I’ve got a new article out in Seed about how religious physicists, in particular, are thinking their way around the theological problems posed by multiverse theory. It’s good, mind-bending stuff.
Scientists now recognize that if space were expanding at a slightly different speed, or if the strong nuclear force were just a little off, our universe would be a hydrogen mush incapable of supporting life. The chances that the cosmic conditions needed for even a single living cell would come about in a random toss-up are astonishingly low, often called the “fine tuning problem.” “The most obvious explanation for fine-tuning is that fine-tuning is real,” argues O’Leary, “that we live in a designed universe.” If, however, we live in a vast and varying multiverse, there could be as many as 10500 different universes in all, making the chance of ours occurring among them comfortably higher.
Unfortunately, a lot of great material from my research didn’t make it into the article. I had the pleasure of sitting down with Brian Greene, author of The Elegant Universe, in Midtown. “There are features of the world that may not be explainable in the conventional sense,” he said, suggestively. “A multiverse is a very different framework to do science in.” We also spoke about the Large Hadron Collider being developed in Europe, which has raised some far-off fears about creating a black hole that will envelop the Earth. Greene told me:
From the public relations standpoint, having the whole black-hole-destroy-the-world thing was very good. I must have done six or eight programs on the LHC, and ultimately that question was, why I was there and why anyone else was there. Is it possible? If you ask me, is it possible that the moon will turn into a big ball of Swiss cheese, I guess it’s possible. It’s so incredibly unlikely that it’s not worth thinking about it or speaking about it, and that’s the kind of possibility we’re talking about.
Though the final version of the article is all about Christianity, I also asked what other traditions might think about multiverses. This brought me to Donald Lopez, a Buddhist studies professor at the University of Michigan, author of Buddhism and Science: A Guide for the Perplexed. “Buddhists are definitely proponents of multiverses,” he says. According to thousand-year-old sutras from the Mahayana school, Buddhas appear not only in our universe but in others, which lie very far away—not unlike an inflationary multiverse.
Speaking of which: it’s important to recognize that there’s more than one kind of multiverse model out there. In an article called “Parallel Universes,” Max Tegmark sets out four separate kinds of multiverse model, each with quite different consequences and connotations. Tegmark thinks it’s possible that all are true. For my Seed piece, I could only deal with one of these, the inflationary multiverse—#2 in Tegmark’s typology. The other multiverse model with major religious implications, the “many-worlds” interpretation of quantum mechanics (Tegmark’s #3), has attracted wide interest particularly among New Agers, meditators, and speculative types who are interested in spiritual theories about the nature of consciousness.
Stanford physicist Leonard Susskind, one of the architects of string theory, offered some humbling parting words (also didn’t make it into the final):
If the multiverse hypothesis proves correct, it will mean that the fine-tunings that drive some people toward a deist view of the universe will have an entirely natural explanation in the laws of physics and probability. … That does not mean that the ultimate origin of the universe is understood.
Comments
14 responses to “The Multiverse Problem”
Nathan Schneider represents my views and my words quite accurately. However, the juxtaposition of my statement about certain approaches being “stupid” and a reference to a book by Wiker and Witt may leave the impression that I am applying that adjective specifically to their book. I emphatically do not.:I find Wiker and Witt’s book to be very intelligent. Let this be a lesson to me. The fault lies not with Nathan Schneider, but with me for (rather stupidly) using the word stupid at all.
As Nathan Schneider correctly states, the multiverse is unobservable and untestable. The reason is that they lie outside our particle horizon, the maximum distance from which particles (i.e. also particles carrying information) could have traveled to the observer in the age of the universe. It represents the portion of the universe which we could have conceivably observed at the present day.
Science is based on observation and experiment. This is the rock-solid foundation for it’s very success in explaining the natural world. The Big Bang and evolution pass the observation/experment test with flying colors, and there is even solid observational evidence relating to an origin of life by natural causes. Yet since it is unobservable and untestable, the multiverse is not science, but philosophy.
This is very well outlined in a presentation by the prominent cosmologist George Ellis:
http://www.st-edmunds.cam.ac.uk/faraday/CIS/Ellis/Ellis_Lecture.ppt
The best that can be hoped for in terms of observability is that a putative multiverse has left an imprint on the cosmic microwave background map. But first, the evidence could probably always be interpreted in a different way, and second, there would still be no way to observe an essential prerequisite for the multiverse hypothesis to work, the idea that all universes within it have different laws of nature.
Certainly, there are theoretical mathematical models that describe the multiverse. But we know from history how well purely theoretical models fare without observational support. Just take the Ptolemaic epicycles that perfectly well explained the movement of the sun and the planets around the Earth, until, low and behold, observational evidence showed that the Earth and the other planets revolve around the sun.
As a biochemist whose work relies on observation and experiment on a daily basis, I am offended by the suggestion that the multiverse might be science.
So, my question to Stephen Barr then would be, what does he mean with his statement regarding the multiverse, “it may turn out someday demonstrable that it’s true”?
Al Moritz
My apologies for the mistakes in English in my previous post. Posting very early in the morning can be tricky, I guess 🙂
In answer to the question posed by Dr. Moritz, I would say that it is hard at the moment to see how a direct test of the multverse idea would be possible. However, it might be possible to get confirmation of the idea indirectly. One can imagine that eventually we will develop a theory that accounts for all observed experimental and observational facts of cosmology and particle physics, passes many experimental tests, leads to many correct predictions, leaves no loose theoretical ends, and has a very tight structure. There may at that point be good grounds for confidence that it constitutes the correct fundamental theory of physics. Conceivably, the equations of that well-tested future theory may imply that the unverse has a multiverse structure. That, in my view, would count as a theoretical demonstration that the multiverse theory is correct.
Admittedly, we are far from that point now, and it is very likely that we may never get to that point. But we might. And already, even at the present state of ur understanding, there are strong theoretical indications that the universe may be a multiverse. If three conditions are satisfied then a multiverse almost certainly results: (a) The theory of physics has many degenerate or nearly degenerate ground states, (b) cosmic inflation happened in the early universe (so that the universe is exponentially larger than our present horizon), and (c) the universe was already so large when there was a selection among the different possible ground states that many regions of the universe were not in causal contact. The arguments for cosmic inflation and the universe being exponentiall larger than our present horizon are very strong. Multiple degenerate ground states are a common feature of theories that are plausible exensions of our current theory — e.g. supersymmetric grand unified theories usually have several degenerate ground states, and superstring theory seems to have at least 10^500 of them. (Another indirect indication of a multiverse structure would the existence of parameters in our theory whose values could be explained with multiverse idea, but which were not anthropically significant — the theta parameter of QCD might turn out to be an example of that. But this requires a long discussion.)
A couple of final remarks. There are many instances in modern physics where very well tested theories predict the existence of phenomena or of regions of spacetime that cannot be directly observed. For example, QCD (the very well tested theory of the strong interactions) predicts the existence of particle called quarks, but also that isolated quarks (“free quarks”) can never be observed directly. Einstein’s theory of gravity predicts the possibility of black holes, and now there is compelling evidence that black holes actually exist. But theory also tells us that we can never see into the interior of a black hole (i.e. into the region within its Schwartzschild radius). We can say certain things about what is likely to be going on inside the Schwartzschild radius and what is extremely unlikely to be going on, even though we cannot ever look into that region: the theory tells us something about it. In a like manner, our theories may be able to tell us something about what is going on outside of our horizon, even though we may not be able to look.
Finally — and this is really the important point theologically — if a person is making arguments for the existence of a cosmic designer that are (a) based on anthropic coincidences and anthropic fine-tuning and (b) require that the universe NOT be a multiverse, then the burden of proof lies on that person to show that the universe is not or is unlikely to be a multiverse. At the moment, we cannot know whether the universe is a multiverse, and we may never know. But everything we do know suggests that it is quite reasonable to suppose that the universe MAY be a multiverse — and that is enough to defeat the kind of design arguments I am talking about. In my view, one CAN nonetheless make good design arguments based on anthropic coincidences and apparent tunings, but to have any force they cannot be based on assumpton (b), i.e. that the universe is not a multiverse. Such arguments are given by me in my book “Modern Physics and Ancient Faith” and by Prof. Robin Collins in his writings.
.
Dr. Barr:
Thank you for your comments.
First-off, since I am a biochemist and not a physicist, if I say something demonstrably wrong, please do not hesitate to correct me. However, as a scientist, and in particular an experimental scientist, I do think I have some points to make regarding what may or may not constitute science from the perspective of what science has been all about until more recent developments and debates.
I do not think that is quite legitimate to compare the multiverse with quarks and black holes, when you make the valid point:
“There are many instances in modern physics where very well tested theories predict the existence of phenomena or of regions of spacetime that cannot be directly observed.”
As far as I understand it, there is very good experimental evidence for the existence of quarks, even though, as you say, the theory predicts that “free quarks” can never be observed directly. There is also solid observational evidence for black holes, even though, as you point out, we will never be able to observe certain things within them. Yet there is no observational evidence whatsoever for the multiverse. Yes, there are theoretical models that some scientists find compelling, but, as I pointed out in my above post with the example of Ptolemaic epicycles, we know from history that theoretical models without observational evidence to support them do not have a tendency to fare well.
In my view only a theory that would satisfy the stringent criteria that you describe above:
“One can imagine that eventually we will develop a theory that accounts for all observed experimental and observational facts of cosmology and particle physics, passes many experimental tests, leads to many correct predictions, leaves no loose theoretical ends, and has a very tight structure. There may at that point be good grounds for confidence that it constitutes the correct fundamental theory of physics. Conceivably, the equations of that well-tested future theory may imply that the universe has a multiverse structure. That, in my view, would count as a theoretical demonstration that the multiverse theory is correct”,
_might_ be worth considering as supporting the multiverse.
And, as you concede, we are far from that point now, and it is very likely that we may never get to that point.
Until then, any discussion of the multiverse as “science” is, in my view, wildly premature. And until then, I do have to consider the multiverse pure metaphysics, and can impossibly call it science — not in the sense of being founded on observation and experiment, which has been the reason for the very success of science in explaining the natural world over the last few centuries. And I cannot subscribe to a radical paradigm shift of what constitutes science — the vigorous propagation of ideas as (possible) facts before there is even remote testability or at least a consistency test as under the stringent conditions you describe (again, previous hypotheses in science have never suffered from a principle testability problem like the multiverse). Therefore, I can also not at all agree to selling the multiverse to the general public as science — certainly not without the caveat that it does not constitute science in the traditional sense (I am quite surprised that you apparently find general audiences with such open ears, but then, most people do not even know about the current internal debate within the science community how science works and what constitutes it). The cosmologist Ellis shares similar concerns as I do about the multiverse presenting philosophy rather than science in the presentation linked to above. And other physicists also cannot subscribe to the idea. In his article, Nathan Schneider quotes Princeton University physicist Paul Steinhardt to have called the multiverse “a dangerous idea that I am simply unwilling to contemplate.” I suppose he calls it a “dangerous idea” in the sense of trying to redefine the essence of science. So yes, the scientific community is deeply divided on the issue.
An important input into the multiverse idea came also from string theory, not just from (eternal) inflation (while it can be argued that there is good observational evidence for inflation, eternal inflation is a purely theoretical construct).
While I can understand why many physicists, including you, find string theory very promising, the theory has been unable so far to shake off some fundamental criticism. The most problematic points are that in 30 years the theory has been unable to make any accurate experimental prediction, which is quite an anomaly in the history of science. And it has been unable to be experimentally confirmed. Even such an “outlandish” theory like General Relativity found a first experimental confirmation within 4 years of its publication.
What about the 10^500 degenerate ground states, resulting in a cosmic landscape of 10^500 universes, which come from calculations within string theory?
As Jeff Murugan suggests in a discussion after Ellis’ presentation:
http://www.st-edmunds.cam.ac.uk/faraday/CIS/Ellis/Ellis_Discussion.pdf
“I have a point regarding Susskind’s statements about the multiverse and so on. When people talk about the 10^500 vacua they are talking about specific solutions to specific couplings fixed and specific fluxes fixed and stuff like that. It’s all very well for them to say there’s 10^500 of these guys but there is, as far as I know, just one explicit construction and it’s an incredibly difficult construction, an entire tour de force that involves algebraic geometry to fix all these fluxes consistently and that’s in a very technical paper by Florien, Deneff and Douglas? To talk about these 10^500 vacua actually showing that one of these vacua solutions itself is a consistent solution is an incredibly difficult problem, so it comes down to what I was saying earlier in that we really don’t fully understand the theory at all. So never mind finding solutions to these things, it’s actually identifying what it is you are working with in the first place that makes it so difficult. So there’s a long way to go before we can start answering questions about what exactly it’s able to tell us about the nature of the Universe.”
The issue of mathematical consistency has been raised by others as well. So we know neither if string theory is correct, nor, if it is correct, that the calculations within it are correct. But to my knowledge string theory is currently the only theory that might suggest sufficiently many ground states as to explain the apparent fine-tuning of the universe by the sheer number of different universes all having different variations of laws of nature within a putative multiverse.
So all in all, I really do think that any presentation of the multiverse as a scientifically explanatory model is premature.
***
I do agree that the multiverse does not really solve the design problem. Your book “Modern Physics and Ancient Faith” argues the point well. As you explain, a multiverse that is taken seriously by physicists is typically not really “many universes”, but a single “many domain universe”, in which all the domains are governed, deep down, by a single set of fundamental laws (cf. p. 152 of the book). If this universe could be such that the laws of physics would be able to vary almost continuously from one domain to another, this in itself would make this universe a very special place. You conclude that “having laws that lead to the existence of domains of a sufficiently rich variety to make life inevitable would _itself_ qualify as an anthropic coincidence. There seems to be no escape. Every way of explaining anthropic coincidences scientifically involves assuming the universe has some sort of very special characteristics that can be thought of as constituting in themselves another set of anthropic coincidences” (p. 154).
You referred to Collins who makes similar points in, for example:
http://home.messiah.edu/~rcollins/muv2.htm
On the other hand, the many-universes multiverse that you criticize is hardly tenable without the idea that everything that might be possible is, in fact, actual. However, if that were the case, then there would also be a universe where fairies, Santa Claus and the Easter Bunny actually do exist! (So much for the credibility of the atheist position, which in the philosophically most silly way argues that the existence of God falls on the same level as these questionable entities.)
***
While I understand the theological points that you make in your post, I cannot quite agree with the argumentation. Not because of theological reasons, but simply because at this point I cannot consider the multiverse science. In apologetics, I tend to say that the multiverse is not science but metaphysics, and that, even if true, it does not solve the design problem (see above) — the latter statement shields apologetics from the remote future possibility that the multiverse might be found to be viable science on the basis of a broad consensus within the scientific community. But for now I cannot possibly take as a starting point that the multiverse might be actually true — even though, certainly, it does not constitute a serious theological difficulty. God could have created as many universes as He wants.
And yes, even though some theistic scientists like you appear to support the multiverse, I do agree with Craig and others that the idea is for a large part also an act of “desperation” on the part of atheist scientists — who, however, do not realize that the many domain universe (e.g. the one from eternal inflation) does not really solve the design problem, as you point out. The atheist desperation seems clear from comments such those in a conversation (available on DVD) between the atheists Steven Weinberg and Richard Dawkins, from which Weinberg was quoted to have said: “If you discovered a really impressive fine-tuning … I think you’d really be left with only two explanations: a benevolent designer or a multiverse.”
To me, the multiverse looks suspiciously like the (failed) steady state theory theory which, as historically can be argued quite well I think, had been formulated on ideological grounds first and on scientific grounds secondarily.
Also you seemed to be suspicious in your book of the multiverse idea on grounds of world view (p. 156/7):
“It is a very curious circumstance that materialists, in an effort to avoid what LaPlace called the unnecessary hypothesis of God, are frequently driven to hypothesize the existence of an infinity of unobservable entities. We saw this before […]. We see it now in the idea of a large and possibly infinite number of domains or universes. […] It seems that to abolish one unobservable God, it takes an infinite number of unobservable substitutes.”
(While you made these comments in the context of the many-universe multiverse, you also mention the domains of the many-domains multiverse.)
Since now you embrace the possibility of a multiverse not just scientifically, but also theologically, this appears to signify somewhat of a shift in your position.
One more thing. I said in my previous post that in my view only a theory that would satisfy the stringent criteria that you describe above:
“One can imagine that eventually we will develop a theory that accounts for all observed experimental and observational facts of cosmology and particle physics, passes many experimental tests, leads to many correct predictions, leaves no loose theoretical ends, and has a very tight structure. There may at that point be good grounds for confidence that it constitutes the correct fundamental theory of physics. Conceivably, the equations of that well-tested future theory may imply that the universe has a multiverse structure. That, in my view, would count as a theoretical demonstration that the multiverse theory is correct”,
_might_ be worth considering as supporting the multiverse.
I overlooked one important fact, however. Such a “final” theory might firmly predict degenerate ground states, the expression of which would result in different laws of nature in different domains of a multiverse. Yet these would be mere possibilities.
We cannot run the risk of confusing possibilities with actualities, a lethal mistake both in science and in philosophy.
Thus, we would need some proof of actualization. However, to prove actualization would require observation. So it seems that we would have to observe the multiverse after all, before we could say that it is “confirmed by science”. Yet such an observation is not possible due to the particle horizon.
Or, at the very bare minimum, we would have to prove a mechanism that would allow these possibilities to transfer to actualities, e.g. eternal inflation or equivalent, by observation. This might not be possible, however, without performing experiments that would mimic universe generation, i.e. without performing experiments that actually do generate (mini) universes!
Dear Dr. Moritz,
Your argument seems to have the following structure: (a) Any hypothesis that is not testable cannot be considered a scientific hypothesis, and thus does not belong to science. (b) The multiverse hypothesis is not testable, and so it is not science. (c) Because the multiverse hypothesis is not science we cannot take seriously the possibility that it might be true. Or to put it another way, in philosophical arguments we should only take seriously hypotheses that are scientific hyotheses, and so we can dismiss the multiverse hypothesis in making philosophical arguments.
In reply, I would make the following points. First, as to point (a), one does not always know whether a hypothesis will turn out to be testable or not. When neutrinos were first hypothesized, it was thought by many people that they would prove to be unobservable. Now, they are not only observable, but people make intense beams of them and observe large numbers of them every day. It is true that general Relativity (GR) made a few predictions that were quickly confirmed. However, the early tests of GR were very few in number (perihelion shift of Mercury, bending of starlight by the Sun) and HARDLY constituted any sort of proof of GR. There were many ways to account for those experimental results besides GR. In fact, there still are. Suppose we had not been lucky? Suppose the planet Mercury did not exist, and that the Moon were a slightly different size so that it did not neatly block out the Sun during total solar eclipses, making possible the bending of starlight measurements? Then it might have seemed almost impossible to test GR back when Einstein proposed it. Would it then not have been a scientific idea? As science progresses, ways of learning about the world that we didn’t suspect often open up, and ideas that seemed untestable turn out to be testable after all.
A second point. On the basis of what we can observe, we can make reasonable inferences about things we cannot observe. Suppose, for example, I were to say that just outside our cosmological horizon (i.e. just beyond the limiting distance of what we can observe, which is about 10 billion light years) space just comes to an end. That is, there is edge to space, beyond which there is nothing, not even space. Can anyone ever test by observation whether there is or is not such an edge to space? No. And so, in a certain sense, you are right that the question of whether or not there is such an edge is not a “scientific” one. Nevertheless, on the basis of what we DO know about science we can say that the idea of an edge to space just outside our horizon is so unlikely as to be almost absurd. We can also say that it is unlikely that there are fire-breathing dragons or unicorns.just outside our horizon. Among untestable possibilities, some are more reasonable, more plausible, more believable than others. This is something we assume also in ordinary life.
Take again the question of what happens inside the Schwartzschild radius of a black hole. After a star collapses and forms a black hole, are there leprechauns and pots of gold inside that black hole? Are there chamber orchestras playing Bach? It is not a “scientific question” in the sense that it is not testable, but that doesn’t mean we can say nothing based on our scientific understanding about what is possibly or probably occuring inside that black hole.
What I would say is that in light of what we know today the idea that the universe has a multiverse structure is a reasonable one. Moreover, as we learn more, it may come to the point that the idea that the universe has a multiverse structure will appear more reasonable to physicists than the idea that it doesn’t. And there may come a time when it becomes highly unreasonable to doubt that there is a multiverse structure.
You say, “But for now I cannot possibly take as a starting point that the multiverse might be actually true.” You are saying that because we cannot now (or perhaps ever) prove that the multiverse idea is true, that we should not take as a possibility that it might actually be true. I suspect you did not mean to speak that strongly, for as stated it is an absurdity. Did Abraham Lincoln wear socks on June 10, 1827? A historian might tell us that it is absolutely impossible for us now to determine the answer to that question. It may be absolutely untestable at this point. Does that mean we cannot accept the possibility “that it might actually be true” that Lincoln wore socks that day? Explain to me again why something has to be testable before we can take seriously the possibility that it might be true!? You are taking positivism to an incredible length here, and I am not sure than positivism is really, in the long run, a friend to religious belief.
There are many things in science and in ordinary life that we not only take as serious possibilities but even as practical certainties whose truth we cannot actually test.
As for the multiverse idea being taken seriously only because atheists need it to escape from design arguments, that is simply not true., I am not an atheist, and I take it seriously. So does the cosmologist Don Page, who is an evangelical Christian. The multiverse idea is being taken with increased seriousness because of superstrings and inflation. (One does not need eternal inflation to have a multiverse, just inflation.) If there are good reasons to take the multiverse seriously, then it does not matter whether there are also a thousand bad reasons. There are bad reasons to believe in God as well as good ones. The fact that there are bad reasons for believing something is not an argument against it, if there are also good reasons.
To ridicule the multiverse idea and dismiss it as a desperate ploy is an unworthy way to argue.
Again: if a theological argument assumes that there is no multiverse, then the onus probandi lies with the person who is making that argument to show that the existence of the multiverse is not a reasonable and plausible possibility.
Some further thoughts. If I am on a jury and the prosecutor proposes to demonstrate by various arguments and evidence that the defendant murdered his wife, the burden of proof is on HIM to rule out other reasonable explanations of the evidence. If I, as a juror, find another explanation that covers the facts and is not unreasonable, then I must vote for acquittal. I have don’t have to prove that the alternative scenario is actually true. It may be impossible to prove that. The alternative scenario may even be absolutely untestable, because evidence has been lost or witnesses have died. But that doesn’t matter, because the burden of proof is on the prosecutor.
The “prosecutor” here is the person arguing from anthropic coincidences in the laws of physics to the conclusion that God exists and designed the world for the purpose of producing living things. The burden is on him to exclude other reasonable explanations of the anthropic coincidences. He has to show that the multiverse idea is not a reasonable scenario. The atheist does not have to prove it is true, only that it IS a reasonable scenario.
And the fact is that from the point of view of theoretical particle physics and cosmology, the multiverse idea IS completely reasonable.
The suggestion that the multiverse idea is “desperate” implies that it is a whacky idea that no one would believe or even contemplate unless he were desperate. It is easy to convince the average person that the multiverse idea is whacky. Many ideas in fundamental physics — extra dimensions of space, particles being in two places at the same time, space-time being curved, black holes, dark energy,and so on — do indeed have a wild and strange sound to them. That does not make them whacky, however. The proverbial “man on the street” is not always a good judge of what ideas in fundamental physics and cosmology are reasonable. Some Christian apologists are playing off of the common-sense prejudices of the ordinary person, which, while reliable on many subjects, are not the best guide in certain areas far removed from common experience.
It is just as easy for atheists like Dawkins to portray certain Christian doctrines as whacky — the Trinity, a person being both God and man, resurrection of the body, the Real Presence in the Eucharist, and so on. Dawkins loves to do that. I think that kind of ignorant jeering on the part of atheists is to be deplored. Unfortunately, some Christian apologists employ essentially the same tactics in reverse. You are certainly NOT ignorant about the physics ideas — your discussion shows that you are very well informed about them. But your saying that the multiverse idea is desperate and not to be taken seriously is only an effective stance to the extent that the people you are addressing have certain intellectual prejudices. Admittedly, many particle physicists and cosmologists have the very same prejudices against the multiverse idea. They have them for the same reason they are prejudiced against religion: it is “untestable”. But they have an excuse: they have a simplistic epistemology that forces them to dismiss as “metaphysics” (which for them means the same things as “irrational nonsense”) ideas that are not scientifically testable. But we religious believe in metaphysics, do we not? Let us say the multiverse is “metaphysics” rather than “science”. So what? How does that mean that it is to be taken less seriously? The context in which we are discussing it is a metaphysical one, after all: we are discussing proofs of God’s existence. metaphysical ideas are completely in order there.
Dear Dr. Barr,
I believe there is a misunderstanding which I will address later, probably tonight. I actually agree with a lot of what you say.
Dear Dr. Barr,
I did not claim that the multiverse idea is being taken seriously _only_ because atheists need it to escape from design arguments — in the same breath I had acknowledged that there are theistic scientists like you who take the idea seriously as well. Initially I thought I had chosen my words carefully, but clearly I was not careful enough, for which I apologize. What I did want to say is that there appears to be an ideological background that transcends science, that I am not comfortable with, and that may have played into the development of the hypothesis. Yet perhaps this background has less of an impact on the scientific reasoning on the matter than I had assumed thus far. I also mentioned the issue in the context of some critical attitude towards the multiverse on grounds of world view that you appeared to express in your book (see my post above).
I think that you made valid points in your post regarding reasons to believe something, and that bad reasons to do so do not invalidate its potential truth.
Now to the main misunderstanding. You said:
“You are saying that because we cannot now (or perhaps ever) prove that the multiverse idea is true, that we should not take as a possibility that it might actually be true.
“Explain to me again why something has to be testable before we can take seriously the possibility that it might be true!? You are taking positivism to an incredible length here, and I am not sure that positivism is really, in the long run, a friend to religious belief.”
I am afraid that you have taken what I said out of its context, which was apologetics. Of course the multiverse may be true, and I indicated that I do not even have theological issues with the notion, since God could create as many universes as He wants. Yes, the multiverse may be true, but I said that, in apologetics, “for now I cannot possibly take as a starting point that the multiverse might be actually true” (the issue here is _starting point_).
I could only do this if I would, indeed, find the multiverse idea already justified from a scientific point of view, which I am not so sure of. Again, I have no theological issues with the idea, while on a philosophical level it may on one hand be a good argument, but my main quarrel with it is that it violates Occam’s razor (“entia non sunt multiplicanda praeter necessitatem”, “entities must not be multiplied beyond necessity”) — now apart from the issue if predictions from certain scientific theories actually may be true and thus Occam’s razor is overruled by what in fact exists. I just see no compelling reason at this point to actually assume the existence of the multiverse — which is different from acknowledging the possibility that it exists.
Furthermore, I suppose when it comes to a many-domain multiverse that can express a sufficient variability of laws of nature to explain the apparent fine-tuning of the laws in our universe as a statistical probability or even necessity, a lot of scientific assumption rests on how confident one is about string theory, which suggests the existence of so many degenerate ground states. I am not as confident or hopeful that string theory is the correct theory as you may be.
The problem in apologetics is for me that atheists constantly proclaim (and they seriously believe it too) that their world view is more “scientific” than that of believers (which is nonsense for a number of reasons). And when it then comes to the multiverse as an answer to the fine-tuning problem, they present it as a scientific alternative to the God hypothesis. And with this I cannot concur. Yes, the multiverse may be considered a good argument, but as of now, for me and many others, it remains a philosophical argument, and not a scientific one, for the cited reasons of observability (also, see below). Perhaps a philosophical argument based on good arguments _from_ science, but nonetheless a philosophical one.
Yet as we both agree, the design argument stands, multi-domain multiverse or not (for the many-universes multiverse the design argument does not apply, but I think we both also agree that the notion of such a multiverse is philosophically quite problematic). So the multiverse doesn’t really help the atheist one bit (it may only make the fine-tuning problem less numerically daunting).
Actually, through this discussion I have now become aware how vital it is in apologetics to make the distinction between the multi-domain multiverse and the many-universes multiverse.
***
Now let me address some other points.
Your argumentation on neutrinos and GR relates to practical observability — and again, you raise some interesting and valid points. Yet the issue seems to be rather observability in principle, not just in practical but also in theoretical terms. In principle we can observe the existence of entities within our own cosmological spacetime. Yet outside of our spacetime observation of the existence of entities is not possible, and the discussion then is essentially about whether we can find legitimate substitutes for this or not.
Yes, certain details about entities in within our spacetime may also escape observation, but that is another issue altogether. While, as you say, we can only develop theories about what happens inside a black hole, at least black holes are in principle observable, which now has worked out in practice as well. It does make sense to construct theories on entities that we do know or can know to actually exist! The multiverse however not just may or may not exist, but its very existence may not be testable in principle, and it therefore poses a fundamentally different epistemological problem than the black hole or the neutrino. This epistemological problem goes to the very heart of what constitutes science.
We will probably not solve our differences of opinion here, but just as the opinions differ between you and me, so do they between other scientists. In fact, as had been pointed out by many sources (and as is also clear from Nathan Schneider’s article), there is a deep controversy within the scientific community on this issue, and neither of us can afford to pretend that this controversy does not exist. While one does not need to become defensive on the issue, I also think it would be a sign of intellectual fairness and openness about the scientific process to mention this controversy in presentations of the multiverse to general audiences (while in the same context it would be worth pointing out that the supposed deep divide within the scientific community about the theory of evolution is entirely made up by creationists — scientists argue about the details of the theory, but not about its principal correctness).
You said:
“Admittedly, many particle physicists and cosmologists have the very same prejudices against the multiverse idea. They have them for the same reason they are prejudiced against religion: it is “untestable”. But they have an excuse: they have a simplistic epistemology that forces them to dismiss as “metaphysics” (which for them means the same things as “irrational nonsense”) ideas that are not scientifically testable.”
The issue is not that simple. The cosmologist George Ellis is a believing Christian and argues against the multiverse on similar scientific grounds as I do (again, see his presentation linked to above, which ends on a rather religious note). Once more, this is a debate on the essence of what constitutes science, and there are theists and atheists on both sides of the debate.
By the way, while the many-universes multiverse most certainly violates Occam’s razor (in a gross way), it may be argued that the many-domain multiverse does not really violate it. After all, it may be seen as just one single universe with many domains, and thus as one single entity. So perhaps, upon closer inspection the many-domain multiverse is philosophically more attractive than I thought.
Sorry, I wish I had the chance to be more involved in this conversation—have been totally engrossed in another writing project. On the question of simplicity (which Occam’s Razor addresses), in his “Does God Love the Multiverse” talk, Don Page argues that the multiverse may actually be simpler than the alternative.
I am not sure if I can agree with Page’s suggestion that the multiverse would be God choosing “elegance over paucity”. I cannot quite see what would be elegant about the vast number of sterile, boring and in many cases short-lived universes that would inevitably be the result of a random variation of the laws of physics. If God really wanted to generate multiple universes it might be more elegant, I would say, to directly select those that would be fruitful in some way. In any case, such aesthetic musings are moot to some extent since in our lifetimes here on earth we will probably not be able to know what God really chose to do.
And Don Page is mistaken about the mutiverse abolishing the fine-tuning argument. Even though in a changed form, the argument still stands, as Stephen Barr and Robin Collins have pointed out.
***
As far as what the many-domain multiverse really presents, upon further thought it might not be fair to say that, if it is not science, then automatically it is metaphysics. One might call it a hypothesis _from_ science. After all, the idea is developed on the basis of scientific, and not philosophical, reasoning (any ideological inclinations that may have played into its development aside). Only once it aims to have any explanatory value with respect to the fine-tuning argument, this would be philosophy. On the other hand, if the many-domain multiverse ever were, on the basis of possible evidence that was discussed here, unanimously to be viewed as science by the scientific community (outsider opinions have little relevance), then any explanatory value that it might have would be scientific as well.
The many-universes multiverse, on the other hand, is philosophy from the start. This includes Everett’s many-worlds hypothesis. While it is born from an interpretation of quantum mechanics, this interpretation is philosophical.
Correction:
“And Don Page is mistaken about the mutiverse abolishing the fine-tuning argument.”
should read:
“…undercutting the design argument”.