Hacker News Reader

Comment by Veelox | Hacker News Reader

>the multiverse emerges as a natural extension of many existing physical theories (string theory, inflation, many-worlds, etc).

I might have looked in the wrong places but I have looked a little why the multiverse is a good explanation but I haven't found anything of the sort.

The best explanations I have seen seem to amount to "we got lucky with our universe, the best way to explain that luck is that there are a lot of universes". I haven't seen any reason why it shouldn't be "we got lucky with our universe, the best way to explain that luck is that it was done on purpose". Would you be willing to explain how you get from observed data or well regarded theories to a multiverse?

View on HN

Occam's Razor, essentially. The basic idea is that most multiverse notions arise naturally out of lower level mechanisms. A theistic explanation requires extra low level mechanics.

Say all of our experiments are explained by Theory X. Then positing Thoery X + God is a model with strictly more moving parts and equal explanatory power as Theory X alone.

On a wider level, explanations like "God did it" tend to operate on the same level as calling the phenomenon "magic"---it's an analysis that doesn't really provide any actionable information on how to influence said phenomenon.

In that way, I personally see "magic" and "God" as operating more in the realm of storytelling. Practically, I find good storytelling is helpful for things like identity-building and theatre, but if I wanna build a bridge then my money's on minimum Kolmogorov complexity, i.e. Occam's Razor.

I see it as a something that arises out of first principles. Everything in our world has a cause, and something always comes from something else. Now a deity, being outside of our universe and separate from it wouldn't be subject to that rule necessarily.

The multiverse is an inductive line of reasoning that infers things to explain away observances. But is it measurable?

I'm not saying we should "God away" the questions, but coming up with inferences that we can't actually measure (and don't answer the question of where the multiverse comes from) really doesn't do much scientifically or philosophically.

Well, it's not quite that simple. A Deist theology basically posits that the universe was created by some entity that exists apart from the universe and doesn't interfere with the workings of the universe. A multiverse theology basically posits that the universe exists in some entity that exists apart from the universe that contains sufficiently many such universes that the improbability of our own universe is immaterial.

In terms of complexity, both theologies necessitate the existence of a space which is inaccessible to our own measurements. You can't favor or disfavor one over the other since there is no distinction in explanatory capabilities, and the complexity exists in an immeasurable state. In a sense, both theologies basically state that the universe is the way it is because it was made that way, and the means of that making is described via effectively bullshittery in both cases.

The difference is that e.g. quantum mechanics already operates on a vector space that contains all possible world-states, so many-worlds doesn't really add anything to the theory, it just interprets them differently. That's similar to the idea of a Platonic universe of mathematical truths, in that it doesn't give you any different predictions, just a different perspective.

What a multiverse "explanation " of the observed constants adds to that is a prior that puts probability mass on every possible state, from which we can get the possibility of our personal universe. (I think that's a bit dubious, since it only shifts around the arbitraryness to some meta-heuristic about priors, thus the scare quotes.)

Similarly, you could imagine a theology that takes all the same laws of physics and interprets them as "God's plan", which again is just a different perspective. But to get additional predictions, you then need to have some idea of what God's plan specifically is, which tends to require thick books of ancient stories that still don't explain everything clearly.

So in the end you could disfavor both theories based on the difficulty of getting them to explain what they are intended to explain.

Your point about a prior distribution over universes is a good one. It just got me thinking about the quantum version of the Sleeping Beauty problem. Apparently, Sean Carroll has used this kind of idea and Everettian mechanics to derive the Born rule.

> In terms of complexity, both theologies necessitate the existence of a space which is inaccessible to our own measurements

One has more side effects than the other. Your requirement that the creating entity exist “apart from the universe” and not “interfere with the workings of the universe” is not generally accepted by deists. It is also not a required consequence of the deist hypothesis.

>The basic idea is that most multiverse notions arise naturally out of lower level mechanisms.

I think this is the part I don't understand. I have heard no evidence-based explanation for what lower level mechanisms cause our universe that imply a multiverse.

>Say all of our experiments are explained by Theory X. Then positing Thoery X + God is a model with strictly more moving parts and equal explanatory power as Theory X alone.

In pricipal I agree with you. In this case I think we disagree. In my mental model, Theory X in this case is the multiverse. At which point you say X + God is more complicated and I agree. I think instead of X + God as the explanation it should instead just be God. In which case I don't see an objective way of saying if Occam's razor would favor multiverse or God.

> I have heard no evidence-based explanation for what lower level mechanisms cause our universe that imply a multiverse.

The example I have in mind is the Everettian interpretation of QM. Formally, we have a Hilbert space and decoherence. We can interpret decohered regions in the Hilbert space as separate universes. These mechanics are an artifact of the mathematical formalism.

Interestingly, the Copenhagen "interpretation" would get ruled out by Occam's Razor since it adds the notion of wave collapse on top of the formalism without experiment being able to differentiate between the two.

> I think instead of X + God as the explanation it should instead just be God.

If that works for your needs, then more power to you. We can also jump a meta-level with Occam's Razor though.

Say we take a firm stance on only wanting to keep around models that are constrained by scientific experiment. Then we need, say, the Standard Model and General Relativity, but I'm not sure the God Did It model would survive said filter.

For scientific progress (i.e. the cooperative human effort of science), such a hard-nosed approach seems to have worked unreasonably well so far. For questions of personal world-view and how that affects your subjective experience, then I doubt such a dogmatic stance is universally ideal.

Anyway, I ramble.

My understanding of the Everettian interpretation is that all worlds that are possible in QM exist and we exist in one of them, the wiki article for this is called many-worlds. I may be being obtuse but this seems like a fundamentally different claim than what I normally see as claimed when talking about the multiverse. The multiverse, in my understanding, claims that there are infinitely many universes, even some that will have different QM from our universe because those universes started differently.

To use a CS analogy, many-worlds says that the tree branches the maximum number of time at each nodes. Multiverse claims there are infinitely many trees. Do I have a different understanding then you when it comes to what constitutes a multiverse?

> Say we take a firm stance on only wanting to keep around models that are constrained by scientific experiment.

My initial asking of this question is that it appears to me that the multiverse is not constrained by scientific experiment. I have read most of the comments and your gets the closest to something predictive but even then, it is still just one interpretation of QM. The other option is that there is something in string theory that implies multiverses but as of yet string theory is just theory and from what I understand string theory doesn't require a multiverse, it requires more dimensions which might cause a multiverse.

I ramble as well.

That about sums it up right there.

Belief is a psychologial attitude towards a claim.

Atheism is a sort of psychological preference for SIMPLER things to be in the role of causes. It simply rebels against what Daniel Dennett calls a "skyhook". Dawkins pretty much straight up admits this in his books when he says he would be fine with an explanation from simple origins.

Also, we want to apply the same rules we have been applying. Dawkins' Ultimate Boeing 747 argument for example, or Occam's Razor.

But as we can see from eg quantum mechanics, our rules for understanding are only useful in a certain context, that includes our observational starting point. We really don't know what kind of rules there might be.

Even the idea of rules implies there is some simple reason for something. But we now know (with AlphaGo for instance) that there may be more complex explanations than can really be understood by a human mind, even in turn based games. Let alone fluid dynamics or a heterogenous multi body gravitational problem, chaos theory and so on.

It is quite arrogant to think that what has worked for us humans so far will let us probe to the end of every inquiry about the universe.

Enter... the multiverse theory. The proud partner of the anthropic principle, for those who care about the frequentist interpretation of probability.

(Because if you don't, then the question doesn't even arise. Just say "because that's the way things are" instead of "because we're in that particular universe".)

It's a bit similar to throwing enough time and space at a problem. For example:

"how did the first replicating organism come about if Kolmogorov complexity of the smallest one is so high in terms of chemistry that it would be astronomically improbable to come about by chance?"

"Panspermia"

"Ok but that just pushes the problem back to other planets. How about the first?"

"Well, primordial soup + lightning + billions of years + trillions of planets ... and maybe buckyballs"

"That is just theories, where is the testing"

"It's the best we got. One day science will tell us the answer. Your theory isn't even scientific."

So we have to be careful not to equate science (an approach) with atheism (a bias).

But here is my question in conclusion:

At the end of the day, given how many times scientific understanding was overturned, what is really the difference between KNLOWLEDGE and BIAS?

We would only say "we got lucky with our universe" if we already presume the existence of other random universes. If there's only one universe that was purposefully created, then luck had nothing to do with it.

So the equivalence you're trying to draw is more of an opposite. Saying "our universe was done on purpose" is the opposite of saying "we got lucky with this universe," in much the same way that carefully placing 4 coins heads-up on a table is the opposite of flipping the same coin 4 times.

Once you're willing to entertain gods as an explanation, then no other explanation is needed, or will suffice. You don't need to go so far as elementary physics before considering a god. You can just look around and say "a god created this one second ago, including our memories."

From the other perspective, if you want to consider the idea of a god scientifically, then the obvious question is, where did the god come from? If they were powerful enough to purposefully create our universe, then they must have REALLY gotten lucky to have evolved that way in their own universe, right?

Sure. For example, string theory posits the existence of an extra 6 dimensions of space, which need to be wrapped up in something called a Calabi-Yau manifold. The exact properties of this manifold determine some of the basic properties of the universe, such as the values of the fundamental constants and the number of elementary particles. The only problem: there are about 10^500 possible Calabi-Yau manifolds, each of which looks just as valid as another. Early string theorists had hoped all the extra manifolds would somehow be ruled out, leaving only one that would uniquely determine the properties of our universe. However, this avenue is now looking pretty hopeless; it seems we're stuck with all 10^500 possibilities. So, what does one do with all these manifolds?

Well, it turns out that the early universe went through a phase of incredible, extraordinary expansion called inflation. The inflation was boosted by something called an inflaton field, and due to quantum properties the inflaton field could spontaneously destabilize and spawn a bubble universe, each of which with a different Calabi-Yau manifold. It's possible that we are inside such a bubble universe, and outside of us the inflaton field is continuing to expand and producing more bubble universes. If this is the case, then it's not much of a surprise why our universe has the properties it does. Out of the 10^500 possibilities, a bubble universe spawned with the right Calabi-Yau manifold for life, and (unsurprisingly) we find ourselves living in that universe. This is the landscape multiverse theory. (I originally read about this theory in Brian Greene's excellent book The Hidden Reality, which describes many other multiverse theories too.)

Now, something important to note is that all this is still highly conjectural. String theory has many pleasing theoretical properties, but about zero physical evidence to support it. What then is the difference between string theory and deism? The difference is that for string theory (and more generally other multiverse theories) we actually have physical and mathematical theories that describe them. Deism doesn't. For string theory, we have papers, conferences, books, equations, and physicists and mathematicians who study this problem. We have something to work with. What do we have for deism? Well, pretty much nothing. No data, no theories, no anything. Deism is a fall-back theory when nothing else works out, and it has no theoretical roots and makes no predictions. This may change in the future, but as of now, deism is a scientific dead end. Of course, I don't know if string theory or another multiverse theory will pan out (indeed, the failure of the LHC to detect supersymmetry is raising some eyebrows), or even if deism actually turns out to be correct. However, on a scientific basis, string theory has more going for it now.

>The best explanations I have seen seem to amount to "we got lucky with our universe, the best way to explain that luck is that there are a lot of universes".

Perhaps you haven't looked that hard?

The idea of multiverse as the "best way to explaining our luck with our universe" was in the bottom of the list of concerns to theorize a multiverse - if it appeared at all:

https://en.wikipedia.org/wiki/Multiverse

I been to the wiki page before. If anything my question is common enough it is discussed in the wiki page which contains this quote.

>Indeed, invoking an infinity of unseen universes to explain the unusual features of the one we do see is just as ad hoc as invoking an unseen Creator. The multiverse theory may be dressed up in scientific language, but in essence it requires the same leap of faith.

— Paul Davies, The New York Times, "A Brief History of the Multiverse"

So while it may be at the bottom of the list, it very much appears.

I think one crucial difference is that this actually has scientific precedence. For example, the Earth lies in an incredible number of "Goldilocks zones", which have just the right conditions for life to develop. For example, the Earth lies at just the right distance from the Sun for water to be in liquid form. To close and it would all evaporate, too far and it would all freeze. Why is this? Early astronomers (Kepler, I think) spent many years trying to discover some physical reason for why the Earth should be at this particular distance. From what physical principles could the distance be deduced? As we know now, this search was in vain, because there was no fundamental reason. Turns out, it was just a coincidence! With the discoveries of the (appropriately) named Kepler telescope, we have discovered thousands of exoplanets in thousands of star systems. With the billions upon billions of stars in the universe, there are likely trillions of planets out there. With odds like that, it is almost certain that a planet with our properties would show up somewhere, which it did. As it turns out, astronomers do invoke an "infinity" of unseen planets to explain the unusual features of this one, and I doubt Paul Davies thinks that's a leap of faith!

I think you raise a good parallel with Earth but I don't think it applies directly to the possibilities for a multiverse because we cannot see other universes but we can see other planets.

Imagine if our solar system was inside a nebula and the only star we could see was Sol. Until we figure a way to send a probe outside the nebula we have no idea if their are more stars or not. Do you see a way to "send a probe outside" and see if there are more universes or not?

Off the top of my head I don't really remember, though Brian Greene does devote a whole chapter to the question of detecting other universes in his book, so that might be a good place to look further. At the moment though, this doesn't really bother me. For example, Giordano Bruno suggested the existence of exoplanets in 1584, and certainly he could not have predicted the modern techniques the Kepler telescope uses to detect them! It may be centuries before we have the theory or technology to determine if other universes are detectable or if they exist. A long time to wait perhaps, but this should not be a reason for ruling them out (though of course, solid evidence would be required before a multiverse theory could be accepted).

Reply on news.ycombinator.com