then if we assume that everything not determined by quantum chance is deterministic, then since the probability of aliens and burglary are both non-zero, then we can’t determine which of those is more likely either. Either it’s impossible that aliens stole your TV, or we have no current way of determining which event is more likely.

• http://thepolemicalmedic.wordpress.com/ Thrasymachus

I am not sure this need hugely trouble someone citing multiverses to respond to fine-tuning. Providing we are using epistemic degree-of-belief probability assignment, we do not need to a frequentist “count up the possible worlds” way of doing probability. Even if we should do that, I am pretty sure there are accounts of infinity we can use to let us compare them to get the ‘right’ probabilistic results.

• Bob (Piemaster1123)

The missing piece to determine that is how the probabilities are assigned and how the multiverse is constructed. If there was a universally accepted way of constructing the multiverse, people could get away with saying “Well since we’re in a multiverse, etc…” However, without a precise description of the multiverse, Andy’s arguments may or may not work.

• Antybu86

As far as I know, there are ways of comparing the infinities but they yield different results… so, there needs to be some underlying method of determining which method of comparison is correct.

• Stefan M.

I don’t get the argument. You seem to conceal the premise that in your world view everything that can be imagined has a probability above 0. Or rather you seem to think that something that is possible in universe X must be possible in this universe. Sorry, but I don’t think either of this is true. (Having the possibility of a world where anything is possible, has not neccessarily a bearing on this world unless you can show that we live in this world.)

• Stefan M.

that world*

• Anonymous

I think the way to think about it is “what are the chances that we are in the universe where X happens,” and not “given we are in universe Y, what are the chances that X happens in this universe.” Things will happen a certain way in this universe, meaning the probability of any event occurring is either 1 or 0, but it’s impossible to know if we’re in a universe where this happens or not, so we calculate the probability that we’re in a universe where X happens.

• Theosucksasswarner

Andy Burke is the biggest stupid ass ever. His asshole is filled with Theo Warner’s semen. Look: http://antybu86.blogspot.com

• Stretmediq

Your argument assumes the many worlds theory is true. Do you have any evidence it is? And if you put forth decoherence to explain away the argument from design (which I also reject but for different reasons) how do you get around the apparent violation of the conservation laws? Unless you can answer these questions your argument must be dismissed as unfounded.

• Antybu86

No… this doesn’t assume that “many worlds theories” / multiverse proposals are true. I’m only saying that IF they are true, THEN this is a problem.

• Stretmediq

And if it’s not true it isn’t a problem. Which begs the question, why did you write it unless you thought there was something to it? You can fashion all kinds of models to prove or disprove anything depending upon your basic assumptions but unless you have something to back it up it’s a useless exercise. Is there any evidence for it at all? Does it match anything we see in the world? Unless you can show it does you’re no different from the Theist who makes unsubstantiated arguments about God.

• Anonymous

I know that I, for one, don’t care to hear anything about any topic that hasn’t been demonstrated as true. I mean, if it’s still speculation, why even talk about it? There’s no reason.

• Antybu86

Of course there is something to it, but that doesn’t necessarily mean that it’s true. There are some really interesting and compelling reasons for thinking that a multiverse – of some sort – might exist. However, nobody yet knows if the multiverse exists, and it’s possible that nobody will ever know..

If it’s something you want to look in to, I’d recommend Brian Greene’s recent book which I mentioned in the main post. It’s called The Hidden Reality.

• Stretmediq

I’ve read Greene and a few others that are a bit more technical. And it is because I have some understanding of the mathematics underlying the theory I tend to doubt the models proposed by that school of thought. I accept the probability of other types of universes with different physics. I just don’t think every variation of a type may be realized because of the conservation laws inherent in any finite universe. The versions that are being touted now are just attempts to get around the Copenhagen interpretation and the fine tuning arguments. But they have nothing to support them but mathematical assumptions. And if you are familiar with things like non-Euclidean geometry or Godel’s incompleteness theorem you know that even mathematical assumptions are still just assumptions. You have to go with the evidence you have. There is no need to invoke unproven speculation to defeat the design argument. Trying to do so in this manner is like watching two people argue about whether the invisible unicorn is pink or blue.

• Stretmediq

BTW what do you mean by “of course there’s something to it but that doesn’t mean it’s true?” Aren’t contradicting yourself there?

• Anonymous

There’s something to the “tree-down” theory of bird evolution and there’s something to the “ground-up” theory of bird evolution. If either of these statements were false, then there wouldn’t be a debate among qualified scientists as to which theory is true, but obviously both can’t be true.

• Stretmediq

That is a false analogy. The evolution of birds is demonstrable. Can you demonstrate the existence of parallel universes?

• http://godlessons.com Godlessons

There is a good bit of evidence that ducks, at least certain groups of ducks, have all the genes necessary to make all the “species” of ducks in the group.

I was actually surprised to see the study, and I had it bookmarked, but I lost the bookmark unfortunately.

Needless to say, that’s why he brought it up, and it’s actually been published in a respectable journal.

• http://godlessons.com Godlessons

I think something else gets missed. People think that because there are an infinite number of things that it automatically makes everything possible. That’s simply not true. You can still have an infinite number of universes without ever creating one that supports life. It’s like the number lines we did in Algebra. (-∞,2] ∪ [7,∞) is an infinite number of numbers, but it doesn’t include 3-6.

The fact is, there can be an infinite number of things that never happen in an infinite number of universes. For instance, in the example above, there are an infinite number of numbers between 2 and 7.

Anyway, just thought I would add that into the mix.